Wetlands ecology, conservation and restoration

WETLANDS: ECOLOGY, CONSERVATION AND RESTORATION

No part of this digital document may be reproduced, stored in a retrieval system or transmitted in any form orby any means. The publisher has taken reasonable care in the preparation of this digital document, but makes noexpressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. Noliability is assumed for incidental or consequential damages in connection with or arising out of informationcontained herein. This digital document is sold with the clear understanding that the publisher is not engaged inrendering legal, medical or any other professional services.

WETLANDS: ECOLOGY, CONSERVATION AND RESTORATION

RAYMUNDO E. RUSSO EDITOR

Nova Science Publishers, Inc. New York

Copyright © 2008 by Nova Science Publishers, Inc. All rights reserved. No part of this book may be reproduced, stored in a retrieval system or transmitted in any form or by any means: electronic, electrostatic, magnetic, tape, mechanical photocopying, recording or otherwise without the written permission of the Publisher. For permission to use material from this book please contact us: Telephone 631-231-7269; Fax 631-231-8175 Web Site: http://www.novapublishers.com

NOTICE TO THE READER The Publisher has taken reasonable care in the preparation of this book, but makes no expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of information contained in this book. The Publisher shall not be liable for any special, consequential, or exemplary damages resulting, in whole or in part, from the readers’ use of, or reliance upon, this material. Independent verification should be sought for any data, advice or recommendations contained in this book. In addition, no responsibility is assumed by the publisher for any injury and/or damage to persons or property arising from any methods, products, instructions, ideas or otherwise contained in this publication. This publication is designed to provide accurate and authoritative information with regard to the subject matter covered herein. It is sold with the clear understanding that the Publisher is not engaged in rendering legal or any other professional services. If legal or any other expert assistance is required, the services of a competent person should be sought. FROM A DECLARATION OF PARTICIPANTS JOINTLY ADOPTED BY A COMMITTEE OF THE AMERICAN BAR ASSOCIATION AND A COMMITTEE OF PUBLISHERS. LIBRARY OF CONGRESS CATALOGING-IN-PUBLICATION DATA Wetlands : ecology, conservation, and restoration / Raymundo E. Russo (editor). p. cm. ISBN 978-1-60876-354-2 (E-Book) 1. Wetland ecology. 2. Wetland conservation. 3. Wetland restoration. I. Russo, Raymundo E. QH541.5.M3W4836 2008 577.68--dc22 2008030635

Published by Nova Science Publishers, Inc. New York

CONTENTS

Preface vii Expert Commentary

Two Alternative Modes for Diffuse Pollution Control by Wetlands 1 Chen Qingfeng, Shan Baoqing and Ma Junjian

Short Communication Multiangular Imaging of Wetlands in New England 7

Lesley-Ann L. Dupigny-Giroux and Eden Furtak-Cole Research and Review Articles

Chapter 1 Wetlands: Water “Living Filters”? 15 Ana Dordio, A. J. Palace Carvalho and Ana Paula Pinto

Chapter 2 Remote Sensing Data for Regional Wetland Mapping in the United States: Trends and Future Prospects 73 Megan W. Lang and Greg W. McCarty

Chapter 3 Transforming Useless Swamps into Valuable Wetlands: Evaluating America’s Policy, 1970-2008 113 Andrea K.Gerlak and Jeanne N. Clarke

Chapter 4 Dynamics of Coastal Wetlands and Land Use Changes in the Watershed: Implications for the Biodiversity 133 Miguel Ángel Esteve, M. Francisca Carreño, Francisco Robledano, Julia Martínez-Fernández and Jesús Miñano

Chapter 5 Pathogen Removal in Constructed Wetlands 177 Kela P. Weber and Raymond L. Legge

Chapter 6 The Role of Harvest and Plant Decomposition in Constructed Wetlands 213 Juan A. Álvarez and Eloy Bécares

Contents vi

Chapter 7 Nutrition and Toxicity of Inorganic Substances from Wastewater in Constructed Wetlands 247 Zhenhua Zhang, Zed Rengel and Kathy Meney

Chapter 8 A Conceptual and Methodological Framework for the Study of Vegetated Fluvial Landscape Evolutionary Trajectories 271 Dov Corenblit, Johannes Steiger, Eric Tabacchi and Angela M. Gurnell

Chapter 9 Macrophyte Morphological Response to the Industrial Effluent Toxicity in a Constructed Wetland 295 H. R. Hadad, M. M. Mufarrege, M. Pinciroli, G. Di Luca, V. del Sastre and M. A. Maine

Chapter 10 Phytoremediation Processes for Water and Air Pollution Control in the Aspects of Nutrient and Carbon Dioxide Removals 325 Jae Seong Rhee, Yonghui Song, Fasheng Li and Janjit Iamchaturapatr

Chapter 11 Phytoplankton Biomass Regulation in Contrasting Environmental States of Temporary Pools 359 Silvia Martín, Marta Rodríguez and David G. Angeler

Chapter 12 Can Tern Migrants Coexist with Urban Development and Estuarine Recreational Activities? 373 Ken Chan, Jill Dening and Marja-Leena Malinen

Chapter 13 Agricultural Wetlands 391 R. Kröger

Chapter 14 Profiling Cover Cycle Dynamics for Prairie Pothole Wetland Landscapes 407 Rebecca L. Phillips and Ofer Beeri

Index 419

PREFACE Wetlands are lands where saturation with water is the dominant factor determining the

nature of soil development and the types of plant and animal communities living in the soil and on its surface. Wetlands vary widely because of regional and local differences in soils, topography, climate, hydrology, water chemistry, vegetation, and other factors, including human disturbance. Indeed, wetlands are found from the tundra to the tropics and on every continent except Antarctica. This new book brings together the latest research in the field.

Short Communication - Multiple view angles (MVA) or multiangular imaging represents a yet to be explored use of the remote sensing of wetlands. The ability to view the landscape off-nadir (traditionally the surface is viewed at right angles) allows for the quantification of moisture stress, species separation and the proportion of vegetation to standing water in these ecosystems. This commentary will focus on the ratio of two broadband wavelengths (near-infrared to blue) derived from multiangular images acquired by the Airborne Multi-angle Imaging SpectroRadiometer (AirMISR) of wetlands across New England. The resulting insights into the photointerpretation, monitoring and mapping of wetlands will be highlighted.

Chapter 1 - Human societies have indirectly used natural wetlands as wastewater discharge sites for many centuries. Observations of the wastewater depuration capacity of natural wetlands have led to a greater understanding of the potential of these ecosystems for pollutant assimilation and have stimulated the development of artificial wetlands systems for treatment of wastewaters from a variety of sources. Constructed wetlands, in contrast to natural wetlands, are human-made systems that are designed, built and operated to emulate wetlands or functions of natural wetlands for human desires or needs. Constructed wetlands have recently received considerable attention as low cost, efficient means to clean-up not only municipal wastewaters but also point and non-point wastewaters, such as acid mine drainage, agricultural effluents, landfill leachates, petrochemicals, as well as industrial effluents. Currently, untreated wastewater discharge in the natural wetlands sites is becoming an increasingly abandoned practice whereas the use of constructed wetlands for treatment of wastewater is an emerging technology worldwide. However, natural wetlands still play an important role in the improvement of water quality as they act as buffer zones surrounding water bodies and as a polishing stage for the effluents from conventional municipal wastewater treatment plants, before they reach the receiving water streams. In fact, one of the emerging issues in environmental science has been the inefficiency of wastewater treatment plants to remove several xenobiotic organic compounds such as pesticides and pharmaceutical residues and consequent contamination of the receiving water bodies. Recent

Raymundo E. Russo viii

studies have shown that wetlands systems were able to efficiently remove many of these compounds, thus reaffirming the importance of the role which can be played by wetlands in water quality preservation.

The aim of this work is to present a review on the application of wetlands as “living filters” for water purification. Emphasis was focused on the removal of micropollutants, especially xenobiotic organic compounds such as pharmaceuticals residues, which are not efficiently removed by conventional municipal wastewater treatment plants. Furthermore, the role of wetlands as protection zones which contribute to the improvement of the aquatic ecosystems’ quality will be discussed.

Chapter 2 - Historically, the biologic, aesthetic, and economic values of wetlands were largely unappreciated. Wetlands within the United States have been and are continuing to disappear rapidly. Efforts are being made to conserve remaining wetlands and many regulatory policies have been adopted in support of this goal. To regulate the loss, preservation, and/or restoration of wetlands and to judge the effectiveness of these regulatory efforts in preserving associated ecosystem services, wetlands must be routinely monitored. Wetland mapping is an essential part of this monitoring program and much effort has been made by the US state and federal governments, as well as other organizations, to provide quality wetland map products. Wetland maps can serve a variety of purposes including regulation and natural resource management. They can also be used to parameterize models that quantify water quality and quantity, as well as the provision of wetland ecosystem services, at the watershed scale. Wetland hydrology is the most important abiotic factor controlling ecosystem function and extent, and it should therefore be a vital part of any wetland mapping or monitoring program. New approaches are needed to not only map wetlands, but also to monitor wetland hydrology as it varies in response to weather, vegetation phenology, surrounding landuse change, and other anthropogenic forces including climate change. Recently developed remote sensing technologies and techniques have the potential to improve the detail and reliability of wetland maps and the ability to monitor important parameters such as hydrology. Various types of remotely sensed data (e.g., aerial photographs, multispectral, hyperspectral, passive microwave, radar, and lidar) have different capabilities with specific advantages and disadvantages for wetland mapping at the regional scale. Although aerial photographs were traditionally used to map wetlands and infer hydrology, fine-resolution optical images are now available more frequently as commercial agencies increase satellite coverage (e.g., Quickbird and IKONOS). However, optical data, such as aerial photographs and multispectral satellite images have limitations, including their inability to detect hydrology below dense vegetative canopies and their limited ability to detect variations in hydrology (i.e., inundation and soil moisture). The restrictions of optical data are increasingly being compensated for with the use of new technologies, including synthetic aperture radar, lidar, and geospatial modeling. The availability of these new data sources is increasing rapidly. For example, many states in the US are now collecting synoptic state-wide coverages of lidar data. The sources, strengths, and limitations of different types of remotely sensed data are reviewed in this chapter, as well as the importance of temporal and spatial resolution necessary for regional scale wetland mapping efforts. The potential of multi-temporal, multi-sensor approaches that capitalize on geospatial modeling are emphasized for meeting current wetland mapping challenges.

Chapter 3 - This paper traces the evolution of America’s wetland policy beginning with passage of the Clean Water Act (CWA) of 1972. This law, for the first time, established a

Preface ix

federal program to protect wetlands, dramatically elevating the value of these ecosystems. However, despite attitudinal changes and new governmental programs, the nation continues to lose its potentially valuable wetlands -- albeit at a slower rate than was the case in the 1970s and prior to the passage of the CWA. This chapter offers an objective evaluation of the federal wetlands protection policy. The authors place this evaluation within a broad societal context, showing that since 1970 there have occurred sweeping demographic, economic, and political changes that clearly have impacted the extent of wetlands in the United States. They argue that Section 404 has failed to reverse the net loss of wetlands in the U.S. Moreover, it has evolved into a policy lightening rod within the water resources arena and been a major factor in Congress’ failure to revise and reauthorize the Clean Water Act. Finally, the authors offer some recommendations designed to improve the policy, arguing for heightened wetlands protection through partnerships and acquisitions.

Chapter 4 - The Mediterranean coastal landscapes have suffered significant changes along the last decades due to the agricultural intensification and tourist development. Such changes have modified the water flows and specifically the hydrological regime of wetlands, as has occurred in the Mar Menor (Southeast Spain). The Mar Menor coastal lagoon and associated wetlands present noticeable ecological and biodiversity values. However, the land-use changes in the watershed and the consequent changes in the water and nutrient flows along the period 1980-2005 are threatening the conservation of these wetlands. A dynamic model has been developed to simulate the key environmental and socio-economic factors driving the export of nutrients to the Mar Menor lagoon and associated wetlands, where some eutrophication processes have appeared.

In this chapter the changes in the vegetal and faunistic assemblages are analysed. Vegetal communities are studied by means of remote sensing techniques, which have provided information about the changes in area and habitat composition of the wetlands along the considered period. This has shown that the habitats more negatively affected by the hydrological changes are those most threatened in the international context and with a highest interest from the point of view of biodiversity conservation. It has also been possible to verify the direct relationships between all these changes at wetlands scale and the agricultural changes at the watershed scale.

Two faunistic communities especially sensitive to these ecosystemic changes have also been studied: i) Wandering beetles and ii) Birds (waterbirds and steppe passerines). Wandering beetles (Coleoptera) were studied with pitfall traps in 1984, 1992 and 2003 and steppe passeriforms with line transects in several years along the period. In both communities evident changes have been observed. Regarding beetles, the most halophilous species have been favoured, some of them especially relevant due to its rarity in the European context. The ratio Carabidae/Tenebrionidae has shown to be a good indicator of the hydrological changes of the wetlands. Waterbirds have shown dramatic changes in their relative abundances within the lagoon, with a long-term decline in the most characteristic original species, increases in generalist piscivores and a recent appearance and rapid growth of the herbivores guild. In the case of steppe passeriforms, this community has been negatively affected, especially some species like Melanocorypha calandra. The family Alaudidae has lost importance to the benefit of the families Turdidae and Fringillidae. These changes can be considered a loss of value in relation with the original passeriform community, since the wetland qualifies as a Specially Protected Area under the EU’s Bird Directive, precisely on the basis of its genuine steppe bird assemblage.

Raymundo E. Russo x

In conclusion, the changes at wetlands scale clearly reflect the hydrological modifications at the watershed scale and have significant effects on the most characteristic biodiversity of the wetlands of coastal arid systems.

Chapter 5 - Conventional secondary and tertiary wastewater treatment methods include activated sludge, trickling filters, slow sand filtration, chlorination, ozonation and UV radiation. Chlorination being the most widely used pathogen disinfection method is presently under scrutiny as chlorination can produce carcinogenic trihalomethanes when natural organic matter is present in the wastewater. Constructed wetlands (CWs) have proven to be an effective treatment alternative for the removal and inactivation of pathogens in wastewaters. Constructed wetlands have low principle and operating costs and are fairly simple to design and implement, making them an attractive wastewater treatment alternative when compared to conventional secondary or tertiary treatment processes. Constructed wetlands designed for pathogen treatment are most often preceded by filtration or sedimentation. Pathogen removal efficiencies upwards of 99.99% have been reported by multiple authors employing many different constructed wetland designs. Constructed wetland design tends to be based largely on rule of thumb sizing, as the specific mechanisms and fundamental variables involved in pathogen removal are only vaguely understood. Suggested mechanisms of pathogen treatment in CWs include but are not restricted to sedimentation, natural die-off, temperature, oxidation, predation, unfavourable water chemistry, biofilm interaction, mechanical filtration, exposure to biocides and UV radiation. Pathogen removal has been shown to correlate well with hydraulic retention time. Use of first order decay kinetics is the preferred method to describe and predict pathogen removal in CWs. A severe lack of attention has been given to the comparative quantification of the specific mechanisms contributing to pathogen treatment in constructed wetlands. Small-scale controllable constructed wetland systems are identified as systems which can be used in conducting well-designed controlled experiments where fundamental mechanisms and variables involved in pathogen removal can be comparatively quantified. It is proposed that if the fundamental mechanisms and variables affecting pathogen removal in constructed wetlands are better understood and quantified the large performance variations reported for similarly designed treatment wetland systems can be better explained, engineered and controlled.

Chapter 6 - Upon decomposition, at the end of the summer and during the autumn, wetland vegetation releases organic carbon into the wetland system. A part of this organic matter remains in the wetland, and is degraded at different rates during the rest of the year. Therefore, litter decomposition has important consequences on constructed wetlands because it is related to the autochthonous production of organic matter, clogging rates in surface-flow wetlands, and terrestrialization in free-water surface wetlands.

The effect of harvest was studied in two free-water surface-flow wetlands. Both wetlands were planted with Typha latifolia with one of the wetlands harvested. On the other hand, decomposition rates of Typha latifolia were quantified during both winter and summer in the non-harvested surface constructed wetland using the litter bag technique. Nutrient concentrations were always lower in the effluent of the harvested wetland, indicating nitrogen and phosphorus release by decomposition of vegetation, in the non-harvested system. In addition, harvesting reduced the effluent TSS and BOD concentrations by 37.3% and 49.2%, respectively, when compared to the non-harvested wetland in spring. Seasonal background concentrations (C*) in the wetlands, increased from winter to spring and decreased again in summer. Organic load and nutrients produced per gram of Typha were evaluated by using in-

Preface xi

situ Typha degradation experiments. Taking into account the experiments of litter bag technique, no significant differences were found in both variables among the different mesh sizes, with the exception of the control bags in winter. Meso or macrofauna did not play any role in plant decomposition. Decomposition rates were significantly different between winter and summer when considering each mesh size separately. Decomposition rates from adjusted exponential models ranged from 0.0014 to 0.0026 d-1 in winter (5ºC), and from 0.0043 to 0.0052 d-1 in summer (20ºC). Typha decomposition rates were compared with others macrophytes. From these decomposition rates, it is estimated that 31% of the initial mass of plant detritus would remain in the system after one year. Based on the research conducted during several experiments, harvesting can be recommended as an operational and management strategy in warm climates and diluted wastewater conditions.

Chapter 7 - The use of constructed wetlands for purification of wastewater has received increasing attention around the world. A variety of wetland plant species (including ornamental ones) as either a monoculture or species mixes are used in constructed wetlands. Plants play an extremely important role in removing pollutants from wastewater. Although there is considerable information on plant productivity, biomass and nutrient dynamics in natural and fertilized wetlands, most studies on constructed wetlands for treatment of wastewaters have only addressed general aspects of plant growth and nutrient accumulation. Nutrition and toxicity of inorganic substances such as nitrogen, sulphur, salts and metals in wastewater on wetland plants has not been fully investigated and their interactive effects and environmental cycling in constructed wetlands remain poorly understood.

Nitrogen nutrition is the most important factor influencing plant performance in constructed wetlands, but higher NH4-N may become toxic to wetland plants. Sulphur is an essential nutrient for plant growth, but under waterlogged conditions sulphate is reduced to hydrogen sulphide that is highly toxic to wetland plants. Many metals in wastewater are essential micronutrients for wetland plants, but become toxic if their concentration exceeds a specific critical point. A proper amount of salts is essential for plant growth, but high concentrations of salts, particularly sodium chloride in wastewater have harmful effects on plant growth.

Wetland plant species have differential capacity to take up nutrients, different preference for nitrogen forms and have evolved various adaptive mechanisms protecting them against toxicity of inorganic substances. Given that plants are an integral part of constructed wetlands, the selection of suitable species, improvement of cultivations and determination of factors affecting growth are needed to produce healthy and effective wetland ecosystems. Understanding biogeochemical cycling in wetlands as well as nutrition and toxicity of inorganic substances from wastewater on plant development and function may help reduce performance variability and enhance pollutant removal in constructed wetlands.

Chapter 8 - This chapter presents a conceptual and methodological framework to study temporal and spatial changes of fluvial landforms and associated plant communities and to identify the underlying causes of either progressive or sudden changes. Mutual interactions and feedbacks between hydrogeomorphic processes, fluvial landforms and vegetation dynamics are considered within this framework, leading to the analysis of biogeomorphic (i.e., landforms and associated vegetation communities) evolution trajectories within the fluvial corridor and to the evaluation of their consequences for ecological and geomorphic forms and processes.

Raymundo E. Russo xii

First, fundamental aspects linked to the conceptual model of Fluvial Biogeomorphic Succession (FBS model) proposed by the authors (cf. Corenblit et al. 2007) are presented. This model describes the most dominant biogeomorphic succession trajectory of temperate rivers under current bioclimatic and anthropogenic conditions, starting from the rejuvenated state (bare sediment within the channel after a destructive flood). This dynamic model involves a characteristic sequence of four biogeomorphic phases where interactions of hydrogeomorphic processes and vegetation dynamics are either strong or weak according to different spatiotemporal configurations. The characteristic evolutionary trajectory corresponds to a progressive shift from the dominance of allogenic (hydrogeomorphic) processes to the dominance of autogenic (ecological) processes. It is marked by a development of specific stabilised vegetated landforms such as banks, islands and floodplains. In particular, the cyclic dynamics of the biogeomorphic succession (i.e., frequency and magnitude of rejuvenation and maturation processes), incorporating critical thresholds are discussed.

Second, a conceptual tool for the description and analysis of potential fluvial landscape evolutionary trajectories is proposed. This conceptual tool is a discrete three dimensional biogeomorphic phase-space composed of five key-stages of vegetation development (bare sediment; seedlings and saplings; adult herbs; adult shrubs; adult trees) within four distinct zones of the river corridor, exposed to four distinct levels of hydrogeomorphic disturbance (permanent submerged area; high flood-frequency area; low flood-frequency floodplain; non-submersible area). The four main processes controlling shifts between biogeomorphic configurations within the phase-space are related to the critical role of pioneer vegetation within fluvial landscape dynamics.

Finally, a methodological basis to test and to refine the model using a probabilistic transition analysis combining the biogeomorphic phase-space, empirical field data, GIS and remote sensing at local and regional scales is proposed and its applications for river management are discussed.

Chapter 9 – This chapter describes the morphological variations of floating and rooted macrophytes growing in a wetland constructed for the treatment of industrial wastewater and in natural wetlands of the Middle Paraná River floodplain, Argentina. Cross-sectional areas (CSA) of the root, stele and of metaxylem vessels and the total metaxylem CSA were measured. In addition, parameters such as dry biomass, chlorophyll concentration, and metal (Cr, Ni and Zn) and nutrient (P) concentrations were compared. During the first months of operation of the constructed wetland, only sewage was poured and floating macrophytes were dominant. After five years of operation, Typha domingensis was the dominant species in the constructed wetland. In this species, biomass and height of the plants at the inlet and outlet were significantly higher than in the natural wetlands. The plants growing at the inlet showed root and stele CSA values significantly higher than those for the plants growing at the outlet and in natural wetlands. The total metaxylem vessels CSA of the inlet plants were significantly higher than those obtained in the outlet and natural wetlands owing to the plants of this site showed the highest number of metaxylem vessels. In order to determine the morphological changes as an adaptive response to the contaminants present in the effluent, greenhouse experiments were carried out with P. stratiotes and E. crassipes. In P. stratiotes, Ni and Cr+Ni+Zn treatments were the most toxic ones, in which biomass, chlorophyll and the internal morphological parameters of roots decreased significantly, while in E. crassipes Ni caused toxic effects in the internal as well as the external morphology. The modifications

Preface xiii

recorded account for the adaptability of T. domingensis to the conditions prevailing in the constructed wetland, which allowed it to become the dominant species. This chapter may contribute to the design and mainteinance of constructed wetlands that include the macrophytes studied.

Chapter 10 - The growth of industries and major agricultural enterprises (especially food industries) supplying the human demands for their increasing population causes an annihilation of water ecosystems and an augmentation of water pollutions. These are the main sources of nutrient supplements in water resources. Excess nutrients led to the eutrophication phenomena and in many cases the deterioration of public health. While the role of carbon dioxide (CO2) gas in global climate change has become well-known, which is one of the most important environmental issues of our day, therefore it is necessary to develop technologies for the minimization of CO2 discharging into the atmosphere. Although CO2 occurs naturally in the atmosphere, its current atmospheric concentrations have been greatly affected by human activities.

One ecological method used for treating polluted water containing high nutrients and encouraging CO2 sequestration is treatment wetlands, where various aquatic plants are used for purifying the water and wastewater from excess nutrients and also withdrawing the anthropogenic CO2 from polluted atmosphere into plant’s biomass by photosynthesis process. Although wetland area around the world has diminished and continues to lose due to economic development, agriculture, and other landscape alterations, recently many of these losses are compensated by construction of new wetlands due to an our increasing understanding of wetland functions and values on global environment.

Chapter 11 - Although abiotic forces play a fundamental role in community and process regulation of disturbed wetland ecosystems, biotic interaction is increasingly recognised for having important regulatory feedback effects. This chapter reports on the context-specific role of biotic and abiotic regulation of phytoplankton biomass in temporary ponds. Contamination of artificial ponds with different application concentrations of a fire retardant resulted in alterations of the trophic status, primary producer and zooplankton communities in treatment ponds. Principal component analyses suggested that facilitation of phytoplankton biomass through cladocerans was the most important controlling factor in nutrient-limited control ponds. These biotic interaction effects disappeared in retardant treatment ponds where phytoplankton biomass was almost exclusively controlled by water depth fluctuation. This context-specific, eutrophication-mediated physical control of algal biomass in treatment ponds adds a new dimension to the traditional perspective of resource and consumer control of phytoplankton in alternative ecosystem states in lakes. The context-dependent interplay of physical and biotic processes in wetlands will likely influence applied issues and challenge wetland management and restoration.

Chapter 12 - Urbanisation and recreational activities are two of the major causes of population declines of species, and throughout the world they continue to spread and intensify at a rapid rate. The two are often linked—an increase in recreational activities is often associated with nearby growth in residential development and vice versa. Developmental growth is greatest in places of high tourism value, such as in coastal areas with sandy shores. Sandy coasts are popular with beach walking and jogging, swimming, off-road vehicles, boating, ecotourism, and other outdoor activities. The most concentrated activities are in estuaries with sandbanks and intertidal flats that are protected from the open ocean. Yet the same estuaries are often sensitive ecosystems, commonly frequented by a variety of resident

Raymundo E. Russo xiv

and migrant birds that use the areas to breed, forage, or roost. Increasing incidents of human disturbance can affect breeding behavior, feeding patterns, opportunities for rest, and decline in estuarine bird abundance. The direct impact on reproduction in breeding birds is obvious, but survival of migratory species is also affected through ineffective build-up of requisite fat reserves to successfully undertake their migratory journey. For both resident and migrant birds, disturbance could result in reduced feeding time, lowering the necessary fat reserves for survival.

Chapter 13 - Increased agricultural production, land drainage and resultant land use changes have increased loads of non-point source pollutants being discharged into aquatic ecosystems. Estimates suggest that non-point source pollution (NPS) contributes over 65% of the total pollution load to inland surface waters, including 332,000 km of rivers, 215,000 ha of lakes and 1.5 x 106 ha of estuaries. There are two types of agricultural wetlands that could mitigate NPS pollution: constructed wetlands and surface drainage ditches. Constructed wetlands are commonly used to mitigate increased nutrient, biological oxygen demand, and pesticide loads prior to entering receiving waters. However, some farmers will forgo the practice of constructing a wetland for routing water because of associated costs of construction, maintenance and loss of land in agricultural production. Agricultural drainage ditches are management tools put in place by farmers to rapidly remove standing water from their farmland. Drainage ditch function is simply one of drainage; however, research has shown that surface vegetated drainage ditches are primary intercept wetlands characterized by an ephemerally inundated hydroperiod, developed hydro-soils and a suite of facultative hydrophytes. Studies in the mid-South US have shown vegetated surface drainage ditches to reduce both pesticide and nutrients loads within the ditch prior to effluent reaching receiving waters. This is increasingly important in today’s landscape where fertilizer and pesticide applications are still high. Pollutant reduction capacity within ditches may be improved with temporal and spatial manipulation of water residence at critical junctions of non-point pollutant loss throughout the year. Primary interception, transformation and mitigation of agricultural pollutants has far reaching consequences for aquatic ecosystem health, downstream eutrophication, and coastal dynamics such as hypoxia, commercial fisheries and economic development.

Chapter 14 - Over 3 million wetlands populate the U.S. portion of the Prairie Pothole Region (PPR), where conservation goals include restoration and preservation of the cover cycle. The cover cycle is characterized by seasonal and annual changes in vegetation and open water and is closely coupled to climate and natural ecosystem functions. A complete cover cycle include periods of time when high waters drown hydric vegetation during deluge and periods where hydric vegetation expands as waters dry-down during drought. Changes in wetland cover may occur on weekly, monthly, or annual time-scales. These dynamics contribute to a rich diversity of habitats that support more waterfowl than any other region in North America. In addition temporal dynamics, PPR wetlands rarely function as single entities because of shared surface and/or groundwater hydrology. This spatial interdependence requires PPR wetland functional assessments represent populations of wetlands, commonly referred to as “profiles.” Synoptic data profiling cover cycle stage and return time for populations of wetlands would scaffold large-scale investigations of ecosystems services, habitat status, and sensitivity to climate change.

This chapter describes application of previously developed tools for synoptic delineation of wetland water and hydric vegetation cover to classify cover cycle for thousands of wetland

Preface xv

basins within a single satellite image (10,000-30,000 km2 of land area). Using satellite data layers in geographic information systems (GIS), wetland profiles developed using current (2007) wetland cover data are compared with profiles developed using National Wetland Inventory (NWI) data from 1980. Results underscore the dynamic nature of these ecosystems and the need for current observations when setting conservation goals, monitoring restoration effectiveness, and evaluating anthropogenic impacts.

In: Wetlands: Ecology, Conservation and Restoration ISBN: 978-1-60456-995-7 Editor: Raymundo E. Russo © 2008 Nova Science Publishers, Inc.

Expert Commentary

TWO ALTERNATIVE MODES FOR DIFFUSE POLLUTION CONTROL BY WETLANDS

Chen Qingfeng∗1, Shan Baoqing2 and Ma Junjian1 1Shandong Analysis and Test Center, Jinan, 250014, China

2Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing, 100085, China

Diffuse pollution has been identified as an important cause of surface water quality

degradation (Novotny, 1999). Some researches indicate that it is responsible for the transport of sediment, nutrients, heavy metals, oils, hydrocarbons and pesticides (Deletic, 1998; Schreiber et al., 2001; Lazzarotto et al., 2005). Furthermore, stormwater pollution can also have a profound effect on the ecological health of streams and reservoirs and is one of the main reasons for ecosystem degradation (Yin and Mao, 2002).

There are many ecological engineering techniques, such as buffer zones, ponds, wetlands and riparian zones currently in use, and wetlands have been shown to be effective in removing pollutants from runoff water (Mitsch et al., 2002). In order to improve treatment efficiency, wetlands can be used as treatment trains. According to the flow route, the control chains can be designed in on-line and off-line treatment trains (Bardin et al., 2001; Michael and John, 2003; Shan et al., 2006; Paolo et al., 2006; Chen et al., 2007). For the on-line treatment train, all of the runoff from a storm routes through all the system structures, which are distributed on the runoff route. The treatment train may have lower pollutant removal efficiency than off-line treatment train if the system storage is not large enough to hold all the runoff from a significant storm event. For the off-line treatment train, the system structures are distributed away from the runoff route. The treatment train is designed to intercept the “first flush”, which has much higher concentration of pollutants in the initial runoff. The later runoff, with lower concentration of pollutants, overflows the catchment directly. The off-line treatment train requires less land area and it is an economical and effective measure for the control of runoff pollution in urban areas. Every mode plays an important role in stabilizing the adjacent ecosystems and reducing the load of runoff pollution.

∗ Email:[email protected]

Chen Qingfeng, Shan Baoqing and Ma Junjian 2

In the process of diffuse pollution control, the selection of the mode is the key step. The selection of the two alternative modes for diffuse pollution control is based on concern with native topography, climate, storm water volume and available land area of the catchment (Figure 1).

Figure 1. The flow chart of the two alternative modes selection for diffuse pollution control.

If there is enough available land area in the catchment, then both modes can be selected. Otherwise, the offline mode may be the only choice for diffuse pollution control. Furthermore, the online mode may be the better choice if reuse of rainwater, additional biologic habitat, and aesthetics value are taken into consideration. In other conditions, the offline mode may be an effective choice for diffuse pollution control. In every mode, many ecological engineering techniques can be included. However, the application of the two modes in urban zoos has received little research attention.

A detailed study was carried out from April 2003 to August 2005 in Wuhan City Zoo, which is surrounded by Moshui Lake. In this study, two catchments were selected to study the characteristics and performances of the online and offline modes in Wuhan City Zoo. For this purpose, an online pond-wetlands system in the Orangutan House Catchment, and an offline filtering ditch-pond system in the Canine House Catchment, were designed to control the small point and diffuse sources of pollution in the urban zoo.

In the Orangutan House Catchment, an online pond wetlands system was used to control pollution from small point and diffuse sources. All the engineering constructions were built to adjust the flow rate of storm water and the kinetic energy of runoff on the runoff route. From upland to downstream, the landscape structures included upland grassland, orangutan house, sediment tank (ST), pond (P), the first wetland (W1) and the second wetland (W2). For the huge storage capacity of the pond-wetlands system (1071m3), most of runoff was able to be stored temporarily and purified by physical, chemical and biologic processes in the wetlands.

Two Alternative Modes for Diffuse Pollution Control by Wetlands 3

The online mode flow of the catchment is shown in Figure 2. Through grids, S2 was initially stored in ST on dry days. During rainfall events, all runoff, coming from S1 and S3, as well as S2, flowed through ST, P, W1 and W2 sequentially and then drained into Moshui Lake. In order to save water, the rainwater, stored in the pond-wetlands system, can be reused for flushing the animal house and irrigating the grassland.

Figure 2. Online mode for diffuse pollution control in the Orangutan House Catchment.

Without enough available land area for water treatment constructions, an offline filtering ditch-pond system was designed to control diffuse pollution in the Canine House Catchment. The off-line treatment train was composed of some pretreatment equipments and a filtering ditch–pond system. The pretreatment equipments include a transport ditch, grids and a sediment tank. The filtering ditch–pond system consists of a filtering ditch and two ponds. This system has a storage capacity of 115m3 and can store the initial 13.7mm runoff depths in a storm. Four species of hydrophytes, including Phragmites communis Trirn., Acorus calamus Linn., Alternanthera philoxeroides and Canna generalis, were planted in the ponds. The landscape structures in the catchment include upland grassland, storm transport ditch (T), Canine House, filtering ditch (FD) and ponds (P) from upland to downstream. FD was underground and rebuilt by an old flue, is 83 m in length, 0.5 m in width, and 1.2 min depth. It has three sections: sediment zone, filtration zone and storage zone. There are 9 subsections in the filtration zone and each subsection is filled with one of the following media: gravel, aluminite stone, bulky sand, cobblestone, ceramic granule, silver sand, turf, steel slag and vermiculite.

All the ecological engineering constructions were finished in April 2004. According to the pollution characteristics, topography, available land area and climate in the catchment, the off-line treatment train was designed to separate the ‘first flush’ from the runoff. The sketch map of the off-line treatment train is shown in Figure 3. Because the main type of land use is upland (61.4%) in the catchment, the off-line treatment train works in a natural process and requires no power. Through the grids, wastewater from flushing the animal houses (S2) was initially stored in the sediment tank (ST) and overflowed to filtering ditch–pond system for decontamination on dry days.

During rainy days, the initial runoff, coming from upland runoff (S1) and roof runoff (S3), as well as wastewater (S2), was diverted to the filtering ditch (FD) for filtration and adsorption. After that, the runoff water overflowed into ponds for further decontamination

Chen Qingfeng, Shan Baoqing and Ma Junjian 4

and then, in the final stage, drained into Lake Moshui. The later runoff, with lower concentration of pollutants, was discharged into the lake directly.

Figure 3. Offline mode for diffuse pollution control in Canine House Catchment.

The results showed that the two modes both improved runoff water quality and had high retention rates for water and pollutants. In the outflows, the event mean concentrations (EMCs) of total suspended solids (TSS), chemical oxygen demand (COD), total nitrogen (TN) and total phosphorus (TP) were reduced by 88%, 59%, 46% and 71% for the online mode, and those were 75%, 50%, 50% and 74% for the offline mode. The annual retention rates of pollutant loads for the online mode were 94.9%–98.5% in the three study years; those for the offline mode were 70.5%–86.4%. Based on calculation, the online mode was able to store the runoff of 66.7 mm rainfall completely, and the offline mode could store that of 31.3 mm rainfall. In addition, the online mode can provide an effective way for rainwater utilization and good habitats for aquatic wildlives, and has an excellent aesthetics value for recreationsal pastimes. The offline mode can save land resources and may be an effective and economical measure for diffuse pollution control in urban areas.

REFERENCES

Bardin, JP, Barraud, S, Chocat, B, 2001. Uncertainty in measuring the event pollutant removal performance of on-line detention tanks with permanent outflow. Urban Water 3, 91–106.

Chen QF, Shan BQ, Yin CQ, Hu CX. An off-line Filtering Ditch-pond system for Diffuse Pollution Control at Wuhan City Zoo. Ecological Engineering, 2007, 30(4):373-380.

Chen QF, Shan BQ, Yin CQ, Hu CX. Two Alternative Modes of Diffuse Pollution Control in an Urban Tourist Area. Journal of Environment Science, 2007, 19(10):1067-1073.

Deletic, A, 1998. The first flush load of urban surface runoff. Water Res. 32 (8), 2462–2470. Lazzarotto, P, Prasuhn, V, Butscher, E., Crespi, C., Flu¨ hler, H., Stamm, C., 2005.

Phosphorus export dynamics from two Swiss grassland catchments. J. Hydrol. 304, 139–150.

Michael Jr., John, H., 2003. Nutrients in salmon hatchery wastewater and its removal through the use of a wetland constructed to treat off-line setting pond effluent. Aquaculture 226, 213–225.

Mitsch, WJ, Lefeuvre, JC, Bouchard, VB, 2002. Ecological engineering—applied to river and wetland restoration. Ecol. Eng. 10, 119–130.

Novotny, V, 1999. Integrating diffuse pollution control and water body restoration into watershed management. J. Am. Water Resour. Assoc. 35 (4), 717–727.

Two Alternative Modes for Diffuse Pollution Control by Wetlands 5

Paolo, SC, Gaspare, V, 2006. Simulation of the operation of detention tanks. Water Res. 40 (1), 83–90.

Schreiber, J.D., Rebich, R.A., Cooper, C.M., 2001. Dynamics of diffuse pollution from US southern catchements. Wat. Res. 35(10), 2534–2542.

Shan BQ, Chen QF, Yin CQ, 2006. On-line control of stormwater pollution by pond-wetlands composite system in urban tourist area[J]. Acta Scientiae Circumstantiae, 26(7): 1068–1075 (in Chinese).

Yin, CQ, Mao, ZP, 2002. Nonpoint pollution control for rural areas of China with ecological engineering technologies. Chin. J. Appl. Ecol. 13 (2), 229–232 (in Chinese).


Short Communication

MULTIANGULAR IMAGING OF WETLANDS IN NEW ENGLAND

Lesley-Ann L. Dupigny-Giroux∗ and Eden Furtak-Cole University of Vermont, Department of Geography

200 Old Mill Building, Burlington, VT 05405-0114, 802-656-2156

ABSTRACT

Multiple view angles (MVA) or multiangular imaging represents a yet to be explored use of the remote sensing of wetlands. The ability to view the landscape off-nadir (traditionally the surface is viewed at right angles) allows for the quantification of moisture stress, species separation and the proportion of vegetation to standing water in these ecosystems. This commentary will focus on the ratio of two broadband wavelengths (near-infrared to blue) derived from multiangular images acquired by the Airborne Multi-angle Imaging SpectroRadiometer (AirMISR) of wetlands across New England. The resulting insights into the photointerpretation, monitoring and mapping of wetlands will be highlighted.

1. INTRODUCTION Multiple view angles (MVA) or multiangular imaging of terrestrial ecosystems has been

shown to provide multispectral data not observed from the nadir or other single view angles only, due to the highly anisotropic reflectance of vegetation (Asner et al., 1998). Vegetation parameters may not be the most sensitive to the nadir view angle (Privette, 1995). Other studies have explored the relationship between a sensor’s field of view and vegetation structure (Widlowski et al., 2004; see Diner et al., 2005 for a full description of these studies), land cover classifications (Hyman and Barnsley, 1997) and the role of sub-pixel heterogeneity (Zhang et al. (2002a, b), as well as view angle and reflectance anisostropy at the red wavelengths (Pinty et al., 2002).

∗ E-mail: [email protected]

Lesley-Ann L. Dupigny-Giroux and Eden Furtak-Cole 8

In a recent study (Dupigny-Giroux, 2007), multiangular images from the Airborne Multi-angle Imaging SpectroRadiometer (AirMISR) of the Howland Forest in Maine, were used for land use/land cover (LULC) separability under varying moisture conditions in the humid, continental environment of central Maine. The study extended original work by Dupigny-Giroux and Lewis (1999) that used the ratio of near-infrared/blue wavelengths plotted against surface temperatures to describe vegetation and moisture stress for the semiarid Brazilian nordeste (northeast). This study complements work by Silva Xavier and Soares Galvão (2005) who used Principal Components Analysis of Multi-angle Imaging SpectroRadiometer (MISR) data from the Amazon to discriminate land cover types.

Results of the Dupigny-Giroux (2007) indicated that the NIR/blue ratio at multiple view angles was able to discriminate variations among wetland, aquatic vegetation and the extent of moisture stress. Contributions of the study included an expansion of the recommended combination of the 15-30˚ solar illumination angle and nadir viewing angle for optimally recording benthic features (Dobson et al., 1995); a sensitivity of the NIR/blue ratio to species type and vigour, water/vegetation proportions and moisture gradients across emergent wetlands; and the distinction between aquatic macrophytes and terrestrial vegetation that are often similar individual wavelengths (Valta-Hulkkonen et al., 2003). The study suggested potential uses of the multi-angular ratio including improved mapping of wetlands in humid temperate regions (Bicheron et al., 1997; Barnsley et al., 1997); the avoidance of false change detection due to drought or water draw down (U.S. Fish and Wildlife Service, 2004) and; the improved photointerpretation of evergreen forested wetlands and more xeric ecosystems (Tiner, 2003).

In this commentary, the methodology of the Dupigny-Giroux (2007) study was applied to wetlands at two other experimental forests in New England to explore the applicability of the technique across disparate wetland types and microclimates.

2. DATA AND METHODOLOGY

2.1. AirMISR Program TheAirMISR instrument is a pushbroom imager that is mounted on the NASA ER-2

aircraft flying at an altitude of 20km over selected temperate and tropical study areas. It used a single camera on a pivoting gimbal mount to collect data at the nine viewing angles used on the spaceborne MISR instrument. These angles are nadir (An), 26.1˚ fore (Af) and aft (Aa), 45.6˚ fore (Bf) and aft (Ba), 60.0˚ fore (Cf) and aft (Ca) and 70.5˚ fore (Df) and aft (Da). The swath width of the imagery varied from 11km at nadir to 32km for the D cameras. Four spectral bands were centered at 446nm, 558nm, 672nm and 867nm (blue, green, red and near-infrared) (Diner et al., 1998).

The data used in this commentary were collected over three experimental forests in New England in August 2003 (Figure 1). The Bartlett Experimental Forest in north-central New Hampshire and the Harvard Forest in western Maine were both flown on 24 August, with data acquisition over Howland Forest in central Maine on 28 August. All three sites are well instrumented with standard meteorological equipment, biomass and carbon sequestration

Multiangular Imaging of Wetlands in New England 9

measurements to support long term experiments including NASA’s Forest Ecosystem Dynamics Project.

Howland ForestBartlett Forest

Harvard Forest

Figure 1. Locations of the three experimental forests in New England.

Only the data from the north-south runs and A-C cameras over each site were used due to data inhomogeneities. Georectified radiance product (L1B2) data were resampled to a 27.5m grid in the UTM (Universal Transverse Mercator) projection and available online from the Langely Distributed Active Archive Center (DAAC). Actual radiances were computed using the AirMISR tool. An IR minimum check was performed for each viewing angle. Ancillary digital data were acquired from the National Wetlands Inventory, Maine GIS, New Hampshire GIS and Massachusetts GIS.

2.2. Wetlands of the Study Sites The wetlands observed at the three study sites varied by extent, species composition, tidal

regimes and permanence of water. The Howland Forest site decreases in elevation from over 120m in the north to about 19m in the south, with palustrine, estuarine, evergreen as well as broad-leaf deciduous and persistent emergent wetlands. To the west, the Bartlett Experimental Forest site located in the White Mountains National Forest ranges in elevation from 59m to 1868 m, an upland area characterized by broadleaf deciduous forests with predominantly palustrine forested broad-leaf deciduous and needle-leaf evergreen wetlands. The Harvard Forest site was lower in elevation (9-548m) and characterized by both palustrine evergreen and freshwater forested shrub wetlands.


3. RESULTS AND APPLICABILITY OF MVA TO WETLAND STUDIES

The relationship between wetlands and view angles can be analyzed by scatterplots (not

shown) of camera pairs on which the 45˚ and best fit lines have been plotted. For the three experimental forests, there was a high degree of correlation between the high camera view angles (An, Af, Aa and Bf). The straight line relationship denotes a moisture gradient from mesic regions (low ratios) to xeric ones (high ones). The relationship is most extensive (with points at both ends of the 1:1 line) for the palustrine, estuarine wetlands of the Howland forest and less so for the other two regions. Differences in species composition and tidal flow regimes were marked across the three forests, influencing the view angles that were most useful for wetland discrimination. For example, at the Howland Forest wetlands, the scatterplot of the Af and Bf forward viewing angles (R2=0.904) was particularly well suited to highlighting moisture stress across forested wetlands, stress that was not observable at nadir (Dupigny-Giroux, 2007). This may be due to the fact that these seasonally flooded wetlands tend to be wetter for shorter durations during the growing season (Tiner, 2003). At the Harvard Forest, both the scatterplots of the high forward view angles (Af and Bf) as well as the nadir (An) and Af pair had the most significant best fit lines (R2 values of 0.946 and 0.957 respectively). The An-Af pairing was marginally better in that it deviated less from the 1:1 line than the Af-Bf pairing. For the Bartlett Forest wetlands, the regression statistics for camera pairs were quite low (R2 <0.3) due to the presence of several outliers which will be discussed shortly.

Figure 2 summarizes various wetland characteristics across view zenith angles at the three study sites. Bowl-shaped and bell-shaped reflectance anisotropy have been observed in other studies of forested areas (e.g. Pinty et al., 2002; Nolin, 2004; Asner, 2000; Goodin et al., 2004). Largest NIR/blue ratios and the most pronounced bowl-shaped anisotropy was observed for stressed estuarine aquatic and emergent vegetation (MF) at Howland Forest related to severe moisture stress in this saltwater habitat, and marked vertical variations in emergent and surface vegetation across the wetland. Almost all of the wetlands sampled at the Harvard Forest site displayed bowl-shaped anisotropy (FR1), peaking at the An view zenith angle and decreasing at the aft view angles. This may be a function of the solar illumination-view angle geometry at these latitudes, as well as the fact that most of wetlands in this region were freshwater in nature.

In contrast, the wetlands at the Bartlett and Howland Forests displayed darkspots or decreases in the NIR/blue ratio at one of the A cameras or the Bf one. At Howland Forest, the Bf darkspot was observed for an upland needle-leaf forested wetland (U), while at Bartlett Forest, it corresponded to a lotic (riverside) upper perennial wetland with an unconsolidated bottom (HW). The dramatic Af darkspot at the Howland site was found in a region of aquatic/emergent vegetation (SA) where the herbaceous species resembled evergreen forests at the other view angles, but residual wetland characteristics showed up as strong NIR absorption at the Af view angle. At the Bartlett Forest, the Af darkspot corresponded to a forested broad-leaf deciduous wetland that was temporally flooded (BD). Larger overall NIR/blue ratios for the latter wetland at all view zenith angles allowed its distinction from the aquatic/emergent species observed at Howland Forest.


0

0.5

1

1.5

2

2.5

Cf Bf Af An Aa Ba Ca

CAMERA

NIR

/blu

e ra

tio

W

MFFR1

AW

SA

HWBD

WA

FR2

DAU D1

D2HA

Figure 2. Variations in NIR/blue ratio as a function of viewing angle for selected wetlands. At Howland Forest - water and persistent emergent wetlands (W); water-dominated intertidal emergent and subtidal aquatic vegetation (WA); drying estuarine subtidal aquatic bed (DA); acid, saturated needle-leaf forested wetland (AW); moisture stressed subtidal aquatic bed/intertidal emergent vegetation (MF); dead forested wetland (D1); transitional subtidal aquatic and intertidal emergent vegetation (SA); and upland needle-leaved evergreen forested wetland (U). At Bartlett Forest - high elevation wetlands with a westerly aspect (HW); upper perennial wetland with an unconsolidated bottom (HA); dead needle-leaf forested wetland (D2); and broad-leaf deciduous wetland (BD). At Harvard Forest - freshwater scrub/shrub wetland (PF01/SS1E) FR1 and freshwater scrub/shrub wetland (PF01/4E) FR2. [Wetland codes according to Cowardin et al., 1979].

Similarly, An darkspots observed at Bartlett and Howland Forests were attributable to different wetland types, with distinct NIR/blue ratios. At Howland Forest, the An darkspot corresponded to upland, lotic, water-dominated subtidal and intertidal wetlands (WA), where the former displayed more vegetation on or below the water surface than the latter. The An darkspot at Bartlett Forest was observed for both a forested, needle-leaf evergreen wetland that was dead (D2), as well as for a scrub-shrub, broad-leaf evergreen wetland that was acidic. NIR/blue ratios were larger for the dead, forested wetland than for the acidic one, and were on the order of those observed for the subtidal wetlands at Howland Forest. Finally, the Aa darkspot observed for the freshwater forested/shrub wetland at Harvard Forest (FR2) was twice the magnitude of that observed for an anomalous acidic, saturated needle-leaved evergreen emergent wetland at Howland Forest (AW).

Two other important across-view angle observations deal with water bodies and the role of topography. Deep “pure” water bodies (W) displayed low, almost constant ratios at all angles, with largest values at the Bf and Cf view angles. High elevation wetlands with a westerly aspect (HW) were characterized by low NIR/blue ratios at the forward view angles, a peak at the Aa view angle and decreasing at the other aft angles.

Two important underlying factors that explain the differences in wetlands across New England were the antecedent moisture conditions and species composition. Precipitation deficits across lowland regions in northern New England (Howland Forest), were neither observed at high elevations of similar latitudes (Bartlett Forest) nor in southern New England (Harvard Forest). Tables 1 and 2 highlight the above normal precipitation received near Bartlett and Harvard Forests that was in stark contrast to the long term dryness that


characterized the Howland Forest (Dupigny-Giroux, 2007). The acceleration of phenology under drought conditions (Dupigny-Giroux, 2001) may have accounted for anomalously high near-infrared radiances observed in the most stressed estuarine wetlands in the Howland Forest. These dry intertidal emergent vegetation regions were characterized by a concentration of organic material near the surface of the water, and displayed the highest NIR/blue ratio of all wetlands sampled.

Table 1. Statewide precipitation received in 2003. Rankings are given in parentheses with 1 being the driest year and 113 being the wettest in the 1895-2007 time frame.

(Data courtesy National Climatic Data Center)

STATE AUGUST 2003 SUMMER 2003 SPRING 2003

Maine 92.7 mm (51) 255.8 (44) 224 mm (38)

New Hampshire 148.3 mm (107) 297.9 mm (66) 257.3 mm (53)

Massachusetts 140.5 mm (98) 346.7 mm (93) 321.3 mm (79)

Table 2. Monthly precipitation received at representative stations close to the study

sites. Monthly normals (statistical averages) for the 1971-2000 are given in parentheses. (Data courtesy National Climatic Data Center)

STATION JULY 2003 AUGUST 2003

Bangor, Maine 22.6 mm (82.3 mm) 50 mm (75.56 mm)

North Conway, NH 79.5 mm (102.1 mm) 187.9 mm (105.4 mm)

Amherst, MA 68.3 mm (100.3 mm) 202.9 mm (104.1 mm)

CONCLUSION Multiple view angles (MVA) or multiangular imaging represents a yet to be explored use

of the remote sensing of wetlands. The application of the NIR/blue ratio to wetlands at three experimental forests in New England revealed that the technique is independent of the vegetation physiology and more a function of the moisture conditions and species composition across the wetland types. Wetlands at the Howland Forest site were the most stressed, reflecting both the long-term and short term moisture deficits across the region. At the Harvard Forest, most of the wetland sampled were of similar vegetation density and health, displaying overlapping bowl-shaped anisotropy. Darkspots were observed at the A and B view angles at the Bartlett and Howland Forest sites, with differences in the magnitude of the NIR/blue allowing for distinction across wetlands of varying species composition, acidity, health and tidal regimes. Finally, the interplay among topographic aspect, solar illumination


and view zenith angle produced low NIR/blue ratios at forward view angles and higher values at aft ones for some high elevation, westerly sites.

ACKNOWLEDGMENTS These data were obtained from the NASA Langley Research Center Atmospheric

Sciences Data Center.

REFERENCES

Asner, G.P., Braswell, B.H., Schmil, D.S. and Wessman, C.A. (1998) Ecological Research Needs from Multiangle Remote Sensing Data, Remote Sensing of Environment, 63(2):155-165.

Asner, G.P. (2000). Contributions of Multi-view Angle Remote Sensing to Land-surface and Biogeochemical Research, Remote Sensing Reviews, 00, 1-26.

Barnsley, M.J., Allison, D. and Lewis, P. (1997). On the information content of multiple view angle (MVA) images, International Journal of Remote Sensing, 18, 1937-1960.

Bicheron, P., Leroy, M., Hautecouer, O., and Breon, F.M. (1997). Enhanced discrimination of boreal forest covers with directional reflectances from the airborne polarization and directionality of Earth reflectances (POLDER) instrument, Journal of Geophysical Research, 102(29), 517-528.

Cowardin, L.M., Carter, V., Golet, F.C. and LaRoe, E.T. (1979). Classification of Wetlands and Deepwater Habitats of the United States, U.S. Fish and Wildlife Service, Washington, DC, FWS/OBS-79-31.

Diner, D.J., Barge, L.M., Bruegge, C.J., Chrien, T.G., Conel, J.E., Eastwood, M.L., Garcia, J.D., Hernandez, M.A., Kurzweil, C.G., Ledeboer, W.C., Pignatano, N.D., Sarture, C.M. and Smith, B.G. (1998). The Airborne Multi-angle Imaging SpectroRadiometer (AirMISR): Instrument Description and First Results, IEEE Transactions on Geoscience and Remote Sensing, 36(4), 1339-1349.

Diner, D.J., Braswell., B.H., Davies, R., Gobron, N., Hu, J., Jin, Y., Kahn, R.A., Knyazikhin, Y., Loeb, N., Muller, J.-P., Nolin, A.W., Pinty, B., Schaaf, C.B., Seiz, G. and Stroeve, J. (2005). The value of multiangle measurements for retrieving structurally and radiatively consistent properties of clouds, aerosols and surfaces, Remote Sensing of Environment, 97, 495-518.

Dobson, J.E., Bright, E.A., Ferguson, R.L., Field, D.W., Wood, L.L., Haddad, K.D., Iredale III, H., Jensen, J.R., Klemas, V.V., Orth, R.J and Thomas, J.P. (1995). NOAA Coastal Change Analysis Program (C-CAP): Guidance for Regional Implementation, NOAA Technical Report NMFS 123, Department of Commerce.

Dupigny-Giroux, L-A. and Lewis, J.E. (1999). A Moisture Index for Surface Characterization over a Semiarid Area, Photogrammetric Engineering and Remote Sensing, 65(8), 937-945.


Dupigny-Giroux, L-A. (2001). Towards characterizing and planning for drought in Vermont. Part I. A Climatological Perspective, Journal of the American Water Resources Association, 37( 3), 505-525.

Dupigny-Giroux, L.-A. (2007). Using AirMISR to explore moisture-driven land use-land cover variations at the Howland Forest, Maine - A case study, Remote Sensing of Environment, 107, 376-384.

Goodin, D.G., Gao, J. and Henebry, G.M (2004). The Effect of Solar Illumination Angle and Sensor View Angle on Observed Patterns of Spatial Structure of Tallgrass Prairie, IEEE Transactions on Geoscience and Remote Sensing, 42(1) 154-165.

Hyman, A.H. and Barnsley, M.J. (1997). On the potential land cover mapping from multiple-view-angle (MVA) remotely-sensed images, International Journal of Remote Sensing, 18(11), 2471-2475.

Nolin, A.W. (2004). Towards retrieval of forest cover density over snow from the Multi-angle Imaging SpectroRadiometer (MISR), Hydrological Processes, 18(18), 3623-3636.

Pinty, B., Widlowski, J-L., Gobron, N., Verstraete, M.M., Diner, D.J. (2002). Uniqueness of Multiangular Measurement - Part I: An Indicator of Subpixel Surface Heterogeneity from MISR, IEEE Transactions on Geoscience and Remote Sensing, 40(7),1561-1573.

Privette, J.L. (1995). Uses of a Bidirectional Reflectance Model with Satellite Remote Sensing Data, Elements of Change 1995, Session I, AGCI.

Tiner, R.W. (2003). Correlating Enhanced National Wetlands Inventory Data with Wetland Functions for Watershed Assessments: A Rationale for Northeastern U.S. Wetlands, U.S. Fish and Wildlife Service, National Wetlands Inventory Program, Region 5, Hadley, MA. 26pp.

U.S. Fish and Wildlife Service (2004). Technical Procedures for Wetlands Status and Trends, Operational Version, Arlington, VA.

Valta-Hulkkonen, K., Pellikka, P., Tanskanen, H., Ustinov, A. and Sandman, O. (2003). Digital false colour aerial photographs for discrimination of aquatic macrophyte species, Aquatic Botany, 75, 71-88.

Widlowski, J.-L., Pinty, B., Gobron, N., Verstraete, M.M., Diner, D.J. and Davis, A.B. (2004). Canopy structure parameters derived from multi-angular remote sensing data for terrestrial carbon studies, Climate Change, 67, 403-415.

Zhang, Y., Tian, Y., Myneni, R.B., Knyazikhin, Y. and Woodcock, C.E. (2002a). Assessing the information content of multiangle satellite data for mapping biomes. I. Statistical analysis, Remote Sensing of Environment, 80, 418-434.

Zhang, Y., Shabanov, N., Knyazikhin, Y. and Myneni, R.B., (2002b). Assessing the information content of multiangle satellite data for mapping biomes. II. Theory, Remote Sensing of Environment, 80, 435-446.


Chapter 1

WETLANDS: WATER “LIVING FILTERS”? Ana Dordio1∗, A. J. Palace Carvalho1,2 and Ana Paula Pinto1,3

1Department of Chemistry, University of Évora, Évora, Portugal 2Centro de Química de Évora, University of Évora, Évora, Portugal

3ICAM – Instituto de Ciências Agrárias e Mediterrâneas, University of Évora, Évora, Portugal

ABSTRACT

Human societies have indirectly used natural wetlands as wastewater discharge sites for many centuries. Observations of the wastewater depuration capacity of natural wetlands have led to a greater understanding of the potential of these ecosystems for pollutant assimilation and have stimulated the development of artificial wetlands systems for treatment of wastewaters from a variety of sources. Constructed wetlands, in contrast to natural wetlands, are human-made systems that are designed, built and operated to emulate wetlands or functions of natural wetlands for human desires or needs. Constructed wetlands have recently received considerable attention as low cost, efficient means to clean-up not only municipal wastewaters but also point and non-point wastewaters, such as acid mine drainage, agricultural effluents, landfill leachates, petrochemicals, as well as industrial effluents. Currently, untreated wastewater discharge in the natural wetlands sites is becoming an increasingly abandoned practice whereas the use of constructed wetlands for treatment of wastewater is an emerging technology worldwide. However, natural wetlands still play an important role in the improvement of water quality as they act as buffer zones surrounding water bodies and as a polishing stage for the effluents from conventional municipal wastewater treatment plants, before they reach the receiving water streams. In fact, one of the emerging issues in environmental science has been the inefficiency of wastewater treatment plants to remove several xenobiotic organic compounds such as pesticides and pharmaceutical residues and consequent contamination of the receiving water bodies. Recent studies have shown that wetlands systems were able to efficiently remove many of these compounds, thus reaffirming the importance of the role which can be played by wetlands in water quality preservation.

∗ Corresponding author: Tel: +351 - 266 745343; E-mail address: [email protected]

Ana Dordio, A. J. Palace Carvalho and Ana Paula Pinto 16

The aim of this work is to present a review on the application of wetlands as “living filters” for water purification. Emphasis was focused on the removal of micropollutants, especially xenobiotic organic compounds such as pharmaceuticals residues, which are not efficiently removed by conventional municipal wastewater treatment plants. Furthermore, the role of wetlands as protection zones which contribute to the improvement of the aquatic ecosystems’ quality will be discussed.

1. INTRODUCTION Wetlands have been recognized throughout human history to be a valuable natural

resource. Their importance has been appreciated in managed forms, for example rice paddies, particularly in South East Asia, but also in their natural state by such people as the Marsh Arabs around the confluence of the rivers Tigris and Euphrates in southern Iraq (Mitsch and Gosselink J.G., 2000). Benefits provided by wetlands include water supply and control, mining, use of plants, wild-life, integrated systems and aquaculture, erosion control, education and training, recreation and reclamation (USEPA and USDA-NRCS, 1995; Cooper et al., 1996; Vymazal et al., 1998a; USEPA, 2000; Sundaravadivel and Vigneswaran, 2001).

The water purification capability of wetlands, in particular, has for long been recognized. Natural wetlands usually improve the quality of water passing through them, acting as “living filters” and serving as transitional zones or “ecotones” between terrestrial and aquatic systems (Mitsch and Gosselink J.G., 2000; Sundaravadivel and Vigneswaran, 2001). They provide several physical, chemical, and biological processes which allow for the depuration of pollutants resulting from point and non-points source, thereby contributing for an improvement of water quality. In many regions, natural wetlands have been used, for centuries, as convenient wastewater discharge sites and sinks for a wide variety of anthropogenic pollutants including toxic organic compounds (Kadlec and Knight, 1996; Vymazal, 1998).

Despite the many benefits offered by wetlands, some of these areas have also been for long regarded as wasted, useless land, unsuitable for agriculture, and subjected to drainage in order to make them available for cultivation. Since the early 20th century, wetland losses attributed to agriculture have been dramatic. Extensive wetland draining in the 1960s and 1970s led to increased available agricultural production acreage. However, in many places, in parallel with this practice, a decline has been observed in the water quality of the neighboring water bodies. An aggravated potential for damage to the aquatic systems, such as increased sedimentation or fish kills, is expected when rivers, streams, and lakes adjacent to the former wetland areas receive runoff following rainfall events. The functional role of wetlands in improving water quality has been, in recent years, a compelling argument for their preservation and study. The realization of the importance of wetlands, adjacent to other water bodies, has resulted in the fact that drainage of wetlands has ceased in many countries and that even previously drained wetlands are restored.

The water treatment capabilities of wetlands are thus generally recognized but the extent of their treatment capacity is largely unassessed. However, studies have led to both a greater understanding of the potential of natural wetland ecosystems for pollutant assimilation and the design of new natural water treatment systems inspired in these natural systems. There remain, however, concerns over the possibility of harmful effects resulting from toxic

Wetlands: Water “Living Filters”? 17

compounds and pathogens that may be present in many wastewater sources. Also, there are concerns that there may be a potential for long-term degradation of natural wetlands due to the addition of nutrients and changes in the natural hydrologic conditions influencing these systems. At least in part due to such concerns, there has been a growing interest in the use of constructed wetlands systems (CWS) for wastewater treatment.

Significant advances have been made in the last years in the engineering knowledge of creating artificial wetlands that can closely imitate the treatment functions that occur in the natural wetland ecosystems. These CWS can be defined as man-made systems that have been designed and constructed to utilize the natural processes involving wetland vegetation, soils, and their associated microbial populations to assist in treating wastewater (Hammer and Bastian, 1989; Vymazal, 1998). They are designed to take advantage of many of the same processes that occur in natural wetlands, but do so within a more controlled environment.

While some CWS have been designed and operated with the sole purpose of treating wastewater, others have been implemented with multiple-use objectives in mind, such as using treated wastewater effluent as a water source for the creation and restoration of wetland habitat for wildlife use and environmental enhancement.

The construction of artificial wetlands for the treatment of wastewater has been developing fast over the last decades and it represents now a widely accepted and an increasingly common treatment alternative. The many advantages offered by CWS such as simplicity of design and lower costs of installation, operation, and maintenance make them an appropriate alternative for both developed and developing countries (USEPA and USDA-NRCS, 1995; Vymazal et al., 1998a; USEPA, 2000).

In the following sections of this text the water purification functions of natural wetlands are presented in greater detail and, afterwards, the constructed wetland systems are described. Studies on the removal of several types of pollutants by these systems, as well as the more recent new trends on the use of CWS for pollutants removal are discussed.

2. NATURAL WETLANDS

2.1. Definition and Characterization Wetlands encompass a broad range of wet environments, ranging from submerged coastal

grass beds to salt marshes, swamp forests and boggy meadows (USEPA and USDA-NRCS, 1995; Kadlec and Knight, 1996; Vymazal et al., 1998a; Sundaravadivel and Vigneswaran, 2001).

A single precise definition and classification for wetlands that can correctly describe them in a comprehensive way for most purposes still has not been developed. In fact, a definition for wetlands has long been the subject of debate, due to the great variety of environments which are, either permanently or seasonally, influenced by water as well as to the specific requirements of the diverse groups of people involved in the study and management of these systems. Terms used in wetlands classification are many and are often confusing. Still, they are important for both the scientific understanding of these systems and for their proper management.


The Ramsar Convention, which in 1971 brought a worldwide attention to wetlands, took a broad approach in its definition in the text of the Convention (Article 1.1) (UNESCO, 1994):

“Wetlands are areas of marsh, fen, peatland or water, whether natural or artificial,

permanent or temporary, with water that is static or flowing, fresh, brackish or salt, including areas of marine water the depth of which at low tide does not exceed six metres”. In addition, for the purpose of protecting coherent sites, the Article 2.1 stipulates in order

to be included in the Ramsar List of internationally important wetlands, they may incorporate:

“riparian and coastal zones adjacent to the wetlands, and islands or bodies of marine water deeper than six metres at low tide lying within the wetlands”. Another definition of wetlands was proposed by the U.S. Fish and Wildfire Services,

which also developed a classification system capable of encompassing and systematically organizing for scientific purposes all types of wetland habitats. According to such view, wetlands are described as the transition areas between terrestrial and aquatic systems, where water is the dominant factor determining soil characteristics and development of associated biological communities. The definition specifies that wetlands need, at least periodically, to fulfill one or more of the following four requirements (Hammer and Bastian, 1989; Cowardin and Golet, 1995; Brady and Weil, 2002):

· areas where the water table is at or near the surface or where the land is covered by

shallow water; · areas supporting predominantly hydrophytes (water-tolerant plant species); · areas with predominantly undrained hydric soils. Hydric soils are those that are

sufficiently wet for long enough to produce anaerobic conditions, thereby limiting the types of plants that can grow on them;

· areas with non-soil substrate (such as rock or gravel) that are saturated or covered by shallow water at some time during the growing season of plants.

The U.S. Fish and Wildlife Service classification system (Cowardin and Golet, 1995) is

much like the hierarchical system used by scientists to classify plants and animals, starting out with five large systems and progressively subdividing into a series of subsystems, classes and subclasses, which are then characterized by examples of dominant types of plants and animals. This system thus provides a consistent standard of terminology to be used among scientists and specialists.

Various legislation and agency regulations define wetlands in more general terms. The U.S. Environmental Protection Agency (EPA) and the U.S. Army Corps of Engineers jointly define wetlands as (Environmental Laboratory, 1987):

“Those areas that are inundated or saturated by surface or ground water at a frequency

and duration sufficient to support, and that under normal circumstances do support, a prevalence of vegetation typically adapted for life in saturated soil conditions. Wetlands generally include swamps, marshes, bogs, and similar areas”.


One of the most recent definitions has been used by the U.S. National Research Council, which defines wetlands as (National Research Council, 1995; Vymazal et al., 1998a):

“an ecosystem that depends on constant or recurrent, shallow inundation or saturation at

or near the surface of the substrate”. Notwithstanding which of the definitions one may use, three major factors are salient in

the characterization of a wetland: water (hydrology), substrate (physico-chemical features) and biota (type of vegetation and microbial activity)(National Research Council, 1995).

The characteristics of all these three components are interdependent and conditioned by each other. It is from the complex interactions among them that results the values and functions of wetlands.

2.2. Natural Wetlands as Natural Water “Living Filters”? Long regarded as wastelands, wetlands are now recognized as important features in the

landscape that provide numerous beneficial functions for people and for fish and wildlife. These services, considered valuable to societies worldwide, are the result of the wetlands’ inherent and unique natural characteristics.

Wetlands support a rich diversity of wildlife and fisheries by being stopping-off points and nesting areas for migratory birds and spawning grounds for fish and shellfish. Those wetlands along the coasts, riverbanks and lakeshores have a valuable role in stabilizing shorelands and protecting them from erosion. One of the greatest benefits of inland wetlands is the natural flood control or buffering provided for downstream areas by slowing the flow of floodwater, desynchronizing the peak contributions of tributary streams and reducing peak flows on main rivers (Hammer and Bastian, 1989; USEPA and USDA-NRCS, 1995). Some wetlands may function as discharge areas for groundwaters, allowing stored groundwater to sustain surface base flow in streams during dry periods (Hammer and Bastian, 1989). Additional benefits provided by these natural systems include mining (peat, sand, gravel), use of plants (staple food plants, grazing land, timber, paper production, roofing, agriculture, horticulture, fodder), integrated systems and aquaculture (e.g. fish cultivation combined with rice production), energy (hydroelectric, solar energy, heat pumps, gas, solid and liquid fuel) education and training, recreation and reclamation (USEPA and USDA-NRCS, 1995; Vymazal et al., 1998a).

Perhaps one of the most important but least understood functions of wetlands is water quality improvement. These systems may well be considered “natural purifiers of water” as they provide an effective treatment for many kinds of water pollution (Hammer and Bastian, 1989; Kadlec and Knight, 1996). In fact, a capacity has already been recognized in wetlands to efficiently reduce or remove large amounts of pollutants from point sources (e.g. municipal and certain industrial effluents) as well as nonpoint sources (e.g. mining, agricultural and urban runoff) including organic matter, suspended solids, excess of nutrients, pathogens, metals and other micropollutants (Hammer and Bastian, 1989).

This pollutants removal is accomplished by the interdependent action of several physical, chemical and biological processes which include sedimentation, filtration, chemical precipitation, sorption, biodegradation, plants uptake among others. The mechanisms and the


interdependences among the wetlands components (water, substrate and biota) are complex and not yet entirely understood, although some progresses have been achieved in the latest years as the awareness to the water depurative functions of wetlands becomes more widespread.

The hydrology of the sites, the soil and the biota (vegetation and microorganisms) are reportedly the main factors influencing water quality in wetlands.

2.3. Factors Influencing Water Quality The hydrological cycle emerges with an underlying major role as it influences the type of

vegetation, microbial activity and biogeochemical cycling of nutrients in soil. Indeed, the biotic status of a wetland is intrinsically linked to the hydrological factors, which affect the nutrient availability as well as physicochemical parameters such as soil and water pH and anaerobiosis within soils. In turn, biotic processes will have an impact upon the hydrological conditions of a wetland (Mitsch and Gosselink J.G., 2000; Kivaisi, 2001).

Soil is one of the most important physical components of natural wetlands. Physical and chemical characteristics of the soil can greatly influence the type of wetland plants that will be prevalent over its surface and the microbial populations that will live on the soil phase. In addition, soil characteristics such as acid-base properties, redox potential and sorption capacity will also determine the type of physical-chemical processes occurring within the aqueous medium that can be responsible for the removal of certain types of pollutants, with a major impact in water quality.

Macrophytes are the dominant vegetation in wetlands. These plant species are typically adapted to water saturated conditions and are able to persist in anaerobic soil conditions as a result of high water content. Compared with the vegetation of well-drained soils, wetland plants have a worldwide similarity which overrides climate and is imposed by the common characteristics of a free water supply and the abnormally hostile chemical environment which plant roots must endure. It is not surprising that plants regularly found in wetlands have evolved functional mechanisms to deal with the environmental stresses.

Virtually all wetland plants have elaborate structural mechanisms to avoid root anoxia. Rhizosphere oxygenation is considered essential for active root function, and also enables the plants to counteract the effects of soluble phytotoxins, including sulfides and metals, which may be present at high concentrations in anoxic substrates.

Microorganisms play a central role in the biogeochemical transformations of nutrients (Cooper et al., 1996; Vymazal et al., 1998b; Stottmeister et al., 2003; Vymazal, 2007), and the metabolism of organic compounds including even some xenobiotic compounds (Machate et al., 1997; Stottmeister et al., 2003). There is a close interdependence between microorganisms and vegetation. Much of the plants nutrients are the result of the mineralization of more complex compounds by the microorganisms, whereas the activity of the latter is stimulated by enzymes released in root exudates. The aerobic conditions provided by the plants in their rhizosphere will also be determinant for the type of microbial populations and bioprocesses available in this region.

The conjugation of all these factors and interactions between different wetlands components and their associated processes lead to essentially different abilities to interact with water pollutants by different types of wetlands. An overview of the abilities by different


types of natural wetlands to cope with various non-point pollution problems along with a summary of each wetland type characteristics is presented in table 1.

Table 1. Characteristics of different types of natural wetlands and their ability to retain

non-point source pollutants (IETC-UNEP, 1999)

Type of Wetland Characteristics Ability to retain non-point source

pollutants Wet meadows Grassland with waterlogged

soil; standing water for part of the year

Denitrification only in standing water; removal of nitrogen and phosphorus by harvest

Fresh water marshes

Reed-grass dominated, often with peat accumulation

High potential for denitrification, which is limited by the hydraulic conductivity

Forested wetlands

Dominated by trees, shrubs; standing water, but not always for the entire year

High potential for denitrification and accumulation of pollutants, provided that standing water is present

Salt water marshes

Herbaceous vegetation, usually with mineral soil

Medium potential for denitrification; harvest possible

Bogs A peat-accumulating wetland with minor flows

High potential for denitrification but limited by small hydraulic conductivity

Shoreline wetlands

Littoral vegetation of significant importance for lakes and reservoirs

High potential for denitrification and accumulation of pollutants, but limited coverage

In spite of the general abilities for water depuration that can be observed for typical

wetland configurations, it is very difficult to predict responses to water pollution by natural wetlands and to translate their behavior from one geographical area to another due to the extreme variability of the functional components that characterizes them. Therefore, natural wetlands can not be viewed as a systematic approach to wastewater treatment. Only in a controlled environment provided by constructed wetlands, can such treatment work in a reliable and reproducible manner.

2.4. Natural Wetlands Preservation through the Construction of Artificial Wetlands

Currently, in many countries, natural wetlands are protected areas, thus meaning that

wastewater application is not permitted. Although studies have shown that natural wetlands are able to provide high levels of wastewater treatment (Hammer and Bastian, 1989), there is concern over possible harmful effects of toxic compounds and pathogens in wastewaters and a long term degradation of wetlands due to additional nutrient and hydraulic loadings from wastewater. These potential benefits and concerns have promoted a growing interest in the use of artificial wetlands for wastewater treatment (USEPA and USDA-NRCS, 1995).

While the application of wastewater to natural wetlands is becoming a deprecated practice worldwide, studies conducted on these systems have led to both a greater understanding of the processes involved in pollutant assimilation and removal as well as


suggesting the design of new natural water treatment systems where these water depuration capabilities would be imitated and improved (Vymazal et al., 1998a). Using the knowledge and the experience with the assimilative capacity of natural wetlands, the construction of man-made wetland systems (constructed wetlands) for wastewater treatment has therefore been proposed. The functional role of wetlands in improving water quality, thus, becomes a compelling argument for the preservation of natural wetlands and in recent years the construction of wetlands systems for wastewater treatment.

Constructed wetlands can be built with a much greater degree of control, using a well-defined composition of substrate, type of vegetation, and flow pattern. In addition, constructed wetlands offer several additional advantages compared to natural wetlands including site selection, flexibility in sizing and, most importantly, control over the hydraulic pathways and retention time (Brix, 1993).

3. CONSTRUCTED WETLANDS SYSTEMS Recent concerns over wetlands losses have generated a need for creation of wetlands,

which are intended to emulate the functions and values of natural wetlands that have been destroyed. Such CWS can be defined as a designed and man-made complex of saturated substrates, emergent and/or submergent vegetation, animal life, and water that simulates natural wetlands for human use and benefits (Hammer and Bastian, 1989). Artificial wetlands are engineered or constructed for one or more of the following purposes as indicated by specific descriptive terminology (Kadlec and Knight, 1996; Sundaravadivel and Vigneswaran, 2001):

creation of wetland habitats to compensate for natural wetlands which have been converted for agriculture and urban development (or help offset their rate of conversion) and hence to conserve native flora and fauna including aquatic plants, fish, water birds, reptiles, amphibians and invertebrates (Constructed habitat wetlands)

• flood control facility (Constructed flood control wetland) • production of food and fiber (Constructed aquaculture wetlands) and • water quality improvement and wastewater treatment system (Constructed treatment

wetlands). Although CWS are being developed in many parts of the world for various functions,

their wastewater treatment capabilities have attracted research efforts for a wide range of treatment applications including domestic wastewaters (Hammer and Bastian, 1989; Cooper et al., 1996; Vymazal et al., 1998a; Kivaisi, 2001; Cooper, 2001; Cameron et al., 2003; Hench et al., 2003; Solano et al., 2004), urban storm-water (Livingston, 1989; USEPA, 1993; Scholes et al., 1998; Carleton et al., 2001; Kohler et al., 2004), agricultural wastewaters (Carty et al., ; Hammer et al., 1989; Cronk, 1996; Knight et al., 2000), landfill leachates (Barr and Robinson, 1999; Nivala et al., 2007), acid mine drainage (Brodie et al., 1989; Ledin and Pedersen, 1996; Sobolewski, 1999) and for polishing advanced treated municipal wastewater for return to freshwater resources. CWS are also used for treating eutrophic lake waters (D'Angelo and Reddy, 1994; Coveney et al., 2002) and for conservation of nature (Hammer


and Bastian, 1989; Brix, 1993). For the specific use of wastewater treatment, the systems are also denominated constructed treatment wetland systems (CTWS) and can be defined as engineered systems that have been designed and constructed to utilize the natural processes involving wetland vegetation, soils and their associated microbial assemblages to assist in treating wastewater (Vymazal, 1998; Vymazal, 2005a). CTWS are designed to take advantage of many of the same processes that occur in natural wetlands, but do so within a more controlled environment.

The first experiments aimed at the possibility of wastewater treatment by wetland vegetation were conducted in the 1950’s by Seidel at the Max Planck Institute in Plön, Germany (Seidel, 1955). During the following two decades, in the 1960’s and 1970’s, Seidel proceeded with numerous experiments on the use of wetland plants for treatment of various types of wastewater (Vymazal et al., 1998a). However, it was not before over 20 years of research that the first operational full-scale constructed wetland for municipal wastewater was built in Europe, in Othfresen, Germany (Vymazal et al., 1998a). Meanwhile, research efforts in the U.S. were developing during the 1970’s and 1980’s. Some systems were installed in the 1970’s with an increasing number in the 1980’s. The 1990’s saw a major increase in the number of these systems as the application expanded for use not only to treat municipal wastewater, but also urban stormwater, industrial, mining and agricultural wastewaters. CWS have since been constructed and used all around the world to treat a wide variety of types of wastewater. A list documenting the history of the first uses (both experimental and operational) of macrophytes and/or CWS for the treatment of different types of pollution up until the first half of the 1990’s was compiled and presented by Vymazal (1998a).

The number of CWS in use has very much increased in the most recent years. The use of CWS in the United States, New Zealand and Australia is gaining rapid interest. Most of these systems are used for tertiary treatment from towns and cities, usually using surface-flow systems to remove low concentrations of nutrients (N and P) and suspend solids. Conversely, in European countries, these CWS are usually used to provide secondary treatment of domestic sewage for village populations. These constructed wetland systems have been seen as an economically attractive and energy-efficient way of providing high standards of wastewater treatment.

3.1. Advantages and Limitations of CWS Constructed wetlands systems offer effective and a reliable treatment to wastewater in a

simple and inexpensive manner. Its application for wastewater treatment offers many advantages in comparison to more conventional wastewater treatment technologies (USEPA and USDA-NRCS, 1995; USEPA, 2000; Sundaravadivel and Vigneswaran, 2001; Haberl et al., 2003). Constructed wetlands:

• can often be less expensive to build than other treatment options; • can be built and operated simply; • utilize natural processes; • their operation and maintenance expenses (energy and supplies) are low; • their operation and maintenance require only periodic, rather than continuous, on-site

labor;


• are able to tolerate fluctuations in flow, (tolerate fluctuations in flow and pollutant concentrations);

• are able to treat wastewaters with very different constituents and concentration; • are characterized by a high process stability (buffering capacity); • are characterized by low excess sludge production; • facilitate water reuse and recycling. Additionally, constructed wetlands (USEPA and USDA-NRCS, 1995; USEPA, 2000;

Sundaravadivel and Vigneswaran, 2001; Haberl et al., 2003): • provide habitat for many wetland organisms; • can be built to fit harmoniously into the landscape; • provide numerous benefits in addition to water quality improvement, such as wildlife

habitat and the aesthetic enhancement of open spaces; • provide recreational and educational opportunities; • are an “environmentally sensitive” approach viewed with favor by the general public. However, there are also some limitations to the use of CWS in treating wastewater

(USEPA and USDA-NRCS, 1995; USEPA, 2000; Sundaravadivel and Vigneswaran, 2001): • they generally require large land area for the same level of treatment by conventional

systems making them unsuitable for the treatment for sources that generate large quantities of wastewater, such as large cities;

• they may be relatively slow to provide treatment compared to more conventional treatment technologies;

• performance may be less consistent than in conventional treatment, CW depend on climate and, thus, may have reduce efficiencies during colder seasons;

• they require long period, typically two or three plant growing seasons, for the vegetation before optimal treatment efficiencies are achieved;

• the process dynamics of the CWS are yet to be clearly understood leading to imprecise design and operating criteria;

• these systems typically lie outdoor and spread over large area, their performance is susceptible to storm, wind, and floods;

• they require a base flow of water, they can tolerate temporary water level drawdowns, but not complete drying;

• there may provide breeding grounds for mosquitoes and other pests, though proper design can reduce this problem;

• long-term maintenance may be required; • the biological components are sensitive to toxic pollutants; • contaminant accumulation must be monitored to maintain ecological health of the

system


3.2. Types of CWS The basic CWS classification is based on the type of water flow regime. Depending on

the level of water column with respect to the substrate bed, two general forms of CWS are used in practice:

• free water surface (FWS) wetlands (also called surface flow (SF) wetlands or aerobic

wetlands), and • subsurface flow (SSF) wetlands (also known as vegetated submerged bed (VSB)

systems). FWS closely resemble natural wetlands in appearance because they contain aquatic plants

that are free floating or rooted in a soil layer on the bottom of wetland. In this type of CWS the water flows horizontally through the leaves and stems of the plants, above the substrate (figs. 1A, 1B and 1C). The near-surface layer of water is aerobic while the deeper waters and the substrate are usually anaerobic (USEPA and USDA-NRCS, 1995).

Free water wetlands can further be sub-classified according to the dominant type of macrophytes growing in the system. This type of wetland systems can be (Brix, 1993; USEPA and USDA-NRCS, 1995; Vymazal et al., 1998b; USEPA, 2000):

• a floating macrophyte system – these systems make use both of floating species that

are rooted in the substrate (e.g. Nymphae spp Nuphar spp. (waterlilies), Potamogeton natans (pondweed), Hydrocotyle vulgaris (pennyworth)) and species which are free floating on water surface (e.g. Eichhornia crassipes (water hyacinth), Lemna spp., Spirodella spp. (duckweed));

• a submerged macrophyte system – the plants used in these systems have their photosynthetic tissue entirely submerged with the flowers being exposed to the atmosphere. Two types of submerged aquatics are usually recognized: the elodeid type (e.g. Elodea spp., Myriophyllum aquaticum (parrot feather), Ceratophyllum spp.) and the isoetid (rosette) type (e.g. Isoetes, Littorella, Lobelia);

• a rooted emergent macrophyte system – these systems use plants which are the dominating form of life in the natural wetlands. Plants grow at well above the water level, producing aerial stems and an extensive root and rhizome system. These comprise species like the Phragmites australis (common reed), Thypha spp. (cattails), Scirpus spp. (bulrushes), Iris spp. (blue and yellow flags) Juncus spp. (rush), Saggitaria latifolia (duck potato), Phalaris arundinocea (reed canary grass), Carex spp. (Sedges), Zizania aquatica (wild rice), Eleocharis spp. (Spikerushes) and Glyceria spp. (mannagrasses).

On the contrary, SSF do not resemble natural wetlands because they do not have standing

water; they are constructed with a substrate (usually small rocks, gravel, sand or soil) which has been planted with aquatic plants, and the wastewater flows beneath the surface of the support matrix but in contact with the plants roots (USEPA and USDA-NRCS, 1995; Vymazal et al., 1998a). SSF systems (which, by definition, must be planted with emergent macrophytes) can be sub-classified according to their water flow patterns (figs. 1D and 1E). (Cooper et al., 1996; Vymazal et al., 1998a):


• horizontal SSF systems • vertical SSF systems ( upstream or downstream characteristics) • hybrid systems In the horizontal SSF (fig. 1 D) the wastewater is fed in at the inlet and flows slowly

through the porous medium under the surface of the bed in a more or less horizontal path (the bottom of the bed will actually have a slight slope to induce the liquid flow) until it reaches the outlet zone where it is collected before leaving via level control arrangement at the outlet. During the passage the wastewater will come into contact with a network of aerobic, anoxic and anaerobic zones. The aerobic zones occur around roots and rhizomes that leak oxygen into the substrate.

In vertical SSF (fig. 1 E) wastewater is applied through different arrangements of wastewater feeding and collection mechanisms to maintain a vertical direction of flow, either descending or ascending, through the substrate. This is achieved either by intermittent wastewater application or by burying inlet pipes into the bed at a depth of 60 to 100 cm. Since the wastewater infiltrates through the substrate bed, this type of wetlands are also called “infiltration wetlands”.

Hybrid systems comprise vertical-flow (vertical SSF) and horizontal-flow (horizontal SSF) systems arranged in a staged manner. There are as yet few of these systems in operation and it is perhaps too early to decide which is the best arrangement.

The SSF type of wetland is thought to have several advantages over the FWS (USEPA and USDA-NRCS, 1995; USEPA, 2000). SSF wetlands show, in general, higher contaminant removal efficiencies than FWS wetlands per unit area of land occupation. In SSF wetlands the substrate provides more surface area for the development of microorganisms which in part is responsible for its increased efficiency. This may allow a smaller area occupation requirement for SSF wetlands in comparison with FWS, in order to achieve a particular level of treatment. As the water surface in a SSF is below the support matrix surface there is also little risk of odors and insect vectors which are associated with the standing wastewaters of the FWS as well as a minimal risk of public or animal exposure.

Figure 1. Different types of CWS (A, FWS with free-floating plants; B, FWS with submerged plants; C, FWS with emergent plants; D, Horizontal SSF; E, Vertical SSF).


The accumulation of plant debris in the surface of the SSF can additionally provide some thermal protection in colder climates. Percolation of the wastewater through the substrate in the SSF systems also provides faster depuration with the respect to the organic pollutants and nutrients (USEPA and USDA-NRCS, 1995; USEPA, 2000). In table 2 a comparison of the advantages and disadvantages for each type of constructed wetlands, according to water flow, is presented.

Table 2. Advantages and disadvantages of free water surface and subsurface flow

wetlands (adapted from Halverson (2004))

Free water surface wetland

Advantages Disadvantages · Less expensive to construct and operate and simpler to

design than SSF wetlands and conventional treatment methods

· Can be used for wastewater with higher suspended solids content

· Offer greater flow control than SSF wetlands · Offer more diverse wildlife habitat · Provides habitat for plants and wildlife

· Lower rates of contaminant removal per unit of land than SSF wetlands, thus they require more land to achieve a particular level of treatment than SSF wetlands

· Requires more land than conventional treatment methods

· Risk of ecological or human exposure to surface-flowing wastewater

· May be slower to provide treatment than conventional treatment

· Odor and insects may be a problem due to the free water surface

Subsurface flow wetland

Advantages Disadvantages · Higher rates of contaminant removal per unit of land

than FWS wetlands, thus they require less land to achieve a particular level of treatment than FWS wetlands

· Lower total lifetime costs and capital costs than conventional treatment systems

· Less expensive to operate than FWS systems · More accessible for maintenance because there is no

standing water · Odors and insects not a problem because the water level

is below the media surface · Provides habitat for plants and wildlife · Minimal ecological risk due to absence of an exposure

pathway

· Requires more land than conventional treatment methods

· May be slower to provide treatment than conventional treatment

· More expensive to construct than FWS wetlands on a cost per acre basis

· Waters containing high suspended solids may cause plugging

There are more than twice as many FWS systems in the United States than SSF systems

(Halverson, 2004). In Europe SSF systems outnumber FWS systems. In general, FWS wetlands require more land to achieve the same pollution reduction as SSF wetlands, but are easier and cheaper to design and build (USEPA, 1993; USEPA and USDA-NRCS, 1995). The SSF technology is based on the work of Seidel (Seidel, 1955). Since then the technology has grown in many European countries and is nowadays applied worldwide.


3.3. CWS Components Constructed wetlands are engineered systems of wastewater treatment that are designed

and constructed to utilize the natural processes that occur in natural wetlands. A constructed wetland consists of a properly-designed basin that contains soils or other selected substrate, water column, and wetland vegetation, as the main elements. Other important components that assist in the treatment of the wastewater, such as the communities of microorganisms, develop naturally. The concerted action of all these components (substrate, macrophytes and microbial population adapted to the wastewater toxicity) through a variety of chemical, physical and biological processes, is responsible for the depuration of wastewaters achieved in a CWS (figure 2).

Aquatic plants play a central role in the depuration mechanisms occurring in a CWS as they provide support for and stimulate the microorganisms’ growth and, as well, promote the removal of a variety of pollutants by their adsorption, uptake or phytodegradation. Their role in the direct degradation and uptake of organic pollutants, in particular xenobiotics of anthropogenic origin, is not as important as the microorganisms’ action, but in many cases is still very significant. Microorganisms have, in fact, the major role in the biodegradation of this type of contaminants, through their metabolic transformations. In addition, they also have a participation in the removal of some inorganic compounds such as in the nitrogen cycle.

Figure 2. Summary of the major physical, chemical and biological processes controlling pollutant removal in CWS.

In the case of a SSF wetland, the substrate component is also fundamental not only because it provides the physical support for the macrophytes and microorganisms


development but also since it can promote a series of chemical and physical processes which additionally contribute to the decontamination the polluted wastewater.

3.4. Pollutant Removal Processes A number of physical, chemical and biological processes operate concurrently in

constructed and natural wetlands to provide the removal of contaminants (Figure 2). These processes include, among others, sedimentation, plant uptake, microbial

degradation, sorption, precipitation and volatilization. Removal of pollutants from water may be accomplished through storage in the wetland substrate and vegetation, or through losses to the atmosphere. A basic understanding of how these processes operate in wetlands is extremely helpful for assessing the potential applications, benefits and limitations of CWS.

3.4.1. Physical Processes

CWS are capable of providing highly efficient physical removal of pollutants associated with particulate matter in the wastewater through sedimentation and filtration. Physical pollutant removal processes in wetland systems occur mainly due to the presence of plant biomass and, in the case of SSF wetlands, also to the substrate media.

Sedimentation of suspended solids is promoted by the low flow velocity and by the fact that the flow is often laminar in wetlands. The pathways of wastewater are retarded in a CWS due to the resistance to flow provided by vegetation, which thereby enhances sedimentation of suspended solids. In addition to serving as sediment traps, a primary role in suspended solids removal by floating plants is also to limit re-suspension of settled particulate matter. On the other hand, the media acts as filter beds, similarly as in filtration processes, and in this manner contribute to the physical removal of suspended solids through straining.

Volatilization, which involves diffusion of a dissolved compound from the water into the atmosphere, is another potential means of contaminant removal in wetlands. Some of the simpler inorganics resulting from mineralization (such as ammonia) and many types of organic compounds are volatile, and are readily lost to the atmosphere from wetlands and other surface waters. Although volatilization can effectively remove certain contaminants from the water, it may prove to be undesirable in some instances, due to the potential for polluting the air with the same contaminants and, subsequently, contaminate neighboring areas through rainfall or dustfall.

3.4.2. Chemical Processes

A wide range of chemical processes are involved in the removal of contaminants in wetlands. The most important chemical removal processes in wetland soils are sorption and precipitation. Other chemical processes such as redox reactions, hydrolysis and complexation, can result in transformations of the pollutants which can be accessory to removal either through precipitation or adsorption. Processes such as photolysis and ionic exchange with mineral components of the substrates can also contribute to the removal of some particular classes of pollutants.

Sorption refers simultaneously to both adsorption and absorption phenomena, and the term is used whenever the extent to which each phenomenon is responsible for the compound’s removal is not clear or well defined. These chemical processes occur at the


surfaces of plants roots and substrate, resulting in a short-term retention or long-term immobilization of the contaminants.

The sorption process is influenced by characteristics of the substrate, such as its texture, content in organic matter and ion exchange properties, by characteristics of the wastewater, such as its dissolved organic matter content, pH and electrolyte composition, and by the properties of the pollutant itself. In the case of pollutants with acid-base properties, the pH in the liquid compartment will determine the form (ionic or neutral) in which they will be present in solution and this aspect may have a strong influence over the extent of sorption to mineral surfaces with ion exchange properties.

Complexation phenomena can also contribute, in some cases, to increase the solubility of some pollutants or, in other cases, to facilitate binding of pollutants to a mineral surface.

Precipitation is also a major process leading to long-term removal of pollutants. The extent of these processes is frequently dependent on pH and redox conditions. Hydrolysis very often leads to the formation of insoluble oxides and hydroxides, while changes in protonation states are a frequent means of converting soluble ionic forms of a compound into a neutral insoluble form. Precursor redox reactions are also commonly associated with precipitation phenomena through the conversion between more soluble oxidation states into oxidation states that easily lead to formation of insoluble compounds. In addition, sorption is a process which, in some cases, also is accessory to co-precipitation phenomena.

3.4.3. Biological Processes

CWS are biological systems in which biological processes play a major role in removal of pollutants. Probably the most widely recognized biological process for inorganic pollutant removal in wetlands is plant uptake. Contaminants that are also forms of essential plant nutrients, such as nitrate, ammonium and phosphate, are readily taken up by wetland plants. However, many wetland plant species are also capable of uptake and even significant accumulation of, certain toxic metals such as cadmium and lead. The rate of contaminant removal by plants varies widely, depending on plant growth rate and concentration of the contaminant in plant tissue.

In wetlands, as in many terrestrial ecosystems, when the plants age and die, the dead plant material, known as detritus or litter, is deposited at the substrate surface. Some of the nutrients, metals or other elements previously removed from the water by plant uptake are lost from the plant detritus by leaching and decomposition, and recycled back into the water and solid media. Water-soluble contaminants are leached more rapidly upon the death of the plant or plant tissue, while a more gradual loss of contaminants occurs during decomposition of detritus by bacteria and other organisms. The recycled contaminants may be flushed from the wetland in the surface water, or may be removed again from the water by biological uptake or other means.

Although microorganisms may also provide a measurable amount of contaminant uptake and storage, it is their metabolic processes that play the most significant role in the decomposition of organic compounds through the transformation of complex molecules into simpler ones. This provides an important biological mechanism for removal of a wide variety of organic compounds, including those found in municipal and industrial wastewater. The efficiency and rate of organic C degradation by microorganisms is highly variable for different types of organic compounds.


Microbial metabolism also affords, in wetlands, for the removal of inorganic nitrogen, i.e., nitrate and ammonium, ultimately resulting in the release of nitrogen gas, in the form of N2, which is subsequently lost to the atmosphere. The coupled processes of nitrification and denitrification are universally important in the cycling and bioavailability of nitrogen in wetland and upland soils (Vymazal, 2007).

An understanding of the basic physical, chemical and biological processes controlling contaminant removal in wetlands will substantially increase the probability of success of treatment wetland applications. Furthermore, a working knowledge of biogeochemical cycling, the movement and transformation of nutrients, metals and organic compounds among the biotic (living) and abiotic (non-living) components of the ecosystem, can provide valuable insight into overall wetland functions and structure. This level of understanding is useful for evaluating the performance of contaminant removal by constructed wetlands and for assessing the functional integrity of human-impacted, restored and mitigation wetlands.

A more detailed discussion follows, describing the mechanisms studied that are involved in contaminant removal by CWS, for each of the most important types of water contaminants.

4. MECHANISMS OF POLLUTANTS REMOVAL BY CWS Wetlands have been found to be effective in treating organic matter, suspended solids and

nutrients as well as for reducing metals, trace organic pollutants and pathogens. The main pollutants removal mechanisms in CWS include biological processes such as microbial metabolic activity and plant uptake as well as physical and chemical processes such as volatilization, sedimentation, filtration, adsorption and precipitation at the water-substrate, root-substrate and plant-water interfaces. The following table provides an overview of the major pollutant removal mechanisms that operate in the wetland environment, which are responsible for the removal/reduction of wastewater contaminants.

Table 3. Pollutant removal mechanisms (Brix, 1993; Cooper et al., 1996; Vymazal et al.,

1998b; Sundaravadivel and Vigneswaran, 2001)

Wastewater pollutant Removal mechanism Total Suspended Solids (TSS) · Sedimentation

· Filtration Soluble Biodegradable Organic Matter (measured as BOD)

· Microbial degradation (aerobic, anoxic and anaerobic)

· Adsorption · Plant uptake

Nutrients: Nitrogen (N) · Ammonification (mineralization)

· Nitrification/ denitrification · Nitrate-ammonification · Plant/microbial uptake · Media adsorption/ion exchange · Ammonia volatilization · ANAMMOX


Table 3. (Continued)

Wastewater pollutant Removal mechanism Phosphorus (P) · Media adsorption

· Plant and microbial uptake · Sedimentation · Precipitation

Metals · Adsorption and cation exchange · Complexation · Precipitation/co-precipitation · Oxidation and hydrolysis · Plant uptake · Microbial oxidation/reduction (microbial-mediated

processes) · Sedimentation and filtration

Pathogens (microbial population) · Sedimentation · Filtration · Natural die-off · Predation · UV irradiation · Excretion of antibiotics form roots by roots of

macrophytes · Adsorption

Organic xenobiotics · Sedimentation · Volatilization · Biodegradation · Adsorption · Plant uptake · Photolysis · Chemical reactions

Constructed wetlands should be engineered to maximize the particular removal

mechanisms required for the treatment of a given wastewater. A good understanding of the complex mechanisms involved in the removal of the different types of pollutants present in the wastewater is therefore essential for the design and operation of any CWS.

4.1. Suspended Solids Removal Municipal and some industrial effluents usually contain large quantities of suspended

solids. Excessive amounts of this suspended particulate matter may cause adverse effects such as reducing the light available to photosynthetic organisms. On settling, it may also alter the characteristics of the river bed, potentially making it a less suitable habitat for the local organisms. However, this problem is amongst the most readily solvable ones of wastewater treatment, and is usually very efficiently addressed in a CWS. A brief review of the literature data reveals usually very high values reported for the removal of suspended solids independently of the CWS type, with efficiencies usually above 90% (Vymazal et al., 1998a;


Gómez Cerezo et al., 2001; Merlin et al., 2002; Ansola et al., 2003; Cameron et al., 2003; Mantovi et al., 2003).

In a CWS the major mechanisms responsible for removal of settleable suspended solids are sedimentation and filtration. Non-settling/colloidal solids are removed, at least partially, by bacterial growth (which results in the settling of some colloidal solids and the microbial decay of others) and collisions with the adsorption to other solids (Vymazal et al., 1998b).

In FWS systems suspended solids are removed in part by sedimentation, enhanced by very low flow velocities and shallow depths, and in part by filtration through the living vegetation and vegetative litter. Additional removal of solids also occurs at the soil interface (Metcalf and Eddy, 1991).

In SSF systems, suspended solids are removed primarily by filtration through the substrate. Thus, there is a tendency for solids to clog or seal the infiltrative surface of the systems. These systems must, then, be designed and operated to minimize loss of infiltrative capacity (Metcalf and Eddy, 1991). In most applications, a sedimentation pond is added upstream of the wetland cells to promote the removal of larger suspended particles and minimize the chance of clogging the wetland cells. The pond can also dilute the raw influent if it is considered too strong. These TSS treatment processes also contribute to a significant portion of the removal of organic matter, nutrients (mostly nitrogen and phosphorus) and pathogens. In SSF systems, the high removal efficiencies of suspended matter have, in some cases, also shown to be less sensitive to seasonal conditions (Merlin et al., 2002).

Frequently, removal of suspended solids is observed to occur mainly within the early stages of treatment (Vymazal et al., 1998a; Gómez Cerezo et al., 2001; Merlin et al., 2002).

4.2. Soluble Organic Matter The biochemical oxygen demand (BOD), which indirectly quantifies the amount of

biodegradable organic matter, provides a broad measure of the effects of organic pollution on a receiving water body.

In general the origins of organic effluents and their composition are extremely diverse and one can expect a similar diversity in the effects they have on receiving waters. Domestic effluent is the greatest source of organic materials discharged to fresh waters. In urban areas, the run-off from houses, factories and roads can also result in severe pollution, especially in storm conditions after periods of dry weather. Industrial effluents are a further source of organic pollution. These may be routed via the wastewater treatment works or they may be released, with or without treatment, directly into a waterway. Among the industries producing effluents containing substantial amounts of organic wastes are the food processing and brewing industries, dairies, abattoirs and tanneries, textile and paper making factories and farms.

One of the major problems caused by effluents with high BOD loads occurs as organic matter is gradually decomposed by the microorganisms in a much similar way to the processes occurring in biological treatment processes in wastewater treatment plants. This microbial process, in which the more complex organic molecules are broken down into simple inorganic molecules, involves a considerable consumption of dissolved oxygen. When high loads of organic pollution are present, oxygen may be used at a higher rate than it can be replenished from the atmosphere or the photosynthetic activity of aquatic plants, thus causing


its depletion with severe consequences for the water stream biota, including reduced fitness and, in extreme cases, asphyxiation (Mason, 2002). The sedimentation of organic waste at the bottom of streams may also alter its characteristics, with potential harmful effects to its biota.

A majority of operating CWS in the world are used for treatment of municipal wastewater in which organic substances constitute the prominent pollutants. Hence, a vast information bank is available on CWS performance of organic matter removal. Wetlands are efficient users of external carbon sources, which is manifested by good reductions in BOD (Hammer and Bastian, 1989; Vymazal et al., 1998b; Sundaravadivel and Vigneswaran, 2001).

Soluble degradable organic matter in wastewater is removed mainly through microbial degradation whereas uptake of degradable organic matter by macrophytes is negligible compared to biological degradation (Metcalf and Eddy, 1991).

The microorganisms responsible for the degradation are generally associated with slimes or films that develop on the surfaces of substrate particles, vegetation and litter (Metcalf and Eddy, 1991; Vymazal et al., 1998b).

Organic compounds may be degraded both aerobically and anaerobically and it is difficult to quantify the ratio between the two types (Ottova et al., 1997). The oxygen required for aerobic degradation is supplied directly from atmosphere by diffusion or oxygen leakage from the macrophyte roots into the rhizosphere.

The efficiency of organic matter removal in CWS is well documented in numerous reports in the literature. In table 4 a sample of some illustrative studies is presented. Soluble organic matter removal efficiency in CWS is generally high, usually exceeding 80%, but the efficiency of organic matter treatment does show some variation according to the CWS type.

Table 4. Organic matter (BOD) removal in several types of CWS

Type of CWS Reduction rate References FWS (pilot-scale) 89% (Chen et al., 2006)

Hierarchical Mosaic of Aquatic Ecosystems (HMAE®)

80-95% (Ansola et al., 2003)

HSSF 86.6% (Vymazal, 1999)

HSSF ≥ 90% (Merlin et al., 2002)

HSSF 94% (Mantovi et al., 2003)

HSSF (review) 85% (Vymazal, 2005c)

SSF 75-79% (Karathanasis et al., 2003)

Multi-stage system 90% (Gómez Cerezo et al., 2001)

Set of single-family CW (review) 70% (Steer et al., 2002)

VSSF 97% (Haberl et al., 2003)

VSSF 96% (Meuleman et al., 2003)

VSSF (pilot-scale) > 98% (Sleytr et al., 2007)


These differences are revealing of the dependence on several design and operational parameters including the type of CWS/vegetation/substrate, the hydraulic and mass loading rates, flow regime, hydraulic residence time, and BOD decay rate, which in turn is a function of temperature (Vymazal, 1999; Vymazal, 2005c). Because of the complexity of these interdependencies, different studies set up under different conditions are not easily comparable. For example, the role of the substrate and the rhizosphere in FWS systems is quite negligible compared with SSF, where long residence times allow extensive interaction with the wastewater.

While resilient, slow growing species with low seasonal biomass turnover, and high root zone aeration capacity may be suitable for FWSs, high productivity species, tolerant to high levels of pollutants and hypertrophic waterlogged conditions may be functionally superior in SSFs (Tanner et al., 1995).

Indeed, one aspect that has been controversial is the role of vegetation and the effects of different plant species as mono or polycultures in promoting treatment. Although some studies (Soto et al., 1999; Vymazal, 1999; Merlin et al., 2002) have documented that macrophytes can improve BOD removal from wastewaters by: i) enhancing sedimentation, ii) providing mechanical filtration, iii) providing nutrient assimilation, and iv) stimulating the microbial activity and its development; others did not detect any significant difference between planted and unplanted systems (Tanner et al., 1995).

Seasonal variations in BOD removal efficiency by CWS in the presence of various macrophytes species have been reported by several investigators, with consistent treatment deterioration being observed in late winter months (Kuehn and Moore, 1995). It is uncertain, however, whether the poor winter performances are due to cold temperatures alone or the combined effect with increased hydraulic loadings, because several other studies have not shown significant treatment effects between winter and summer (Neralla et al., 2000).

4.3. Nutrients Removal The major sources of nutrients pollution (primarily, sources of nitrogen and phosphorus)

in water bodies are derived, in urban areas, from domestic wastewaters, industrial wastewaters and storm drainage whereas rural sources of nutrients include those from agriculture, from forest management and from rural dwellings.

Often, either nitrogen or phosphorus is the limiting factor in an aquatic system, and the addition of excess nutrients may cause eutrophication in rivers, streams, estuaries and coastal waters or other undesirable changes in the ecological community such as ammonia’s toxicity to aquatic biota and problems caused by excessive levels of nitrates in drinking water which may be a public health issue.

The problems caused by excess of nutrients can also directly affect human activities and these problems can be summarized into three main areas: those associated with quality of water supply, those affecting aesthetic and recreational activities, and those causing difficulties with the management of water courses and lakes (Mason, 2002).

A wetland system offers excellent ecological possibilities to act as a sink for nutrients. Nutrient cycling in CWS includes pathways of nutrient transformation between several different forms and transfer between the different compartments such as water, substrate, litter, plants, etc.


4.3.1. Nitrogen Removal in CWS Nitrogen is a major component of most wastewaters and all nitrogen forms, including

nitrogen gas (N2), are biochemically interconvertible and are components of the nitrogen cycle. The dominant forms of nitrogen in wetlands that are of importance to wastewater treatment include organic nitrogen, ammonia, ammonium, nitrate, nitrite, and nitrogen gas. The relative proportion of each depends on the type of wastewater and pre-treatment. Organic forms are present in dissolved and particulate forms, while inorganic N is present in dissolved forms. Particulate forms are removed though settling and burial, while the removal of dissolved forms is regulated by various biogeochemical reactions functioning in the substrate and water column (Fig. 3). Relative rates of these processes are affected by physico-chemical and biological characteristics of substrate, water column and vegetation type.

The transformation and removal of nitrogen in CWS involves a complex set of physical, chemical and biological processes (figure 3). The mechanisms involved in the removal of nitrogen from wastewater depend on the major forms in which the nitrogen is present. Mineralization, biological uptake, nitrification and nitrate reduction to ammonium are processes that transform one form of nitrogen to another (figure 3). In terms of the nitrogen cycle, these mechanisms are conservative processes and operate to cycle nitrogen within a system. Denitrification and ammonia volatilization are export processes and result in the net loss of nitrogen from the system. Conversely, nitrogen fixation returns nitrogen from the atmosphere to the CWS.

Figure 3. Nitrogen transformation/removal mechanisms in CWS (adapted from Cooper et al. (1996)).


4.3.1.1. Nitrogen Conservative Transformations The major processes of nitrogen transformations which do not lead to nitrogen release

from the system are enumerated in Table 5. The various forms of nitrogen are continually involved in chemical transformations from inorganic to organic compoundHs and back from organic to inorganic. All of these transformations are necessary for wetland ecosystems to function successfully, and most chemical changes are controlled through the production of enzymes and catalysts by the living organisms they benefit (Vymazal, 2007).

Table 5. The major nitrogen conservative transformations and retention processes in

CWS (adapted from Vymazal (2007)) Process Transformation Ammonification (mineralisation) Organic-N ammonia-N Nitrification Ammonia-N nitrite-N nitrate-N Nitrate-ammonification Nitrate-N ammonia-N Plant/microbial uptake (assimilation) Ammonia-N, nitrite-N, nitrate-N organic-N Ammonia adsorption Organic nitrogen burial Ammonification (mineralization) – is the process where organic nitrogen is biologically

converted into inorganic nitrogen, especially ammonium (Vymazal et al., 1998b; Vymazal, 2007). Ammonium is converted from organic forms through a complex biochemical process involving the catabolism of amino acids with a net release of energy (Vymazal, 2007). In some cases, this energy is used by microorganisms for growth, and ammonia is directly incorporated into microbial biomass (Vymazal, 2007).

The process can occur under aerobic or anaerobic conditions but, in comparison to facultative and obligate anaerobic mineralization, the contribution of aerobic mineralization to the overall N mineralization would be very small (Vymazal et al., 1998b).

A wide range of ammonification rates are reported in the literature, with values ranging between 0.004 and 0.53 g N m−2 d−1 (Reddy and D'Angelo, 1997; Tanner et al., 2002). Net release of NH4

+ is determined by the balance between ammonification and immobilization (conversion of inorganic nitrogen ions into organic forms) which is controlled by the N requirements of microorganisms involved, nature of organic N, temperature, pH, redox conditions, C/N ratio, available nutrients concentration and substrate conditions such as texture and structure (Vymazal, 2007). The optimal ammonification temperature is reported to be 40–60 °C while optimal pH is between 6.5 and 8.5 (Vymazal et al., 1998b; Vymazal, 2007).

Nitrification – is usually defined as the biological oxidation of ammonium to nitrate with nitrite as an intermediate in the reaction sequence (Vymazal et al., 1998b; Vymazal, 2007). Nitrification is performed primarily by chemoautotrophic bacteria, which derive energy from the oxidation of ammonia and/or nitrite and carbon dioxide is used as a carbon source for synthesis of new cells (Vymazal et al., 1998b; Vymazal, 2007). The process consists of two main sequential steps (Vymazal et al., 1998b): NH4

+ + 3/2 O2 → NO2- + 2 H+ + H2O (1)

NO2

- + 1/2 O2 → NO3- (2)


NH4+ + 2 O2 → NO3

- + 2 H+ + H2O (3) The first step, the oxidation of ammonium to nitrite, is executed by strictly

chemolithotrophic (strictly aerobic) bacteria which are entirely dependent on the oxidation of ammonia for the generation of energy for growth. In soil, species belonging to the genera Nitrosospira, Nitrosovibrio, Nitrosolobus, Nitrosococcus and Nitrosomonas have been identified (Vymazal et al., 1998b; Brady and Weil, 2002; Vymazal, 2007).

The second step in the process of nitrification, the oxidation of nitrite to nitrate, is performed by facultative chemolithotrophic bacteria, Nitrobacter, which can also use organic compounds, in addition to nitrite, for the generation of energy for growth (Vymazal, 2007).

Nitrification rates in wetlands were reported to be in the range of 0.01–2.15 g N m−2 d−1 with the mean value of 0.048 g N m−2 d−1 (Reddy and D'Angelo, 1997; Tanner et al., 2002). Nitrification is influenced by concentrations of ammonium-N, BOD concentration, temperature, pH value, alkalinity of the water, inorganic C source, moisture, microbial population, dissolved oxygen and potential for toxic compounds.

Nitrification is strictly an aerobic process in which the end product is nitrate; this process is limited when anaerobic conditions prevail.

Nitrate-ammonification – refers to a process of anoxic reduction of nitrate to ammonia. This is the first anoxic process to occur after oxygen depletion in the reduction of nitrate to molecular nitrogen or ammonia. This is performed by two different groups of nitrate-reducing bacteria: the denitrifying bacteria which produce N2O and N2 as major reduction products (denitrification), and the nitrate-ammonifying bacteria which produce NH4

+ as the major end product of the reduction of nitrate (Vymazal, 2007).

The nitrate-ammonification rates is influenced by environmental factors such as absence of O2, redox potential, temperature, pH value, organic matter, etc.

Nitrogen assimilation (plant/microbial uptake) – refers to a variety of biological processes that convert inorganic nitrogen forms into organic compounds that serve as building blocks for cells and tissues. The two forms of nitrogen generally used for assimilation are ammonium and nitrate nitrogen (Vymazal, 2007). Macrophyte growth is not the only potential biological nitrogen assimilation process in wetlands, microorganisms and algae also utilize nitrogen.

The potential rate of nutrient uptake and storage (standing stock) by plant is limited by its net productivity (growth rate) and the concentration of nutrients in the plant tissue. Therefore, desirable traits of a plant used for nutrient assimilation and storage would include rapid growth, high tissue nutrient content, and the capability to attain a high standing crop.

In the literature, there are many reviews on nitrogen concentrations in plant tissue as well as nitrogen standing stocks for plants found in natural stands and constructed wetlands (Mitsch and Gosselink J.G., 2000; Vymazal, 2007). However, if the wetland vegetation is not harvested, the vast majority of the nutrients that have been incorporated into the plant tissue will be returned to the water as result of decomposition processes.

Ammonia adsorption/ionic exchange – the reduced state ammoniacal N is stable but can be adsorbed onto active sites of the substrate. However, the ion exchange of NH4

+-N on cation exchanging sites of the substrate is not considered to be a long-term sink for NH4

+-N removal but rather is presumed to be rapidly reversible (Vymazal et al., 1998b).


In a continuous-flow system, under stable conditions, the sorbed NH4+-N will be in

equilibrium with NH4+-N in solution. An intermittently loaded system will display rapid

removal of NH4+-N by adsorption mechanisms and a gradual release during rest periods.

The rate and extent of these processes are reported to be influenced by several factors, such as nature and amount of clays, nature and amount of soil organic matter, periods of submergence and drying, and presence of vegetation.

Organic nitrogen burial – some fractions of the organic nitrogen incorporated in detritus in a wetland system may eventually become unavailable for additional nutrient cycling through the process of peat formation and burial. The values of organic nitrogen burial have been reported for various natural wetlands, however, in constructed wetlands there are practically no data available (Vymazal, 2007).

4.3.1.2. Removal/retention mechanisms

In addition to the processes just described, some of which actually remove nitrogen from the wastewater into some component of the wetland, further removal may consist of nitrogen release off the system through ammonia volatilization, denitrification and anaerobic ammonium oxidation. Conversely, normal operation of CWS may include the natural incorporation of nitrogen in the system through fixation.

Ammonia volatilization – is a physicochemical process where ammonium-N is in equilibrium with gaseous and hydroxyl forms (Vymazal et al., 1998b; Brady and Weil, 2002; Vymazal, 2007):

NH4

+ + OH- → H2O + NH3 (4) As can be seen, ammonia volatilization will be more pronounced at higher pH levels.

These high pH values which facilitate volatilization are created during the day, in wetland, by the photosynthesis of submerged macrophytes and algae.

The volatilization rate is controlled by the NH4+ concentration in water, pH values,

temperature, wind velocity, solar radiation, the nature and density of vegetation and the capacity of the system to change the pH value in diurnal cycles (absence of CO2 increases volatilization).

The volatilization of ammonia in CWS can result in nitrogen removal rates as high as 2.2 g N m−2 d− 1 (Stowell et al., 1981).

Denitrification – is a bacterial facilitated anoxic process of nitrate reduction that may ultimately produce molecular nitrogen (N2) through a series of intermediate gaseous nitrogen oxide products. This respiratory process reduces oxidized forms of nitrogen in response to the oxidation of an electron donor such as organic matter (Brady and Weil, 2002).

The organisms that carry out this process are commonly present in large numbers and are mostly facultative anaerobic bacteria in genera, such as Pseudomonas, Bacillus, Micrococcus and Achromobacter. These denitrifying bacteria are chemoheterotrophs, which obtain their energy and carbon from the oxidation of organic compounds. Other organisms are autotrophs, such as Thiobacillus denitrificans, which obtain their energy from the oxidation of sulfide. The exact mechanisms vary depending on the conditions and organism involved (Brady and Weil, 2002).

It is generally agreed that the actual sequence of biochemical changes from nitrate to elemental gaseous nitrogen is (Vymazal, 2007):


2 NO3- => 2 NO2

- => 2 NO => N2O => N2 (5) Environmental factors known to influence denitrification rates include the absence of O2,

redox potential, substrate moisture, temperature, pH value, presence of denitrifiers, substrate type, organic matter, nitrate concentration and the presence of overlying water (Vymazal, 2007).

The estimation of denitrification rates in CWS varies widely in the literature between 0.003 and 1.02 g N m−2 d−1 (Reddy and D'Angelo, 1997; Vymazal, 2007).

Denitrification is the predominant microbial process that modifies the chemical composition of nitrogen in a wetland system and the major process whereby elemental nitrogen is returned to the atmosphere.

Anaerobic ammonium oxidation (ANAMMOX) – is the anaerobic conversion of NO2− and

NH4+ to N2. It was demonstrated that in an ANAMMOX process, nitrate is used as an electron

acceptor (Vymazal, 2007). Not all the transformation/removal processes of nitrogen occur in all types of constructed

wetlands and the magnitude of individual processes varies among types of CWS. Volatilization may be a significant route for nitrogen removal in constructed wetlands

with open water surface where algal assemblages can create high pH values during the day through their photosynthetic activity (Vymazal et al., 1998b; Vymazal, 2007).

Nitrification occurs in all types of CWS, however, the extent of this process is determined by oxygen availability. Nitrification takes place when oxygen is present in concentrations high enough to support the growth of strictly aerobic nitrifying bacteria. Nitrification per se is a limiting process for nitrogen removal from most types of CWS because ammonia is the dominant species of nitrogen in sewage and many other wastewaters (Vymazal, 2007). On the other hand, denitrification is considered as a major removal mechanism for nitrogen in most types of constructed wetlands (Vymazal, 2007). The concentrations of nitrate, however, are usually very low in wastewater (the exception is drainage water from the agriculture and some industrial wastewaters) which also limits the extent to which this process can occur. However, the coupling between the nitrification and denitrification processes provides the conditions for this to be the major nitrogen removal process in many treatment wetlands. Different requirements for the presence of oxygen for nitrification and denitrification are the major obstacle in many treatment wetlands for achieving higher nitrogen removal.

Nitrate-ammonification occurs under conditions of low redox potential values and, therefore, there is a potential that this process may be important in treatment wetlands where anaerobic conditions do occur, e.g., in treatment wetlands with horizontal sub-surface flow (Vymazal, 2007).

Ammonium adsorption is limited to CWS with sub-surface flow where the contact between substrate and wastewater is efficient.

In Table 6 are displayed the removals of various forms of nitrogen by different types of CWS which can be found in the literature.

The removal efficiency of several nitrogen forms in CWS is well documented in numerous reports in the literature. In table 6 a sample of some illustrative studies is presented.


Table 6. Nitrogen removal in several types of CWS

Type of CWS Reduction rate References Created wetland 97% (NO3

-/NO2 --N)

100% (NH3-N) (Kohler et al., 2004)

CWS (pilot scale) 85% (NO2 --N)

88% (NO3 --N)

13% (NH4 +-N)

(Maine et al., 2007)

CWS (Single-family) 56% (NH4 +-N) (Steer et al., 2002)

FWS 53% (inorganic-N) 60% (NO2

--N) 70% (NO3

--N)

(Maine et al., 2006)

FWS (review) 41.2% (total-N) 55.1% (NH4

+-N) 60.7% (NO3

--N)

(Vymazal, 2007)

FWS 3 to 15% (total-N) (Braskerud, 2002)

FWS (pilot-scale) 56% (NH3-N) (Chen et al., 2006)


98% (NH4 +-N) (Ansola et al., 2003)

HSSF 50% (total-N) 79% (organic-N)

(Mantovi et al., 2003)

HSSF (review) 42% (total-N) 48% (NH4

+-N) 35% (NO3

--N)

(Vymazal, 2005c)

HSSF (review) 42.3% (total-N) 48.3% (NH4

+-N) 38.5% (NO3

--N)

(Vymazal, 2007)

Small-scale wetland mesocosms

55%( Kjeldahl-N) (Hench et al., 2003)

VSSF 30% (total-N) (Meuleman et al., 2003)

VSSF (review) 44.6% (total-N) 84.2% (NH4

+-N) (Vymazal, 2007)

VSSF (pilot-scale) 99.5 % (NH4 +-N) (Sleytr et al., 2007)


Due to the varied forms in which nitrogen may be present in a wastewater, several very different mechanisms will be operative in their transformation or removal. Furthermore, the efficiencies of the different mechanisms are very dependent on the particular conditions. As such, comparisons are quite difficult to establish among several studies and the reported total nitrogen removed will vary significantly.

The most effective removal of nitrogen results from volatilization of ammonia and the coupled nitrification-denitrification process with elimination of gaseous nitrogen. The absence of a free water surface in SSF systems is a limiting factor for the efficiency of the volatilization process in this type of systems, and thus the efficiency of this process may be considerable only in FWS. On the other hand, a single-stage system hardly provides both aerobic as well as anoxic zones for adequate nitrification and denitrification in simultaneity. Multi-stage systems have been suggested as possibly being able to improve the conditions for the coupled nitrification-denitrification process.

Some environmental conditions, such as temperature and dissolved oxygen also have a significant influence in the overall efficiencies.

Importance of harvesting in the case of nitrogen removal is not as critical as in other cases, for instance the removal of phosphorus.

4.3.2. Phosphorus Removal in CWS

Phosphorus is another nutrient which, when available in excessive amounts, also causes eutrophication in water bodies. Removal of phosphorus tends not to be as high as nitrogen removal in wetlands as no direct metabolic pathway is provided for its biological removal. Phosphorus removal in wetlands is achieved by a variety of physical, chemical and biological processes.

Phosphorus is typically present in wastewaters as organic phosphorus and in inorganic form, predominantly as orthophosphate but also as polyphosphates (Cooper et al., 1996; Vymazal et al., 1998b). Organically-bound phosphorus is present e.g., in phospholipids, nucleic acids, nucleoproteins, phosphorylated sugars or organic condensed polyphosphates (coenzymes, ATP, ADP) (Vymazal, 2007). Organic phosphorus is not readily bioavailable until it undergoes enzyme hydrolysis. Free orthophosphate is the only form of phosphorus believed to be utilized directly by algae and macrophytes and thus represents a major link between organic and inorganic phosphorus cycling in wetlands (Vymazal, 2007). Phosphorus availability from organic phosphorus depends on the rate of mineralization and biodegradability, which increases with the nutrient loading in CWS.

Wetlands provide an environment for the inter-conversion of all forms of phosphorus, which are cycled in the soil-water-plant components.

4.3.2.1. Removal/Retention Mechanisms

Phosphorus removal in CWS can be described as a combination of both short-term (uptake by plants and microorganisms) and long-term storage processes (burial in peat/soil). Direct assimilation of wastewater phosphorus by macrophytes represents in general only a minor contribution to the overall phosphorus removal. Adsorption and precipitation can initially only be considered as short-term storage. However, once this material is buried within a layer of peat, it becomes part of the long-term “sink” storage. If the settled particulate matters remain insoluble, the phosphorus removal by sedimentation is permanent.


However, if the particulate material is biodegradable organics, after degradation, phosphorus will be released back to the water column.

Phosphorus transformations and removal in CWS, thus, involve a variety of processes which include sedimentation and burial in soil, adsorption, precipitation, mineralization and plant/microbial uptake (Vymazal, 2007).

Sedimentation and burial – most studies on phosphorus cycling in wetlands have shown that soil/peat accumulation is the major long-term phosphorus sink (Richardson and Marshall, 1986). Over long-term periods, significant portions of organic P remain in soil as part of peat build-up and, under anaerobic conditions, these forms of phosphorus are relatively resistant to microbial breakdown. In nutrient-enriched wetlands long-term P peat accretion can reach nearly 1 gm−2 yr−1 (Craft and Richardson, 1993).

Adsorption – adsorption of phosphorus to the substrate has been recognized as one of the most important removal mechanisms (Vymazal, 2007). Traditionally, locally available materials such as sand and soils have been used as substrate for P-removal. In many cases, these substrates have been used without any knowledge of the P-retaining capacities (Brix et al., 2001) even though greater attention has been given more recently to the use of materials with well-tested P adsorptive qualities, among which light expanded clay aggregates (LECA) is one the most popular choices (Brix et al., 2001). Retention by adsorption is strongly related to the amounts of Fe (oxides), Al and Ca minerals and frequently is associated with posterior longer-term removal by precipitation and co-precipitation. The rate of adsorption is controlled by soil pH and Eh, adsorptive surface area and temperature (Reddy and D'Angelo, 1997; Vymazal et al., 1998b).

Precipitation – in acid soils, inorganic P is adsorbed on oxides and hydroxides of iron and aluminum and may precipitate as insoluble phosphates (Vymazal et al., 1998b). On the other hand, in the alkaline, calcareous soils precipitation by calcium carbonate as insoluble Ca mineral bound-P is the dominant removal process (Vymazal et al., 1998b; Brix et al., 2001). Thus, precipitation can refer to the reaction of phosphate ions with metallic cations, especially Fe, Al, Ca or Mg, forming amorphous or poorly crystalline solids.

In addition to direct chemical reaction, phosphorus can co-precipitate with other minerals, such as ferric oxyhydroxide and the carbonate minerals, such as calcite (calcium carbonate).

Plant uptake – the phosphorus uptake capacity of macrophytes is lower as compared with nitrogen uptake. Thus, phosphorus concentration in plant tissues are much lower than that of nitrogen (Vymazal et al., 1998b; Brix et al., 2001). The concentration of phosphorus in the plant tissue varies among species and sites and also during the seasons. Phosphorus uptake by macrophytes is usually highest during the beginning of the growing season (in most regions during the early spring), before maximum growth rate is attained (Vymazal, 2007).

Most of the phosphorus taken up by the herbaceous vegetation of wetlands is obtained through their roots from the soil porewater and translocated to the aboveground growing tissues (Reddy and D'Angelo, 1997; Vymazal, 2007). Upon maturity and senescence, a substantial portion of the phosphorus in above-ground tissues is translocated and stored into below-ground biomass (roots and rhizomes). Depending on the type of macrophyte, in nutrient-rich systems most of the remaining aboveground-stored phosphorus is released back into the water as detrital tissue either by initial leaching or as result of decomposition.

Phosphorus storage in vegetation can range from short- to long-term, depending on type of vegetation, litter decomposition rates, leaching of P from detrital tissue, and translocation


of P from shoots to roots tissues. Phosphorus storage in aboveground biomass of emergent macrophytes is usually short-term, with a large amount of P being released during the decomposition of litter (Vymazal, 2007). In order to remove phosphorus from CWS it is necessary to harvest macrophyte biomass. This is especially necessary for free-floating plants.

Microbiota uptake – the phosphorus uptake by microbiota (bacteria, fungi, algae, microinvertebrates, etc.) is rapid because these organisms grow and multiply at high rates. It seems that the amount of microbial storage depends also on trophic status of the wetland (Vymazal, 2007). In less enriched sites the microbial uptake may store more phosphorus as compared to more eutrophic sites.

Although bacteria are generally considered decomposers that simply mineralize organic P, they have also been shown to regulate the P flux across the sediment–water interface (Kleeberg and Schlungbaum, 1993)and contribute to terminal P burial through production of refractory organic compounds.

Of importance, and seldom recognized, is the amount of P that can be sequestered by the algal component of wetlands, especially in areas with open water (Vymazal, 2007). Algae and algal assemblages can affect phosphorus cycling either directly (uptake, release) or indirectly through photosynthesis-induced changes in water and soil/water interface (e.g pH, dissolved oxygen). In fact, diurnal fluctuations of pH (increasing during daytime due to photosynthetic activity and decreasing during the night due to respiration) can produce cycles of precipitation and resolubilization.

Table 7 presents some results about the reduction rates (%) of phosphorus in different wetland systems found in literature.

Removal of total phosphorus is reported in a range of efficiencies varying between 20% and 90% showing a marked dependence on fundamental aspects such as the CWS type, the vegetation and especially the substrate used, water flow regime, hydraulic and mass loads, hydraulic residence time and environmental conditions.

The CWS generally have a greater potential to remove nitrogen than phosphorus because nitrogen can be converted to nitrogen gas and be emitted to the atmosphere as a consequence of the coupled nitrification-denitrification process. The only sustainable removal mechanism for phosphorus in a CWS is plant uptake and subsequent harvesting and that is an especially important route of phosphorus removal in FWS systems with free-floating macrophytes. However, it is important to develop an efficient harvesting frequency in order to keep macrophytes at the optimum growth stage to ensure optimum phosphorus removal (Vymazal, 2007).

Still, the amount of phosphorus that can be removed by harvesting the plant biomass usually constitutes only a small fraction of the amount of phosphorus loaded into the system with the wastewater except in CWS which have low inflow loadings.

Microbial uptake is considered in all treatment systems only as temporary storage of phosphorus with very short turnover rate. Phosphorus which is taken up by microbiota is released back to the water after the decay of the organisms.


Table 7. Phosphorus removal in several types of CWS

Type of CWS Reduction rate References

Created wetland 74% (Kohler et al., 2004)

CWS (Single-family) 80% (Steer et al., 2002)

FWS 70% (Hadad et al., 2006)

FWS 21 – 44% (Braskerud, 2002)

FWS (pilot-scale) 35% (Chen et al., 2006)

FWS (review) 48.8% (Vymazal, 2007)


93 % (Ansola et al., 2003)

Hybrid system 89% (Oovel et al., 2007)

HSSF 60 % (Mantovi et al., 2003)

HSSF (review) 41.1% (Vymazal, 2007)

HSSF (review) 32% (Vymazal, 2005c)

VSSF >80% (Brooks et al., 2000)

VSSF 24% (Meuleman et al., 2003)

VSSF 59.5% (Vymazal, 2007)

VSSF (pilot-scale) < 47.4 % (Sleytr et al., 2007)

Phosphorus may also be bound to the substrate of the SSF mainly as a consequence of

adsorption and precipitation reactions with calcium, aluminum and iron in the substrate. The capacity of a CWS to remove phosphorus from wastewaters may then be dependent of substrate characteristics (contents in Al, Fe and Ca ions, grain size distributions, pH and specific area). Various artificial media have been tested in order to improve the P-removal in CWS among which are, for example, light expanded clay aggregates (LECA), wollastonite, vermiculite, diatomaceous earth, blast furnace slag and limestone (Brooks et al., 2000; Johansson and Gustafsson, 2000; Brix et al., 2001; Oovel et al., 2007). The removal of phosphorus through adsorption and precipitation can be significant (Vohla et al., 2005) but it is important to realize that these processes are saturable and adsorption decreases over time (Vymazal, 2007). In addition, daily pH variations due to the respiration/photosynthesis cycles may be responsible for cycles of phosphorus precipitation/resolubilization.

4.4. Pathogen Removal Waterborne diseases remain a major hazard in many parts of the world. The important

organisms from a public health point of view are the pathogenic bacteria and viruses. Protozoan pathogens and helminth worms are also of particular importance in tropical and subtropical countries (Rivera et al., 1995; Cooper et al., 1996; Vymazal et al., 1998b).


Wastewater discharges are the major source of contamination by faecal pathogenic microorganisms in rivers and coastal waters posing a risk to public health (Mason, 2002; Sleytr et al., 2007).

The treatment of wastewater pathogens in CWS is essentially a two stage process. Most pathogens are particles ranging from very small viruses to the large eggs and cysts of helminths. One of the stages of pathogen treatment is particle removal. This occurs via the same processes as for removal of suspended solids, namely sedimentation, filtration, surface adhesion and aggregation.

A series of other processes are important in influencing the viability of pathogens as infectious agents which may occur in a stage either before or after pathogenic particles removal. The major mechanisms in this stage are the hostility of the environmental conditions (temperature, pH, dissolved oxygen concentration, redox potential, salinity, turbity), predation by nematodes, protists and zooplankton and infection by other organisms, antibiosis, exposure to UV radiation and natural die-off (Metcalf and Eddy, 1991; Cooper et al., 1996; Kadlec and Knight, 1996; Ottova et al., 1997; Vymazal et al., 1998a).

The efficiency of CWS concerning the removal of microorganisms, especially indicator microorganisms like coliforms and enterococci, is a topic that has been thoroughly investigated (Kadlec and Knight, 1996; Perkins and Hunter, 2000; Langergraber and Haberl, 2001; Hench et al., 2003). In table 8, a small sample of such studies is presented as an illustrative display of the typical efficiencies achieved by several types of CWS.

Reported faecal bacteria removal efficiency in CWS is generally high, usually exceeding 85%, and is usually higher for faecal coliforms and somewhat lower for faecal streptococci (Vymazal, 2005b). Is should however be noted that, in spite of high removal efficiencies, if the number of bacteria at the inflow is very large, at the outflow bacteria number may still be too high to meet wastewater quality criteria.

Treatment efficiencies depend on several design and operational parameters including the type of CWS, hydraulic regime, type of vegetation, hydraulic residence time, hydraulic and mass loading rate, substrate, and temperature.

The efficiency of pathogens treatment does show some variation according to the CWS type and observed efficiencies can in most cases be ranked in the order: hybrid wetlands > SSF wetlands > FWS wetlands (Vymazal, 2005b). These differences may be related to the larger contact area among water, bacteria and substrate, which is much bigger in SSF constructed wetlands compared to FWS (Sleytr et al., 2007) therefore enhancing the process rates of the system (Langergraber and Haberl, 2001).

Wetland vegetation plays a crucial role in increasing the efficiency removal of pathogen in CWS. Wetland vegetation improves the trapping efficiency for small particles like viruses by increasing the surface area of biofilms in the flow path. Many species can also release exudates having antimicrobial properties or which can enhance the development in the rhizosphere of populations of bacteria with antibiotic activity (e.g. Pseudomonas).


Table 8. Removal of pathogens by different types of CWS

Indicator microorganisms

Type of CWS Removal efficiencies

References

Faecal coliform (FC) Hybrid systems (review) 99.4 % (Vymazal, 2005b)

FWS 85 – 94 % (Perkins and Hunter, 2000)

FWS 52 % (Cameron et al., 2003)

FWS > 99 % (García et al., 2008)

FWS (review) 85.6 % (Vymazal, 2005b)


99.997 % (Ansola et al., 2003)

Set of single-family constructed wetland (review) 88 % (Steer et al., 2002)

SSF 92 % (Vymazal, 2005c)

SSF 91 % (Mashauri et al., 2000)

SSF 93 – 98 % (Karathanasis et al., 2003)

SSF 99 % (García et al., 2008)

SSF 99.999 % (Soto et al., 1999)

SSF (review) 91.5 % (Vymazal, 2005b)

SSF > 99 % (Hench et al., 2003)

Faecal streptococci (FS)

Hybrid systems (review) 97.7 % (Vymazal, 2005b)

FWS 82 – 90 % (Perkins and Hunter, 2000)

FWS 99 % (García et al., 2008)

FWS (review) 84 % (Vymazal, 2005b)

SSF 99 % (García et al., 2008)

SSF > 98 % (Mantovi et al., 2003)

SSF 93 - 98 % (Karathanasis et al., 2003)

SSF 83 – 90 % (Perkins and Hunter, 2000)

SSF (review) 92.6 % (Vymazal, 2005b)

SSF 99.999 % (Soto et al., 1999)


For example, it has been shown that root excretions of species such as Scirpus lacustris and Phragmites australis can diminish the populations of faecal indicators and pathogenic bacteria (Vymazal, 2005b). In addition, the presence of oxygen in the water column (produced by photosynthetic activity of submerged plants and algae in FWS, or released in the rhizosphere through the roots of macrophytes in SSF) creates unfavorable life conditions for enteric bacteria which are either facultative or obligate anaerobic.

Effects of the hydraulic retention time are very simple: the longer the wastewater remains in the system, the longer bacteria remains exposed to unfavorable conditions.

4.5. Metals Removal Metals are naturally present in the environment. However, human activities are

responsible for a significant increase in their concentration levels up to a point where they begin to pose an environmental and public health problem. Beyond the natural sources, contamination with metals is mainly associated with such activities as soil disturbance, mining, manufacturing, urbanization, burning of fossil fuels and use of manufactured products such as paints, pesticides, sacrificial anodes and anti-foulants.

In small doses some metals are, in fact, essential to some biological processes (e.g. copper, chromium, nickel, zinc). For example, at low concentrations, copper is a micronutrient of plants essential to the photosynthetic electron transport system. However, at higher concentrations, it is marketed as an effective herbicide. In addition to concentration, the chemical form is also associated with a greater metal toxicity. For example, the methylated form of mercury is much more toxic than inorganic mercury (Mitra, 1986).

Toxic effects by metals are varied and sometimes diffuse and difficult to characterize. In some cases, for very high metal concentrations, toxicity may be acute and ultimately lethal. However, usually toxicity by metals will cause chronic effects resulting from a long-term exposure. Examples of chronic health effects include cancer, disruption of the endocrine system, liver and kidney damage, disorders of the nervous system, damage to the immune systems, and birth defects.

Some metals are not easily eliminated by the organisms and, therefore, they have the potential for bioaccumulation and biomagnification. This constitutes one of the major problems with metal contamination, with the potentiation of the metals' chronic toxicity along the food chain.

In CWS a variety of processes may provide routes for metal retention in the CWS components and their elimination from the wastewater. The main mechanisms occurring in each of the compartments (solid medium, aqueous medium and vegetation) are illustrated in figure 4.

The substrate is generally considered to be a sink for metals anthropogenically introduced in the environment. A major fraction of the elements entering the CWS will rapidly be adsorbed onto the solid phase, where a number of physical and chemical processes will determine the strength of metal retention. A small proportion of the metals can, however, remain dissolved and become available for plant uptake. In CWS another important role in metal removal is played by plants through several processes which include filtration, adsorption, cation exchange, uptake, and root-induced chemical changes in the rhizosphere (Dunbabin and Bowmer, 1992; Chen et al., 2000; Vandecasteele et al., 2005).


Figure 4. Metal removal mechanisms in CWS (adapted from Cooper et al. (1996)).

Long-term metal sequestration by plants depends on the rate of uptake, rates of translocation and retention within individual tissue types, and the rate and mode of tissue decomposition (Kadlec and Knight, 1996). Studies report the highest amounts of metals in the roots, while leaf tissue has the second highest concentrations followed by stems and rhizomes (Burke et al., 2000).

Microorganisms may also play a relevant role in heavy metal removal. Such contribution may occur through their metabolism with the modification of the oxidation states of metals which in turn may lead to other transformations, such as precipitation, that effectively remove them from the wastewater.

A more detailed account follows of the several physical, chemical and biological processes which concur for heavy metal removal in a CWS.

4.5.1. Physical Removal Processes

Sedimentation – this has long been recognized as one of the main processes in removal of heavy metals from wastewater in natural and constructed wetlands (Kadlec and Knight, 1996; Hammer, 1997; ITRC, 2003). Sedimentation is a physical process which follows other mechanisms (precipitation/co-precipitation and flocculation) whereby heavy metals aggregate into particles large enough to sink (Walker and Hurl, 2002). In this way heavy metals are removed from wastewater and trapped in the wetland sediments, thus protecting the ultimate receiving water bodies, i.e. aquatic ecosystem (Sheoran and Sheoran, 2006). Efficiency of sedimentation is proportional to the particle settling velocity and the length of wetland.

4.5.2. Chemical Removal Processes

In addition to physical removal processes a wide range of chemical processes are involved in the removal of heavy metals in the wetlands:


Sorption – among the chemical processes, sorption is one of the most important removal processes in wetlands, which results in the transfer of ions from water to the soil/sediments and a short-term retention or long-term immobilization of several classes of contaminants (Sheoran and Sheoran, 2006).

In sediments heavy metals are adsorbed by either cation exchange or chemisorption (Sheoran and Sheoran, 2006). In the former process the metal cation will exchange with other small cations (such as, Na+, K+, NH4

+, etc) in their positions in the mineral structures of clays and negatively charged groups of humic acids. Therefore the capacity of soils for retention of metal cations, expressed as cation exchange capacity (CEC) increases with increasing content in clay and organic matter. Chemisorption is a process which involves the formation of chemical bonds with the surface, frequently through complexation/chelation phenomena.

The adsorption capacity by cation exchange or non-specific adsorption depends upon the physico-chemical environment of the medium (e.g. pH, the properties of the metals concerned and the concentration and properties of other metals and soluble ligands present) (Debusk et al., 1996; Sheoran and Sheoran, 2006). Therefore, heavy metals speciation may change with time as the sediment conditions change (Groudev et al., 1999; Wiebner et al., 2005; Sheoran and Sheoran, 2006).

Much of the heavy metals can be easily adsorbed onto particulate matter in the wetland and subsequently be removed from the water by sedimentation. Lead and copper in general tends to be adsorbed most strongly while zinc, nickel and cadmium are usually held weakly which implies that these metals are likely to be more labile and bio-available (Sheoran and Sheoran, 2006).The adsorption of metals varies with the fluctuation of pH in the outflow water (Machemer and Wildeman, 1992). Precipitated hydroxides may also act as adsorption sites for phytotoxic metals present in the water compartment of the wetland (Wood, 1990).

Oxidation and hydrolysis of metals – The states of oxidation of a metal will have a marked influence in its chemical behavior in water. In particular, under some oxidation states a metal may typically hydrolyze to form insoluble oxides or hydroxides whereas in other oxidation states it can be more soluble. Such is the case, for example, of iron, aluminum and manganese which can form insoluble compounds through hydrolysis (sometimes following oxidation processes). This leads to the formation of a variety of oxides, hydroxides and oxyhydroxides (Woulds and Ngwenya, 2004; Sheoran and Sheoran, 2006)

Iron removal depends on pH, redox potential and the presence of various anions. In alkaline conditions Fe2+ is a highly soluble cation in water with low content of dissolved oxygen. On the other hand, the form Fe3+ is insoluble except in very acid conditions (pH < 3.5).

Manganese removal is the most difficult to be achieved because its oxidation takes place at a pH close to 8 (Stumm and Morgan, 1981). In this case bacteria play an important role in the oxidation of Mn by catalyzing the oxidation of Mn2+ to Mn4+.

On the other hand, aluminum removal is purely governed by pH. Aluminum hydroxides will precipitate at pH above 5.0-6.0.

Precipitation and co-precipitation – is a major process of heavy metals removal in wetland sediments. The formation of insoluble metal precipitates is one of many factors limiting the bioavailability of heavy metals to many aquatic ecosystems. Precipitation depends on the solubility product (Ks) of the metal involved, pH of the wetland and concentration of metal ions and relevant anions (Brady and Weil, 2002).


Co-precipitation is an adsorptive phenomenon also frequent in wetland sediments. Heavy metals commonly co-precipitate with secondary minerals. Copper, nickel, manganese, and other metals are co-precipitated in Fe oxides and cobalt, iron and nickel are co-precipitated in manganese oxides (Stumm and Morgan, 1981). In addition arsenic and zinc were reported to be retained on iron plaques at the surface of plant roots (Otte et al., 1995). Oxiferric hydroxide surfaces are positively charged under acidic pH conditions and negatively charged under alkaline pH conditions. Thus, adsorption and removal of oxyanions such as arsenic, antimony and selenium, through iron co-precipitation, is favored under acidic pH conditions (Brix, 1993). Alkaline conditions are favorable for co-precipitation of cationic metals such as copper, zinc, nickel and cadmium. Thus metals may become associated with iron and manganese oxides as a result of co-precipitation and adsorption phenomena (Stumm and Morgan, 1981). The process is presumed not to be very important in long-term removal and retention of metals because iron and manganese oxides, being redox sensitive, many re-dissolve following changes in oxygen concentration (Sheoran and Sheoran, 2006).

In addition to oxides, hydroxides and oxihydroxides resulting from hydrolysis, other typical insoluble metal compounds include carbonates and sulfides. Conditions exist for precipitation of heavy metal carbonates when the bicarbonate concentration in water is high. Carbonate formation can take place when bacterial production of bicarbonate alkalinity in wetland sediments is substantial (ITRC, 2003). Carbonate precipitation is especially effective for the removal of lead and nickel (Lin, 1995), but Sobolewski (1999) some authors reports significant quantities of copper and manganese carbonates accumulated in some natural wetlands.

Wetlands with appropriate substrate may promote the growth of sulfate reducing bacteria in anaerobic conditions. These bacteria will generate hydrogen sulfide which reacts with most heavy metals leading to formation of highly insoluble metal sulfides (Stumm and Morgan, 1981). These provide for long-term metal removal, remaining permanently in wetland sediments as long as they are not re-oxidized (Sobolewski, 1999). Metals such as copper, lead, zinc, cadmium, and arsenic may form highly insoluble sulfides in contact with low concentration of H2S (ITRC, 2003).

Field results suggest that upon start up of a constructed wetland, the adsorption of dissolved metals onto organic sites in the substrate material will be an important process but over time sulfide precipitation becomes the dominant process for metal removal (Machemer and Wildeman, 1992).

4.5.3. Biological Removal Processes

Biological removal is perhaps the most important pathway for heavy metal removal in the wetlands where plant uptake plays probably the most widely recognized role. While sediments of wetlands form primary sinks for heavy metals (Gray et al., 2000), macrophytes may absorb heavy metals through roots and also shoots.

It has been proposed that the processes used by plants are not necessarily the same for different species and for different metals. Submerged rooted plants may have high potential for the metals phytoextraction from sediments as well as water, while floating plants can extract metals only from water (Sriyaraj and Shutes, 2001). Among such processes can be mentioned: sorption by roots (a combination of physical and chemical processes such as chelation, ion exchange and chemical precipitation), and the biological processes including


translocation to the aerial part and precipitation induced by root exudates or by microorganisms.

The rate of metal removal by plants varies widely, depending on plant growth rate and concentration of the heavy metals in plant tissue. The rate of metal uptake per unit area of the wetland is often much higher for herbaceous plants, or macrophytes such as duckweed (Lemna minor) (Zayed et al., 1998), salix (Stoltz and Greger, 2002), cattail (Typha latifolia) and common reed (Phragmites australis) (Sheoran and Sheoran, 2006). Some of these species can tolerate high concentrations of several metals in their body mass without showing negative effects on the growth (Sheoran and Sheoran, 2006).

There are also some examples in the literature indicating that some species may have the ability to accumulate only specific heavy metals, e.g. the Spirodela polyrhiza for Zn (Markert, 1993). Constructed wetlands with well grown Cyperus alternifolius and Vallarsia exaltata have been reported to be an effective tool in phytoremediation of cadmium, copper, manganese, zinc and lead (Cheng et al., 2002).

Microorganisms also provide a measurable amount of heavy metal uptake and storage; it is their metabolic processes that play the most significant role in removal of heavy metals (Ledin and Pedersen, 1996; Russell et al., 2003; Hallberg and Johnson, 2005; Sheoran and Sheoran, 2006).

Reduction of metals to non-mobile forms by microbial activity in wetlands has been reported by Sobolewski (1999). Metals like chromium and uranium become immobilized when reduced through processes biologically catalyzed by microorganisms (Fude et al., 1994).

In table 9 are presented illustrative results found in the literature concerning the removal efficiencies for several heavy metals obtained in different constructed wetlands systems.

Table 9. Heavy metals removal efficiencies in different types of CWS

Pollutant Type of CWS Reduction rate References

CWS (small-scale plot) ≥ 90% (Liu et al., 2007) Lead

HSSF ≥ 70% (Mantovi et al., 2003)

FWS 67 % (Maine et al., 2006)


Grenhouse experiment > 43 % (Hadad et al., 2007)


HSSF 49 % (Lesage et al., 2007)

Nickel

VSSF ≥ 80% (Lee and Scholz, 2007)

HSSF 79 % (Mantovi et al., 2003) Copper

VSSF ≥ 95 % (Lee and Scholz, 2007)

Iron FWS 95 % (Maine et al., 2006)

Aluminum HSSF 93% (Lesage et al., 2007)


Pollutant Type of CWS Reduction rate References



Grenhouse experiment 100% (Hadad et al., 2007)

Chromium

HSSF 52% (Mantovi et al., 2003)

CWS (small-scale plot) ≥ 90% (Liu et al., 2007)

Grenhouse experiment ≥ 35% (Hadad et al., 2007)

Zinc


CWS (small-scale plot) ≥ 90% (Liu et al., 2007) Cadmium


Removal efficiencies reported in CWS studies present some variation, from quite low

values (~ 25%) to nearly complete removal of some metals. In general, however, the efficiencies are high (> 70%) but these will depend, as usual, on varied factors such as the influent metal loads, the type of vegetation used, the CWS type and on environmental conditions. Obviously, better removals will be achieved when the systems are specifically designed and optimized to solve well-defined metal contamination problems such as mine drainage, where well-known metal accumulator plants will be used preferably, in comparison with systems designed for broader treatment targets where metals are only possibly one among several types of pollutants to remove from wastewaters.

As elemental substances, metal cations are naturally non-biodegradable and, for their permanent elimination from the system, the portion of metals removed by plant uptake will require a periodic plant harvesting. For CWS designed to treat high loads of metal inputs, the harvested plant biomass should afterwards be disposed as hazardous waste and receive appropriate treatment.

4.6. Organic Xenobiotics Removal Organic xenobiotics include a large range of synthetic organic compounds, such as

phthalates, polychlorinated biphenyls (PCBs), dioxins, polycyclic aromatic hydrocarbons (PAHs), pesticides, sulfonated azo dyes, alkylphenols, bisphenols and pharmaceuticals and personal care products (PPCPs) (Wu, 1999; Mason, 2002; Fent et al., 2006). Several of these substances have been released in increasing amounts in the environment since decades, and, due to the low degradation rate of many of these compounds, a significant increase of their background concentrations has been observed in the different environmental compartments (Tyler et al., 1998; Skakkebaeck et al., 2000).

A growing environmental concern has been emerging in recent years, because of the high toxicity and high persistence of most of these substances in the environment and in biological systems. Even though they occur only at very low concentrations in the environment, and their threats to aquatic life and public health are still not completely understood, nevertheless, sub-lethal effects of these compounds over long-term exposure may cause significant damage to aquatic life, particularly considering that some of these compounds may cause significant


endocrine disruption, impair reproduction functions of animals or even be carcinogenic, mutagenic or teratogenic (Wu, 1999; Mason, 2002). Furthermore, the high lipophilicity of many of these xenobiotics greatly enhances their biomagnification, thereby posing potential health hazards on predators at higher trophic levels (including human beings).

The major ecological concern of xenobiotics is their ability to impair reproductive functions and subsequently threaten survival of the species. In fact, there is growing evidence from laboratory and field studies showing that exposure to trace amounts (µg/L – ng/L level) of certain xenobiotic organic compounds (e.g. halogenated hydrocarbons, PCBs, DDT, TBT) may cause reduced gonad development, disruption of normal metabolism of sex hormones (including gonadotropins), arrest of sperm maturation and block a variety of “oestrogen-like” effects on female reproductive systems in fish, birds, reptiles and mammals. This in turn, may lead to reproductive dysfunction such as delayed sex maturity, reduced fertility and hatch rate, depression in secondary sexual characteristics, alternation of sex behavior and viability of offspring (Wu, 1999).

Due to long environmental and biological half lives, recovery from the effects of many xenobiotic compounds is expected to be slow. Indeed, it has been shown that some 15 years were required to remove the negative effects of DDT on reproduction of the white tail eagles in the Baltics, and another 10 years for the population to recover (HELCOM, 1996). Furthermore, despite a decrease in environmental concentrations, the adverse effects may remain in the ecosystem for a much longer period. In the Baltics, DDE decline to 10% of the original levels in 1984, but increased again afterwards, and the egg shells of fish eating birds, which had begun to return to normal, have recently become thinner again. Thus, the downward trend was halted after the ban, and may be due to the recycling of persistent chemicals in sediment (HELCOM, 1996).

A variety of sources may be the origin for the presence of organic xenobiotics in water bodies. A number of xenobiotics classes (phthalates, pesticides, PCBs and bisphenols) are industrial products, used worldwide in several applications and are therefore ubiquitous pollutants (Safe, 1994; Stales et al., 1997; Mason, 2002). Other kinds of compounds (dioxins and PAHs) are not commercial products, but are formed as by-products of various industrial and combustion processes; they are transported from atmosphere to soil and water bodies by the atmospheric runoff or deposited on the soil during the dry period and then go through the water cycle by land runoff (Birkett and Lester, 2003).

Several studies have shown that a vast range of these xenobiotics are present in the effluents from domestic and industrial conventional wastewater treatment plants (WWTPs) (Birkett and Lester, 2003) which indicates that they resist removal by conventional wastewater treatment processes and may persist in the environment even after going through WWTPs. These are designed to deal with bulk substances that arrive regularly in large quantities (TSS, organic matter and nutrients) and many of these organic xenobiotics show a different chemical behavior for which the conventional processes are not well-suited. The different studies show that the WWTPs removal rates vary according to compound nature, WWTPs overall performance, and environmental conditions.

This is also of potential concern about treated wastewater reuse for non industrial applications, such as irrigation of crops and aquaculture, since these pollutants may become a source of contamination of the food chain.


4.6.1. Removal Processes in CWS Organic xenobiotics removal by CWS involves several interdependent processes which

may be classified as abiotic (physical or chemical) or biotic (microbial or phytological). The primary abiotic and biotic processes that participate in removing organic xenobiotics

from contaminated water in a CWS are described in table 10 (Evans and Furlong, 2003; Pilon-Smits, 2005).

Table 10. Abiotic and biotic processes involved in xenobiotics removal in CWS

Processes Description Abiotic Sorption Including adsorption and absorption, the chemical

processes occurring at the surfaces of plants and substrate that result in a short-term retention or long-term immobilization of xenobiotics

Hydrolysis The chemical breakdown of organics by the action of water, a process which frequently is pH-dependent

Photodegradation/ oxidation Degradation/oxidation of organic xenobiotics by the action of sunlight

Oxidation/reduction Modification, which sometimes may be quite substantial, of the xenobiotics due to the action of oxidizing (frequently dissolved oxygen) or reducing agents. Sometimes a redox reaction is a first step leading to removal by other processes, such as precipitation or volatilization. Redox reactions are also frequently brought about by biotic agents such as bacteria, or enzymatically catalyzed

Precipitation Many organic compounds have low water solubility and, especially those exhibiting acid-base properties, may convert into insoluble forms by pH changes

Settling and sedimentation Removal of particulate matter and suspended solids Volatilization Release of some organic xenobiotics, as vapors, which

occurs when these compounds have significant vapor pressures

Biotic Aerobic/anaerobic

biodegradation Metabolic processes of microorganisms, which play a significant role in organic xenobiotics removal in CWS

Phytodegradation Breakdown of organic xenobiotic, either internally, having first been taken up by the plants, or externally, using enzymes excreted by them

Rhizodegradation Plants provide root exudates that enhance microbial degradation of some organic xenobiotics

Phytovolatilization/ evapotranspiration

Uptake and transpiration of volatile organic xenobiotics through the leaves

The contribution of each process to the overall efficiency of the system will be very

dependent on a wide variety of factors relative not only to each CWS component characteristics but also to the properties of the organic xenobiotics, the characteristics of the wastewater and the environmental conditions.


4.6.1.1. Factors Affecting Organic Xenobiotics Removal Efficiency in a CWS The degree to which each process will contribute to the overall removal of the organic

xenobiotics from contaminated waters in CWS is in turn dependent on the physico-chemical properties of these compounds (e.g., water solubility, sorptive affinity), characteristics of the substrate (e.g., pH, organic matter content, redox status), the plants species, effluent characteristics (e.g., pH, dissolved organic matter, electrolyte composition) as well as other environmental conditions (e.g., temperature, moisture).

Some of the most important organic xenobiotics properties that affect their behavior and removal in CWS are its molecular structure, polarity, ionization constant (pKa), water solubility, sorption coefficient (Kd), octanol-water partition coefficient (Kow), volatility and chemical stability.

The texture of a soil or substrate is an extremely important characteristic in the sorption process. If the substrate is made up of mostly clay and organic matter a significant amount of sorption will take place. Clay, especially intermixed with organic particles, by far adsorbs the most out of the main types of texture (e.g. silt and sand) because of its small particle size, high surface area and high surface charge.

The content in organic matter also has a strong influence in the sorptive properties of the mineral media, mainly due to the presence of humic acids which form a large portion of their composition (Brady and Weil, 2002). These huge organic molecules are characterized by hydrophobic regions suitable for the adsorption of non-polar or weakly polar hydrophobic xenobiotics. However, they also contain numerous chemically active functional groups some of which confer or enrich the ion exchange properties, both cationic and anionic, and surface charge to the substrate which makes it also suitable for the adsorption of polar compounds (Brady and Weil, 2002). The sorption of polar or ionic xenobiotics is also very significant on some silicate clay minerals having substantial surface charge and ion exchange properties. In these materials the pH-changes can affect the sorption processes due to impacts on the pH-dependent ion exchange capacities of these variable charge components (humic organic matter included).

Some of the wastewater characteristics which will strongly influence how the organic xenobiotics will distribute between the aqueous medium and the matrix include the dissolved organic matter (DOM) content, the wastewater’s pH and its electrolyte composition. The amount and composition of the DOM may have an important influence on the compound’s solubility, and conversely on its sorption by the matrix. DOM may form complexes with the xenobiotic and enhance to its water affinity or it can compete for sorption sites in the substrate’s surface (Muller et al., 2007).

In the case of organic xenobiotics with acid-base properties, the pH of the wastewater will determine the form in which they will be present in solution (either ionic or neutral) which, in turn, will affect its water solubility and sorptive affinity for the substrate. The importance of the wastewater’s pH is not restricted to the influence it has on the protonation state of the organic xenobiotics but it will also affect the surface charge of the substrate and, thus, influence the electrostatics-based sorption of ionic forms of the compounds.

The presence in the wastewater of certain inorganic ions may introduce additional enhancement or disfavor of the organic xenobiotics sorption onto the matrix. Certain inorganic ions such as phosphate will frequently compete for ion exchange sorption sites in the mineral matrix. On the other hand, complexation phenomena can either contribute, in


some cases, to increase the solubility of some xenobiotics or, in other cases, to facilitate binding of the compounds to a mineral surface.

4.6.1.2. Organic Xenobiotics Removal by the Substrate

Many phenomena relevant for organic xenobiotics removal occur within the substrate compartment or is influenced by the media characteristics. However, only in sorption processes does the substrate play an actually active role.

Latest research on CW systems has been focusing on its ability for removing xenobiotic compounds like which are, in general, not very efficiently removed in the conventional treatment processes of the wastewater treatment plants (Moore et al., 2000; Schulz and Peall, 2001; Hannink et al., 2002; Cheng et al., 2002; Stearman et al., 2003; Haberl et al., 2003; Schulz, 2004; Matamoros et al., 2007). However, the media used have been mainly gravel, sand or local soils (Schulz and Peall, 2001; Stearman et al., 2003; Matamoros et al., 2007) and little importance has been given to the selection of more appropriate materials which can enhance the removal of this type of compounds by sorption.

Nevertheless, some materials like LECA, which is being increasingly used as substrate in CWs have already been tested with success for its sorption capacity of some organic xenobiotics like MCPA, clofibric acid, ibuprofen and carbamazepine (Dordio et al., 2007; Dordio et al., 2008a).

Clay minerals and organic matter are generally identified as the two most important solid matrix constituents that confer the xenobiotics retention properties to the substrate. However, substantially different sorption mechanisms are responsible for the sorption of non-polar and polar/ionic xenobiotics.

4.6.1.3. Organic xenobiotics removal by the microorganisms

Biological removal of organic pollutants in a CWS depends on their bioavailability, which depends on their chemical properties, the CWS physico-chemical characteristics and the environmental conditions (Pilon-Smits, 2005; Collins et al., 2006).

Microorganisms usually play an important role in the removal mechanisms of some organic xenobiotics in the CWS The degradation rate of organic xenobiotic and the extent of microbial growth during degradation is highly influenced by the xenobiotic chemical structure (Dua et al., 2002). Structurally simple compounds with high water solubility and low adsorptivity are usually more similar to the naturally occurring substances which are usually used as energy sources by the microorganisms and are easily degraded. In contrast, xenobiotics with chemical structures very different from the naturally occurring compounds are often degraded slowly since microorganisms do not possess suitable degrading genes. In these cases degradation by non-specific enzymes may still occur but at a slow rate by non-specific reactions which do not support microbial growth (co-metabolism) (Seffernick and Wackett, 2001).

Microbial xenobiotic degradation is also strongly influenced by the support medium where the microorganisms are. Temperature, pH, oxygen, presence of toxic substances and nutrients available within the CWS are expected to play a very important role in the removal efficiency.

Microorganisms with the ability to degrade a wide variety of compounds, like benzene, phenol, naphthalene, atrazine, nitroaromatics, biphenyls, polychlorinated biphenyls (PCBs) and chlorobenzoates, have been isolated and characterized (Dua et al., 2002). Although


simple aromatic compounds are biodegradable by a variety of degradative pathways, their halogenated counterparts are more resistant to bacterial attacks and often necessitate the evolution of novel pathways (Dua et al., 2002). Among the halogenated compounds, the chlorinated compounds are the most extensively studied (Dua et al., 2002). The presence of chlorine atoms on the aromatic nucleus is known to greatly retard the rate of degradation.

Most of the information available on the biodegradation of chlorinated compounds is on oxidative degradation, since aerobic culture techniques are relatively simple, compared with anaerobic culture methods. Also, aerobic processes are considered the most efficient and generally applicable (Dua et al., 2002). Aerobic degradation is dependent on the presence of molecular oxygen and is catalyzed by enzymes that have evolved for the catabolism of natural substrates and exhibit low specificities.

4.6.1.4. Organic Xenobiotics Removal by the Plants

Plants have an important role in the biotic processes of organics removal in CWS, involving numerous biological processes of which many details still remain to be known or understood. Depending on the pollutant properties, organics may be degraded in the roots zone through the plant’s stimulation of microbial activity or by direct uptake by the plant, followed by degradation, sequestration or volatilization (Evans and Furlong, 2003; Pilon-Smits, 2005).

Plants action in pollutant removal takes different names depending on where the processes predominantly take place. If the organic pollutants are degraded by the microorganisms within the plants’ rhizosphere (and under its influence) the process is called phytostimulation or rhizodegradation. If the plants degrade the organic pollutants directly within their tissues via their own enzymatic activities, then the process is called phytodegradation; other pollutants can leave the plant in a volatile form, and this is called phytovolatilization (Evans and Furlong, 2003; Pilon-Smits, 2005).

Many organic pollutants can be readily taken up by plants but, as consequence of many of them being xenobiotic, there are no specific transporters for these compounds in plant membranes. Therefore, organic xenobiotic move into and within plant tissues via diffusion (passive uptake) through cell walls and membranes (Dietz and Schnoor, 2001; Pilon-Smits, 2005). The flux is driven by the water potential gradient created throughout the plant during transpiration, which depends on the plants characteristics and the CWS environmental conditions. Translocation of the compounds is highly dependent on their water solubility and hydrophobicity. There may exist an optimal hydrophobicity that allows the organic compound to bind to the lipid bilayer of the membrane but not too strongly so that transport can still be facilitated.

Direct uptake is usually an efficient removal mechanism for moderately hydrophobic organic chemicals (log Kow = 0.5 – 3). These include most chlorinated solvents, BTEX (benzene, toluene, ethylbenzene and xylene), many pesticides, and short-chain aliphatic chemicals. Hydrophobic chemicals (log Kow > 3) are bound so strongly to the surface of roots and soils that they cannot be translocated easily within the plant, and chemicals that are quite water soluble (log Kow < 0.5) are not sufficiently sorbed to roots nor actively transported through plant membranes (Dietz and Schnoor, 2001; Pilon-Smits, 2005). Hydrophobic chemicals (log Kow > 3) are candidates for phytostabilization and/or rhizosphere bioremediation by virtue of their long residence times in the root zone (Dietz and Schnoor, 2001; Pilon-Smits, 2005).


Organic compounds which are sorbed to roots can be translocated to other plant tissues and, subsequently, they may be volatilized, they may undergo partial/complete degradation, or they may be transformed to less toxic, especially less phytotoxic, compounds and bound in plant tissues. In general, most organics appear to undergo some degree of transformation in plant cells before being sequestered in vacuoles or bound to insoluble cellular structures such as lignin. Metabolism of pesticides was extensively studied many years ago. More recently, the metabolism of nonagricultural xenobiotics such as trichloroethylene (TCE), TNT, glyceroltrinitrate (GTN), PAHs, PCBs and other chlorinated compounds has also been studied. It was shown that most of these compounds are metabolized but only a few chemicals appear to be fully mineralized. Some plant metabolites of pollutants may be more toxic than the original compounds, making plants less attractive compared with bacteria, which totally degrade organic pollutants.

In the rhizosphere take place important physiological and biochemical processes involved in the removal of contaminants which are induced by the interaction between plants, microorganisms and the solid matrix. Plants sustain large microbial populations in the rhizosphere by secreting a variety of products (exudates, mucigels, dead cell material, etc), which is known as rhizodeposition (Stottmeister et al., 2003). Consequently, favorable plant characteristics for phytoremediation are large and dense root systems which also have high levels of degrading enzymes.

The chemical composition of the exudates is very diverse and varies with the plant species, but they normally include sugars, organic acids and vitamins. It has been shown that this availability of nutrient in the immediate proximity of the roots makes the microbial population much larger in the rhizosphere than in the bulk soil, and that these larger populations increase the degradation of organic compounds (Yu et al., 2003; Sun et al., 2004). Classes of organic compounds that are more rapidly degraded in the rhizosphere than in bulk soil include polycyclic aromatic hydrocarbons (PAHs), total petroleum hydrocarbons, chlorinated pesticides as well as other chlorinated compounds like polychlorinated biphenyls (PCBs), explosives such as TNT and RDX, organophosphate insecticides, and surfactants.

One of the most important characteristics of the macrophyte plants is their ability to supply their root system with oxygen from the atmosphere. The oxygen is released mainly around the root tips promoting detoxification of harmful organic pollutants by oxidation. While these oxidative reactions take place in the immediate surroundings of the plants roots, anaerobic reactions are taking place just a few microns away (Sundaravadivel and Vigneswaran, 2001). The oxidative protective film formed on the root surface protects these sensitive areas from being damaged by toxic wastewater components in the anoxic, usually extremely reduced, rhizosphere (Stottmeister et al., 2003).

For the purpose of phytoremediation, plant species have to be selected based on the criteria of their tolerance to the pollutants toxicity and their capacity to lower the pollutants concentration in water through the variety of the processes which have been described above. Aquatic macrophytes seem to be especially resistant to a great variety of these substances in concentrations normally encountered in typical wastewater compositions. In addition, in numerous studies did these plant species exhibit the capability to reduce the concentrations of various xenobiotics in water. For example, the uptake of explosives such as RDX and TNT from the environment has been observed in several macrophytes (Typha spp., Carex spp., Scirpus spp., Myriophyllum aquaticum, Phalaris spp.) in numerous studies (Best et al., 1997; Best et al., 1999). Typha spp. and Phragmites spp. are also known to uptake petroleum


hydrocarbons (Haberl et al., 2003) and extensive studies have been conducted on pesticides uptake (Dordio et al., 2008b).

4.6.1.5. CWS Developed for Organic Xenobiotics Removal

Many constructed wetlands applications have been consisting of domestic wastewater treatment where the BOD and COD parameters have been used as a cumulative measure of the amounts of organic compounds. However, a more specialized use for the removal of specific organic compounds or classes of compounds has been developing as a growing type of CWS applications.

A significant experience already exists with wastewater from the petroleum industry, food processing industry, pesticide contaminated agricultural runoff, landfill leachates, and waters containing surfactants, solvents, and mineral oils. Specific compounds which have received successful treatment in CWS include a range of petroleum hydrocarbons, among which are BTEX and PAHs, organic solvents (in particular the more problematic chlorinated compounds), explosives such as TNT and RDX, PCBs and textile dyes.

In table 11, an overview of recently published studies using fully assembled CWS serves to show the vitality and diversity of work that is currently being done in this field.

Table 11. Several classes of organic xenobiotics removal in CWS

Organic xenobiotic Type of CW Vegetation References

Petroleum hydrocarbons/ PAHs

Benzene VSSF Phragmites australis (Eke and Scholz, 2008) FWS Typha latifolia (Haberl et al., 2003) BTEX VSSF Emergent plants (Bedessem et al., 2007) HSSF Phragmites australis (Giraud et al., 2001) PAHs Hybrid system Typha spp., Scirpus lacustris (Machate et al., 1997) FWS Typha latifolia, Schoenoplectus

tabernaemontani (Gessner et al., 2005) Petroleum

hydrocarbons VSSF (pilot-scale)

Salix spp. , Scirpus spp., Juncus spp., Phragmites australis

(Haberl et al., 2003)

Chlorinated solvents

Monochlorobenzene HSSF Phragmites australis (Braeckevelt et al., 2007)

TCE and PCE VSSF Common fen plants, winter-tolerant sedges and rushes

(Amon et al., 2007)

TCE Wetland microcosms

Populus deltoides, Typha latifolia

(Bankston et al., 2002)

Explosives RDX SSF mesocosms Typha laxmanil (Low et al., 2008)

FWS Submergent and emergent plants (Best et al., 2001) TNT HSSF (pilot-scale)

Phragmites australis (Haberl et al., 2003)

Pesticides Alachlor SSF Phragmites australis (Matamoros et al.,

2007)


Organic xenobiotic Type of CW Vegetation References

CW cells Typha latifolia (Runes et al., 2003) CW mesocosms Primarily Juncus spp. (Moore et al., 2000)

Atrazine

VSSF Schoenoplectus lacustris, Typha latifolia, Iris pseudacorus and Phragmites australis

(McKinlay and Kasperek, 1999)

Azinphos-methyl CW Typha capensis, Juncus kraussii, and Cyperus dives

(Schulz et al., 2003)

Endosulfan Flow-trough wetlands

Typha capensis, Juncus kraussii and Cyperus dives

(Schulz and Peall, 2001)

CW Sparganium erectum, Phragmites australis, Phalaris arundinacea, Myostis scorpioides, Urtica dioica

(Braskerud and Haarstad, 2003)

Mecoprop

SSF Phragmites australis (Matamoros et al., 2007)

Methyl parathion CW mesocosms Juncus effusus and Leersia oryzoides

(Moore et al., 2006)

Simazine CW cells Scirpus validus (Stearman et al., 2003) Others Acid orange 7 (AO7) dye

VSSF Phragmites australis (Davies et al., 2005)

Benzoic acid SSF Scirpus validus (Zachritz et al., 1996) PCBs FWS Typha latifolia (Haberl et al., 2003) Several organic solvents

SSF Juncus effusus, Carex lurida, Iris pseudocarus, Pondeteric cordata

(Grove and Stein, 2005)

Hybrid system Phragmites australis (Bulc and Ojstrsek, 2008)

Surfactants

SSF Zantedeschia aethiopica (Belmont and Metcalfe, 2003)

The case studies for treatment of waters contaminated with special organic compounds

display a variety of applications for constructed wetlands. The reported removal efficiencies of organic xenobiotics show in many cases surprisingly good results, with values above 70% being frequently achieved in these studies. However, very little is commonly known about the exact pathways of the xenobiotics removal. Given the diversity of chemical characteristics of these compounds, which despite being classified under a common designation of xenobiotics are in fact formed by widely unrelated families of chemical substances, it is of no surprise that several very different mechanisms are responsible for removal of different xenobiotic. A comprehensive description of xenobiotic removal in CWS is therefore not an easy task to accomplish and these systems are still largely operated as a “black box”.

Although much of the design of CWS in the past has been done with little knowledge of the roles played by each component and how its function could be optimized, treating the system primarily as a black box, nowadays the study base that has been accumulating is beginning to be applied. We now can see a much greater variety of plant species, matrix materials and wetlands designs being introduced. The goals of the target contaminants to remove in CWS are also becoming progressively more ambitious.


5. NEW TRENDS IN CONSTRUCTED WETLANDS SYSTEMS APPLICATIONS

In the past, CWS have been used mainly as wastewater treatment alternatives or

complementary to the conventional treatment for domestic wastewaters of small communities. Thus, CWS have been mostly applied in the reduction of TSS, organic matter, excess of nutrients and faecal microorganisms. This is somewhat overlapping with the design goals of WWTPs, but with the benefits of low cost and low maintenance which are characteristic of CWS.

More recently, CWS applications to deal with more specific types of pollution (such as that caused by xenobiotics, as described in the previous section) have been meeting a larger interest and have been the subject of an increasing number of studies. CWS have been proving to be efficient and cost-effective solutions for the removal of some pollutants, where the use of conventional wastewater treatment processes does not satisfactorily solve the problem and a resort to more advanced treatments are too costly to become viable alternatives on a larger scale.

Among the several types of organic xenobiotics, pharmaceuticals have been recently attracting much attention of the international scientific community and emerging as a new important class of environmental contaminants. Compounds used in human and veterinary medicine are being continuously introduced in the environment, mainly due to improper disposal of unused or expired drugs, and through metabolic excretion. Some of these pharmaceutical residues are discharged directly in the environment without going through appropriate treatment, but even those receiving appropriate disposal in WWTPs in many cases are not effectively removed by the conventional wastewater treatment processes (Halling-Sørensen et al., 1998; Daughton and Ternes, 1999; Heberer, 2002; Fent et al., 2006).

In spite of being contaminants of many water bodies for already quite a long time (Garrison et al., 1976; Hignite and Azarnoff, 1977), environmental issues related with pharmaceuticals contamination are only recently becoming subject of more intensive study (Daughton and Ternes, 1999; Petrovic et al., 2003; Fent et al., 2006; Kemper, 2008). The low concentrations of these compounds in the environment, which are typically at trace levels (ng/L to low μg/L), associated to the unavailability, until recently, of suitably sensitive methods of analysis for these low concentration ranges has been the main reason for the late interest on the environmental problems raised by these compounds. Over the latest years, in numerous monitoring studies, residues of lipid regulating drugs, analgesics and anti-inflammatory drugs, antibiotics, hormones, chemotherapy agents, antidiabetics, neuroactive compounds and beta-blocking heart drugs have all been detected in wastewaters, surfacewaters and even groundwaters worldwide (Daughton and Ternes, 1999; Fent et al., 2006; Hernando et al., 2006).

Despite the trace level concentrations of pharmaceutical residues, such low amounts can sometimes be large enough to induce toxic effects on organisms (Halling-Sørensen et al., 1998; Ferrari et al., 2003; Fent et al., 2006; Crane et al., 2006; Hernando et al., 2006). A major problem is caused by the very nature of some pharmaceuticals since these compounds are designed to have very specific modes of action and biological effects, and many are persistent in the body. Because of their physicochemical and biological properties, when released into environment, it may be possible for them to cause serious impacts on non-target


species, e.g., aquatic and terrestrial organisms. Concern has been mostly focused on antibiotics that may cause resistance among bacteria or steroids that can induce estrogenic effects on aquatic species (Halling-Sørensen et al., 1998; Fent et al., 2006). It was a surprise when in 2004, diclofenac, an analgesic and anti-inflammatory drug frequently detected in WWTPs effluents and water courses was found to be the culprit of the nearly extinction of some species of Asian vultures (Oaks et al., 2004). The metabolites and degradation by-products of pharmaceuticals are also of concern, because many of them have a toxicity which in many cases is similar to or even higher than the parent compounds (Fent et al., 2006).

Besides toxicity, the element of persistence is of particular importance when considering the environmental significance of pharmaceuticals. Unlike persistent organic pollutants like pesticides, many pharmaceuticals are not lipophilic, so they do not bioaccumulate in the environment. However, some of those are “pseudo persistent pollutants” due to their continuous introduction in the environment. While not persistent in terms of a long half-life, these chemicals are constantly entering the environment, resulting in long-term exposure for the aquatic ecosystem. Potential synergetic and chronic effects have been mostly overlooked in the past, but recent ecotoxicological studies indicate that pharmaceutical residues pose a major threat especially for the aquatic species due to the continuous life-long exposure to which they are subjected (Fent et al., 2006) . The possible development of antibiotic-resistant bacteria, the genotoxic effects of some drugs, and endocrine disruption by therapeutically administered synthetic and natural hormones have all been discussed (Halling-Sørensen et al., 1998; Daughton and Ternes, 1999; Fent et al., 2006), but very little is known about possible long-term subtle effects on non-target organisms.

The foreseeable environmental consequences of high environmental loads of pharmaceuticals points out to the urgent need of finding ways to retain and remove these pollutants before they reach the waterbodies. Optimization of the WWTP processes has been tried by increasing sludge residence times, and some advanced technologies have been evaluated to decrease their discharge into water bodies, e.g. oxidative processes, activated carbon and membrane filtration (Andreozzi et al., 2002; Fent et al., 2006; Esplugas et al., 2007; Kim et al., 2007; Snyder et al., 2007). However, despite the sometimes high removal efficiencies attained, these processes are generally not cost-effective on a large scale (Fent et al., 2006) .

Due to the recent emergence of this environmental problem, CWS have not yet been fully evaluated for the removal of pharmaceutical residues. While they are being used efficiently (as shown in the previous section) for removing several other types of organic xenobiotics (Williams, 2002; Hannink et al., 2002; Braskerud and Haarstad, 2003; Haberl et al., 2003), fewer studies exist on their use for pharmaceuticals depuration (Gross et al., 2004; Matamoros et al., 2005; Matamoros and Bayona, 2006; Matamoros et al., 2007; Dordio et al., 2008a; Matamoros et al., 2008b).

Pharmaceutical substances span a wide range of chemical behaviors: a diversity of acid-base properties provide for both easily ionizable as well as neutral compounds; while some compounds are very water soluble and hydrophilic, others are significantly hydrophobic/lipophilic; some compounds are very stable to a wide range of redox conditions whereas others are readily oxidizable or reduced. Conversely, the effective mechanisms of pharmaceutical removal in a CWS are also varied.

Easily bio-degradable pharmaceuticals are most effectively removed by the biotic components of a CWS. The action of the rhizostimulated microbial populations on one hand,


and uptake by plant on the other, are both possible mechanisms (as described in the previous section, only now acting on much lower concentration levels of pollutants which is per se a relevant issue for the system’s efficiency) that can be responsible for pharmaceuticals removal, depending on the compounds properties. An aspect to consider when designing a CWS for treatment of these types of substances and when selecting the type of biotic components is the assessment of the long-term tolerance to possible toxic effects which may be caused by pharmaceutical residues as well as other xenobiotics.

Removal of more recalcitrant pharmaceuticals has also been observed in CWS (Gross et al., 2004; Matamoros et al., 2005; Matamoros and Bayona, 2006; Matamoros et al., 2008a; Matamoros et al., 2008b). Sorption by the substrate plays a major role in such cases. However, the characteristics of the substrate are crucial for the effectiveness of this process, and some materials do provide for substantial removal efficiencies (Dordio et al., 2007; Dordio et al., 2008a) whereas other do not (Matamoros et al., 2005; Matamoros et al., 2008a; Matamoros et al., 2008b). Therefore, a careful selection of the type of substrate which displays great affinity for the type of compounds to be removed by the CWS is an essential step in the system’s optimization and the achievement of acceptable efficiencies.

Ultimately, the optimization of a CWS for the removal of more specific targets requires a basic knowledge of the processes involved in the pollutants removal and the interactions between the CWS components. New trends in CWS research are moving towards the study of such processes and interactions and focus on the selection and optimization of the components for more specific applications.

CONCLUSION Wetlands have long been known to have water depuration capabilities and, in fact, have

been frequently used as wastewater discharge sites. However, in spite of these characteristics, such practice has led over the years to a substantial degradation of many of these ecosystems and in many countries the discharge of wastewaters in natural wetlands is no longer allowed. Furthermore, drainage of wetlands for agricultural purposes has also contributed to the decline in the quality of these regions. For the purpose of the restoration of some of these ecosystems or creation of habitats for wetlands wildlife, the construction of artificial wetlands has been carried out in many former natural wetlands areas. In addition, constructed wetlands have also been developed to imitate the functions provided by natural wetlands, in particular their wastewater treatment capabilities, which can also contribute to the preservation of natural wetlands and neighboring water bodies, and as well perform wastewater treatment in a more controlled and optimized environment to achieve better wastewater treatment efficiencies.

Constructed treatment wetlands systems are a biotechnology for wastewater treatment that is becoming increasingly popular as an alternative to conventional treatments or integrated in conventional systems as a secondary or tertiary treatment step. Low cost and low maintenance are some of its most attractive characteristics along with the aesthetic and landscape enhancement qualities, although they do have the disadvantages of requiring substantially more land area occupation and showing a less reproducible behavior than conventional technologies.


These systems are now becoming a mature technology for the removal of suspended solids, organic matter and nutrients. Focus is now moving into the removal of more specific and recalcitrant compounds for which the conventional treatment systems are not effective. CWS are nowadays being increasingly used for the cleanup of specific pollutant types such as organic xenobiotics and new challenges have been emerging such as the removal of pharmaceuticals and other micropollutants which present new problems to be solved.

Often CWS have been looked at as a “black box” where only influent and effluent pollutants concentrations where measured and where no more in-depth investigations were run. To use CWS as an efficient response to these new challenges this “black-box” approach to CWS operation has to be abandoned and a thorough understanding of the variety of processes involved in pollutants removal in CWS as well as the way the CWS components interact is direly needed. This has been in fact an effort which has been increasingly undertaken in the most recent years in the area of CWS research and development, not only in field studies but also in numerous lab studies as well.

Already a considerable base of knowledge exists about the CWS components and the removal processes for which they are responsible. Still, there is plenty of room for further research, as is the case of the most advanced genetic engineering technology applied both to the vegetation as to the microbial components, but also in the more traditional research subjects such as the advantages of polyculture vegetation or use of composite substrates. Additional areas to explore may be how the components may interact to provide synergetic enhancements for pollutants removal.

In conclusion, the results obtained in the research currently being conducted in this area are mostly very positive and show the potential of this promising technology even for the type of wastewater contamination that the conventional systems have not been able to cope with. Naturally constructed treatment wetlands systems have their own shortcomings and these will not be the one-size-fits-all solution to every wastewater contamination problem but it can certainly be in some cases an alternative, with advantages, to conventional wastewater treatment (e.g., for the case of single-family or small community municipal wastewaters) and in other cases complimentary to such conventional treatment.

Ultimately, the services provided by constructed wetlands may also contribute to a wider recognition of the value of natural wetlands and the role they play in the preservation of water quality in their surroundings.

REFERENCES

Amon, J. P.; Agrawal, A.; Shelley, M. L.; Opperman, B. C.; Enright, M. P.; Clemmer, N. D.; Slusser, T.; Lach, J.; Sobolewski, T.; Gruner, W.; Entingh, A. C. Ecol. Eng. 2007, 30, 51-66.

Andreozzi, R.; Marotta, R.; Pinto, G.; Pollio, A. Water Res. 2002, 36, 2869-2877. Ansola, G.; González, J. M.; Cortijo, R.; de Luis, E. Ecol. Eng. 2003, 21, 43-52. Bankston, J. L.; Sola, D. L.; Komor, A. T.; Dwyer, D. F. Water Res. 2002, 36, 1539-1546. Barr, M. J.; Robinson, H. D. Waste Manage. Res. 1999, 17, 498-504. Bedessem, M. E.; Ferro, A. M.; Hiegel, T. Water Environ. Res. 2007, 79, 581-586. Belmont, M. A.; Metcalfe, C. D. Ecol. Eng. 2003, 21, 233-247.


Best, E. P. H.; Miller, J. L.; Larson, S. L. Water Sci. Technol. 2001, 44, 515-521. Best, E. P. H.; Sprecher, S. L.; Larson, S. L.; Fredrickson, H. L.; Bader, D. F. Chemosphere.

1999, 38, 3383-3396. Best, E. P. H.; Zappi, M. E.; Fredrickson, H. L.; Sprecher, S. L.; Larson, S. L.; Ochman, M.

Ann. NY Acad. Sci. 1997, 829, 179-194. Birkett, J. W.; Lester, J. N. Endocrine Disruptors in Wastewater and Sludge Treatment

Processes; CRC Press: Boca Raton, FL, USA, 2003. Brady, N. C.; Weil, R. R. The Nature and Properties of Soils; Prentice Hall: New Jersey, NJ,

USA, 2002. Braeckevelt, M.; Rokadia, H.; Imfeld, G.; Stelzer, N.; Paschke, H.; Kuschk, P.; Kastner, M.;

Richnow, H. H.; Weber, S. Environ. Pollut. 2007, 148, 428-437. Braskerud, B. C. Ecol. Eng. 2002, 18, 351-370. Braskerud, B. C.; Haarstad, K. Water Sci. Technol. 2003, 48, 267-274. Brix, H. In Constructed Wetlands for Water Quality Improvement; Moshiri, G. A.; Ed.; Lewis

Publishers: Boca Raton, FL, USA, 1993, pp 9-22. Brix, H.; Arias, C. A.; del Bubba, M. Water Sci. Technol. 2001, 44, 47-54. Brodie, G. A.; Hammer, D. A.; Tomijanovich, D. A. In Constructed Wetlands for Wastewater

Treatment; Hammer, D. A.; Ed.; Lewis Publishers: Chelsea, MI, USA, 1989. Brooks, A. S.; Rozenwald, M. N.; Geohring, L. D.; Lion, L. W.; Steenhuis, T. S. Ecol. Eng.

2000, 15, 121-132. Bulc, T. G.; Ojstrsek, A. J. Hazard. Mater. 2008, In Press, Corrected Proof. Burke, D. J.; Weis, J. S.; Weis, P. Estuar. Coast. Shelf Sci. 2000, 51, 153-159. Cameron, K.; Madramootoo, C.; Crolla, A.; Kinsley, C. Water Res. 2003, 37, 2803-2812. Carleton, J. N.; Grizzard, T. J.; Godrej, A. N.; Post, H. E. Water Res. 2001, 35, 1552-1562. Carty, A.; Scholz, M.; Heal, K.; Gouriveau, F.; Mustafa, A. Bioresour. Technol. In Press,

Corrected Proof. Chen, H. M.; Zheng, C. R.; Tu, C.; Shen, Z. G. Chemosphere. 2000, 41, 229-234. Chen, T. Y.; Kao, C. M.; Yeh, T. Y.; Chien, H. Y.; Chao, A. C. Chemosphere. 2006, 64, 497-

502. Cheng, S. P.; Grosse, W.; Karrenbrock, F.; Thoennessen, M. Ecol. Eng. 2002, 18, 317-325. Collins, C.; Fryer, M.; Grosso, A. Environ. Sci. Technol. 2006, 40, 45-52. Cooper, P. J. Chart. Inst. Water Environ. Manage. 2001, 15, 79-85. Cooper, P. F.; Job, G. D.; Green, M. B.; Shutes, R. B. E. Reed Beds and Constructed

Wetlands for Wastewater Treatment; WRc Publications: Medmenham, Marlow, UK, 1996.

Coveney, M. F.; Stites, D. L.; Lowe, E. F.; Battoe, L. E.; Conrow, R. Ecol. Eng. 2002, 19, 141-159.

Cowardin, L. M.; Golet, F. C. Vegetatio. 1995, 118, 139-152. Craft, C. B.; Richardson, C. J. Biogeochemistry. 1993, 22, 133-156. Crane, M.; Watts, C.; Boucard, T. Sci. Total Environ. 2006, 367, 23-41. Cronk, J. K. Agric. Ecosyst. Environ. 1996, 58, 97-114. D'Angelo, E. M.; Reddy, K. R. J. Environ. Qual. 1994, 23, 937-943. Daughton, C. G.; Ternes, T. A. Environ. Health Perspect. 1999, 107, 907-938. Davies, L. C.; Carias, C. C.; Novais, J. M.; Martins-Dias, S. Ecol. Eng. 2005, 25, 594-605. Debusk, T. A.; Laughlin, R. B.; Schwartz, L. N. Water Res. 1996, 30, 2707-2716. Dietz, A. C.; Schnoor, J. L. Environ. Health Perspect. 2001, 109, 163-168.


Dordio, A. V.; Candeias, A. J. E.; Pinto, A. P.; Costa, C. T.; Carvalho, A. J. P. Ecol. Eng. 2008a, In Press, Corrected Proof.

Dordio, A. V.; Carvalho, A. J. P.; Pinto, A. P.; Costa, C. T. In Pesticide Research Trends; Tennefy, A. B.; Ed.; Nova Science Publishers: Hauppauge, NY, USA, 2008b.

Dordio, A. V.; Teimão, J.; Ramalho, I.; Carvalho, A. J. P.; Candeias, A. J. E. Sci. Total Environ. 2007, 380, 237-246.

Dua, M.; Singh, A.; Sethunathan, N.; Johri, A. Appl. Microbiol. Biotechnol. 2002, 59, 143-152.

Dunbabin, J. S.; Bowmer, K. H. Sci. Total Environ. 1992, 111, 151-168. Eke, P. E.; Scholz, M. J. Chem. Technol. Biotechnol. 2008, 83, 55-63. Environmental Laboratory. Corps of Engineers Wetland Delineation Manual; Technical

Report Y-87-1; US Army Engineer Waterways Experiment Station: Vicksburg, MS, USA, 1987.

Esplugas, S.; Bila, D. M.; Krause, L. G.; Dezotti, M. J. Hazard. Mater. 2007, 149, 631-642. Evans, G. M.; Furlong, J. C. Environmental Biotechnology: Theory and Application; Wiley:

Chichester, UK, 2003. Fent, K.; Weston, A. A.; Caminada, D. Aquat. Toxicol. 2006, 76, 122-159. Ferrari, B.; Paxéus, N.; Lo Giudice, R.; Pollio, A.; Garric, J. Ecotox. Environ. Safe. 2003, 55,

359-370. Fude, L.; Harris, B.; Urrutia, M. M.; Beveridge, T. J. Appl. Environ. Microbiol. 1994, 60,

1525-1531. García, M.; Soto, F.; González, J. M.; Bécares, E. Ecol. Eng. 2008, 32, 238-243. Garrison, A. W.; Pope, J. D.; Allen, F. R. In Identification and Analysis of Organic Pollutants

in Water; Keith, L. H.; Ed.; Ann Arbor Science Publishers: Ann Arbor, MI, USA, 1976, pp 517-556.

Gessner, T. P.; Kadlec, R. H.; Reaves, R. P. Ecol. Eng. 2005, 25, 457-469. Giraud, F.; Guiraud, P.; Kadri, M.; Blake, G.; Steiman, R. Water Res. 2001, 35, 4126-4136. Gómez Cerezo, R.; Suárez, M. L.; Vidal-Abarca, M. R. Ecol. Eng. 2001, 16, 501-517. Gray, S.; Kinross, J.; Read, P.; Marland, A. Water Res. 2000, 34, 2183-2190. Gross, B.; Montgomery-Brown, J.; Naumann, A.; Reinhard, M. Environ. Toxicol. Chem.

2004, 23, 2074-2083. Groudev, S. N.; Bratcova, S. G.; Komnitsas, K. Miner. Eng. 1999, 12, 261-270. Grove, J. K.; Stein, O. R. Water Res. 2005, 39, 4040-4050. Haberl, R.; Grego, S.; Langergraber, G.; Kadlec, R. H.; Cicalini, A. R.; Martins-Dias, S.;

Novais, J. M.; Aubert, S.; Gerth, A.; Thomas, H.; Hebner, A. J. Soil Sediment. 2003, 3, 109-124.

Hadad, H. R.; Maine, M. A.; Bonetto, C. A. Chemosphere. 2006, 63, 1744-1753. Hadad, H. R.; Maine, M. A.; Natale, G. S.; Bonetto, C. Ecol. Eng. 2007, 31, 122-131. Hallberg, K. B.; Johnson, D. B. Sci. Total Environ. 2005, 338, 115-124. Halling-Sørensen, B.; Nielsen, S. N.; Lanzky, P. F.; Ingerslev, F.; Lützhøft, H. C. H.;

Jørgensen, S. E. Chemosphere. 1998, 36, 357-394. Halverson, N. V. Review of Constructed Subsurface Flow vs. Surface Flow Wetlands;

WSRC-TR-2004-00509; Westinghouse Savannah River Company: Aiken, SC, USA, 2004.

Hammer, D. A. Creating Freshwater Wetlands; CRC Press: Boca Raton, FL,USA, 1997.


Hammer, D. A.; Bastian, R. K. In Constructed Wetlands for Wastewater Treatment; Hammer, D. A.; Ed.; Lewis Publishers: Chelsea, MI, USA, 1989, pp 5-19.

Hammer, D. A.; Pullin, B. P.; McCaskey, T. A.; Eason, J.; Payne, V. W. E. In Constructed Wetlands for Wastewater Treatment; Hammer, D. A.; Ed.; Lewis Publishers: Chelsea, MI, USA, 1989.

Hannink, N. K.; Rosser, S. J.; Bruce, N. C. Crit. Rev. Plant Sci. 2002, 21, 511-538. Heberer, T. Toxicol. Lett. 2002, 131, 5-17. HELCOM. Baltic Sea Environment Proceedings No. 64B. Third Periodic Assessment on the

State of the Marine Environment of the Baltic Sea, 1989-1993; Background Document.; Helsinki Commission, Baltic Marine Environment Protection Commission: Helsinki, Finland, 1996.

Hench, K. R.; Bissonnette, G. K.; Sexstone, A. J.; Coleman, J. G.; Garbutt, K.; Skousen, J. G. Water Res. 2003, 37, 921-927.

Hernando, M. D.; Mezcua, M.; Fernandez-Alba, A. R.; Barceló, D. Talanta. 2006, 69, 334-342.

Hignite, C.; Azarnoff, D. L. Life Sci. 1977, 20, 337-341. IETC-UNEP. Planning and Management of Lakes and Reservoirs: An Integrated Approach to

Eutrophication; 1999. ITRC. Technical and regulatory guidance document for constructed treatment wetlands; The

Interstate Technology and Regulatory Council Wetlands Team: New Jersey, NJ, USA, 2003.

Johansson, L.; Gustafsson, J. P. Water Res. 2000, 34, 259-265. Kadlec, H. R.; Knight, R. L. Treatment Wetlands; Lewis Publishers: Boca Raton, FL, USA,

1996. Karathanasis, A. D.; Potter, C. L.; Coyne, M. S. Ecol. Eng. 2003, 20, 157-169. Kemper, N. Ecol. Indic. 2008, 8, 1-13. Kim, S. D.; Cho, J.; Kim, I. S.; Vanderford, B. J.; Snyder, S. A. Water Res. 2007, 41, 1013-

1021. Kivaisi, A. K. Ecol. Eng. 2001, 16, 545-560. Kleeberg, A.; Schlungbaum, G. Hydrobiologia. 1993, 253, 263-274. Knight, R. L.; Payne, V. W. E.; Borer, R. E.; Clarke, R. A.; Pries, J. H. Ecol. Eng. 2000, 15,

41-55. Kohler, E. A.; Poole, V. L.; Reicher, Z. J.; Turco, R. F. Ecol. Eng. 2004, 23, 285-298. Kuehn, E.; Moore, J. A. Water Sci. Technol. 1995, 32, 241-250. Langergraber, G.; Haberl, R. Minerva Biotechnol. 2001, 13, 123-134. Ledin, M.; Pedersen, K. Earth-Sci. Rev. 1996, 41, 67-108. Lee, B. h.; Scholz, M. Ecol. Eng. 2007, 29, 87-95. Lesage, E.; Rousseau, D. P. L.; Meers, E.; Tack, F. M. G.; De Pauw, N. Sci. Total Environ.

2007, 380, 102-115. Lin, L. Y. In Encyclopedia of environmental biology; Nierenberg, W. A.; Ed.; Academic

Press: San Diego, CA, USA, 1995; Vol. 3, pp 479-484. Liu, J.; Dong, Y.; Xu, H.; Wang, D.; Xu, J. J. Hazard. Mater. 2007, 147, 947-953. Livingston, E. H. In Constructed Wetlands for Wastewater Treatment; Hammer, D. A.; Ed.;

Lewis Publishers: Chelsea, MI, USA, 1989, pp 253-264. Low, D.; Tan, K.; Anderson, T.; Cobb, G. P.; Liu, J.; Jackson, W. A. Ecol. Eng. 2008, 32, 72-

80.


Machate, T.; Noll, H.; Behrens, H.; Kettrup, A. Water Res. 1997, 31, 554-560. Machemer, S. D.; Wildeman, T. R. J. Contam. Hydrol. 1992, 9, 115-131. Maine, M. A.; Suñe, N.; Hadad, H.; Sánchez, G.; Bonetto, C. Ecol. Eng. 2006, 26, 341-347. Maine, M. A.; Suñe, N.; Hadad, H.; Sánchez, G.; Bonetto, C. Chemosphere. 2007, 68, 1105-

1113. Mantovi, P.; Marmiroli, M.; Maestri, E.; Tagliavini, S.; Piccinini, S.; Marmiroli, N.

Bioresour. Technol. 2003, 88, 85-94. Markert, B. Plants as biomonitors: indicators for heavy metals in the terrestrial environment;

VCH: Weinheim, Germany, 1993. Mashauri, D. A.; Mulungu, D. M. M.; Abdulhussein, B. S. Water Res. 2000, 34, 1135-1144. Mason, C. Biology of freshwater pollution; Pearson Education: Harlow, UK, 2002. Matamoros, V.; Bayona, J. M. Environ. Sci. Technol. 2006, 40, 5811-5816. Matamoros, V.; García, J.; Bayona, J. M. Environ. Sci. Technol. 2005, 39, 5449-5454. Matamoros, V.; Puigagut, J.; Garcia, J.; Bayona, J. M. Chemosphere. 2007, 69, 1374-1380. Matamoros, V.; Caselles-Osorio, A.; Garcia, J.; Bayona, J. M. Sci. Total Environ. 2008a, 394,

171-176. Matamoros, V.; García, J.; Bayona, J. M. Water Res. 2008b, 42, 653-660. McKinlay, R. G.; Kasperek, K. Water Res. 1999, 33, 505-511. Merlin, G.; Pajean, J. L.; Lissolo, T. Hydrobiologia. 2002, 469, 87-98. Metcalf; Eddy Wastewater engineering: treatment, disposal and reuse; McGraw Hill: New

York, NY, USA, 1991. Meuleman, A. F. M.; van Logtestijn, R.; Rijs, G. B. J.; Verhoeven, J. T. A. Ecol. Eng. 2003,

20, 31-44. Mitra, S. Mercury in the ecosystem; Trans Tech Publications: Aedermannsdorf, Switzerland,

1986. Mitsch, W. J.; Gosselink J.G. Wetlands; John Wiley and Sons: New York, NY, USA, 2000. Moore, M. T.; Bennett, E. R.; Cooper, C. M.; Smith, S.; Farris, J. L.; Drouillard, K. G.;

Schulz, R. Environ. Pollut. 2006, 142, 288-294. Moore, M. T.; Rodgers, J. H.; Cooper, C. M.; Smith, S. Environ. Pollut. 2000, 110, 393-399. Muller, K.; Magesan, G. N.; Bolan, N. S. Agric. Ecosyst. Environ. 2007, 120, 93-116. National Research Council Wetlands: Characteristics and Boundaries; National Academy

Press: Washington, DC, USA, 1995. Neralla, S.; Weaver, R. W.; Lesikar, B. J.; Persyn, R. A. Bioresour. Technol. 2000, 75, 19-25. Nivala, J.; Hoos, M. B.; Cross, C.; Wallace, S.; Parkin, G. Sci. Total Environ. 2007, 380, 19-

27. Oaks, J. L.; Gilbert, M.; Virani, M. Z.; Watson, R. T.; Meteyer, C. U.; Rideout, B. A.;

Shivaprasad, H. L.; Ahmed, S.; Chaudhry, M. J. I.; Arshad, M.; Mahmood, S.; Ali, A.; Khan, A. A. Nature. 2004, 427, 630-633.

Oovel, M.; Tooming, A.; Mauring, T.; Mander, U. Ecol. Eng. 2007, 29, 17-26. Otte, M. L.; Kearns, C. C.; Doyle, M. O. Bull. Environ. Contam. Toxicol. 1995, 55, 154-161. Ottova, V.; Balcarova, J.; Vymazal, J. Water Sci. Technol. 1997, 35, 117-123. Perkins, J.; Hunter, C. Water Res. 2000, 34, 1941-1947. Petrovic, M.; Gonzalez, S.; Barceló, D. Trac-Trends Anal. Chem. 2003, 22, 685-696. Pilon-Smits, E. Annu. Rev. Plant Biol. 2005, 56, 15-39. Reddy, K. R.; D'Angelo, E. M. Water Sci. Technol. 1997, 35, 1-10. Richardson, C. J.; Marshall, P. E. Ecol. Monogr. 1986, 56, 279-302.


Rivera, F.; Warren, A.; Ramirez, E.; Decamp, O.; Bonilla, P.; Gallegos, E.; Calderon, A.; Sanchez, J. T. Water Sci. Technol. 1995, 32, 211-218.

Runes, H. B.; Jenkins, J. J.; Moore, J. A.; Bottomley, P. J.; Wilson, B. D. Water Res. 2003, 37, 539-550.

Russell, R. A.; Holden, P. J.; Wilde, K. L.; Neilan, B. A. Hydrometallurgy. 2003, 71, 227-234.

Safe, S. H. Crit. Rev. Toxicol. 1994, 24, 87-149. Scholes, L.; Shutes, R. B. E.; Revitt, D. M.; Forshaw, M.; Purchase, D. Sci. Total Environ.

1998, 214, 211-219. Schulz, R. J. Environ. Qual. 2004, 33, 419-448. Schulz, R.; Hahn, C.; Bennett, E. R.; Dabrowski, J. M.; Thiere, G.; Peall, S. K. C. Environ.

Sci. Technol. 2003, 37, 2139-2144. Schulz, R.; Peall, S. K. C. Environ. Sci. Technol. 2001, 35, 422-426. Seffernick, J. L.; Wackett, L. P. Biochemistry. 2001, 40, 12747-12753. Seidel, K. In Ökologie, Morphologie und Entwicklung, ihre Stellung bei den Volkern und ihre

wirtschaftliche Bedeutung; Sweizerbart'sche Verlgsbuchhandlung: Stuttgart, Germany, 1955, pp 37-52.

Sheoran, A. S.; Sheoran, V. Miner. Eng. 2006, 19, 105-116. Skakkebaeck, N. E.; Leffers, H.; Rajpert-De Meyts, E.; Carlsen, E.; Grigor, K. M. Trends

Endocrinol. Metab. 2000, 11, 291-293. Sleytr, K.; Tietz, A.; Langergraber, G.; Haberl, R. Sci. Total Environ. 2007, 380, 173-180. Snyder, S. A.; Adham, S.; Redding, A. M.; Cannon, F. S.; DeCarolis, J.; Oppenheimer, J.;

Wert, E. C.; Yoon, Y. Desalination. 2007, 202, 156-181. Sobolewski, A. Int. J. Phytoremediat. 1999, 1, 19-51. Solano, M. L.; Soriano, P.; Ciria, M. P. Biosyst. Eng. 2004, 87, 109-118. Soto, F.; Garcia, M.; de Luis, E.; Bécares, E. Water Sci. Technol. 1999, 40, 241-247. Sriyaraj, K.; Shutes, R. B. E. Environ. Int. 2001, 26, 433-439. Stales, C. A.; Peterson, D. R.; Parkerton, T. F.; Adams, W. J. Chemosphere. 1997, 35, 667-

749. Stearman, G. K.; George, D. B.; Carlson, K.; Lansford, S. J. Environ. Qual. 2003, 32, 1548-

1556. Steer, D.; Fraser, L.; Boddy, J.; Seibert, B. Ecol. Eng. 2002, 18, 429-440. Stottmeister, U.; Wiessner, A.; Kuschk, P.; Kappelmeyer, U.; Kastner, M.; Bederski, O.;

Muller, R. A.; Moormann, H. Biotechnol. Adv. 2003, 22, 93-117. Stowell, R.; Ludwig, R.; Colt, J.; Tchobanoglous, G. J. Environ. Eng. -ASCE. 1981, 107, 919-

940. Stumm, W.; Morgan, J. J. Aquatic Chemistry: An Introduction Emphasizing Chemical

Equilibra In Natural Waters; John Wiley and Sons: New York, NY, USA, 1981. Sun, H.; Xu, J.; Yang, S.; Liu, G.; Dai, S. Chemosphere. 2004, 54, 569-574. Sundaravadivel, M.; Vigneswaran, S. Crit. Rev. Environ. Sci. Technol. 2001, 31, 351-409. Tanner, C. C.; Clayton, J. S.; Upsdell, M. P. Water Res. 1995, 29, 17-26. Tanner, C. C.; Kadlec, R. H.; Gibbs, M. M.; Sukias, J. P. S.; Nguyen, M. L. Ecol. Eng. 2002,

18, 499-520. Tyler, C. R.; Jobling, S.; Sumpter, J. P. Crit. Rev. Toxicol. 1998, 28, 319-361.


UNESCO. Convention on Wetlands of International Importance especially as Waterfowl Habitat; Ramsar, Iran, 1994. Available online at: http://www.ramsar.org/key_ conv_e.htm.

USEPA. Constructed wetlands for wastewater treatment and wild life habitat: 17 Case Studies; EPA 832-R-93-005; Office of Research and Development: Cincinnati, OH, USA, 1993.

USEPA. Constructed Wetlands Treatment of Municipal Wastewaters; EPA/625/R-99/010; Office of Research and Development: Cincinnati, OH, USA, 2000.

USEPA; USDA-NRCS. A Handbook of Constructed Wetlands. Volume 1: General Considerations; USEPA Region III with USDA-NRCS: Washington, DC, USA, 1995.

Vandecasteele, B.; Meers, E.; Vervaeke, P.; De Vos, B.; Quataert, P.; Tack, F. M. G. Chemosphere. 2005, 58, 995-1002.

Vohla, C.; Poldvere, E.; Noorvee, A.; Kuusemets, V.; Mander, U. J. Environ. Sci. Health Part A-Toxic/Hazard. Subst. Environ. Eng. 2005, 40, 1251-1264.

Vymazal, J. Water Sci. Technol. 1999, 40, 133-138. Vymazal, J. Ecol. Eng. 2005a, 25, 475-477. Vymazal, J. J. Environ. Sci. Health Part A-Toxic/Hazard. Subst. Environ. Eng. 2005b, 40,

1355-1367. Vymazal, J. Sci. Total Environ. 2007, 380, 48-65. Vymazal, J. In Constructed wetlands for wastewater treatment in Europe; Vymazal, J.; Brix,

H.; Cooper, P. F.; Green, M. B.; Haberl, R.; Eds.; Backhuys Publishers: Leiden, The Netherlands, 1998, pp 1-15.

Vymazal, J. Ecol. Eng. 2005c, 25, 478-490. Vymazal, J.; Brix, H.; Cooper, P. F.; Green, M. B.; Haberl, R. Constructed wetlands for

wastewater treatment in Europe; Backhuys Publishers: Leiden, The Netherlands, 1998a. Vymazal, J.; Brix, H.; Cooper, P. F.; Haberl, R.; Perfler, R.; Laber, J. In Constructed

wetlands for wastewater treatment in Europe; Vymazal, J.; Brix, H.; Cooper, P. F.; Green, M. B.; Haberl, R.; Eds.; Backhuys Publishers: Leiden, The Netherlands, 1998b, pp 17-66.

Walker, D. J.; Hurl, S. Ecol. Eng. 2002, 18, 407-414. Wiebner, A.; Kappelmeyer, U.; Kuschk, P.; Kastner, M. Water Res. 2005, 39, 248-256. Williams, J. B. Crit. Rev. Plant Sci. 2002, 21, 607-635. Wood, A. In Constructed wetlands in water pollution control; Cooper, P. F.; Findlater, B. C.;

Eds.; Pergamon Press: Oxford, UK, 1990. Woulds, C.; Ngwenya, B. T. Appl. Geochem. 2004, 19, 1773-1783. Wu, R. S. S. Mar. Pollut. Bull. 1999, 39, 11-22. Yu, Y. L.; Chen, Y. X.; Luo, Y. M.; Pan, X. D.; He, Y. F.; Wong, M. H. Chemosphere. 2003,

50, 771-774. Zachritz, W. H.; Lundie, L. L.; Wang, H. Ecol. Eng. 1996, 7, 105-116.


Chapter 2

REMOTE SENSING DATA FOR REGIONAL WETLAND MAPPING IN THE UNITED STATES: TRENDS AND FUTURE PROSPECTS

Megan W. Lang and Greg W. McCarty USDA-ARS, Remote Sensing and Hydrology Laboratory,

Beltsville Agricultural Research Center, Bldg 007, Rm 104, 10300 Baltimore Ave., Beltsville, MD 20705

ABSTRACT

Historically, the biologic, aesthetic, and economic values of wetlands were largely unappreciated. Wetlands within the United States have been and are continuing to disappear at a rapid rate. Efforts are being made to conserve remaining wetlands and many regulatory policies have been adopted in support of this goal. To regulate the loss, preservation, and/or restoration of wetlands and to judge the effectiveness of these regulatory efforts in preserving associated ecosystem services, wetlands must be routinely monitored. Wetland mapping is an essential part of this monitoring program and much effort has been made by the US state and federal governments, as well as other organizations, to provide quality map products. Wetland maps can serve a variety of purposes including regulation, natural resource management, and input for models that quantify water quality and quantity as well as the provision of wetland ecosystem services at the watershed scale. Wetland hydrology is the most important abiotic factor controlling ecosystem function and extent, and it should therefore be a vital part of any wetland mapping or monitoring program. New approaches are needed to not only map wetlands, but also to monitor wetland hydrology as it varies in response to weather, vegetation phenology, surrounding landuse change, and other anthropogenic forces including climate change. Recently developed remote sensing technologies and techniques have the potential to improve the detail and reliability of wetland maps and the ability to monitor important parameters such as hydrology. Various types of remotely sensed data (e.g., aerial photographs, multispectral, hyperspectral, passive microwave, radar, and lidar) have different capabilities with specific advantages and disadvantages for wetland mapping at the regional scale. Although aerial photographs were traditionally used to map wetlands and infer hydrology, fine-resolution optical images are now

Megan W. Lang and Greg W. McCarty 74

available more frequently as commercial agencies increase satellite coverage (e.g., Quickbird and IKONOS). However, optical data, such as aerial photographs and multispectral satellite images have limitations, including their inability to detect hydrology below dense vegetative canopies and their limited ability to detect variations in hydrology (i.e., inundation and soil moisture). The restrictions of optical data are increasingly being compensated for with the use of new technologies, including synthetic aperture radar, lidar, and geospatial modeling. The availability of these new data sources is increasing rapidly. For example, many states in the US are now collecting synoptic state-wide coverages of lidar data. The sources, strengths, and limitations of different types of remotely sensed data are reviewed in this paper, as well as the importance of temporal and spatial resolution necessary for regional scale wetland mapping efforts. The potential of multi-temporal, multi-sensor approaches that capitalize on geospatial modeling are emphasized for meeting current wetland mapping challenges.

1. INTRODUCTION Until recently, the biologic, aesthetic, and economic values of wetlands were largely

unappreciated. United States federal legislation, such as the Swamp Lands Act of 1850 and farming innovations, encouraged the draining of wetlands. In the 1930’s and 1940’s, the US government provided free engineering services to farmers who wanted to drain their lands (Dahl and Johnson 1991). Much ecosystem damage has been done and the conterminous United States has lost over half of its wetlands (Dahl and Johnson 1991). Some states have lost more wetlands than others. For example, California has lost over 91% of its wetlands whereas Maryland has lost over 64% (Dahl and Johnson 1991; Tiner 2005). In the last half of the 20th century, public recognition of the value of wetlands along with concern over accelerating rates of wetland loss led to a change in society’s perception of wetlands. Interest in wetland conservation has increased and new governmental policies (e.g., the Highly Erodible Land Conservation and Wetland Conservation Compliance (Swampbuster) provisions that were introduced in the 1985 US Farm Bill) have removed most of the wetland drainage incentives for agricultural lands. Other US legislative acts, such as the Clean Water Act (which grew out of the Federal Water Pollution Control Act of 1948 as amended in 1972 and 1977), are now used to protect wetlands. Wetland protection is further supported by the US federal “no net wetland loss” policy (Federal Geographic Data Committee 1994). Efforts are being made to conserve remaining wetlands and many policies have been adopted in support of this goal. Accurate and up to date wetland maps assist in the proper implementation of wetland regulations. For the United States to effectively manage its remaining wetlands, their abundance, distribution, boundaries, and quality must be better understood.

The way in which society interacts with wetlands is largely dependent on what value they assign to these natural resources. This value is subjective and is often measured by the function that the resource serves. Therefore, a better understanding of wetland ecosystem functioning within the landscape is necessary in order for society to place a value on wetlands (National Research Council 1995). As natural resource management becomes more holistic and moves towards ecosystem management, the synoptic view that remotely sensed data provides will become increasingly important (Finlayson and van der Valk 1995; Klemas 2001). Remote observation of wetlands is particularly necessary because they are often

Remote Sensing Data for Regional Wetland Mapping in the United States 75

difficult to access on the ground, and on-site mapping at the landscape scale is usually cost prohibitive (Harvey and Hill 2001; Rundquist et al. 2001; Baker et al. 2006; Silva et al. 2008). Remotely sensed imagery has the potential to add to our understanding of wetlands within the wider landscape setting and to better ensure their preservation via an increased understanding, more informed management practices, and a heightened appreciation of this unique resource. Remotely sensed data can be used to create not only maps of wetland presence, but also estimates of wetland condition and function (Rosenqvist et al. 2007).

Significant effort has been made by scientists and managers to provide quality map products and recently developed remote sensing technologies have the potential to further improve their detail and reliability. The diversity of remotely sensed data and the techniques available to process these data have increased rapidly since the 1970’s, when the US first began to systemically map national wetland resources. Still, the dynamic nature of these ecosystems (e.g., ephemeral hydrology), their diversity (e.g., variations in plant structure and phenology), and the often small proportion of the landscape that they occupy challenge existing sensors. Recent advances in the quality and availability of radar, lidar, and other types of data, as well as the introduction of new processing methods hold great potential for the furthering of regional wetland mapping and monitoring efforts. These different geospatial datasets provide complementary information about wetland presence and function. This paper reviews the advantages and disadvantages of applying different types of remotely sensed data, including aerial photographs, multi-spectral, hyperspectral, passive microwave, radar, and lidar data, to the mapping and monitoring of wetlands.

2. CURRENT US REGIONAL WETLAND MAPPING PROGRAMS Multiple state and federal agencies have produced regional wetland maps over the last

thirty-five years and these maps have developed rapidly. One of the most common wetland mapping methods uses both remotely sensed and field data. Natural resource managers have been using a combination of remotely sensed imagery and field data since the 1970’s and the techniques and the quality of data used have been rapidly evolving. In the past, most maps were produced by government agencies, such as the United States Fish and Wildlife Service (FWS) and National Oceanic and Atmospheric Administration (NOAA), but today private organizations are also contributing to the effort. Two commonly relied upon US federal wetland mapping programs are discussed below.

2.1. US Fish and Wildlife Service National Wetland Inventory One of the earliest and most commonly relied upon mapping efforts was the US FWS

National Wetland Inventory (NWI). This Congressionally mandated wetland mapping and monitoring effort was initiated in 1974. NWI maps are primarily produced using mid- to high altitude aerial photographs, photointerpretation techniques, field verification, and some collateral data sources (Federal Geographic Data Committee 1994). The oldest maps are based on 1:80,000 black and white aerial photographs and more recent maps use 1:40,000 scale color infrared aerial photographs. Regardless of the type of aerial photograph used,


maps are based on aerial photographs collected at one point in time, usually during the early spring. The NWI maps are created to provide natural resource managers with the information necessary to make well-informed decisions regarding the future of wetlands (Tiner 1999). Maps can be downloaded online at the NWI website (www.nwi.fws.gov).

A new classification scheme was developed so that NWI could provide a nationally consistent product (Cowardin and Golet 1995). This new classification, often termed the Cowardin et al. classification, is a hierarchical system based on ecosystem properties. The “system” is the broadest level of the classification and it includes the estuarine, lacustrine, marine, palustrine, and riverine wetland groups. These systems are divided further based on vegetation, hydrologic regime, salinity, ecology, and the shape and location of wetlands (Cowardin et al. 1979).

Although great care has been taken in the production of these maps and they are relied upon by numerous scientists and managers (Kudray and Gale 2000; Pantaleoni 2007), challenges to the cartographic process and product delivery remain. Mapping is most accurate when there is a rapid change between vegetation, hydrology, or soil at the wetland boundary (National Research Council 1995). Other wetland types, such as forested, especially evergreen forested wetlands, and temporarily flooded wetlands, are mapped more conservatively (Tiner 1997b; Tiner 2003; Kudray and Gale 2000). NWI maps usually err less by commission and more by omission. Therefore, if a wetland is indicated on an NWI map there is a high probability that one exists or that it did at the time that the aerial photograph was acquired (Nichols 1994; Stolt and Baker 1995; Tiner 1997b). Estimates of the extent of NWI’s forested wetland omission errors vary widely but omission errors can be substantial (Swartwout et al. 1981; Stolt and Baker 1995; Rolband 1995; Kudray and Gale 2000; Wright and Gallant 2007). Due to improvements in technology and expertise as well as possible land cover changes, the newer NWI maps are likely to be more reliable. However, even the newer maps are based on aerial photographs collected at one point in time and the reliability of maps can be affected by weather conditions (e.g., drought or flood) prior to and during aerial photograph collection. In addition to cartographic challenges, it has also been a challenge for NWI to deliver wetland maps, particularly digital maps, and status reports in a timely manner and for those maps and reports to be updated fast enough to keep up with landscape changes (Pantaleoni 2007; Wright and Gallant 2007). According to the NWI website (www.nwi.fws.gov), significant portions of the US have not even been mapped once and large areas are only available in hardcopy, not digital. It is uncertain whether areas of the US currently without NWI maps will be mapped in the future (Wright and Gallant 2007). Meanwhile many NWI maps are out of date or are rapidly becoming so.

The US FWS has recently begun to estimate wetland function, primarily based on landscape position, landform, water body types, and hydrologic flow paths as illustrated in NWI maps, topographic maps, and aerial photographs. Estimated functions include: 1) nutrient and sediment retention, 2) preservation of biodiversity, 3) provision of habitat, 4) surface water retention, 5) maintenance of stream flow, 6) shoreline stabilization, and 7) coastal storm-surge protection. This additional information could help to better characterize ecosystem service studies such as nutrient cycling, water storage, contaminant removal, carbon storage and export, and more (Tiner 1997c; Tiner 2003; Tiner 2005).


2.2. NOAA Coastal Change Analysis Program The NOAA Coastal Change Analysis Program (C-CAP) is a nationally standardized

database of land cover and habitat change produced for the coastal areas of the United States (NOAA 2008). This regularly updated database was established in the 1990’s to assist scientists, mangers and regulators in better understanding the coastal environment, the interactions between major land cover types (e.g., wetlands and uplands), and the impact of those interactions on marine organisms (Dobson et al. 1995). In doing so, C-CAP promotes an ecosystem-based approach to natural resource management. Unlike NWI, C-CAP maps are created in a digital environment using image processing software. It primarily relies on digital satellite imagery, field data, and other ancillary data in a geographic information system (GIS) format (Dobson et al 1995). C-CAP maps have been produced for multiple dates, at 30 m spatial resolution. Current efforts are focused on adding additional dates and production of finer resolution products in specific regions (Nate Herald 2008 personal communication). Data summaries, reports, and digital maps are available online (www.csc.noaa.gov /landcover).

C-CAP is produced in coordination with several other federal agencies as part of the National Land Cover Database (NLCD), facilitated by the Multi-Resolution Land Characteristics Consortium, and serves as the coastal portion of NLCD. Map production methods are nearly identical to those used to produce NLCD data (www.mrlc.gov) for the interior portion of the US, with the largest difference being that C-CAP includes additional wetland classes which are not mapped in the interior of the US. Similar to NLCD, C-CAP primarily relies on Landsat Thematic Mapper (TM) and Landsat Enhanced Thematic Mapper Plus (ETM+) data, although additional satellite imagery is used in some regions. These additional imagery sources include Landsat Multispectral Scanner (MSS), Systeme Pour l’Observation de la Terre (SPOT) High Resolution Visible (HRV) data, IKONOS, and Quickbird (Dobson et al. 1995; NOAA 2008). The finer resolution imagery, such as IKONOS and Quickbird, are used to gather more detailed information for training or in areas of enhanced interest (e.g., areas of rapid change; Nate Herold 2008 personal communication). In addition to the standard NLCD land cover classes, C-CAP also maps palustrine forested, palustrine scrub/shrub, palustrine emergent, estuarine forested, estuarine scrub/shrub, and estuarine emergent wetlands, as well as palustrine aquatic beds, and estuarine aquatic beds. In this way, C-CAP contains more detailed wetland classes than NLCD created for the interior of the country but it is less detailed than NWI, providing less information about hydrologic regime and other parameters. C-CAP products from different years are compared to determine the amount, rate and types of changes to land cover which have occurred between map dates. By doing so, C-CAP can be used to assess cumulative impacts of land cover change on ecosystem health (e.g., water quality and habitat quality).

Although its coarser spatial resolution makes C-CAP more suitable for regional rather than finer scale applications, it has been found to be relatively accurate. In addition its standardized, digital production aids in timely distribution of consistent, up to date maps. In 1991, field data were used to investigate the reliability of NWI and C-CAP maps. Random locations were examined and NWI was found to be 88-100% accurate while C-CAP was found to be 63-97% accurate (Burgess 1995). Henderson et al. (1998) found that C-CAP data were useful for parameterizing non-point source pollution water quality models. The integration of wetland and upland land cover categories makes C-CAP particularly well


suited for parameterizing water quality models since many models have been designed to import land cover data but few ingest wetland specific maps. Although the models may not distinguish between all wetland categories present in C-CAP, the fact that they are available, readily ingested and easily collapsible into more generalized NLCD categories makes their use more likely in the future. Both C-CAP and NWI are exploring different methods for rapidly updating their maps. One of these methods is cross-correlation analysis (CCA), a technique developed to detect changes in land cover using multispectral satellite data (Koeln and Bissonnette 1999). Houhoulis and Michener (2000) have developed a similar process that can also be used to update wetland maps although this particular method is not currently used by C-CAP. An important advantage to these methods is that coarser resolution digital data can be used to rapidly screen for changes in areas which were originally mapped with more costly, finer resolution imagery.

2.3. Other Wetland Mapping Efforts NWI and C-CAP are two examples of regional wetland maps being used today. Other

federal efforts to map or monitor wetlands include the US Environmental Protection Agency’s Environmental Monitoring and Assessment Program and the United States Department of Agriculture (USDA) Natural Resources Conservation Service (NRCS) National Resource Inventory. NRCS also produces the Soil Survey Geographic Database (SSURGO). Although SSURGO is not a map of current wetlands, it does contain information on hydric or wetland soils which can be used to infer either current or historic wetland presence. If a hydric soil is present, this usually indicates that a wetland either is or was once present at the location. In addition, some states are also producing their own wetland maps. For example, the Maryland Department of Natural Resources produces detailed wetland maps using a methodology similar to the US FWS NWI program (Burgess 1995). The Wisconsin Department of Natural Resources and the New York State Department of Environmental Conservation also produces wetland inventory maps based on photointerpretation of aerial photographs (Johnston and Meysembourg 2002; New York State Department of Environmental Conservation 2008).

As an increasing number of federal, state, and local agencies within the US have produced their own wetland maps, it has become increasingly important to develop a wetland mapping standard so that these maps can be compared across jurisdictional lines and dates. Since 1993, the Wetlands Subcommittee of the Federal Geographic Data Committee has been working to coordinate and integrate the collection of wetlands data by various US government agencies. This integration is divided into four increments: 1) standardize terminology, definitions, and classification schemes used by government agencies to gather information concerning wetlands; 2) coordinate the collection of wetlands data by government agencies; 3) evaluate the consistency of wetlands data and statistical results collected by government agencies; and 4) analyze the public policy implications and feasibility of the coordination of wetlands information. This program has made improvements on the coordination of government agencies involved in wetlands data collection (Federal Geographic Data Committee 1996; Shapiro 1995; Federal Geographic Data Committee 1994) and a draft wetland mapping standard was available online (www.fgdc.gov) at the time this chapter was written. The purpose of this standard is to enhance current and future, but not


necessarily historic, wetland mapping efforts. The standard specifies minimum data quality standards for inclusion in the US National Spatial Data Infrastructure when these activities are conducted or funded by the US federal government (Federal Geographic Committee 2007).

3. WETLAND HYDROPERIOD: A KEY WETLAND MAPPING PARAMETER

By basic definition, wetlands are simply wet-lands (Reed 1993). For that reason, one of

the main parameters that should be considered when mapping wetlands is wetland hydrology. Fluctuations in the areal extent, duration, and frequency of wetland flooding and soil saturation are called wetland hydropattern or hydroperiod (National Research Council 1995; Mitsch and Gosselink 2007). Wetland hydroperiod results from all transfers of water into and out of a wetland (National Research Council 1995; Mitsch and Gosselink 2007). Although numerous factors interact to influence wetlands, hydroperiod is the single most important force controlling not only the formation, and therefore the location, but also the functioning of a wetland (Reed 1993; Nestler and Long 1997; Mitsch and Gosselink 2000; Baghdadi et al. 2001; Töyrä et al. 2002; Bartsch et al. 2007;

Beeri and Phillips 2007; Martinez and Le Toan 2007). Since wetlands are often situated at a hydrologic edge (Chopra et al. 2001; Chiu and Couloigner 2006), small changes in hydrologic regime can cause substantial changes in ecosystem characteristics and function (Mitsch and Gosselink 2000). Hydrologic conditions control important abiotic factors, which in turn influence wetland biogeochemistry, soils, and vegetation. Wetland chemistry, soils, and vegetation are therefore unique and they interact to serve a variety of functions valued by society, such as pollution reduction and flood control (Whitehead and Thompson 1993; Richardson 1994; Hruby et al. 1995; Whigham 1996; Poor 1999; Woodward and Wui 2001; Baker et al. 2006). Approximately half of all animal and plant species included on the US threatened or endangered species list depend on wetlands during a portion of their lives (US Fish and Wildlife Service 2002). Information on hydroperiod can be used to infer the types of functions that may be served by a wetland (Nestler and Long 1997; Cole and Brooks 2000; Mitsch and Gosselink 2007). Wetland soils (hydric soils) and vegetation (hydrophilic vegetation) can often be distinguished from upland soils and vegetation on remotely sensed images. Additionally, remotely sensed data can discern subtle topographic features that greatly influence the function of wetlands. Certain conditions (e.g., topographic depressions) can encourage the concentration of water and therefore the origination of wetland hydrology. Moreover, coarser landforms that favor the formation of wetlands, such as the margins of stream, can also provide important clues to the location of wetlands. For the above stated reasons, remote sensors (i.e., people who analyze remotely sensed images) monitor wetland hydrology, hydric soils, hydrophytic vegetation, and/or conditions which encourage the concentration of water to map wetlands.


4. DIFFICULTIES INHERENT IN WETLAND MAPPING Regardless of the methodology used to map wetlands, there are numerous difficulties

inherent in the mapping process. These limitations may be related to a variety of factors including the character of the wetland being mapped, the map scale, the natural conditions present at the time of data collection, the spectral and spatial resolution of the imagery used, the software and hardware used to process the data, and the skill of the image processor (Tiner 1999). It should be noted that the specifications necessary to produce a “good” wetland map depend heavily on what that map is being used for. For example, a relatively fine scale wetland map like NWI may not be necessary when assessing wetland abundance at a global scale but it would be helpful when attempting to determine exactly where to locate a wetland mitigation site. Similarly, it may be less important to use a wetland map that accurately delineates forested wetlands if a researcher or natural resource manager is trying to locate potential habitat for an endangered bird which only lives in marshes, wetlands with non-woody vegetation.

4.1. Temporal Considerations The timing of image collection is vital to the accurate mapping of wetlands (Klemas

2001). Ideal times can vary by hour, day, month, or year and by wetland type. Generally, the more water present in an area, either in the form of soil saturation or flooding, the more unique an area will appear as compared to surrounding uplands, and the easier it will be to identify the area using remotely sensed data. However, hydroperiod is a highly dynamic parameter in many wetlands and these wetlands must be imaged at specific times to capture water levels at their peak. For example, moisture levels can vary due to tides, rainfall events, the melting of snow and ice, the release of upstream dams, or changing levels of evapotranspiration (a combination of evaporation and transpiration primarily controlled by temperature and plant physiology). It is also important to consider the fact that precipitation and wetness conditions change from year to year (e.g., drought or flood years) as well as between seasons. For that reason, images captured during years of average weather conditions will best represent the location of wetlands. However, in some circumstances the collection of imagery during weather extremes may aid the mapping of wetlands. For example, Munyati (2000) found that some wetlands in Africa were readily apparent during droughts when wetland vegetation was the only green vegetation present in the landscape.

Ideal timing of imagery collection for wetland map creation depends on the type of wetland being mapped. Deciduous forested wetlands are best mapped during the spring before leaf-out. At this time, the tree canopy interferes less with the detection of the ground and water tables tend to be high due to snowmelt and/or an extended period of low evapotranspiration rates. Because evergreen vegetation retains its leaves year round, evergreen forested wetlands do not have an optimal time for imagery collection based on vegetative state. Submerged aquatic vegetation (SAV), on the other hand, is best mapped during maximum biomass, low turbidity, low wave/wind action, and during low tide. When collecting remotely sensed data over tidal areas, it is difficult to obtain images that are recorded at the same tidal level. This is not only due to the difficulty of obtaining adjacent


scenes at the same tidal level. It is also due to tidal flooding that proceeds gradually upstream so that wetlands within a large estuarine system are in different stages of flooding at one point in time (Tiner 1999).

4.2. Spatial Considerations The spatial resolution or scale of imagery affects what can and cannot be interpreted in a

wetland map. All maps have a minimum mapping unit (MMU) and a target mapping unit (TMU). The MMU represents the smallest wetland that appears on the map, while the TMU estimates the smallest wetland that is consistently mapped. The TMU will differ not only based on the spatial resolution of the imagery used to create the map but also based on the type of wetland being mapped (Tiner 1999). For example, easily identified wetlands, such as coastal marshes and flooded basins, will have a smaller TMU than wetlands that are more difficult to identify, like drier or evergreen wetlands. To map wetlands at a certain MMU, requires an appropriate scale of imagery. For example, when mapping wetlands in Maryland using color infrared aerial photography Tiner and Smith (1992) recommended imagery with a spatial scale of 1:58,000 to produce maps with a MMU of 0.4 ha and 1:24,000 imagery to produce maps with a MMU of 0.1 to 0.2 ha. The use of finer resolution imagery, however, requires a greater mapping effort. A spatial resolution of 1:24,000 or better is recommended for local wetland mapping when boundaries need to be precise and smaller wetlands need to be identified. Whereas, broader scale imagery, such as 1:58,000, are more appropriate for national mapping efforts (Tiner 1999). Generally, as the resolution of imagery used to map wetlands becomes finer, wetland boundaries are more distinct and refined, smaller wetlands are mapped, stereoscopic details become obvious, and difficult to identify wetlands become more easily identified. However, finer scale imagery only increases the quality of wetland maps to a point. For example, wetlands are easier to delineate at 1:12,000 than 1:5,760. This is due to extremely fine details, like tree branches, that make it difficult to delineate wetlands at this scale (Tiner and Smith 1992). It could be said at this point that the “forest (or forested wetland) cannot be seen through the trees” at this finer scale.

4.3. Wetland Type Considerations The most difficult types of wetlands to map are palustrine forested wetlands (Jacobson et

al. 1987; Tiner 1990; Sader et al. 1995; Kudray and Gale 2000; Klemas 2001; Wright and Gallant 2007), evergreen wetlands, and wetlands with ephemeral hydrology (Augusteijn and Warrender 1998; Kudray and Gale 2000). Palustrine forested wetlands are difficult to map because their canopies often prevent viewing of the ground’s surface and the trees found in this type of wetland are identical or spectrally similar to those found in upland forests (Sader et al. 1995; Augusteijn and Warrender 1998; Klemas 2001). Moreover, relatively small variations in topography are capable of forming these types of wetlands. As a result, all forested wetlands, but especially small, isolated forested wetlands (e.g, vernal pools) are one of the most difficult types of wetlands to detect due to a combination of the presence of the forest canopy, often ephemerally hydrology, and in the case of vernal pools their size. Although they are difficult to detect using remotely sensed images, the mapping of palustrine


forested wetlands is vital because they represent approximately half of all wetlands found in the US (Sader et al. 1995) and they are the most likely type of wetland to be lost in the future (US Fish and Wildlife Service 2002). New techniques need to be developed to remotely sense forested wetlands. Although forested wetlands are often more readily identified with finer resolution imagery, this is only true to a degree. Wilen and Smith (1996) argue that the challenge to identifying forested wetlands lies not in spatial resolution but in spectral resolution. Promising technologies for meeting the forested wetland mapping challenge include radar and lidar. In addition, better quantification of the influence of hydrology on the spectral signature of forested wetlands using hyperspectral data would be helpful. Rundquist et al. (2001) called for the increased characterization of the composite vegetation-water spectral signature of emergent vegetation. For forested wetlands, the impact of hydrology on the spectral signature of the leaf layer after it has been exposed to flooding (leaves on the ground) may also be helpful if it is the leaf layer and not soils that are being remotely sensed. Since evergreen wetlands retain their leaves year-round, it is difficult to determine hydroperiod or soil type beneath them. Sometimes evergreen wetlands provide hints regarding their status. Evergreens growing in wetlands may be shorter than those growing elsewhere or they may show signs of chlorosis due to water stress (Tiner 1999). With deciduous or evergreen forested wetlands, it may be possible to view saturated soils or understory wetland characteristics through canopy openings. Wetlands exist along a continuum of soil moisture conditions, between permanently inundated, deepwater habitats and drier areas that do not sustain the soil moisture levels needed to support anaerobic soils. Many of the drier wetlands are difficult to identify on the ground yet alone with remotely sensed data. The more difficult it is to map a wetland on the ground, the more conservatively it will be mapped with remotely sensed data (Tiner 1999).

5. THE POTENTIAL OF DIFFERENT SENSORS FOR WETLAND MAPPING

Aerial photographs have traditionally been used to map wetlands and some current

wetland mapping programs use multispectral images (Baghdadi et al. 2001), but recent research suggests that other types of remotely sensed imagery have promise for wetland delineation. The National Research Council (1995) concluded that certain types of imagery may provide enhanced wetland mapping capabilities over traditional photointerpretation methods. This section investigates the advantages and disadvantages of using aerial photography and multispectral, hyperspectral, passive microwave, radar, and lidar imagery for wetland mapping. This is not an exhaustive review of all types of remotely sensed imagery. Instead it represents the types of imagery that are most common and that researchers have suggested have the greatest potential to improve wetlands mapping.

This section discusses both spaceborne and airborne imaging systems. It is important to not only differentiate between different types of imagery but also between the different platforms used to collect those images since there are advantages and disadvantages inherent to these vantage points. The advantages of using satellite imagery for wetland mapping include coverage, timeliness, lower costs, and the ease with which it can be integrated with other types of data in a GIS (Dobson et al. 1995; Federal Geographic Data Committee 1992;


Papa et al. 2006). But the use of satellite data also implies certain limitations. These limitations include the often greater interference of weather with satellite data collection. In the past, the use of satellite data has meant broader spatial resolutions but this in no longer the case with new generation sensors (e.g., IKONOS and Quickbird) that collect finer spatial resolution data.

5.1. Aerial Photography The combination of aerial photographs and manual photointerpretation has been and

continues to be the most commonly relied upon wetland mapping method (Lyon and McCarthy 1995; National Research Council 1995; Phinn et al. 1999; Harvey and Hill 2001). This is primarily due to the use of well-established methods and protocols, the availability of current and historic data over broad areas (Baker et al. 2006), the comparatively fine spatial resolution of aerial photographs (Klemas 2001), and of course the relatively reliable results obtained from such efforts (Rutchey and Vilchek 1999; Harvey and Hill 2001; Wright and Gallant 2007). In addition, the aerial platform provides flexibility with the timing of data collection that the satellite platform lacks. For example, airplanes can collect images at certain times of the day that correspond with optimal tidal levels whereas the overpass time of satellites is usually predetermined. Historic aerial photographs are available for much if not all of the US providing a valuable baseline for change detection analysis (Baker et al. 2006). One of the most common sources of aerial photography used for wetland mapping in the US is the US National Aerial Photography Program (NAPP) and the National High Altitude Aerial Photography (NHAP) program, both directed by the US Geological Survey (USGS).

One of the major differences between aerial photography and satellite data lies not in the character of the data but with the methods that have traditionally been used to process those data. Aerial photography is often classified by a photointerpreter without computer automation of the mapping process. Photointerpretation relies on the human eye to make qualitative decisions based on the image’s tone, color, spatial patterns, texture, height/topography, associations, and other characteristics. The accuracy of these decisions can benefit from the experience and judgment of the photointerpreter. The interpretation of aerial photography also benefits from the use of stereoscopic viewing. Stereoscopic viewing helps the photointerpreter determine relative variations in elevation and vegetation height. Topographic variations provide clues as to where wetlands would likely form and information on vegetation height aids in the determination of wetland type (e.g., forested versus shrub-scrub). Changes in plant height may also indicate soil moisture and other abiotic conditions (National Research Council 1995).

Studies have compared the use of aerial photographs and aerial photointerpretation to the use of multispectral data for wetland mapping in different environments (Rutchey and Vilchek 1999; Harvey and Hill 2001). One of the major disadvantages of aerial photographs as compared to multispectral data is the relatively lower spectral resolution of the aerial photographs but the superior spatial resolution of aerial photographs can compensate for this disadvantage. A study conducted in the Florida Everglades found that SPOT data overestimated the amount of cattail present in an impounded wetland due to confounding factors, such as, fire, hydrology, periphyton species composition, and macrophyte morphology. Aerial photography was better able to overcome these difficulties and map


cattail in these areas due to the experience and reasoning of the photointerpreter (Rutchey and Vilchek 1999). Harvey and Hill (2001) compared the ability of multispectral data (SPOT and Landsat TM) to that of aerial photographs and manual interpretation techniques for mapping an Australian freshwater swamp. Although fourteen land cover types could be delineated from 1:15,000 aerial photographs with 89% accuracy, only three broad land cover types could be distinguished using the multispectral data while meeting minimum accuracy requirements (≤80%). The difference in accuracy between the aerial photographs and multispectral data was, in part, attributed to the contextual and textural information used by the photointerpreter (Harvey and Hill 2001).

Although the manual interpretation of aerial photography is the most commonly used method for detailed wetland mapping, this method is subjective (Finlayson and van der Valk 1995; Augusteijn and Warrender 1998; Baker et al. 2006), time consuming and therefore cannot be quickly updated (Phinn et al. 1999; Harvey and Hill 2001; Wright and Gallant 2007), is relatively expensive (Dobson et al. 1995; Mumby et al. 1999; Lunetta and Balogh 1999; Phinn et al. 1999; Harvey and Hill 2001; Klemas 2001; Hirano et al. 2003), and has difficulty mapping certain wetland types (see section 2.1). There are inherent limitations to the use of human vision to classify a map (Rutchey and Vilchek 1999) and these decisions are inherently subjective so resultant maps may vary according to the person who creates them (Finlayson and van der Valk 1995; Augusteijn and Warrender 1998; Baker et al. 2006). Mumby et al. (1999) estimate that the creation of maps for a relatively small area (~150 km2) may take six times as long using aerial photointerpretation rather than automated interpretation of digital airborne imagery and that the cost of manual interpretation of aerial photographs will increase at a faster rate than the cost of digital interpretation. The inability of the US FWS to use aerial photointerpretation methods to map wetlands for the entire US (starting in the 1970s) illustrates the cumbersome nature of aerial photointerpretation for broad scale mapping.

The type of aerial photograph used to create a wetland map has the potential to affect map accuracy. Although photointerpretation has traditionally been a manual process, which utilized analog photographs rather than digital images, analog aerial photographs can be converted to digital. Still the challenge of scanning in large numbers of photographs and normalizing photographs for variations in illumination, angle of acquisition, and sometimes even phenology and atmospheric conditions remains a challenge (Phinn et al. 1999). Digital cameras are an improvement over analog aerial photographs and can be processed using methods similar to those used for processing multispectral images (Phinn et al. 1999), although the lack of spectral information is still a limitation (Becker et al. 2007; Silva et al. 2008). If possible, it is usually advantageous to use color or color infrared aerial photographs to map wetlands, rather than black and white images. Color infrared film is usually preferred for wetlands mapping because it includes more spectral information than true color film and allows for greater contrast between different plant communities (Tiner 1999; Federal Geographic Data Committee 1992). Dale et al. (1996) found that color infrared aerial photography was a valuable tool in the identification of wetlands. A major exception to this statement occurs when mapping submerged aquatic vegetation (SAV). The relatively low spectral resolution of aerial photographs significantly limits their use for the detection of SAV in general (Silva et al. 2008). However, the use of true color film for SAV detection is preferable to the use of color infrared film due to the superior ability of shorter wavelengths


energy (the visible bands which compose true color imagery) to penetrate the water column (Tiner 1999).

5.2. Multispectral Multispectral sensors (Table 1) provide information on solar reflectance from materials

on the Earth’s surface in multiple portions of the electromagnetic spectrum (e.g., visible, near infrared, mid-infrared, and thermal), at relatively fine (1 – 5 m) to medium spatial resolutions (10 – 30 m) and some also provide thermal data. Sensors with coarser spatial resolutions, such as the AVHRR (Advanced Very High Resolution Radiometer) with approximate 1 km resolution, are usually not considered viable options for wetland mapping at the local to regional scale (Munyati 2000; Ringrose et al. 2003), so they are not discussed below. Relative to aerial photographs, satellite borne multispectral sensors provide enhanced spectral resolution (multiple bands) and are capable of imaging a larger area at once (Rundquist et al. 2001; Li and Chen 2005), which simplifies image interpretation. Studies have found that the increased spectral resolution of multispectral data enhances wetland mapping (Federal Geographic Data Committee 1992; Phinn et al. 1999; Harvey and Hill 2001; Töyrä et al. 2002) and, in some cases, helps compensate for reduced spatial resolution (Harvey and Hill 2001). In addition, if fine detail is not required these images are usually less expensive than aerial photographs (Mumby et al. 1999) and they can be collected regularly, as dictated by satellite orbits (Rundquist et al. 2001; Li and Chen 2005). Therefore, these sensors provide not only increased spectral but also increased temporal resolution over aerial photographs, both of which are important for distinguishing different wetland classes. The comparatively fine temporal resolution of these multispectral sensors is complemented by a relatively long historic record (e.g., Landsat MSS first launched in 1972). Finally, their digital format and standardized imagery allow for automated, repeatable classification of wetlands and other land cover types (Houhoulis and Mitchner 2000; Ausseil et al. 2007).

The infrared portion of the electromagnetic spectrum, which is often collected by multispectral sensors (Table 1), is considered to be particularly well suited for wetland mapping applications (Federal Geographic Data Committee 1992; Phinn et al. 1999; Munyati 2000). This is due to the high spectral response of vegetation in the near-infrared coupled with the strong absorption of water in the same region, which makes for a sharp contrast between water and vegetation in the near-infrared (Lyon and McCarthy 1995) and water absorption features (portions of the electromagnetic spectrum with distinctly lower reflectance when water is present) in the mid-infrared. Both multispectral and hyperspectral data can take advantage of this contrast with individual bands over these regions. The use of color-infrared (CIR) aerial photos can also improve wetland detection via information on the near infrared portion of the electromagnetic spectrum. However, the use of color infrared photographs is not as advantageous as the isolation of these regions through the use of distinct bands by way of digital multispectral images (Lyon and McCarthy 1995).

Although many of the older satellite systems, such as Landsat MSS, Landsat TM, and SPOT, offer improved spectral resolution, their spatial resolution is often insufficient for mapping smaller or more complex wetlands (Welch et al. 1999; Ramsey and Laine 1997; Ausseil et al. 2007; Wright and Gallant 2007).


Table 1. Commonly used satellite borne multispectral sensors Sensor Resolution (m) Bands Width Repeat Life-Span Information

Landsat MSS 80 (240 thermal) G, R, NIR, MIR 185 km 18 days 1972 - 1992 edc.usgs.govLandsat TM 30 (120 thermal) B, G, R, NIR, MIRx2, TH 185 km 16 days 1982 - Pres. edc.usgs.gov

Landsat ETM+ 30 (60 thermal; 15 pan) P, B, G, R, NIR, MIRx2, TH 185 km 16/18 days 1999 - Pres. edc.usgs.govIRS LISS-III 24 G, R, NIR, MIR 141 km 24 days 2003 - Pres. nrsa.gov.inSPOT 1,2,3 20 (10 pan) P, G, R, NIR 60 km var 1986 - Pres. spot.com

SPOT 4 20 (10 mono) M, G, R, NIR, MIR 60 km var 1998 - Pres. spot.comSPOT 5 10 (20 MIR; 2.5 or 5 pan) P, G, R, NIR, MIR 60 km var 2002 - Pres. spot.comIKONOS 4 (1 pan) P, B, G, R, NIR ~12 km var 1999 - Pres. geoeye.com

Quickbird 2.4 (61 cm) P, B, G, R, NIR 16.5 km var 2001 - Pres.digitalglobe.comOrbView-3 4 (1 pan) P, B, G, R, NIR 8 km var 2003 .geoeye.com

Note that many of the more recent commercial satellites are pointable and therefore the repeat frequency can vary. This list is not comprehensive. Multiple bands in one portion of the electromagnetic spectrum are denoted with an “x” and then the number of bands (e.g., x2). Sensor name (Sensor), spatial resolution in meters (Resolution), the portion of the electromagnetic spectrum sampled by each band (Bands), swath width (Width), temporal resolution or number of days between overpasses (Repeat), the years the sensor has operated (Life-Span; Present abbreviated as Pres.), and websites where more information about the sensors can be found are provided in the table. Phinn et al. 1999 suggests that current multispectral sensors (Landsat TM, SPOT XS, and

IRS-1C) are limited to mapping wetlands with a MMU greater than 9 ha, although other studies estimate that Landsat TM is capable of mapping wetlands at much finer MMU (0.8 – 1.0 ha; Federal Geographic Data Committee 1992; Lunetta and Balogh 1999; Wright and Gallant 2007). Landsat MSS data (80 m spatial resolution) have been used to map wetlands with varying levels of success (Moore and North 1974; Severs et al. 1974; Bennet 1987). Severs et al. (1974) were able to use Landsat MSS bands 5 (red) and 7 (near infrared) to classify wetland patches of ~4 ha or more into four broad categories (i.e., marsh, seasonally flooded depressions, meadow, and open water). However, in another case, researchers had difficulties using Landsat MSS to map large, homogenous coastal wetlands, which are among the easiest wetland types to map (Ramsey and Laine 1997). The use of Landsat TM for inventorying wetlands is also limited due to the lack of spatial detail (Ramsey and Laine 1997; Wilen and Tiner 1989). Newer satellite data, such as IKONOS, provide finer spatial resolution multispectral data (Table 1). For example, IKONOS collects 4 m resolution data over the blue, green, red, and near-infrared areas of the electromagnetic spectrum and Quickbird collects 2.4 m data in the blue, green, red, and near-infrared portions of the electromagnetic spectrum. It has been predicted that newly available finer resolution multispectral sensors (e.g., IKONOS, Quickbird, and Orbview) should provide an increase in wetland mapping capability (Phinn et al. 1999).

As with aerial photography, multispectral images can only be collected during the day and their collection is limited by cloud cover and atmospheric conditions (e.g., haze). These limitations may preclude the regular and repetitive collection of data that is required for certain wetland studies (Baghdadi et al. 2001; Costa 2004; Costa and Telmer 2007). This is especially true if combined with infrequent temporal coverage (e.g., satellite repeat frequency). Unfortunately, the optimal period for data collection may be short. For example, in many forested regions of the US there is a short time period between ice/snow melt and the leafing-out of trees, after which features below the forest canopy are usually obscured. More than one study has cited the presence of a vegetative canopy as one of the main deterrents to


accurate wetland mapping using multispectral data (Moore and North 1974; Carter 1982; Alsdorf et al. 2007; Costa and Telmer 2007). Unless significant gaps in the canopy are present, the hydrologic conditions on the ground cannot be observed. In the sub-tropics and tropics optimal time for image collection often corresponds with the periods of increased cloudiness (precipitation events). Since Landsat TM can only be acquired every 16 days, the chance of collecting a relatively cloud-free image during this optimal period is small (Federal Geographic Data Committee 1992). The ability of some sensors (e.g., SPOT and Quickbird) to collect data at multiple view angles has the potential to reduce the time between satellite overpasses and therefore increase the likelihood that images can be collected during times when conditions on the ground are optimal.

When using multispectral and even hyperspectral imagery, there is often a trade-off between spectral and spatial resolution (Harvey and Hill 2001; Becker et al. 2007), with finer spatial resolution sensors having fewer bands (e.g., IKONOS, Quickbird, and OrbView) and sensors with more bands having coarser spatial resolution (e.g., Moderate Resolution Imaging Spectroradiometer [MODIS]). Adequate spatial resolution is necessary since wetland patches are often small. However spectral resolution is also important because the spectral character of different land cover classes is often similar and adequate spectral resolution is necessary to distinguish different land cover types (Harvey and Hill 2001). Wetlands that are spectrally distinct from surrounding land cover types will be mapped more accurately with multispectral images than those which are not (McCarthy et al. 2005). Therefore no one existing multispectral sensor is optimal for wetland mapping (Becker et al. 2007) in all situations and the success of wetland mapping efforts often depends just as much on the type of wetland being mapped as the type of imagery being used. However, the literature does support some generalizations regarding the utility of available multispectral data.

Two of the most commonly used groups of multispectral sensors include SPOT (Jensen et al. 1991; Rutchey and Vilcheck 1999; Ringrose et al. 2003; Töyrä and Pietroniro 2005) and Landsat (Sader et al. 1995; Baghdadi et al. 2001; Townsend and Walsh 1998; Töyrä et al. 2002; Baker et al. 2006; Wright and Gallant 2007). Landsat data have been found to be preferable to other multispectral satellite data for wetland detection (Hewitt 1990; Bolstad and Lillesand 1992; Harvey and Hill 2001) and the recent announcement that it will be distributed free of charge after January of 2009 makes its use even more attractive. Although Harvey and Hill (2001) found that both Landsat TM and SPOT XS were adequate for mapping broad categories of wetland types, the coarser spatial resolution seven band Landsat TM data were preferable to the finer resolution four band SPOT data for mapping wetlands in a freshwater Australian swamp. Longer wavelength bands, especially the mid-infrared Landsat TM band (band 5; 1.55 – 1.75 μm), have been found to be especially useful for the detection of water (Federal Geographic Data Committee 1992; Phinn et al. 1999; Harvey and Hill 2001; Töyrä et al. 2002). The Federal Geographic Data Committee (1992) identified Landsat TM bands 4 and 5 as most effective for wetland delineation. Landsat TM bands 2, 3, and 4 have been found to be helpful for identifying the presence of understory vegetation (Congalton et al. 1993), which is helpful for mapping some forested wetlands (Harvey and Hill 2001). Sader et al. (1995) found that Landsat TM could be used to distinguish between forested wetland, other wetlands, forested uplands and other uplands with an accuracy of approximately 80%. Temporal resolution was found to be an important variable when using Landsat TM to map a diverse group of wetlands on the border between Maryland and Delaware. When the accuracy of a wetlands map produced with single date leaf-on (June) TM


imagery was compared to one that was produced using multi-temporal TM data (leaf-on and leaf-off [April]), the multi-temporal map was found to be more reliable (Lunetta and Balogh 1999). The leaf-on data were used primarily to produce a land cover map and the leaf-off data were used to detect wetland hydrology. The accuracy of the multi-temporal wetlands map was superior to that of the single date map, 88% accuracy versus 69% accuracy respectively. Forested wetlands were found to be especially difficult to map without the use of multi-date imagery. On the other hand, open water areas are relatively easy to identify using Landsat data. Beeri and Phillips (2007) used Landsat TM and ETM+ to estimate hydroperiod in wetlands with open water with high accuracy (96% detection of water bodies greater than ~15 m). They suggest the use of SAR data to improve the detection of water beneath emergent vegetation and fine resolution elevation data to locate areas of likely inundation since the presence of vegetation (Moore and North 1974; Carter 1982) and sediment (Engman and Gurney 1991) reduce the effectiveness of multispectral data for the detection of inundation. SPOT data have also been useful for wetland mapping, especially when differentiating between medium to small wetland patches which are spectrally distinct from other land cover types. Töyrä et al. (2002) found that SPOT was capable of detecting flooded wetlands with between 66% and 80% accuracy in freshwater wetlands with a combination of forests and non-woody (i.e., herbaceous) vegetation. The Indian Remote Sensing Satellites (IRS) may be used more frequently for land cover mapping in the future as the availability of current high quality Landsat images becomes less certain due to the instrument malfunction which occurred in the Landsat 7 (ETM+) sensor in July of 2003 and the uncertain future of Landsat 5 (TM) due to its advanced age. However, the lack of a mid-infrared band on the IRS-Linear Imaging Self Scanning Sensor (LISS)-II sensor was found to limit wetland detection (Johnston and Barson 1993; Mahlke 1996).

In complex wetland environments, more traditional statistical classifications commonly used with multispectral imagery (e.g., supervised and unsupervised classification) may have limited utility (McCarthy et al. 2005). In these cases and when greater accuracy is required in general, the use of different classification techniques can improve the accuracy of resultant maps. For example, the use of indexes, such as the normalized difference vegetation index (cite) or transformations, such as the tassel-cap transformation may improve the discrimination of different wetland types (Sader et al. 1995; McCarthy et al. 2005; Wright and Gallant 2007). Huguenin et al. (1997) found that subpixel spectral analysis was significantly better than traditional classification methods (ISODATA, maximum likelihood, and minimum distance) for mapping bald cypress trees and tupelo gum trees in Georgia and South Carolina when using Landsat TM imagery. Subpixel spectral analysis identified cypress with 89% accuracy and tupelo gum trees with 91% accuracy when they occurred in pure stands or mixed with other species (Huguenin et al. 1997). Classification trees have proved helpful for wetland mapping (Wright and Gallant 2007), partially because they can be used to incorporate data from a variety of sources (Wright and Gallant 2007). When using Landsat ETM+ images to identify wetlands, a Stochastic Gradient Boosting decision tree technique was found to be more accurate (86.0%) than a Classification Tree Analysis (73.1%; Baker et al. 2006). Chiu and Couloigner (2006) improved classification accuracy by using a Fuzzy C-Means classifier to represent the boundary between different wetland types. This method represents the transition between different wetland types as it often is on the ground – gradual.


In addition to the improvement of wetland maps via the use of different classification methods, acquiring multispectral imagery at multiple view angles may enhance the discrimination of different land cover types, including wetlands (Vanderbilt et al. 2002; Dupigny-Giroux 2007). This enhanced discrimination is due to the varying appearance of wetland vegetation and water signatures when viewed from different angles and is sometimes called a directional signature (Dupigny-Giroux 2007). POLDER (Polarization and Directionality of Earth’s Reflectance) data were found to reliably discriminate three cover types (open water, emergent vegetation above inundation, and non-inundated cover types) when images were collected at angles which maximize specular reflectance of the water surface (i.e., maximize the interception of sunglint by the sensor; Vanderbilt et al. 2002). Pinty et al. (2002) found that multi-directional signatures could be used to distinguish between inundation and dark soils, which often appear similar on multispectral images. Another study found that ratios of AirMISR data collected at different angles [fore and aft of nadir (26.1o, 45.6o, and 60.0o)] were capable of distinguishing not only different wetland types in a forested area of Maine but also moisture gradients in emergent wetlands, species type and vigor, the relative proportion of water and vegetation, and areas of vegetation undergoing moisture stress (Dupigny-Giroux 2007). Dupigny-Giroux (2007) postulates that a multi-angle approach may help map evergreen wetlands. Another study found that a combination data collected to maximize sun glint and spectral mixture analysis could be used to detect areas of wetland inundation, with accuracy of within pixel estimates of inundation increasing with increasing pixel size (Vanderbilt et al. 2007).

5.3. Hyperspectral Hyperspectral data are characterized by numerous, narrow spectral bands collected in the

visible, near-infrared, mid-infrared, and sometimes thermal portions of the electromagnetic spectrum. For example, the US National Aeronautics and Space Administration’s (NASA) satellite borne Hyperion sensor provides imagery with 220 spectral bands at a spatial resolution of 30 m. NASA’s Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) has been flown on a variety of different aircraft and has 224 narrow, contiguous spectral bands from 400 to 2500 nm (10 nm wide bands) and a pixel size of 20 m (Vane et al., 1993). These numerous, finely segregated spectral allow analysts to identify different materials based on their “spectral signature” or diagnostic patterns in absorption and reflection usually associated with the molecular and/or cellular properties of the material (Kokaly et al. 2003; Schmidt and Skidmore 2003). Although the use of spectral signatures can be very helpful in imagery analysis, these signatures can vary temporally with phenology and environmental conditions (Judd et al. 2007; Silva et al. 2008) making generalizations difficult, and therefore mapping through time and space challenging (Schmidt and Skidmore 2003). In addition, the importance of texture and context should not be discounted (Hirano et al. 2003) even when spectral resolution is extremely high. Similar to other types of imagery, hyperspectral data are capable of detecting parameters that can be used to infer wetland function. However, in addition to more obvious parameters (e.g., habitat fragmentation) hyperspectral data have the potential to detect biochemical properties such as nutrient and chlorophyll content (Schmidt and Skidmore 2003; Judd et al. 2007). Although the detection of these properties is currently


unlikely to be part of a local to regional wetland mapping program, this may change in the future.

The drawbacks of hyperspectral data include those of all optical data, such as an inability to penetrate vegetation cover, sensitivity to clouds and other weather events, and restricted use to daylight hours. Additional restrictions of hyperspectral data include large data volume (Phinn et al. 1999; Hirano 2003; Laba et al. 2005; Becker et al. 2007), less developed/more complex image processing techniques (Phinn et al. 1999; Klemas 2001; Hirano et al. 2003; Laba et al. 2005), and the relatively poor availability of hyperspectral data in general and high price of commercial data. All of these drawbacks should be lessened as technologies mature. At present, fine to moderate resolution hyperspectral data are primarily available from airborne sensors, although an experimental hyperspectral satellite (Hyperion) collects a limited amount of medium resolution (30 m) data. As of 2008, Hyperion has greatly exceeded its life expectancy and shows signs of decline. MODIS, another satellite borne hyperspectral sensor with 36 bands, is less promising for local to regional wetland mapping due to its relatively coarse spatial resolution (250 to 1000 m). Although MODIS’s broad spatial resolution is not ideal for most regional wetlands mapping, its launch does provide hope for the availability of finer resolution hyperspectral satellite data in the future. Even the spatial resolution of some commonly available finer resolution datasets (e.g., Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) and Hyperion) may be too coarse for some wetland mapping applications (Hirano et al. 2003; Becker et al. 2007).

Until hyperspectral data become more available, their use in operational wetland mapping programs is unlikely (Phinn et al. 1999; Govender et al. 2007). As more hyperspectral sensors begin to collect data, data will become more available and processing techniques will advance. With time the ever-increasing capabilities of new hardware will be able to better handle hyperspectral’s large datasets. When these problems are solved, hyperspectral data could be used to supplement other types of remotely sensed data in regional wetlands mapping, especially if more specific information on plant species presence or other factors is necessary. For now, it is more likely that hyperspectral images will continue to be used to identify narrowly defined wetland classes and plant species of interest (Sahagian and Melack 1996; Klemas 2001; Schmidt and Skidmore 2003; McCarthy et al. 2005), including invasive species (Laba et al. 2005).

Although the availability of satellite derived hyperspectral data and even publicly available US government sponsored airborne data (AVIRIS) is currently limited and fine spatial resolution data are not available, the availability of finer spatial resolution commercial data is increasing. There are currently several hyperspectral sensors being commercially produced and flown aboard aircraft including the CASI 1500, AISA Eagle, and AISA Hawk. The CASI 1500 is capable of collecting up to 288 bands between 380 and 1050 nm at a spatial resolution as high as .25 m (www.itres.com). The AISA Hawk and Eagle are often used together, since they collect information from different portions of the electromagnetic spectrum. The Eagle is capable of collecting up to 488 bands between the wavelengths of 400 and 970 nm, while the Hawk is capable of collecting 320 bands between the wavelengths of 970 and 2450 nm (www.specim.fi). The spatial resolution of the Hawk and Eagle sensors will vary based on platform and sensor parameters.

Hyperspectral sensors have primarily been used to map wetland plant species (Hirano et al. 2003; Kokaly et al. 2003; Schmidt and Skidmore 2003; McCarthy et al. 2005; Filippi and Jensen 2006; Judd et al. 2007), although they have also been used to monitor other wetland


parameters, such as the condition of wetland hydrology (Anderson and Perry 1996). A review of several research projects involving the use of hyperspectral data for mapping aquatic vegetation found that accuracies ranged from 70 to 96% (Silva et al. 2008). Schmidt and Skidmore (2003) found that the spectral signatures of different salt marsh vegetation types were significantly different and they propose that these differences should allow identification of individual species using hyperspecrtal imagery. Other studies have found that the mapping of various marsh vegetation species is possible (Judd et al. 2007). Hirano et al. (2003) found higher accuracies when mapping spike rush (100%) as compared to red mangroves (40%). Although hyperspectral data, like all optical data, are limited by the tree canopy during much of the year, hyperspectral data have been used to detect forested wetlands during the leaf-on period. This can be accomplished by detecting water stress in some tree species, because when some tree species are exposed to flooding they have elevated reflectance in the green (550 nm) and near-infrared (770 nm) portions of the electromagnetic spectrum. This approach has been shown to be effective in monospecific stands of facultative wetlands trees (Anderson and Perry 1996). Hyperspectral data are particularly well suited for the mapping of SAV, since the water column reduces reflectance from the vegetation itself making detailed examination of spectral characteristics necessary to identify SAV (Silva et al. 2008).

Similar to the analysis of multispectral data, spectral unmixing methods to can be used with hyperspectral data to estimate the fraction of endmembers included in each pixel (Roberts et al. 1993; Lillesand and Kiefer 1994; Rosso et al. 2005; Judd et al. 2007). This type of analysis has the potential to be more precise when using the numerous bands available with hyperspectral images, although the selection of truly pure (unmixed) endmembers can be a challenge (Filippi and Jensen 2006; Judd et al. 2007). By estimating the fraction of plant species per pixel, spectral unmixing methods can be used produce a more accurate wetland vegetation map (McCarthy et al. 2005). Rosso et al. (2005) applied spectral mixture analysis and multiple endmember spectral mixture analysis to AVIRIS data in order to map marsh vegetation. Both approaches were found to be suitable for mapping marshes, although the multiple endmember spectral mixture analysis had the advantage of incorporating more than one endmember per class. Other classification approaches, such as neural networks, have also proved beneficial. Filippi and Jensen (2006) found that neural networks performed better than end member based approaches. This neural network approach has the additional advantage of not requiring the initial input of spectral endmembers.

5.4. Passive Microwave Passive microwave radiometers operate in the same spectral range as active radar sensors,

such as synthetic aperture radars, but they sense microwave energy that is naturally emitted from objects. The emission of this energy is dependent on numerous factors including, surface roughness, temperature, soil and vegetation water content, bulk density, and soil texture. Passive sensors can be used to measure soil moisture, depth to shallow water table, and biomass, which are all important characteristics for locating wetlands (Shutko et al. 1997; Jackson et al. 1999). Indeed, sensitivity of passive microwave systems to soil moisture has been well established (Du et al. 2000; Shi et al.2006). Passive microwave sensors usually detect energy between 0.5 to 30 cm. Below 0.5 cm the sensor is sensitive to clouds and rain and above 30 cm radar, television, and other forms of radiation will interfere with microwave


reception. It is also between 0.5 and 30 cm that passive microwave sensors are most receptive to soil moisture. To best detect soil moisture, energy should be collected from 2 or 3 radiometers in different parts of the electromagnetic spectrum between 0.5 and 30 cm (Wilen and Smith, 1996).

The broad spatial resolution of most passive microwave sensors, such as the Scanning Multichannel Microwave Radiometer (SMMR), the Special Sensor Microwave/Imager (SSM/I), the Electronically Scanned Thinned Array Radiometer (ESTAR), and the Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), limits their use to large geographic regions (Martinez and Le Toan 2007). Even if synthetic aperture technology is used, spaceborne passive microwave sensors can still only achieve a spatial resolution of between 10 and 30 km, a scale that is insufficient for regional wetlands mapping (Jackson et al. 1999). This resolution can be partially remedied by the use of spectral mixture models, which can detect subpixel endmembers. Even with spectral unmixing, spaceborne microwave radiometers are still limited by their coarse spatial resolution (Smith 1997). Photo Science and Geoinformatic, two commercial companies, tested airborne passive microwave sensors for their sensitivity to wetlands and also found that spatial resolution was a problem. It was not felt that these sensors would provide improvements over present methods of regional wetland mapping (Wilen and Smith 1996).

5.5. Radar Imaging radars (radio detection and ranging) provide information that is fundamentally

different from sensors that operate in the visible and infrared portions of the electromagnetic spectrum. This is primarily due to the much longer wavelengths used by SAR sensors and the fact that they send out and receive their own energy (i.e., active sensors). One of the most common types of imaging radar is synthetic aperture radar (SAR). SAR technology provides the increased spatial resolution that is necessary in regional wetland mapping and SAR data has been used extensively for this purpose. For these reasons the following discussion will focus exclusively on SAR sensors and data. SAR sensors have different operating parameters, including not only multiple wavelengths, but also polarizations and incidence angles. It is important to appreciate differences in these instrument specifications because the selection of optimal wavelength, polarization, and incidence angle is vital for successful wetland mapping (Harris and Digby-Arbus 1986; Baghdadi et al. 2001; Costa and Telmer 2007). Optimal specifications vary based on the goals of the mapping project and the environment in which it is being carried out.

Microwave wavelengths commonly used for remote sensing include: X-band (2.4 – 3.8 cm), C-band (3.9 – 7.5 cm), L-band (15 – 30 cm), and P-band (30 – 100 cm; Jensen 2000). Electromagnetic energy transmitted from the SAR sensor towards the surface of the Earth is composed of an electric and a magnetic component. These two components travel, at the speed of light (≈3 x 108 m s-1), orthogonal to one another. The orientation of the electric component of electromagnetic energy (perpendicular to the direction of travel) determines the polarization of that energy. In all but the newest satellite based sensors (i.e., RADARSAT-2 and Phased Array type L-band Synthetic Aperture Radar [PALSAR]) the energy is either transmitted or received horizontally (H) or vertically (V), relative to the surface of the Earth. SAR bands are often described by their wavelength (e.g., X, C, L, or P) and polarization (e.g.,


HH = horizontally transmitted and received and VV = vertically transmitted and received). The energy from SAR sensors is also transmitted and received at different angles relative to the Earth’s surface. These incidence angles are measured relative to an imaginary line perpendicular to the surface of the Earth, with smaller angles being closer to perpendicular to the terrain and larger angles being closer to parallel.

It is primarily the wavelength, polarization, and incidence angle of the microwave energy in combination with certain key characteristics of Earth’s surface (dielectric property, size/roughness, and structure) that determine the amount of energy reflected in the direction of the sensor and therefore received by the sensor. Water content determines the dielectric property of most natural materials. Typically, the higher the water content, the higher the dielectric constant (a measure of the aptitude of a substance to conduct electrical energy) of the material and therefore the greater the amount of incident energy returned from the material (Jensen 2000).

When mapping and monitoring wetland ecosystems, imaging radars have many advantages over sensors that operate in the visible and infrared portions of the electromagnetic spectrum. Microwave energy is sensitive to variations in soil moisture and inundation, and is only partially attenuated by vegetation canopies, especially in areas of lower biomass (Kasischke and Bourgeau-Chavez 1997; Kasischke et al. 1997a; Kasischke et al. 1997b; Townsend and Walsh 1998; Baghdadi et al. 2001; Townsend 2001; Townsend 2002; Rosenqvist et al. 2007; Lang and Kasischke 2008) or when using data collected at longer wavelengths (Hess et al. 1990; Hess et al. 1995; Martinez and Le Toan 2007). The sensitivity of microwave energy to water and its ability to penetrate vegetative canopies, make SAR ideal for the detection of hydrologic features below vegetation (Hall 1996; Kasischke et al. 1997; Kasischke and Bourgeau-Chavez 1997b; Phinn et al. 1999; Rao et al. 1999; Wilson and Rashid 2005). SAR data can even be used to detect freeze/thaw events (Bartsh et al. 2007) because the dielectric of ice is much lower than that of water. The presence of standing water interacts with the radar signal differently depending on the dominant vegetation type/structure (Hess et al. 1995) as well as the biomass and condition of vegetation (Töyrä et al. 2002; Costa and Telmer 2007). When exposed to open water without vegetation, specular reflection occurs and a dark signal (weak or no return) is observed (Dwivedi et al. 1999), making the detection of open water relatively simple. The radar signal is often reduced in wetlands dominated by lower biomass herbaceous vegetation when a layer of water is present due largely to specular reflectance (Kasischke et al. 1997a). Conversely, the radar signal is often increased in forested wetlands when standing water is present due to the double-bounce effect (Harris and Digby-Arbus 1986; Dwivedi et al. 1999). This occurs in flooded forests when energy that is sent out by the sensor is reflected strongly by the water surface away from the sensor (specular reflectance) but is then redirected back towards the sensor by a second reflection from a nearby tree trunk. The use of small incidence angles (closer to nadir) is thought to enhance the ability to map hydrology beneath the forest canopy via increased penetration of the forest canopy (Hess et al. 1990; Bourgeau-Chavez et al. 2001; Toyra et al. 2001).

The temporal resolution of available radar data is relatively high for a variety of reasons. Radar sensors can collect data regardless of solar illumination, cloud cover, and most rain events (shorter wavelength [e.g., X-band] radar can be sensitive to intense rainstorms). For example, the use of radar data is particularly helpful in areas of northern Alaska where daylight can be limited and cloud cover prevalent (Evans 1995; Baghdadi et al. 2001). The


temporal resolution of SAR data is also increased by the ability of certain satellite borne sensors (e.g., Advanced Synthetic Aperture Radar (ASAR), RADARSAT-1, and RADARSAT-2) to collect data at multiple incidence angles, thus reducing the time between satellite overpasses. The availability of multiple well-calibrated (Martinez and Le Toan 2007) satellite borne sensors further increases the temporal resolution of SAR data because those data can be used to gather information about the same area of interest.

Although knowledge of radar data characteristics and processing methods have developed considerably, further research is required in order to capitalize on all of the potential benefits that SAR data can provide to the wetland mapping process (Horritt et al. 2003; Costa 2004). The interpretation of SAR data is less intuitive than that of optical imagery (Silva et al. 2008) and the methods and software used to process SAR data are less developed than those for optical data. In addition, the sensitivity of SAR data to variations in elevation often requires radiometric and geometric corrections that utilize a digital elevation map. Finally, until recently only moderate resolution (30 m) satellite borne SAR data were available, although RADARSAT-2 data are now available (2008) at a spatial resolution of 3 m.

Seasat, launched in 1978, was one of the first imaging radars to be used to study wetlands and other ecosystems (Place 1985; Pope et al., 1997; Ramsey et al. 1998). Researchers found that the L-HH microwave energy transmitted by Seasat was particularly sensitive to flooding, even below forest canopies due to the increase in backscatter caused by double-bounce scattering between tree trunks and the flooded surface (Krohn et al. 1983; Place 1985; Hess et al. 1990; Pope et al. 1997). In addition to Seasat, the ability of L-HH SAR to map inundation in forested wetlands has been well documented with a variety of other sensors (Table 2) including AIRSAR (an airborne sensor), Spaceborne Imaging Radar (SIR)-C, and Japanese Earth Resources Satellite (JERS)-1 (available from 1992 – 1998; Ormsby et al. 1985; Hess et al. 1995; Wilen and Smith 1996; Townsend and Walsh 1998; Martinez and Le Toan 2007).

After the successful launch of European Remote Sensing satellite (ERS-1) (1991 to 2000) and ERS-2 (1995 to present) and RADARSAT-1 (1995 to present), C-band data were increasingly available (Table 2) and wetland studies using these shorter wavelength data were initiated. Researchers found that although C-HH band radar data were not as well suited for forested wetland studies as those from L-HH SARs, they could be used to monitor inundation patterns, especially in areas of lower biomass (Townsend and Walsh 1998; Townsend 2000; Costa 2004; Lang and Kasischke 2008). C-VV data from the ERS systems (from which the longest continuous record of SAR observations exist) have primarily been used to study herbaceous vegetation (e.g., Kasischke et al. 2003; Grings et al. 2006), but have also been successful in detecting inundation under forest canopies during the leaf-off period (Kasischke et al. 1997b; Townsend 2002).

Although SIR-C/X-SAR (deployed onboard NASA’s Space Shuttle) only collected data during the spring and fall of 1994, it provided new details regarding the advantages and limitations of spaceborne multi-wavelength (X, C, and L bands), polarimetric (capable of sending and receiving multiple polarizations at the same time) SAR (Hall 1996). This advancement led to the ability to detect finer details of vegetation structure and therefore improved the accuracy of wetland classifications from space (Pope et al. 1994; Hess et al. 1995; Hall 1996; Smith 1997; Bourgeau-Chavez et al. 2001).


Table 2. Commonly used spaceborne SAR sensors Sensor Platform Resolution Bands Incidence Polarization Width Life-Span InformationJERS Satellite 18m L 39o HH 75 km 1992 -1998 jaxa.jp/index_e.htmlERS-1 Satellite 30m C 23o VV 100 km 1991 -2000 earth.esa.intSIR-C Shuttle 10 - 200m X, C, L 20 - 50o Polarimteric 15 - 90 km 1994 jpl.nasa.govX-SAR Shuttle 10 - 200m X 20 - 50 VV 15 - 40 km 1994 jpl.nasa.gov

RADARSAT-1 Satellite 8 - 100m C 10 - 59o HH 50 - 500 km 1995 - Pres. space.gc.caERS-2 Satellite 30m C 23o VV 100 km 1995 - Pres. earth.esa.intASAR Satellite 30m - 1km C 15 - 45o HH, VV, VH, HV 56 - 400 km 2002 - Pres. earth.esa.int

Terra SAR-X Satellite 1 - 16m X 15 - 60o HH, VV, VH, HV 10 - 100 km 2006 - Pres. dlr.de/enPALSAR Satellite 10 - 100m L 8 - 60o Polarimeteric 70 - 250 km 2006 - Pres. jaxa.jp/index_e.html

RADARSAT-2 Satellite 3 - 100m C 70 - 60o Polarimteric 10 - 500 km 2007 - Pres. radarsat2.info Please note that specifications are approximate and not all specifications are available at the same time.

Sensor name (Sensor), platform type (Platform), spatial resolution (Resolution), wavelength band (Bands), incidence angle (Incidence), polarization, swath width (Width), the years the sensor has operated (Life-Span; Present abbreviated as Pres.), and websites where more information about the sensors can be found are provided in the table. For example, Hess et al. (1995) and Pope et al. (1994) used SIR-C data to identify

various wetland vegetation types and found that the discrimination of these classes was best when using multiple wavelengths and polarizations. Recently SAR sensors have been launched which offer multiple polarizations (ASAR; launched 2002) and even polarimetric data (PALSAR; 2006 to present and RADARSAT-2; 2007 to present). The additional information offered by these sensors is predicted to further improve the utility of SAR data for wetland mapping applications (Pope et al. 1997; Baghdadi et al. 2001; Sokol et al. 2004; Touzi et al. 2007). There are now more opportunities to use and fully explore the potential of SAR data for wetland mapping due to the relatively recent launch of ALOS PALSAR and RADARSAT-2. Both PALSAR and RADARST-2 are capable of providing data with polarimetric data which has been found to improve the detection of wetlands (Horritt et al. 2003; Baghdadi et al. 2001). Additional SAR sensors are currently being planned (e.g., Radar Imaging Satellite). These extra satellites will not only increase the amount of data available for analysis, they will also increase collection frequency, and the variety of polarizations and frequencies available (Ramsey 1997).

The launch of additional SAR satellites by the US and other countries will increase SAR capabilities and applications. NASA is working towards the launch of multiple radar satellites, including the Deformation, Ecosystem Structure, and Dynamics of Ice (DESDynI), the Soil Moisture Active-Passive (SMAP), the Surface Water and Ocean Topography (SWOT) sensors. The DESDynI and SMAP sensors are scheduled to be developed and launched (2010 – 2013) before SWOT (2013 - 2016). The DESDynI sensor will contain an interferometric L-band SAR (<35 m resolution) and non-imaging lidar (~1064 nm laser; ~25 m horizontal, and ~ 1 m vertical spatial resolution) component. DESDynI will, in part, be designed to map and monitor ecosystem characteristics (e.g., biomass and tree height) particularly in forests. The SMAP sensor will contain a SAR and microwave radiometer and is designed to monitor soil moisture and freeze-thaw events. Although the coarse resolution of the sensor (>1 km) will limit its ability to map wetlands at a regional scale, its fine temporal resolution (2-3 day) would provide valuable information concerning the provision of wetland ecosystem services as they are influenced by hydroperiod. SWOT will, in part, be designed to inventory freshwater storage in water bodies, such as wetlands. It will contain a ku-band (~2 cm) near-nadir SAR interferometer, a multiple frequency microwave radiometer,


and a ku-band nadir-looking radar altimeter (vertical resolution of a few cm). These details and more regarding the sensors discussed above can be found in the US National Research Council (2007) publication entitled “Earth Science and Applications from Space: National Imperatives for the Next Decade and Beyond.” In addition to the new SAR satellites planned by the US, the governments of other countries are also planning to launch new SAR satellites in the future. These planned satellites include the Radar Imaging Satellite (RISAT; ≥3m spatial resolution, multiple polarization C-band; India; Misra et al. 2005) and the Multi-Application Purpose SAR (MAPSAR; ≥3m spatial resolution, multiple polarization L-band, Germany and Brazil; Schröder et al. 2005).

Radar not only has potential to detect different types of wetlands; it can also be used to study the condition and function of these valuable areas. Recent advances in radar remote sensing data and processing have made the estimation of biomass and other forest parameters possible on a landscape scale (Le Toan et al. 1992; Kellndorfer et al.1998; Kasischke et al. 1997b; Mougin et al. 1999; Martinez and Le Toan 2007). The ability of radar to map flooding under forest canopies, where optical data would not be able to detect inundation, is particularly promising (Kasischke and Bourgeau-Chavez 1997). Although microwave sensors can currently only be used to estimate soil moisture within the first 10 cm of soil, new models are being developed to help extend this estimate (Li et al. 1998). Additionally interferometric techniques can be used to create extremely accurate digital elevation models using radar imagery. These models have been used to monitor wetland vegetation height and biomass (Simard et al. 2006) and the impact of management practices on wetland hydrology (Wdowinski et al. 2008).

Methods that use a combination of different bands and polarizations or synergistic approaches that use imagery from multiple radar instruments as well as optical data often provide superior results (Sahagian and Melack 1996; Smith 1997; Augusteijn and Warrender 1998; Töyrä et al. 2002; Li and Chen 2005; Töyrä and Pietroniro 2005) because they bring different pieces of information to the process. Decision tree analysis is a good option when combining different types of SAR (Baghdadi et al. 2001) or SAR and optical data (Li and Chen 2005). The use of multi-temporal radar data combined via an intensity, hue, and saturation (IHS) transformation has also been found to improve wetland mapping (Kushwaha et al. 2000).

5.6. Lidar Similar to radar sensors, lidars (light detection and ranging) are also active sensors,

sending and receiving energy produced by the sensor. However, lidars use energy with much shorter wavelengths (VIS and near-infrared) than radars (microwave). There are multiple types of lidar data including waveform and more readily available discrete point return lidar. Waveform lidars sample the entire laser pulse return (echo) whereas discrete point lidars only record a certain number returns (e.g., 2 or 4). This discussion will focus on discrete point return lidar data which are often available from state or local governments (Roger Barlow – personal communication; Vierling et al. 2008) or are readily available from commercial mapping companies (Rosso et al. 2006). The vast majority of lidar data used to inform wetland mapping has been discrete return lidar data. However, it should be noted that waveform lidar technology is rapidly advancing and waveform lidar data are presently being


collected by some companies. Therefore the use of waveform lidar for wetland mapping and monitoring purposes will likely increase in the future.

Lidars emit short pulses of energy, often from the blue-green (bathymetric lidar) or the infrared portion of the electromagnetic spectrum (standard for terrestrial applications), and these pulses illuminate very small portions of the land’s surface (Goodwin et al. 2006; Vierling et al. 2008). A recent survey of airborne lidar sensors (Lemmens 2007) found that most terrestrial lidar sensors operate in the 900 to 1550 nm range. Lidar data can be used to calculate highly accurate x,y,z locations through the use of an onboard Global Positioning and Inertial Navigation System and by calculating the distance to an object by recording the amount of time it takes for an emitted pulse, or a portion of that pulse, to return to the sensor (Vierling et al. 2008). Discrete point lidars usually detect between two and four returns per laser pulse and the strength of those returns is largely determined by the types of materials that the pulse interacts with on the ground (similar to all optical data). For example, pure water strongly absorbs near-infrared energy and green leaves strongly reflect near-infrared energy. Therefore the intensity of near infrared lidar returns from water are normally weaker than those coming from a tree canopy.

Lidar derived digital elevation maps (DEMs) are often used to enhance wetland mapping based on optical (Lichvar et al. 2006; Vierling et al. 2008) data, radar data, or both (Li and Chen 2005; Töyrä and Pietroniro 2005). In this way, landscape position (e.g., slope, depression, or peak) can be made part of the wetland mapping process similar to the information provided by stereoscopic viewing of aerial photographs. This additional information can aid in the detection of wetlands which are normally difficult to identify, such as vernal pools (Lichvar et al. 2006). Although topographic information is commonly available (e.g., photogrametrically derived data and USGS topographic maps) for the United States, the spatial resolution of these data is often not sufficient for wetland identification, especially in areas of subtle topographic change. In general conventional, non-lidar derived DEMs have much coarser vertical accuracies (1 – 10 m) than those derived from lidar (15 cm – 1 m; Murphy et al. 2007b). Lidar derived DEMs can be used to detect subtle variations in topography (Töyrä and Pietroniro 2005) that are often not visible when using the finest resolution optical data. Although the automated processes typically used to analyze digital data do not benefit as directly from the judgment of the analyst as they do in manual photogrammetry, digital DEMs can be used to estimate wetness indices based on slope and contributing area (Tenenbaum et al. 2006; Murphy et al. 2007b). In this way the lidar data are being used to provide some of the same information that is often ascertained by photointerpreters via stereoscopic analysis. Areas with higher wetness indices are more likely to contain wetlands (Murphy et al. 2007b). In addition, DEMs can be used to ascertain information on the functions that wetland serve by providing information on hydrologic connectivity and surface flow pathways (Sorensen et al. 2006; Murphy et al. 2007a). However, lidar provides information not only on elevation, but also on the intensity of the lidar return. The intensity of the lidar return may be helpful for estimating wetland hydrology (Brzank et al. 2004) below the surface of vegetative canopies, provided those canopies have gaps at the time when the lidar data were collected. The increasing availability of lidar sensors that use various wavelength lasers (Lemmens 2007), likely means that proper instrument selection will be increasingly important in a wetland mapping process. The relationship between laser wavelength and water absorption features in the near infrared region maybe critical. Although lidar derived DEMs and other lidar data provide


opportunities for improved wetland mapping, these data do have restrictions. Similar to radar, the processing methods for lidar vary significantly from those developed for more traditional optical sensors (e.g., multispectral). A large reason for this is the fact that lidar data is often presented as point clouds which must then be interpolated to create DEM grids. Software has been developed to process these point data, but often it must be purchased separately from software developed to analyze other types of remotely sensed data. In addition, lidar data should be collected to different specifications based on their application and data collected for one application may not be suitable for another. For example, higher point densities may be needed from mapping in forested areas of relatively subtle topographic change. Point density can be improved by increasing the number of pulses, flying the platform at a lower altitude, decreasing the beam divergence angle, or simply collecting data over the same area twice (Goodwin et al. 2006). Collecting data from high platform altitudes (>3000 m) may compromise the ability of the dataset to resolve the Earth’s surface below forest canopy by weakening the return signal since only signals above a certain threshold level are detected (Goodwin et al. 2006). The utility of lidar data is also limited in certain vegetation types. For example, Rossso et al. (2006) found that lidar could not be used to accurately map ground elevation under dense herbaceous vegetation. Some researchers have found that the accuracy reports provided by commercial vendors are insufficient (Rosso et al. 2006). Finally, although the price of obtaining and processing commercial lidar should decrease in the future, it is still relatively high (Vierling et al. 2008).

CONCLUSION In the past, researchers have debated as to whether spatial or spectral resolution is more

important to the mapping of wetlands. It appears from available research that a combination of spatial and spectral resolutions is needed to map wetlands with remotely sensed data. The necessary spatial resolution, of course, depends upon the size of the wetland patch that is being mapped. For example, as a general rule, it takes an area three by three pixels wide to identify an object on the ground. The more similar a cover type is to adjacent cover types, the larger the area that will be needed (Federal Geographic Data Committee 1992). Although a minimum spatial resolution is needed to map wetlands, spectral information is also vital, especially when mapping wetlands with indistinct spectral signatures, such as forested wetlands. It is this optimal combination of spatial and spectral resolution that is necessary to best map wetlands. If available, the incorporation of finer resolution optical data into the mapping process can be used to effectively increase the spatial resolution of coarser resolution optical imagery (Dymond and Shepherd 2004). This process, often called pan-sharpening when panchromatic data are used, increases the ability to detect smaller wetlands.

Different types of data are sensitive to different components of the landscape. These components (e.g., soil moisture, presence or absence of standing water, biomass, vegetation height, cellular structure, and more) can then be synthesized to compose a superior wetland map. This data fusion is aimed at reducing classification error by incorporating more spectral information. Since radar and optical data are sensitive to very different landscape characteristics, the combination of radar and optical data can significantly improve wetland mapping (Sahagian and Melack 1996; Kushwaha et al., 2000; Ramsey et al. 1998; Lyon and


McCarthy 1995; Töyrä et al. 2002; Silva et al. 2008) and provide a superior land cover map (Rignot 1997; Ramsey et al. 1998). Optical-radar fusion has been shown to improve the accuracy of land cover mapping by as much as 10% (Haack and Bechdol 2000). Radar data not only bring to classifications information on inundation, plant structure, and biomass; the data have also been found to improve soil moisture estimation, even in forested wetlands (Lang and Kasischke 2008). Although optical sensors may be sensitive to soil moisture, some researchers believe that they are less suitable for this application due to their relatively low signal to noise ratio (Neusch and Sties 1999). Multispectral and hyperspectral data do, however, provide superior information concerning the identity of vegetation communities derived mainly from the molecular and cellular structure of the plants.

The potential of optical-radar fusion for wetland mapping is furthered by recent increases in the availability of multiple wavelength (e.g., launch of Radarsat-2, PALSAR, and Terra SAR-X) and multiple polarization radar data and the improvement of the spatial resolution of both optical (e.g., Quickbird, 2.4 m) and radar (i.e., Radarsat-2, 3 m) data. The availability of radar data and techniques to process radar data have improved to the point that the Canadian government is using SAR data as part of their national wetland mapping program (Milton et al. 2003; Li and Chen 2005). The US government (NASA) is currently working towards the launch of new radar satellites (e.g., DESDynI, SMAP, and SWOT), as are the governments of other countries including Germany and Brazil (MAPSAR) and India (RISAT). Companies currently collecting fine resolution optical data are planning to expand their capabilities which will make the acquisition of these datasets during optimal time periods easier. For example, DigitalGlobe which currently provides QuickBird data is expanding their data delivery options with the addition of WorldView-1 (0.5 m pachromatic) which was launched in fall of 2007 and WorldView-2 (0.46 m panchromatic and 1.8 m 8 band multispectral data) which is scheduled to be launched in early 2009. The combination of the three DigitalGlobe satellites will provide daily <1 m data over the entire globe (personal communication -Brett Thomassie; information also available at www.digitalglobe.com).

Newer technologies (e.g., hyperspectral, SAR, and lidar) have the potential to solve many of our most intractable wetland mapping challenges. Hyperspectral data can be used to identify wetland patches which are spectrally indiscernible using multispectral data and are often better at identifying individual plant species (e.g., invasive plants) than multispectral data. SAR data can be used to reveal subtle patterns in hydrology which indicate the presence of wetlands that are normally difficult to identify (e.g., forested wetlands) and the functioning of all wetlands. Lidar data can be used to locate low lying areas which often harbor wetlands, especially when they have a large upland contributing area. Fine resolution lidar data can also be used to locate vernal pools which are often very small and flooded for relatively short lengths of time. The identification of vernal pools is important since these small areas often have a disproportionate impact on biodiversity, especially when compared to the small area that they occupy on the landscape. In addition, their ability to provide habitat is particularly vulnerable to climate change. For example, small changes in hydroperiod can lead to their failure to provide breeding habitat for many amphibian species. Lidar data can also be used to map hydrologic flow pathways which regulate the ability of wetlands to provide valuable ecosystem services (e.g., water quality). The future holds promise for an expanded wetland mapping toolbox with greater availability of currently used remotely sensed data, new remotely sensed data, more robust hardware, and new processing capabilities. The spatial and spectral resolutions of the available sensors are expected to increase rapidly along with the


accessibility of the data (Glackin 1998). Technology assessments in 1996 concluded that the full potential of then current remote sensing imagery had not been satisfactorily explored (Sahagian and Melack 1996; Wilen and Smith 1996) but now, recently available imagery provides an even greater opportunity for advances in wetland mapping.

In addition to spectral resolution, temporal resolution is also vital to the advancement of wetland mapping. Multitemporal imagery can improve wetland mapping capabilities (Töyrä et al. 2001; Li and Chen 2005), especially when images are collected during different seasons and moisture regimes (Kushwaha et al. 2000). Wetlands are inherently dynamic systems and multi-temporal data are often needed to detect these changes. Changes in plant phenology (Silva et al. 2008) and biomass, as well as changes in soil moisture and flooding patterns, are seen throughout the year (Lyon and McCarthy 1995). Multitemporal SAR data can be used to map hydroperiod as it changes in response to seasonal variations in weather and phenology and periods of extreme weather (i.e., floods and droughts; Lang et al. 2008). These hydroperiod maps can not only be used to infer wetland function (e.g., denitrification), they can also be used to update wetland boundaries as they shift in response to climate and land cover change and to identify lands that are transitional between wetlands and uplands. These transitional lands which are not flooded or saturated long enough to be considered wetlands may still serve important ecosystem services which are considered to be typical of wetlands (e.g., habitat provision and biogeochemical transformations). One example of these transitional areas is prior converted wetlands presently on agricultural fields. Many have been drained enough to allow for cultivation but remain flooded or saturated for considerable portions of the year. It should also be noted that the collection of imagery during drought, flood, or normal years can have a large impact on wetland mapping outcomes and ideal time of collection depends on the ultimate goal of the mapping project. Selection of the ideal spatial, spectral, and temporal resolution of remotely sensed imagery, as well as acquisition date will of course depend on the specific goals of the mapping project and the resources available to carry-out the project.

Selection of the appropriate type(s) of imagery is vital to enhanced wetland mapping, but the addition of ancillary data (Sader et al. 1995; Li and Chen 2005) and the use of geographic information systems (GIS) and hydrologic models can also greatly benefit the mapping process. Ancillary datasets that are beneficial to the wetland mapping process include (but are not limited to) information on soils, water bodies, topography, current and prior land use, tides, and weather. Decision support systems (e.g., decision trees, GIS, and hydrologic models) can then be used to extract valuable information from these ancillary datasets in combination with remotely sensed data. The National Research Council (1995) found that GIS holds great possibilities for the study of wetlands and that models make wetland delineation more successful and expeditious. For example, a modified version of a water management simulation model (DRAINMOD) was used to aid in the definition and identification of wetlands via their hydrology (Skaggs et al. 1994). Doren et al. (1999) hypothesized that in the future a combination of aerial photography, rapidly advancing satellite imagery (e.g., hyperspectral data and fine spatial resolution multispectral data), and GIS will allow the detailed mapping of wetland vegetation comparable to those created with detailed ground surveys. Others offer a more cautious predication. Cowardin and Golet (1995) argued that although remotely sensed data in conjunction with GIS and modeling hold promise for improved wetland mapping, wetland maps will never be perfect as they are an


attempt to put artificial boundaries on natural gradations. This ecological reality can never be fully reconciled, but wetland scientists can develop improved maps to reduce errors.

The rapidly updateable maps that digital data and especially satellite images provide support adaptive management of wetland ecosystem services and other natural resources. The value of a wetland map can be increased dramatically when it is used in a modeling framework or a GIS. Spatially explicit information concerning the size, type, location, and other details concerning wetlands can be used in models and other decision support systems. Wetland function may be inferred by some of the characteristics that can be illustrated with a wetland map, such as hydrology, biomass, area, shape, location, and relation to other landscape features, like streams and agricultural fields. For example, the frequency and extent of flooding greatly affects many wetland functions including organic accumulation and storage, primary productivity, and nutrient cycling (Mitsch and Gosselink 2007). Wetland area, location, and number can give us information concerning flood and pollution control. As updated maps demonstrate that wetlands are disappearing, a loss of function is often inferred. But this loss is not purely through the reduction in wetland area. It is also due to a reduction in linkage, average size, and other attributes that can be assessed with a map (National Research Council 1995). Ultimately, this characterization of the loss of wetland function may help the public understand the consequences of wetland destruction (Tiner 1997a).

REFERENCES

Alsdorf, D.E., E. Rodríguez, and D.P. Lettenmaier. 2007. Measuring surface water from space. Reviews of Geophysics 45:RG2002, doi:10.1029/2006RG000197.

Anderson, J. and J. Perry. 1996. Characterization of wetland plant stress using leaf spectral reflectance: implications for wetland remote sensing. Wetlands 16: 477-487.

Augusteijn, M.F. and C.E. Warrender. 1998. Wetland classification using optical and radar data and neural network classification. International Journal of Remote Sensing 19:1545-1560.

Ausseil, A.G.E., J.R. Dymond, and J.D. Shepherd. 2007. Rapid mapping and prioritization of wetland sites in the Manawatu-Wanganui region, New Zealand. Environmental Management 39:316-325.

Baghdadi, N., M. Bernier, R. Gauthier, and I. Neeson. 2001. Evaluation of C-band SAR data for wetlands mapping. International Journal of Remote Sensing 22:71-88.

Baker, C., R. Lawrence, C. Montagne, and D. Patten. 2006. Mapping wetlands and riparian areas using Landsat ETM+ imagery and decision-tree-based models. Wetlands 26:465-474.

Bartsch, A., R.A. Kidd, C. Pathe, K. Scipal, and W. Wagner. 2007. Satellite radar imagery for monitoring inland wetlands in boreal and sub-arctic environments. Aquatic Conservation: Marine and Freshwater Ecosystems 17:305-317.

Becker, B.L., D.P. Lusch, and J. Qi. 2007. A classification-based assessment of the optimal spectral and spatial resolutions for Great Lakes coastal wetland imagery. Remote Sensing of Environment 108:111-120.


Bennett, M.W.A. 1987. Rapid monitoring of wetland water status using density slicing. Proceedings of the 4th Australiasian Remote Sensing Conference, Adelaide, 14-18 September 1987, pp. 682-691.

Beeri, O. and R.L. Phillips. 2007. Tracking palustrine water seasonal and annual variability in agricultural wetland landscapes using Landsat from 1997 to 2005. Global Change Biology 13:897-912.

Bolstad, P.V. and T.M. Lillesand. 1992. Improved classification of forest vegetation in northern Wisconsin through a rule-based combination of soils, terrain, and Landsat Thematic Mapper Data. Forest Science 38:5-20.

Bourgeau-Chavez, L.L., E.S. Kasischke, S.M. Brunzell, J.P. Mudd, K.B. Smith and A.L. Frick. 2001. Analysis of space-borne SAR data for wetland mapping in Virginia riparian ecosystems. International Journal of Remote Sensing 22:3665-3687.

Brazank, A., J. Gopfert, and P. Lohmann. 2005. Aspects of Lidar processing in Coastal Areas. In International Archives of Photogrammetry and Remote Sensing, Vol. XXXVI Part I/W3, Hanover, Germany, 6 pages, CD.

Burgess, W. 1995. Regulatory wetlands map workshop: a report of activities and findings, Maryland Department of Natural Resources, Salisbury, MD.

Carter, V. 1982. Applications of remote sensing to wetlands. In: Remote Sensing of Resource Management, C.J. Johannsen and J.L. Sanders (eds), Soil Conservation Society of America: Ankeny Iowa, pp. 284-300.

Chiu, W.Y. and I. Couloigner. 2006. Modeling fuzzy c-means classification technique for mapping vague wetlands using Landsat ETM+ imagery. Hydrological Processes 20:3634-3634.

Chopra, R., V.K. Verma, and P.K. Sharma. 2001. Mapping, monitoring and conservation of Harike wetland ecosystem, Punjab, India, through remote sensing. International Journal of Remote Sensing 22:89-98.

Cole, C.A. and R.P. Brooks. 2000. Patterns of wetland hydrology in the Ridge and Valley Province, Pennsylvania, USA. Wetlands 20:438-447.

Congalton, R.G., K. Green, and J. Teply. 1993. Mapping old growth forests on national forest and park lands in the Pacific-Northwest from remotely sensed data. Photogrammetric Engineering and Remote Sensing 59:529-535.

Costa, M.P.F. and K.H. Telmer. 2007. Mapping and monitoring lakes in the Brazilian Pantanal wetland using synthetic aperture radar imagery. Aquatic Conservation: Marine and Freshwater Ecosystems 17:277-288.

Costa, M.P.F. 2004. Use of SAR satellites for mapping zonation of vegetation communities in the Amazon floodplain. International Journal of Remote Sensing 25:1817-1835.

Cowardin, L., V. Carter, F. Golet, and E. LaRoe. 1979. Classification of wetlands and deepwater habitats of the United States, US Fish and Wildlife Service, Washington, D.C.

Cowardin, L. and F. Golet. 1995. US Fish and Wildlife Service 1979 wetland classification: a review. Vegetatio 118:139-152.

Dahl, T. and C. Johnson. 1991. Wetlands: status and trends in the conterminous United States mid-1970's to mid-1980's, US Department of the Interior, Fish and Wildlife Service, Washington, D.C.

Dale, P., A. Chandica, and M. Evans. 1996. Using image subtraction and classification to evaluate change in sub-tropical intertidal wetlands. International Journal of Remote Sensing 17:703-719.


Dobson, J. E., E. A. Bright, R. L. Ferguson, D. W. Field, L. L. Wood, K. D. Haddad, H. Iredale III, J. R. Jensen, V. V. Klemas, R. J. Orth, and J. P. Thomas. 1995. NOAA coastal change analysis program (C-CAP) guidance for regional implementation, NOAA Technical Report NMFS 123 – US Department of Commerce, National Oceanic and Atmospheric Administration, Seattle, WA.

Doren, R., K. Rutchey, and R. Welch. 1999. The Everglades: a perspective on the requirements and applications for vegetation map and database products. Photogrammetric Engineering and Remote Sensing 65:155-161.

Du, Y., F. Ulaby, and M. Dobson. 2000. Sensitivity to soil moisture by active and passive microwave sensors. IEEE Transactions on Geoscience and Remote Sensing 38:105-113.

Dupigny-Giroux, L.A.L. 2007. Using AirMISR data to explore moisture-driven land use-land cover variations at the Howland Forest, Maine – A case study. Remote Sensing of Environment 107:376-384.

Dwivedi, R., B. Rao, and S. Bhattacharya. 1999. Mapping wetlands of the Sundaban Delta and it's environs using ERS-1 SAR data. International Journal of Remote Sensing 20:2235-2247.

Dymond, J.R., and J.D. Shepherd. 2004. The spatial distribution of indigenous forest and its composition in the Wellington region, New Zealand, from ETM+ satellite imagery. Remote Sensing of Environment 90:116-125.

Eagman, E.T., and R.J. Gurney. 1991. Water resources management and monitoring. In: Remote Sensing in Hydrology, E.T. Eagman and R.J. Gurney (eds). Chapman and Hall: London pp. 193-210.

Evans, D.L. (ed.) 1995. Spaceborne synthetic aperture radar: current status and future directions. A Report to the Committee on Earth Sciences, Space Studies Board, National Research Council, NASA Technical Memorandum 4679, April 1995.

Federal Geographic Data Committee, 1992. Application of satellite data for mapping and monitoring wetlands - facts finding report: Technical Report 1, Wetlands Subcommittee, Federal Geographic Data Committee, Washington, D.C.

Federal Geographic Data Committee, 1996. Federal Geographic Data Committee Wetlands Subcommittee issues progress report. Federal Geographic Data Committee Newsletter, March 1996.

Federal Geographic Data Committee, Wetlands Subcommittee, 1994. Strategic interagency approach to developing a national digital wetlands data base (second approximation), Federal Geographic Data Committee, Washington, D.C.

Federal Geographic Data Committee, Wetlands Subcommittee. 2007. FGDC Working Draft Wetland Mapping Standard. Federal Geographic Data Committee, Washington, D.C.

Filippi, A.M. and J.R. Jensen. 2006. Fuzzy learning vector quantization for hyperspectral coastal vegetation classification. Remote Sensing of Environment 100:512-530.

Finlayson, C. and A. van der Valk. 1995. Wetland classification and inventory: a summary. Vegetatio 118:185-192.

Goodwin, N.R., N.C. Coops, and D.S. Culvenor. 2006. Assessment of forest structure with airborne LiDAR and the effects of platform altitude. Remote Sensing of Environment 103:140-152.

Govender, M., K. Chetty, and H. Bulcock. 2007. A review of hyperspectral remote sensing and its application in vegetation and water resources studies. Water SA 33:145-151.


Grings, F.M., P. Ferrazzoli, J.C. Jacobo-Berlles, H. Karszenbaum, J. Tiffenberg, P. Pratolongo and P. Kandus. 2006. Monitoring flood condition in marshes using EM models and Envisat ASAR observations. IEEE Transactions on Geoscience and Remote Sensing 44:936-942.

Haack, B and Bechdo, M. 2000. Integrating multisensor data and radar texture measures for land cover mapping. Computers and Geosciences 26:411-421.

Hall, D.K. 1996. Remote sensing applications to hydrology: Imaging radar. Hydrological Sciences Journal 41:609-624.

Harris, J. and S. Digby-Argus 1986. The detection of wetlands on Radar imagery. Proceedings of the tenth Canadian Symposium on Remote Sensing Edmonton, Alberta.

Harvey, K.R. and G.J.E. Hill. 2001. Vegetation mapping of a tropical freshwater swamp in the Northern Territory, Australia: a comparison of aerial photography, Landsat TM and SPOT satellite imagery. International Journal of Remote Sensing 22:2911-2925.

Henderson, F.M., T.F. Hart, Jr., L. Orlando, B. Heaton, J. Portolese, and R. Chasan. 1998. Application of C-CAP Protocol Land-Cover Data to Nonpoint Source Water Pollution Potential Spatial Models in a Coastal Environment. Photogrammetric Engineering and Remote Sensing 64:1015-1020.

Hess, L., J. Melack, S. Filoso, and Y. Wang. 1995. Delineation of inundated area and vegetation along the Amazon floodplain with the SIR-C synthetic aperture radar. IEEE Transactions on Geoscience and Remote Sensing 33: 896-904.

Hess, L., J. Melack, and D. Simonett. 1990. Radar detection of flooding beneath the forest canopy: a review. International Journal of Remote Sensing 11:1313-1325.

Hewitt, M.J. 1990. Synoptic inventory of riparian ecosystems: The utility of Landsat Thematic Mapper data. Forest Ecology and Management 33/34:605-620.

Hirano, A., M. Madden, and R. Welch. 2003. Hyperspectral image data for mapping wetland vegetation. Wetlands 23:436-448.

Horritt, M.S., D.C. Mason, D.M. Cobby, I.J. Davenport, and P.D. Bates. 2003. Waterline mapping in flooded vegetation from airborne SAR imagery. Remote Sensing of Environment 85:271-281.

Houhoulis, P. and W. Michener. 2000. Detecting wetland change: A rule-based approach using NWI and SPOT-XS data. Photogrammetric Engineering and Remote Sensing 66:205-211.

Hruby, T., W.E. Cesanek and K.E. Miller 1995. Estimating relative wetland values for regional-planning. Wetlands 15:93-107.

Huguenin, R.L., M.A. Karaska, D. van Blaricom, and J.R. Jensen. 1997. Subpixel classification of bald cypress and tupelo gum trees in thematic mapper imagery. Photogrammetric Engineering and Remote Sensing 63:717-725.

Jacobson, J.E., R.A. Ritter, and G.T. Koeln. 1987. Accuracy of thematic mapper derived wetlands as based on National Wetlands Inventory data, ASPRS/ACSM/WFPLS 1987 Fall Convention, 4-9 October 1987 at Reno, NV USA American Society for Photogrammetry and Remote Sensing, Falls Church, VA, pp. 109-118.

Jackson, T.J., D.M. Le Vine, A.Y. Hsu, A. Oldak, P.J. Starks, C.T. Swift, J.D. Isham, and M. Haken. 1999. Soil moisture mapping at regional scales using microwave radiometry: the southern Great Plains hydrology experiment. IEEE Transactions on Geoscience and Remote Sensing 37:2136-2150.


Jensen, J.R. 2000. Remote Sensing of the Environment: An Earth Resource Perspective. Upper Saddle River, N.J., Prentice-Hall, Inc.

Jensen, J.R., H. Lin, Y. Yang, E. Ramsey, B.A. Davis and C.W. Thoemke. 1991. The measurement of mangrove characteristics in southwest Florida using SPOT multispectral data. Geocarto International 6:13-21.

Johnston, C.A. and P. Meysembourg. 2002. Comparison of the Wisconsin and national wetlands inventories. Wetlands 22:386-405.

Johnston, R.M. and M.M. Barson. 1993. Remote sensing of Australian wetlands: An evaluation of Landsat TM data for inventory and classification. Australian Journal of Marine and Freshwater Resources 44:235-252.

Judd, C., S. Steinberg, F. Shaughnessy, and G. Crawford. 2007. Mapping salt marsh vegetation using aerial hyperspectral imagery and linear unmixing in Humboldt Bay California. Wetlands 27:114-1152.

Justice, C.O., E. Vermote, J.R.G. Townshend, R. Defries, D.P. Roy, D.K. Hall, V.V. Salomonson, J.L. Privette, G. Riggs, A. Strahler, W. Lucht, R.B. Myneni, Y. Knyazikhin, S.W. Running, R.R. Nemani, W. Zhengming; A.R. Huete, W. van Leeuwen, R.E. Wolfe, L. Giglio, J. Muller, P. Lewis, and M.J. Barnsley. 1998. The moderate resolution imaging spectroradiometer (MODIS): land remote sensing for global change research. IEEE Transactions on Geoscience and Remote Sensing 36:1228-1246.

Kasischke, E. and L. Bourgeau-Chavez. 1997. Monitoring South Florida wetlands using ERS-1 SAR imagery. Photogrammetric Engineering and Remote Sensing 63:281-291.

Kasischke, E., L. Bourgeau-Chavez, K. Smith, E. Romanowicz, and C. Richardson. 1997a. Monitoring hydropatterns in South Florida ecosystems using ERS SAR data, 3rd ERS Symposium on Space at the Service of our Environment, Florence, Italy, pp. 71-76.

Kasischke, E., J. Melack, and M. Dobson. 1997b. The use of imaging radars for ecological applications-a review. Remote Sensing of Environment 59:141-156.

Kasischke, E.S., K.B. Smith, L.L. Bourgeau-Chavez, E.A. Romanowicz, S. Brunzell, and C.J. Richardson. 2003. Effects of seasonal hydrologic patterns in south Florida wetlands on radar backscatter measured from ERS-2 SAR imagery. Remote Sensing of Environment 88(4):423-441.

Kellndorfer, J., L. Pierce, M. Dobson, and F. Ulaby. 1998. Toward consistent regional-to-global vegetation characterization using orbital SAR systems. IEEE Transactions on Geoscience and Remote Sensing 36:1396-1411.

Klemas, V.V. 2001. Remote sensing of landscape-level coastal environmental indicators. Environmental Management 27:47-57.

Koeln, G. and J. Bissonnette. 1999. Cross-correlation analysis: mapping landcover changes with a historic landcover database and a recent, single-date, multispectral image, Earth Satellite Corporation, Rockville, MD.

Kokaly, R.F., D.G. Despain, R.N. Clarka, and K.E. Livoa. 2003. Mapping vegetation in Yellowstone National Park using spectral feature analysis of AVIRIS data. Remote Sensing of Environment 84:437-456.

Krohn, M.D., N.M. Milton, and D.B. Segal. 1983. Seasat synthetic aperture radar (SAR) Response to lowland vegetation types in eastern Maryland and Virginia. Journal of Geophysical Research-Oceans and Atmospheres 88:1937-1952.

Kudray, G.M. and M.R. Gale. 2000. Evaluation of national wetland inventory maps in a heavily forested region in the upper great lakes. Wetlands 20:581-587.


Kushwaha, S., R. Dwivedi, and B. Rao. 2000. Evaluation of various digital image processing techniques for detection of coastal wetlands using ERS-1 SAR data. International Journal of Remote Sensing 21:565-579.

Laba, M., F. Tsal, D. Ogurcak, S. Smith, and M.E. Richmond. 2005. Field determination of optimal dates for the discrimination of invasive wetland plant species using derivative spectral analysis. Photogrammetric Engineering and Remote Sensing 71:603-611.

Lang, M., G. McCarty, and M. Anderson. 2008. Monitoring wetland hydrology at a watershed Scale: dynamic information for adaptive management. Journal of Soil and Water Conservation 63:49A.

Lang, M.W. and E.S. Kasischke. 2008. Using C-band synthetic aperture radar data to monitor forested wetland hydrology in Maryland's Coastal Plain, USA. IEEE Transactions on Geoscience and Remote Sensing 46:535-546.

Le Toan, T., A. Beaudoin, J. Riom, and D. Guyon. 1992. Relating forest biomass to SAR data. IEEE Transactions on Geoscience and Remote Sensing 30:403-411.

Lemmens, M. 2007. Airborne lidar sensors. GIM International 21: February 2007. Li, K., R. Jong, and J. Boisvert. 1998. Towards estimating soil moisture in the root zone using

remotely sensing surface data. Canadian Journal of Remote Sensing 24:255-263. Li, J., and W. Chen. 2005. A rule-based method for mapping Canada’s wetlands using

optical, radar, and DEM data. International Journal of Remote Sensing 26:5051-5069. Lichvar, R.W., D.C. Finnegan, S. Newman, and W. Ochs. 2006. Delineating and Evaluating

Vegetation Conditions of Vernal Pools Using Spaceborn and Airborne Remote Sensing Techniques, Beale Air Force Base, CA. US Army Corps of Engineers Engineer Research and Development Center, ERDC\CRREL TM-06-3.

Lillesand, T. and R. Kiefer. 1994. Remote Sensing and Image Interpretation. John Wiley and Sons, Inc., New York, NY.

Lunetta, R. and M. Balogh. 1999. Application of multi-temporal Landsat 5 TM imagery for wetland identification. Photogrammetric Engineering and Remote Sensing 65:1303-1310.

Lyon, J. and J. McCarthy. 1995. Wetland and Environmental Applications of GIS. Lewis Publishers, New York, NY.

Mahlke, J. 1996. Characterization of Oklahoma Reservoir wetlands for preliminary change detection mapping using IRS-1B Satellite imagery. IGARSS 1996: 1996 International Geoscience and Remote Sensing Symposium, pp. 1769-1771.

Martinez, J.M. and T. Le Toan. 2007. Mapping of flood dynamics and spatial distribution of vegetation in the Amazon floodplain using multitemporal SAR data. Remote Sensing Environment 108:209-223.

McCarthy, J., T. Gumbricht, and T.S. McCarthy. 2005. Ecoregion classification in the Okavango Delta, Botswana from multitemporal remote sensing. International Journal of Remote Sensing 26:4339-4357.

Milton, G. R., L. Belanger, Y. Crevier, R. Helie, J. Hurley, and B.H. Kazmerik. 2003. Development of a Remote-Sensing Wetland Inventory and Classification System for Canada. backscatter Volume 14, Number 1, Winter 2003. Halifax, Nova Scotia p. 3.

T. Misra, S. S. Rana, and K. N. Shankara. 2005. Synthetic aperture radar payload of Radar Imaging Satellite (RISAT) of ISRO. Proceedings of the XXVIIIth International Union of Radio Science General Assembly, New Delhi, India.

Mitsch, W.J. and J.G. Gosselink. 2007. Wetlands, 4th ed., John Wiley and Sons, Inc., New York.


Moore, G.K. and G.W. North. 1974. Flood inundation in the southeastern United States from aircraft and satellite imagery. Water Resources Bulletin 10:1082-1096.

Mougin, E., C. Proisy, G. Marty, F. Fromard, H. Puig, J.L. Betoulle, and J.P. Rudant. 1999. Multifrequency and multipolarization radar backscattering from mangrove forests. IEEE Transactions on Geoscience and Remote Sensing 37:94-102.

Mumby, P.J., E.P. Green, A.J. Edwards and C.D. Clark. 1999. The cost-effectiveness of remote sensing for tropical coastal resources assessment and management. Journal of Environmental Management 55:157-166.

Munyati, C. 2000. Wetland change detection on the Kafue Flats, Zambia, by classification of a multitemporal remote sensing image dataset. International Journal of Remote Sensing 21:1787-1806.

Murphy, P.N.C., J. Ogilvie, F.R. Meng and P. Arp. 2007a. Stream network modeling using lidar and photogrammetric digital elevation models: a comparison and field verification. Hydrological Processes DOI: 10.1002/hyp.6770.

Murphy, P.N.C., J. Ogilvie, K. Connor, and P.A. Arp. 2007b. Mapping wetlands: A comparison of two different approaches for New Brunswick, Canada. Wetlands 27:845-854.

National Research Council, Committee of Earth Sciences and Applications from Space. 2007. Earth Science and Applications from Space: National Imperatives for the Next Decade and Beyond. National Academy Press, Washington, DC.

National Research Council, Committee on the Characterization of Wetlands. 1995. Wetlands: Characteristics and Boundaries. National Academy Press, Washington, DC.

Nestler, J.H. and K.S. Long. 1997. Development of hydrological indices to aid cumulative impact analysis of riverine wetlands. Regulated Rivers: Research and Management 13:317-334.

Neusch, T. and M. Sties. 1999. Application of the dubois-model using experimental synthetic aperture radar data for the determination of soil moisture and surface roughness. Journal of Photogrammetry and Remote Sensing 54:273-278.

New York State Department of Environmental Conservation. 2008. Freshwater Wetlands Mapping. http://www.dec.ny.gov/lands/5124.html#how.

Nichols, C. 1994. Map accuracy of National Wetlands Inventory Maps for areas subject to Maine Land Use Regulation Commission Jurisdiction. U.S. Fish and Wildlife Service, Hadley, ME, USA. Ecological Services Report R5–94/6.

NOAA. 2008. NOAA Coastal Change Analysis Program. http://www.csc.noaa.gov/crs /lca/ccap.html

Ormsby, J.P., B.J. Blanchard, and A.J. Blanchard (1985). Detection of lowland flooding using active microwave systems. Photogrammetric Engineering and Remote Sensing 51:317-328.

Pantaleoni, E. 2007. Assessing Coastal Plain Wetland Composition using Advanced Spaceborne Thermal Emission and Reflection Radiometer Imagery. Dissertation Submitted to the Faculty of Virginia Polytechnic Institute and State University.

Papa, F., C. Prigent, F. Durand, and W.B. Rossow. 2006. Wetlands dynamics using a suite of satellite observations: A case study of application and evaluation for the Indian Subcontinent. Geophysical Research Letters, 33, LO8401, doi:10.1029/2006GL025767

Phinn, S., L. Hess and C.M. Finlayson. 1999. An assessment of the usefulness of remote sensing for wetland inventory and monitoring in Australia. Finlayson C.M. and Spiers


A.G. (eds) Techniques for enhanced wetland inventory and monitoring. Supervising Scientist Report 147, Supervising Scientist, Canberra.

Pinty, B., J.L. Widlowski, N. Gobron, M.M. Verstraete, and D.J. Diner. 2002. Uniqueness of multiangular measurement – Part I: An indicator of subpixel surface heterogeneity from MISR. IEEE Transactions on Geoscience and Remote Sensing 40:1561-1573.

Place, J.L. 1985. Mapping of forested wetland: use of Seasat radar images to complement conventional sources. Professional Geographer 37:463-469.

Poor, P.J. 1999. The value of additional central flyway wetlands: The case of Nebraska's Rainwater Basin wetlands. Journal of Agricultural and Resource Economics 24: 253-265.

Pope, K., E. Rejmankova, J. Paris, and R. Woodruff. 1997. Detecting seasonal flooding cycles in marshes of the Yucatan Peninsula with SIR-C polarimetric radar imagery. Remote Sensing of Environment 59:157-166.

Pope, K., J. Rey-Benayas, and J. Paris. 1994. Radar remote sensing of forest and wetland ecosystems in the Central American tropics. Remote Sensing of Environment 48: 205-219.

Ramsey, E., 1997. Dynamic flood monitoring in coastal environments. In: O. Huh (Ed.), Natural Coastal Environmental Hazards, Louisiana State University, pp. 27-38.

Ramsey, E. and S. Laine. 1997. Comparison of Landsat Thematic Mapper and high resolution photographs to identify change in complex coastal wetlands. Journal of Coastal Research 13:281-292.

Ramsey, E.W., III, S.C. Laine, G.A. Nelson, S.K. Sapkota, M.L. Strong, J.K. Wooderson, R.H. Day, R.E. Spell, D.K. Chappell, T.L. Stoute, R.G. Kirkman, and M.A. Books. 1998. Identifying wetland zonation and inundation extent by using satellite remote sensing and ground-based measurements. In G.R. Guntenspergen and B.A. Vairin (eds). Vulnerability of Coastal Wetlands in the Southeastern United States: Climate Change Research Results, 1992-97. U.S. Geological Survey, Biological Resources Division Biological Science Report USGS/BRD/BSR--1998-0002. pp. 83-92.

Rao, B.R.M., R.S. Dwivedi, S.P.S. Kushwaha, S.N. Bhattacharya, J.B. Anand, and S. Dasgupta. 1999. Monitoring the spatial extent of coastal wetlands using ERS-1 SAR data. International Journal of Remote Sensing 20: 2509-2517.

Reed, D.J. 1993. Hydrology of temperate wetlands. Progress in Physical Geography 17:20-31.

Richardson, C.J. 1994. Ecological Functions and Human-Values in Wetlands - a Framework for Assessing Forestry Impacts. Wetlands 14:1-9.

Rignot, E., W. Salas, and D. Skole. 1997. Mapping deforestation and secondary growth in Rondonia, Brazil, using imaging radar and Thematic Mapper data. Remote Sensing of Environment 59:167-179.

Ringrose, S., C. Vanderpost, and W. Matheson. 2003. Mapping ecological conditions in the Okavango delta, Botswana using fine and coarse resolution systems including simulated SPOT vegetation imagery. International Journal of Remote Sensing 24:1029-1052.

Roberts, D.A., M.O. Smith and J.B. Adams. 1993. Green vegetation, non-photosynthetic vegetation, and soils in AVIRIS data. Remote Sensing of Environment 44:255-269.

Rolband, M.S. 1995. A comparison of wetland areas in northern Virginia: National Wetlands Inventory maps versus field delineated wetlands under the 1987 manual. Wetland Journal 7:10–14.


Rosenqvist, A., C.M. Finlayson, J. Lowry, and D. Taylor. 2007. The potential of long-wavelength satellite-borne radar to support implementation of the Ramsar Wetland Convention. Aquatic Conservation: Marine and Freshwater Ecosystems 17:229-244.

Rosso, P.H., S.L. Ustin, and A. Hastings. 2005. Mapping marshland vegetation of San Francisco Bay, California, using hyperspectral data. International Journal of Remote Sensing 26:5169-5191.

Rosso, P.H., S.L. Ustin, and A. Hastings. 2006. Use of lidar to study changes associated with Spartina invasion in San Francisco Bay marshes. Remote Sensing of Environment 100:295-306.

Rundouist, D., S. Narumalani, and R. Narayanan. 2001. A review of wetland remote sensing and defining new considerations. Remote Sensing Reviews 20:207-226.

Rutchey, K. and L. Vilchek. 1999. Air photointerpretation and satellite imagery analysis techniques for mapping cattail coverage in a northern Everglades impoundment. Photogrammetric Engineering and Remote Sensing 65:185-191.

Sahagian, D. and J. Melack. 1996. Global wetland distribution and functional characterization: trace gases and the hydrologic cycle. IGBP Report 46, International Geosphere-Biosphere Programme (IGBP).

Sader, S.A., D. Ahl, and W.S. Liou. 1995. Accuracy of Landsat-TM and GIS rule-based methods for forest wetland classification in Maine. Remote Sensing of Environment 53:133-144.

Schmidt, K.S. and A.K. Skidmore. 2003. Spectral discrimination of vegetation types in a coastal wetland. Remote Sensing of Environment 85:92-108.

Schröder, R., J. Puls, I. Hajnsek, F. Jochim, M.M. Quitino da Silva, W.R. Paradella, and M.A. Chamon. 2005. The MAPSAR Mission: objectives, design and status. Proceedings of the XII Annual Brazilian Symposium on Remote Sensing. Goiania, Brazil.

Seevers, P., R. Peterson, R. Mahoney, D. Maroney, and D. Rundquist. 1975. An inventory of Nebraska wetlands with the use of imagery from the Earth Resources Technology Satellite. Proceedings of the Fourth Annual Remote Sensing of Earth Resources Conference. University of Tennessee Space Institute. pp. 281-292.

Shapiro, R. 1995. Coordination and integration of wetland data for status and trends and inventory estimates. Technical Report 2, Federal Geographic Data Committee Wetlands Subcommittee.

Shi, J., L. Jiang, L. Zhang, K.S. Chen, J.P. Wigneron, A. Chanzy, and T.J. Jackson. 2006. Physically based estimation of bare-surface soil moisture with the passive radiometers. IEEE Transactions on Geoscience and Remote Sensing 44:3145-3153.

Shutko, A., A. Haldin, E. Novichikhin, G. Yazerian, G. Chukhray, E. Vorobeichik, V. Agura, S. Kalashnik, V. Sarkisjants, N. Sklonnaja, B. Logan, and E.W. Ramsey III. 1997. Application of microwave radiometers for wetlands and estuaries monitoring. Proceedings of the Fourth International Conference on Remote Sensing for Marine and Coastal Environments, Volume 1, March 17-19, 1997, Orlando, FL. pp. 553-561.

Silva, T.S.F., Costa, M. P. F., Melack, J.M., and Novo, E.M.L.M. 2008. Remote sensing of aquatic vegetation: theory and applications. Environmental Monitoring and Assessment 140:131-145.

Simard, M., K. Zhang, V.H. Rivers-Monroy, M.S. Ross, P.L. Ruiz, E. Castañeda Moya, R.R. Twilley and E. Rodriguez. 2006. Mapping height and biomass of mangrove forests in


Everglades National Park with SRTM elevation data. Photogrammetric Engineering and Remote Sensing 72:299-311.

Sokol, J., H. NcNairn, and T.J. Pultz. 2004. Case studies demonstrating the hydrological applications of C-band multipolarized and polarimetric SAR. Canadian Journal of Remote Sensing 30:470-483.

Skaggs, R., D. Amatya, R. Evans, and J. Parsons. 1994. Characterization and evaluation of proposed hydrologic criteria for wetlands. Journal of Soil and Water Conservation 49:501-510.

Smith, L. 1997. Satellite remote sensing of river inundation area, stage, and discharge: a review. Hydrological Processes 11:1427-1439.

Spanglet, H., S. Ustin, and E. Rejmankova. 1998. Spectral reflectance characteristics of California subalpine marsh plant communities. Wetlands 18:307-319.

Sorensen, R., U. Zinko, and J. Seibert. 2006. On the calculation of the topographic wetness index: evaluation of different methods based on field observations. Hydrology and Earth System Sciences 10:101-112.

Stolt, M.H., and J.C. Baker. 1995. Evaluation of National Wetland Inventory Maps to inventory wetlands in the southern Blue Ridge of Virginia. Wetlands 15:346–353.

Swarthout, D.J., W.P. MacConnell, and J.T. Finn. 1981. An evaluation of the National Wetland Inventory in Massachusetts. In-place resource inventories: Principles and practices, proceedings of a national workshop. Orono, ME: University of Maine. pp. 685-691.

Tenenbaum, D.E., L.E. Band, S.T. Kenworthy, and C.L. Tague, 2006 Analysis of soil moisture patterns in forested and suburban catchments in Baltimore, Maryland, using high-resolution photogrammetric and LIDAR digital elevation datasets. Hydrological Processes 20: 219-240.

Tiner, R.W. 1990. Use of high-altitude aerial photography for inventorying forested wetlands in the United States, Forest Ecology and Management 33:593-604.

Tiner, R. 1997a. Adapting the NWI for preliminary assessment of wetland functions, The Future of Wetland Assessment: Applying Science through the Hydrogeomorphic Assessment Approach and Other Approaches. The Association of State Wetland Managers Institute for Wetland Science and Public Policy, Parole, MD, pp. 105-106.

Tiner, R. 1997b. NWI maps: what they tell us. National Wetlands Newsletter 19:7-12. Tiner, R. 1997c. Piloting a more descriptive NWI. National Wetlands Newsletter 19:14-16. Tiner, R. 1999. Wetland Indicators: A Guide to Wetland Identification, Delineation,

Classification, and Mapping. Lewis Publishers, Washington, DC. Tiner, R.W. 2004. Remotely-sensed indicators for monitoring the general condition of

“natural habitat” in watersheds: an application for Delaware’s Nanticoke River watershed. Ecological Indicators 4:227-243.

Tiner, R.W. 2003. Estimated extent of geographically isolated wetlands in selected areas of the United States. Wetlands 23:636-652.

Tiner, R.W. 2004. Remotely-sensed indicators for monitoring the general condition of “natural habitat” in watersheds: an application for Delaware’s Nanticoke River watershed. Ecological Indicators 4:227–243.

Tiner, R.W. 2005. Assessing cumulative loss of wetland functions in the Nanticoke River watershed using enhanced National Wetlands Inventory data. Wetlands 25:405-419.


Tiner, R. and G. Smith. 1992. Comparisons of four scales of color infrared photography for wetland mapping in Maryland. National Wetlands Inventory Report R5-92/03, US Fish and Wildlife Service, Region 5, Newton, MA.

Touzi, R., A. Deschamps, and G. Rother. 2007. Wetland characterization using polarimetric RADARSAT-2 capability. Canadian Journal of Remote Sensing 33:S56-S67 Supplement: Suppl. 1

Townsend, P.A. 2000. A quantitative fuzzy approach to assess mapped vegetation classifications for ecological applications. Remote Sensing of Environment 72: 253-267.

Townsend, P.A. 2002. Relationships between forest structure and the detection of flood inundation in forested wetlands using C-band SAR. International Journal of Remote Sensing 23:443-460.

Townsend, P. and S. Walsh. 1998. Modeling floodplain inundation using an integrated GIS with radar and optical remote sensing. Geomorphology 21:295-312.

Töyrä, J., A. Pietroniro, L.W. Martz, and T.D. Prowse. 2002. A multi-sensor approach to wetland flood monitoring. Hydrological Processes 16:1569-1581.

Töyrä, J., and A. Pietroniro. 2005. Towards operational monitoring of a northern wetland using geomatics-based techniques. Remote Sensing Environment 97:174-191.

Töyrä, J., A. Pietroniro, and L. W. Martz. 2001. Multisensor hydrologic assessment of a freshwater wetland. Remote Sensing Environment 75:162-173.

U.S. Fish and Wildlife Service. 2002. National wetlands inventory: A strategy for the 21st century. Washington, DC: US Department of the Interior, Fish and Wildlife Service.

Vanderbilt, V.C., S. Khanna, S.L. Ustin. 2007. Impact of pixel size on mapping surface water in subsolar imagery. Remote Sensing Environment 109:1-9.

Vanderbilt,V.C., G.L. Perry, G.P. Livingston, S.L. Ustin, M.C. Diaz Barrios, F.M. Bréon, M.M. Leroy, J.Y. Balois, L.A. Morrissey, S.R. Shewchuk, J.A. Stearn, S.E. Zedler, J.L. Syder, S. Bouffies-Cloche, and M. Herman. 2002. Inundation discriminated using sun glint. IEEE Transactions on Geoscience and Remote Sensing 40:1279-1287.

Vane, G., R.O. Green, T.G. Chiren, H.T. Enmark, E.G. Hansen, and W.M. Porter. 1993. The airborne visible/infrared imaging spectrometer (AVIRIS). Remote Sensing of Environment 44:127-143.

Vierling, K., L. Vierling, W. Gould, S. Martinuzzi, and R. Clawges. 2008. Lidar: Shedding new light on habitat characterization and modeling. Frontiers in Ecology and the Environment 6:90-98.

Wdowinski, S., S.W. Kim, F. Amelung, T.H. Dixon, F. Miralles-Wilhelm, and R. Sonenshein. 2008. Space-based detection of wetlands’ surface water level changes: from L-band SAR interferometry. Remote Sensing of Environment 112:681-696.

Welch, R., M. Madden, and R. Doren. 1999. Mapping the Everglades. Photogrammetric Engineering and Remote Sensing 65:163-170.

Whigham, D.F. 1996. Ecosystem functions and ecosystem values. In: Ecosystem Functions and Human Activities. R. D. Simpson and N. L. Christensen (eds). New York, NY, International Thomson Publishing pp. 225-239.

Whitehead, J.C. and C.Y. Thompson. 1993. Environmental preservation demand - altruistic, bequest, and intrinsic motives. American Journal of Economics and Sociology 52:19-30.

Wilen, B. and G. Smith. 1996. Assessment of remote sensing/GIS technologies to improve National Wetlands Inventory maps, Sixth Biennial Forest Service Remote Sensing Applications Conference, Denver, CO, pp. 50-64.


Wilen, B. and R. Tiner. 1989. The National Wetlands Inventory - the first ten years. American Water Resources Association, September: 1-12 1989.

Wilson, B.A. and H. Rashid. 2005. Monitoring the 1997 flood in the Red River Valley using hydrologic regimes and RADARSAT imagery. Canadian Geographer 49:100-109.

Woodward, R.T. and Y.S. Wui. 2001. The economic value of wetland services: a meta-analysis. Ecological Economics 37:257-270.

Wright, C. and A. Gallant. 2007. Improved wetland remote sensing in Yellowstone National Park using classification trees to combine TM imagery and ancillary environmental data. Remote Sensing of Environment 107:582-605. Reviewed by: Dr. Dianna Hogan Research Physical Scientist Eastern Geographic Science Center U.S. Geological Survey 12201 Sunrise Valley Drive, MSN 521 Reston, VA 20192


Chapter 3

TRANSFORMING USELESS SWAMPS INTO VALUABLE WETLANDS:

EVALUATING AMERICA’S POLICY, 1970-2008

Andrea K. Gerlak∗ and Jeanne N. Clarke University of Arizona, Department of Political Science

324 Social Sciences, Tucson, Arizona 85721

ABSTRACT

This paper traces the evolution of America’s wetland policy beginning with passage of the Clean Water Act (CWA) of 1972. This law, for the first time, established a federal program to protect wetlands, dramatically elevating the value of these ecosystems. However, despite attitudinal changes and new governmental programs, the nation continues to lose its potentially valuable wetlands -- albeit at a slower rate than was the case in the 1970s and prior to the passage of the CWA. This paper offers an objective evaluation of the federal wetlands protection policy. We place this evaluation within a broad societal context, showing that since 1970 there have occurred sweeping demographic, economic, and political changes that clearly have impacted the extent of wetlands in the United States. We argue that Section 404 has failed to reverse the net loss of wetlands in the U.S. Moreover, it has evolved into a policy lightening rod within the water resources arena and been a major factor in Congress’ failure to revise and reauthorize the Clean Water Act. Finally, we offer some recommendations designed to improve the policy, arguing for heightened wetlands protection through partnerships and acquisitions. Keywords: Wetlands, Clean Water Act, Section 404.

∗ (520)621-7600 (office); (520)621-5780 (fax); [email protected]

Andrea K. Gerlak and Jeanne N. Clarke 114

INTRODUCTION During the last several years, wetland protection has attracted national attention. On Earth

Day 2004, President Bush called for a new commitment to increase the nation’s wetlands. In 2006, the United States Supreme Court handed down decisions regarding the scope of the Clean Water Act of 1972 (CWA) in highly publicized cases involving land use and federal regulatory authority. No longer do wetlands evoke reactions of “disgust at their sight and smell, fear of malaria and yellow fever, and unease about rich resources running to waste within them” (Meyer 1994, 109). Rather, the public interest in wetlands has changed dramatically, from a desire to get rid of them to one of protection, preservation, and even re-creation (Vileisis 1997). Despite attitudinal changes and new governmental programs, however, the nation continues to lose its valuable wetlands -- albeit at a slower rate than was the case in the 1970s and prior to the passage of the CWA. Today, wetlands are valued for numerous functions and values, including flood control and water quality benefits, groundwater recharge, and wildlife habitat. On a per-acre basis, they provide immense ecosystem services (Constanza et al. 1997). Presently, the largest ecosystem restoration project in human history is underway: Restoration of the Florida Everglades and its wetland ecosystem.

In this paper we trace the evolution of the Section 404 program of the 1972 CWA, including its growing controversy and its widening institutional reach: Decision makers in all branches of government and at all levels of government for decades have participated in the debate over the federal protection of wetlands. But to what effect? We thus ask the following questions: What have been the overall results of the thirty-five year effort at protecting wetlands? Have we in fact saved wetlands in the United States? Can wetlands policy be improved, and if so, how?

We place this evaluation within a broad societal context, showing that since 1970 there have occurred sweeping demographic, economic, and political changes that clearly have impacted the extent of wetlands in the United States. We consider to what extent federal policy has contributed to reducing the loss of wetlands versus how much can be attributed to macro-level social change. We conclude that Section 404 has not been an effective tool at reversing the net loss of wetlands in the U.S. Instead, it has evolved into a policy lightning rod within the water resources arena and has been a major factor in Congress’ failure to revise and reauthorize the Clean Water Act. We offer some recommendations designed to improve its track record inasmuch as federal action at preserving wetlands remains necessary.

THE CREATION OF A MULTI-FACETED NATIONAL WETLANDS

PROTECTION POLICY, 1972-1988 Following two years of intense work and a congressional override of a presidential veto,

Congress passed the Federal Water Pollution Control Act Amendments (Clean Water Act) in 1972. Representing the “most ambitious and expensive pollution control requirements promulgated into law” up to that time, the wastewater treatment program contained in the Act initially attracted the greatest attention and the most federal funding (Lieber 1975, 11). Yet, its Section 404 arguably was the most novel part of the Act. Interestingly, the term wetland is

Transforming Useless Swamps into Valuable Wetlands 115

not expressly mentioned in the Clean Water Act of 1972, but rather the term “waters of the United States” was intended to include wetlands. At just one page in length, Section 404 is brief, but it is the source of what has evolved into a complex, multi-jurisdictional wetlands protection policy. Its statement that the “guiding principle should be that degradation of special sites may represent an irreversible loss of valuable aquatic resources” Section 404 signaled a dramatic shift in our collective attitude towards wetlands: But the legislation went further by setting up a national regulatory program designed to determine if and when the alteration of a wetland represented an “irreversible loss.” Two federal agencies, the Army Corps of Engineers (Corps) and the newly created Environmental Protection Agency (EPA), were given lead authority in making those decisions.

The selection of those two agencies underscores the adage that politics makes strange bedfellows. The Corps was selected because it had a long-standing interest in dredge and fill activities. The 1899 Rivers and Harbors Act, often referred to as the Refuse Act, prohibited “the discharge of dredged or fill material” into navigable waterways of the United States without first obtaining a permit from the Corps. The agency’s supporters in Congress argued that the 404 permitting process was a logical extension of the Corps’ authority in the water resource regulatory arena. However, there also existed within Congress members who were suspicious about the Corps’ commitment to environmental protection and who wanted to see this authority given to the newly created Environmental Protection Agency. A compromise was reached wherein the Corps would be the agency that processed the permits and made a preliminary decision, which then would be reviewed by the EPA. In effect, the EPA would exercise a veto over the Corps’ decisions.

Not surprisingly, given the very different organizational cultures of the two agencies, the first several years of Section 404 implementation were characterized by considerable conflict between the Corps and the EPA. Getting the staff in these two federal agencies to cooperate was a task unto itself. Officials in the EPA, for instance, pushed for an expansive authority to regulate all wetlands in the U.S., while Corps personnel insisted on a more limited approach, arguing that Section 404 was intended to apply only to interstate navigable waters. It wasn’t long before these interagency conflicts spilled over into the judicial arena. In a 1975 lawsuit brought by an environmental organization, Natural Resources Defense Council v. Calloway [392 F. Sup. 685 (D.D.C. 1975)], the U.S. District Court invalidated the Corps’ regulations on wetlands permitting, finding that the agency had interpreted Section 404 too narrowly. The Corps had adopted an administrative definition of navigable waters limited to waters subject to the ebb and flow of the tide, waters presently or in the past used for interstate commerce, and waters susceptible in the future for use for interstate commerce purposes. The District Court decision resulted in the Corps’ revising its regulations to include significant intrastate waters, along with interstate water bodies.

Nevertheless, the issue of which wetlands were or weren’t covered under Section 404 persisted. By 1977 special interests mobilized on both sides. Environmentalists sought broader jurisdiction over navigable waters while resource users such as ranchers, farmers, miners, and loggers feared burdensome federal regulations. With the 1977 Amendments to the CWA Congress resolved any debate about the statute’s original intent to protect wetlands. According to Senator Edmund Muskie, Chair of the Senate Subcommittee on Environmental Pollution and a primary sponsor of the CWA, “There is no question that the systematic destruction of the Nation’s wetlands is causing serious, permanent ecological damage…The unregulated destruction of these areas is a matter which needs to be corrected and which


implementation of section 404 has attempted to achieve” [123 Cong. Rec. S26,697 (daily ed. August 4, 1977)]. While the 1977 Amendments retained a broad jurisdictional approach in regulating dredge and fill activities it also exempted certain routine land-use activities, such as farming and forestry, from the permit process. The Amendments didn’t settle the contentious issues of what defined a wetland, however, leaving it up to the implementing agencies to promulgate the necessary regulations and the courts to interpret them. Congress’ inability to settle on a uniform wetland definition would plague wetland protection policy in years to come.

In 1979 President Carter’s Attorney General, Benjamin Civiletti, attempted to resolve interagency disputes when he determined that the EPA enjoyed ultimate authority to decide the scope of wetlands regulation because that agency carried the responsibility for implementing the other sections of the 1972 CWA [43 Op. Attorney General 15]. In a 1982 court case, the Corps’ very authority to administer (or co-administer, as was the case) Section 404 was challenged. In Buttrey v. United States [690 F. 2d 1186 (5th Cir., 1982), cert. denied, 461 U.S. 927 (1983)] petitioners claimed, unsuccessfully, that Section 404 was an unconstitutional delegation of regulatory authority to the Corps. When some private land developers took the Corps to court, the Supreme Court issued a landmark decision in wetlands policy. Its 1985 decision in U.S. v Riverside Bayview Homes [474 U.S. 121] found that the Corps (and the EPA) can regulate wetlands even in instances where no surface water or other apparent hydrological connection exists between the wetland and an adjacent water body. In a unanimous opinion, the Court determined that Congress, in 1972, chose to define the waters covered in the Act broadly, and so the Court, too, could discover no practical boundaries to federal jurisdiction over wetlands.

If Congress initially had created a system of shared management authority, involving only the Corps and the EPA in protecting wetlands, that would have been one matter. But at about the same time as it enacted the CWA, Congress passed legislation that brought still more federal agencies and departments into the wetlands policy arena. A 1970 statute, the Water Bank Act, created a federal program which paid farmers to preserve wetlands habitat for waterfowl. Administered by the Department of Agriculture’s Soil Conservation Service (now the Natural Resources Conservation Service), this program has grown significantly over the past thirty-five years. Employing incentives (i.e., federal monies) as opposed to regulation, this type of wetlands protection program understandably has been much more popular with the public than has the Section 404 permit process.

In 1973 still another dimension was added when Congress passed the Endangered Species Act (ESA). Administered by the U.S. Fish and Wildlife Service (FWS), this highly controversial statute added another federal player to the list of agencies implementing the wetlands protection policy. Consultation with the FWS became the norm since so many wetlands are home to endangered and threatened species. In some situations, the FWS has been given the authority to manage sensitive wetland areas. Thus, over the last thirty years this agency, too, has become an integral component of the national wetlands policy.

By the end of the “environmental decade” of the 1970s, at least four federal agencies had major responsibilities for protecting wetlands. A number of other federal agencies, including the Forest Service, the Bureau of Reclamation, the National Marine Fisheries Service, and the Bureau of Land Management, had ancillary responsibilities. Other legislation added several more bureaucratic participants to an already crowded landscape. For example, the Farm Services Agency in the Department of Agriculture came to administer and enforce the


Swampbuster provision contained in the 1985 Farm Bill. The Swampbuster provision denies federal farm program benefits to producers who planted an agricultural commodity on wetlands that were converted after December 23, 1985. Figure 1 shows how wetland preservation legislation has broadened beyond the regulatory program of Section 404 of the Clean Water Act to include a variety of acquisition, restoration, and incentive programs. Table 1 displays the primary federal agencies involved in wetland protection activities. Of the thirty-six federal agencies involved in wetland-related activities, six agencies account for more than 70% of the funding and 65% of the staffing associated with such activities (U.S. GAO 1998).

Table 1. Primary Federal Agencies Involved in Wetland Protection Activities

Federal Agency Role in Wetlands Protection Department of Defense, Army Corps of Engineers

Administers wetland permitting and enforcement under Section 404 of the Clean Water.

Department of Agriculture, Farm Service Agency

Manages voluntary programs to help protect and restore wetlands associated with farmland

Department of Agriculture, Natural Resources Conservation Service

Delineates wetlands under farm bill legislation and provides technical and financial assistance to landowners to restore and enhance wetlands.

Department of Commerce, National Oceanic and Atmospheric Administration

Initiates wetland restoration activities associated with marine ecosystems.

Department of the Interior, Fish and Wildlife Service

Reviews Section 404 permit applications and makes recommendations to the Corps and promotes voluntary restoration programs.

Environmental Protection Agency

Performs oversight over Corps permitting process and enforcement.

Source: U.S. General Accounting Office, Wetlands Overview: Problems with Acreage Data Persist (Washington, DC, General Accounting Office, 1998).

Adapted from: U.S. General Accounting Office, Wetlands Overview: Federal and State Policies,

Legislation, and Programs (Washington, DC, General Accounting Office, 1991).

Figure 1. Major Wetland-related Laws by Purpose.


This list of interested federal parties doesn’t include state, regional, tribal, and local agencies that had an historic interest in either draining or less frequently, preserving wetlands. For example, several states have taken the lead and adopted expansive wetland protection laws, although state regulatory programs and statutes vary substantially (Association of State Wetland Managers 2007). When these non-federal parties are added, the number of participants goes up dramatically. And so does political conflict, stonewalling, and passing the buck over wetlands protection. Thus, despite favorable federal court rulings in the 1970s and 1980s and the increase in the number of governmental agencies involved in wetlands protection, the total acreage of wetlands in the U.S. continued to shrink. This was taken by some as proof that the multi-jurisdictional approach which emphasized the federal role in wetlands protection was not achieving the intended results.

ESCALATING POLITICS CONFLICT OVER WETLANDS, 1988-2008 Despite what appeared to be considerable governmental activity in the wetlands policy

arena from 1972 through the 1980s, the federal wetlands program enjoyed a relatively low public profile. Only a handful of Washington insiders, some environmental activists, and a small number of homebuilders paid much attention to how well or how poorly the permit program was working. That condition of benign neglect changed in 1988 when Vice-President George Bush, campaigning for the presidency, announced his "No-Net-Loss" policy. With that announcement of a new national wetlands policy there came an increase in public visibility and an escalation in political conflict over it. Indeed, environmental policy generally became much more contentious as the decade of the 1980s drew to a close.

Upon entering office in 1989, the Bush White House was lobbied by developers and property owners. Administration officials were inundated with horror stories about uncertainties and long delays in the implementation of the Section 404 program. Among the voices it heard from was the newly formed National Wetlands Coalition, which consisted of a group of primarily Southern-based energy companies intent on protecting its rights to drill for oil and gas in wetlands (National Wetlands Coalition 2004).

A Council on Competitiveness, headed by Vice President Dan Quayle, engaged the debate by reviewing the Corps’ and the EPA’s new, 1989, wetlands delineation manual. The manual attempted to eliminate inconsistent policies across federal agencies, but in doing so the new guidelines slightly expanded the definition of a wetland that had been contained in a 1987 manual. For example, the Corps Manual had required the visible presence of water, vegetation, and soils whereas the Joint Manual allowed, under some circumstances, that one or more of these three parameters be assumed from the presence of others. Public hearings were held across the country to help define what Congress, in the original legislation and in a subsequent re-authorization, had not defined. The Farm Bureau Federation submitted over 300 complaints to the Senate Environment and Public Works Committee. The mass media focused on seemingly innocent victims of the new rules. The journal Forbes published maps showing that the new regulations turned virtually all of Dorchester County, Maryland, into wetlands (Steinhart 1993).

After two years of controversy, in 1991 the White House offered a proposal that restricted the Corps’ and EPA’s jurisdiction by narrowing the definition of a wetland. The


outcry from environmentalists to that proposal was heard across the country (Lemonick 1991, 53). They found objectionable most of the Administration's new wetlands policy, including the requirement that water must be found at the surface for ten to twenty consecutive days during the growing season in order for an area to be designated a wetland. The proposed 1991 manual also shortened the growing season, and it reduced the variety of plants that previously had qualified an area for wetland status. Critics charged that the Bush manual wrote off one-third of the nation's remaining waterfowl nesting areas, floodplains, vernal pools, playas, and prairie potholes. In fact, critics claimed, approximately 30 million acres – an area the size of Florida – were removed from federal protection on the basis that they were too small, or not sufficiently wet for sufficiently long, to qualify as wetlands (Pope 1991).

In the midst of the controversy involving the Bush Administration’s proposal, Congress intervened by passing an amendment to the 1992 Energy and Water Development Appropriations Act. Referred to as the Johnston Amendment (after Louisiana Senator J. Bennett Johnston), P.L. 101-104 prohibited the Corps and the EPA from using the more inclusive 1989 Joint Manual in reviewing permit applications. Consequently, the agencies returned to using the 1987 manual for the critical process of wetland delineation.

The debate over a politically acceptable, and scientifically supportable definition of wetlands, which plagued the Bush Administration through its entire four years in office, was never resolved. By January 1993, EPA officials claimed they had received more than 80,000 formal comments concerning the “dueling manuals” issue. Mired in gridlock, environmentalists and developers alike urged the Bush Administration to move the wetlands delineation issue out of the political arena and into the scientific arena. Although the President did not do so, Congress took the lead in this request. Among more than twenty wetlands-related bills introduced during 1992, the only one that passed Congress authorized a study of wetlands by the prestigious National Research Council (NRC) (National Wetlands Newsletter 1993).

Both the 103rd Congress and the incoming Clinton Administration initially worked for wetlands reform. Numerous bills to re-authorize the Clean Water Act were introduced, which would have amended the contentious and confusing Section 404 program. None passed (Zinn and Copeland 1993). At the other end of Pennsylvania Avenue Vice President Al Gore was put in charge of the Administration's "reinventing government" initiative. Within that broad mandate, an inter-agency task force was set up to work on wetlands regulatory reform. An interim new policy was announced in August, 1993. Titled, "Protecting America's Wetlands: A Fair, Flexible, and Effective Approach," it included numerous changes to the program, including the wider use of mitigation banking, an exemption for small landowners who wanted to alter 1/2 acre or less of non-tidal wetlands, a streamlined appeals process, and an exemption for previously converted croplands (White House Office of Environmental Policy 1993). It also embraced the "No-Net-Loss" policy of the previous Administration.

Viewed in their entirety, and with hindsight, the 1993 reforms represented incremental reform. The Clinton Administration “muddled through” a politically difficult situation when what might have been accomplished at this juncture – with a new Democratic Administration and a Democratic majority in Congress – was comprehensive reform of a stalled program (Lindblom 1959). To the consternation of the environmental community at the time, environmental reform simply was not a high priority of the Clinton Administration in its first terms: “growing the economy” was.


Two years into the Clinton Presidency, the National Research Council released its report. Titled, “Wetlands: Characteristics and Boundaries,” it contained a definition of a wetland that was grounded more in science than in politics (NRC 1995). The report also made eighty recommendations for improving the numerous federal programs relating to wetlands. But the publication of this important document coincided with the changing of the guard in Congress. The Republicans’ agenda, called “The Contract with America,” doomed action on the report. It still awaits serious attention by Congress.

In 1997, on the twenty-fifth anniversary of the CWA, Vice President Gore announced a new “Clean Water Initiative.” At the heart of Gore’s plan was the goal, starting in 2005, of a net annual gain of up to 100,000 acres of wetlands in the United States. President Clinton mentioned the Clean Water Action Plan in his 1998 State of the Union address as the centerpiece of a five-year, $2.3 billion initiative to speed up restoration of the nation’s waterways. Watershed management was at the core of the Clinton Administration’s clean water goals. Because the problems and inadequacies of the Section 404 program proved to be politically intractable, absent any legislative leadership, it appeared as though the Clinton Administration decided on a strategy of shifting public attention to broader natural resource management issues. Conflicts over wetlands then would be subsumed under the formation regional watershed management plans.

Despite these actions, the scope of Section 404 continues to be litigated and debated in the twenty-first century. With its 2001 decision, Solid Waste Agency of Northern Cook County v. U.S. Army Corps of Engineers [121 S.Ct. 675] (also known as the SWANCC decision) the Supreme Court “reversed elements of more than fifteen years of federal jurisdiction when it ruled that the federal government overstepped its authority in regulating certain isolated wetlands” (National Wetlands Newsletter 2001, 2). In this case, a county solid waste agency sought permission from the Corps to develop a disposal site on 17 acres of land that contained isolated, seasonal ponds used by migratory birds. In its 5-4 decision, the Court ruled that the Corps could no longer use the “Migratory Bird Rule” to regulate isolated, non-navigable, intrastate ponds. Concurring with the Court’s ruling, Bush Administration officials questioned “whether the CWA should apply to non-navigable tributaries of navigable waters, intermittent and ephemeral streams, human-constructed watercourses connecting these waters, and wetlands adjacent to such waters” (Izzo 2002). Subsequently, a January 2003 Corps-EPA regulatory guidance letter required their regional offices to withhold protections from isolated, non-navigable intrastate waters and to seek formal, project-specific, approval from their headquarters (EPA 2003; Pianin 2003a, A05).

But after receiving more than 133,000 comments opposing efforts to narrow the CWA’s scope, the Bush Administration reversed its 2002 decision to consider new rules based on a broad interpretation of the SWANCC decision (Pianin 2003b, A20). By doing so, President Bush, like his predecessor, embraced the “No-Net-Loss” policy first promulgated by his father in 1988. The current Bush Administration emphasizes Corps’ district flexibility and allows for trade-offs related to wetland functions, a marked distinction from the previous administration’s mitigation policy which emphasized a minimum replacement ratio of one acre of new wetlands for every acre destroyed (Corps 2002). On Earth Day 2004, President Bush called for a new commitment to attain an overall increase in the quality and quantity of wetlands with the specific goal of restoring, improving, and protecting three million acres of wetlands over the next five years (Sanger and Halbfinger 2004). The most recent wetland report released by the White House Council on Environmental Quality (CEQ) in April 2005


illustrates these efforts. The report, titled Conserving America’s Wetlands, Implementing the President’s Goal, calls for a new commitment to attain an overall increase in the quantity of wetlands in the U.S. (CEQ 2005). The President’s goals include restoring and creating at least one million acres, improving or enhancing at least one million acres, and protecting at least one million acres.

In June 2006, the U.S. Supreme Court again handed down decisions regarding the scope of the Clean Water Act and the Section 404 Program. The two cases from Michigan, Carabell v. U.S. Army Corps of Engineers and Rapanos v. United States, were consolidated for argument before the Court. In both cases, developers challenged federal regulators’ refusal to permit development activities in wetlands. Both cases involved the flow of water off the proposed development sites and the connection to federal navigable waters which triggers federal regulatory authority under the CWA. The Supreme Court issued a split 4-1-4 decision, with four justices agreeing that the Clean Water Act called for a restrictive view of the scope of federal jurisdiction to reach remote wetlands, and four other justices concluding that the statute permits the government to take upstream actions to prevent downstream degradation of federal water resources. Justice Kennedy was the critical fifth vote restricting federal authority and sending the cases back to the lower courts. With the Supreme Court failing to offer any additional guidance for government agencies implementing Section 404, it seems then that the district and circuit federal courts will have to decide these issues on a case-by-case basis (Richey 2006). Such uncertainty negatively impacts enforcement. EPA’s regional offices are not pursuing CWA violations because of the uncertainty regarding their jurisdiction (Eilperin 2008).

EVALUATING U.S. WETLANDS POTECTION POLICY An analysis of the thirty-five year history of U.S. wetlands policy, beginning with Section

404 of the Clean Water Act, yields some interesting findings and some conventional wisdom. To the original question, “Has it worked?,” the answer is, “No,” but with some important qualifications to bear in mind. It is “No” for three principal reasons: (1) The Section 404 program has evolved into a policy lightning rod within the water resources arena, with development and environmental interests taking extreme and/or unreasonable positions on its worth; (2) despite its modest size and budget, the 404 program has been a major factor in Congress’ failure to revise and reauthorize the Clean Water Act; (3) and, most importantly, the program has not reversed the net loss of wetlands in the United States. Development interests still find it burdensome and an infringement on their property rights while environmental interests think it is too weak -- but better than nothing. It appears to have few genuine supporters. A 1992 statement by the U.S. Advisory Committee on Intergovernmental Affairs (ACIR) about the program remains applicable today: It is illustrative of “all the problems concerning federal environmental decision-making in state and local public works projects” (ACIR 1992, 48).

In recent years, government and academic reports have highlighted wetland data problems and inconsistencies in agency jurisdiction and mitigation practices. The General Accounting Office (GAO) reported that wetland acreage data are unreliable and inconsistent; it found poor reporting practices demonstrated by a lack of consistency in use of terms, the


inclusion of non-wetlands acreage in wetland project totals, and the double counting of accomplishments (GAO 1998, 9). The GAO also reported mismanagement of wetland mitigation replacement requirements and a lack of effective oversight by the Army Corps of Engineers (GAO 2001). More recently, it documented troubling inconsistencies in the Corps’ federal regulatory jurisdiction determinations under the Clean Water Act (GAO 2004).

Criticisms persist despite the fact that the present program covers a relatively small percentage of extant wetlands -- although as we pointed out its coverage has waxed and waned over the past thirty-five years, depending on how wetlands are defined. Given the congressionally authorized exemptions for ongoing farming, forestry, and ranching activities, it is estimated that Section 404 regulates only about 20 percent of the activities that continue to drain wetlands (Bianucci and Goodenow 1991). For those actions that do require a permit, virtually all of them are granted by the Corps and the EPA. The agencies claimed that in an average year fewer than 1% of permit applications are denied (Davis 1997; EPA 1991; Corps 2006).

Finally, there is the relatively new federal wetlands mitigation banking program that allows developers to pay a fee for the right to develop a wetland, with that payment going into a fund which then pays for the creation and/or restoration of wetlands elsewhere. In theory, mitigation banking has the potential to provide both flexibility for developers and promote wetland restoration and creation. Considerable scholarly attention has been devoted to wetland mitigation practices with several studies investigating wetland mitigation compliance and effectiveness in specific regions of the country (Allen and Feddema 1996; Cole and Shafer 2002; Sudol and Ambrose 2002; Hornyak and Halvorsen 2003). The National Research Council raised concerns over inadequate mitigation monitoring and enforcement (NRC 2001). Further, there is concern that mitigation banks alone cannot achieve no-net-loss (Brown and Lant 1999). In early 2006, EPA and the Corps released draft mitigation banking rules which address many concerns raised by the NRC, including issues related to monitoring and the use of science-based assessment procedures (Department of Defense and EPA 2006). They await public comment.

Given all of the exemptions and modifications to the original program, one must conclude that it has not worked as originally intended in 1972. But do many federal programs work exactly as intended? The answer to that question is of course “No” (Pressman and Wildavsky 1973). Therefore an evaluation must include the unintended consequences of the Section 404 program, and also take into account the significant economic and demographic growth that has occurred over its lifespan—recalling that those two factors historically have posed the greatest threat to wetlands preservation. It is a fact that despite a growing gross domestic product, and a 28% increase in population, wetlands destruction has in fact slowed considerably since 1972. The regulatory program has worked to the extent that it has raised Americans’ consciousness about the value of wetlands, which tellingly are no longer called swamps. For instance, public opinion polls from the early 1990s showed that a majority of respondents thought the government was not doing enough to protect wetlands, and they would be willing to pay additional taxes to protect wetlands (Environment Opinion Survey 1991; National Wildlife Federation 1989). In a more recent poll, three-fourths of respondents (hunters and anglers in Iowa) said that the U.S should be a world leader in global warming issues. Of particular concern to them was the impact of warmer temperatures on wetland habitat (ENS, 2007).


Section 404 also offers a public review procedure that allows interested parties to comment on potential adverse impacts from proposed wetland conversion (Alvayay and Baen 1990). Permit applications now are routinely published in local newspapers. One result of this raised visibility is that the American public for the most part no longer questions the value of wetlands but rather debates how best to protect them. This is not an insignificant development.

Over the life of the regulatory program, the debate in Washington has moved from the legislative arena to the executive and judicial branches, and to some extent back to the state and local levels of government. Each of the three most recent chief executives – George H.W. Bush, Bill Clinton, and George W. Bush -- formulated a “new” wetlands policy for the nation. Federal agencies and their state and local counterparts then wrestle with implementing the new policy. At the same time federal courts continue to be very much involved in the issue, as evidenced by the 2001 and 2006 Supreme Court decisions.

Again, this is typical of the evolution of much federal legislation. While Congress has found itself bogged down in gridlock over revising Section 404 and the Clean Water Act generally, other political actors fill the vacuum. But, administrative and judicial fine-tuning cannot forever take the place of a thorough legislative re-examination of the program: Recall that the last time Congress seriously amended Section 404 of the CWA was in 1977, some thirty years ago.

In effect, for the past twenty-five years the Congress has bypassed the contentious regulatory issue by changing the subject. With Republican majorities in the House and Senate for much of this period, legislators have actively pursued other approaches to wetlands protection. Building on land acquisition and incentives programs, such as the Water Bank Act and the Land and Water Conservation Fund, recent legislation has targeted coastal wetlands protection through funding federal and state coastal zone management programs. Similarly, Congress has nudged the Department of Agriculture towards greater wetlands protection with incentives-based programs encouraging farmers to rehabilitate wetlands and to control agricultural runoff with the creation of the Wetlands Reserve Program. Although these programs appear to be working better than the 404 regulatory program, and are certainly more popular with the affected parties, it is very likely that, without the power of the Section 404 federal stick, we wouldn’t have so many federal carrots.

Viewed broadly, the good news is that the numerous and sundry governmental programs, coupled with changing economic, demographic, and social conditions, have slowed the annual rate of wetland loss. But which is the more important cause – federal action or broad societal changes – is virtually impossible to determine with any degree of reliability. The two have occurred simultaneously so that assigning causality is based as much on informed judgment as it is on quantitative measures. What we do know, however, is that wetland conversion rates dropped from nearly half a million acres per year during the 1950s-1970s to 58,000 acres per year in the 1980s-1990s. (See Figure 2). An 80% reduction in average annual loss from prior decades is a critical change to some scholars (Dahl 2000, 9). Most recently, the U.S. Fish and Wildlife Service reported an average annual net gain of 32,000 acres of wetlands between 1998 and 2004 (Dahl 2006, 16). The creation of artificial freshwater ponds, designed and maintained as open water basins for water retention and ornamentation and largely lacking in vegetation, account for the net gain (Dahl 2006, 74). Without them, wetland losses would have surpassed gains.


Note: Wetlands loss figures are reported for the mid-1950s to mid-1970s, mid-1970s to mid-1980s, and 1986-1997. T.E. Dahl, Status and Trends of Wetlands in the Conterminous United States, 1986-1997 (Washington, DC, United States Department of the Interior, Fish and Wildlife Service, 2000).

Figure 2. Average Annual Net Wetland Loss.

Table 2 places the recent slowdown in wetlands conversion into an economic and demographic context. It shows that the nation’s annual wetlands loss has decreased in spite of population increase and robust economic growth. It also shows that total farmland acreage has declined by about 15% since 1970, with the result that the agricultural sector’s share of gross wetlands conversion dropped from more than 80% in 1954-1974 to 20% in 1982-1992 (Heimlich et al. 1998, 18).

Table 2. Population and Farmland Trends, 1970-2000

Year Total U.S.

Population* Urban Population* (% of total)

Rural Population* (% of total)

Farmland Acreage*

Farm GDP** (as % of total GDP)

1970 203.3 149.6 (74%) 53.6 (26%) 1,102.4 26.2 (2.5%)

1980 226.5 167.1 (74%) 59.5 (26%) 1,038.9 56.1 (2.0%)

1990 248.7 187.1 (75%) 61.7 (25%) 986.9 79.6 (1.4%)

2000 281.4 222.4 (79%) 59.1 (21%) 943.1 71.0 (0.7%)

% change 1970 - 2000 +28% +33% +10% - 15% - 68% (as % of total GDP)

* in millions. ** in billions. Sources: U.S. Bureau of Census, 1980 Census of Population, Number of Inhabitants (Washington, DC, U.S. Department of Commerce, June 1983; U.S. Bureau of Census, 1990 Census of Population, General Population Characteristics (Washington, DC, U.S. Department of Commerce, November 1992; U.S. Census Bureau Census 2000 Summary File1 (Washington, DC, U.S. Department of Commerce; U.S. Department of Agriculture, 2002); National Agricultural Statistics, Historical Data at http://www.nass.usda.gov:81/ipedb/, accessed 26 July; Bureau of Economic Analysis, National Income and Product Account Tables at http:// www.bea.doc.gov/bea/dn/nipaweb/SelectTable.asp?Selected=N, accessed 11 March, 2003; Council of Economic Advisors, Economic Report of the President, January 2001, Table B-10: Gross Domestic Product by Sector, 1959-2000 at http://w3.access.gpo.gov/usbudget/ fy2002/sheets/b10.xls, accessed 21 February 2004.


Declining agricultural prices helped to reduce wetland conversion in the 1980s at about the same time as federal programs designed to conserve agricultural wetlands were enacted. The farm percentage of the GDP continues to decrease, as does farmland acreage. (It has been estimated that the number of farmers in America today has shrunk to about the number the country had in 1850.) Finally, Table 2 shows that the nation’s population continues its century-long march from the country to the city, with no signs of stopping. Urban and suburban development has replaced agriculture and heavy industry as the major threats to wetlands in 21st century America.

While there is good news with respect to wetlands protection there also is bad news. The bad news is that we still have not accomplished President Bush’s 1989 goal of “No-Net-Loss,” and are making miniscule steps toward meeting Vice-President Gore’s 1997 “Net-Gain” goal and President Bush’s 2004 Earth Day call. And while the 2005 CEQ report is a notable attempt at standardizing wetland data reporting protocols across federal agencies, it also raises some important questions related to wetland data. First, there is the serious problem of counting acreage restored multiple times under the current measurement practices. Such “double-counting” is common because of the sheer number of wetland programs and agencies involved. One state resource manager described the problem of double-counting this way:

“Imagine a farmer in Iowa who enrolls in the Conservation Reserve Program and restores

twenty acres of wetlands on his property. The twenty acres of wetlands are tallied as restored wetlands. Fifteen years later when the program contract expires, he then enrolls the Wetland Reserve Program and the twenty acres are counted yet again” (Bishop 2006). The 2006 CEQ report also demonstrates a broader problem with how wetlands

management has been framed. Since President George H.W. Bush’s 1988 campaign pledge of “no-net-loss,” through the present Bush Administration’s actions, the wetlands debate has centered on wetland acreage – on quantity. Yet, wetland net loss numbers provide an incomplete picture of wetlands loss and conversion nationally. They ignore the qualitative dimension. For example, imagine that a valuable urban wetland in a suburb of Chicago is drained and filled for housing development. Ten acres of valuable wetlands which provide important water quality and habitat functions are now lost. Meanwhile, an agricultural wetland that provides far fewer ecosystem benefits is “re-established” in North Carolina. Fifteen acres of wetlands are now calculated as “new” wetlands, producing a five-acre net gain. This gain is then interpreted as apolicy success, without regard for wetland functions and values. All of the federal agency reports tracking dollars spent and acres “restored,” including the most recent CEQ report, fail to capture this. So too does the recently released U.S. Fish and Wildlife Service (FWS) Status and Trends Report, which consistently admits that it’s reports do not account for wetland quality (Dahl 2006, 74). Indeed, it is easier to measure acreage than to measure changes in wetland functions. Bill Wilen, head of the U.S. FWS’s National Wetlands Inventory, writes:

“Somehow, we have lost sight of the fact that wetlands are being lost in one part of the

country and restored in another. We have also lost sight of the fact that losses in certain types of wetlands are being balanced in gains in other types of wetlands. Most important, gains and losses in wetland acreage do not reflect the changes in wetland functions that result from restoration or the losses in quality that result from discharge of pollutants, vegetation removal,


hydrologic modifications, subsidence, fragmentation, sedimentation, or sediment diversion” (Wilen 1995, 10). This overemphasis on numbers tallies clouds policy evaluation and performance

measures (Dale and Gerlak 2007). Indeed, the wetlands numbers game has some serious policy implications. First, it makes adversaries of the agencies generating the wetlands acreage figures and the people conducting the restoration work. This is because, according to Wilen, the people generating the acreage estimates are not finding the new acreages implied by the restoration numbers on the ground (Wilen 1995). Second, incomplete and unreliable acreage data pose serious challenges for local and state decision-makers who struggle to make better land-use decisions and comply with federal environmental regulations. Because of significant congressional cuts in the U.S. Fish and Wildlife Service is budget, state and local resource managers must often rely on old FWS National Wetlands Inventory maps that do not correctly capture wetland location and loss. This provides obstacles for developers or state officials seeking to mitigate for wetlands loss in one location by restoring a wetland in another location. Indeed, reliable, complete data on wetland location are a necessary condition of mitigation. For mitigation to be successful, we need not just wetland acreage numbers, but also a measure of wetland function and quality across the nation. It is a huge task that not surprisingly has not been undertaken to date.

REFORMING U.S. WETLANDS POLICY: SOME MODEST SUGGESTIONS

Presidents, executive branch agencies, and the federal courts have done as much as they

can do to protect wetlands, absent new legislation. When Congress decides to act, the appropriate place to begin is with the National Research Council’s 1995 report. With its emphasis on regionalism and ecosystems management, and a definition of wetlands based on science, the report offers an objective basis for resolving the many jurisdictional and delineation issues that have hampered the program from the beginning (NRC 1995). Inasmuch as Congress authorized the report in 1993, it should seriously consider it in rewriting the Clean Water Act and its contentious Section 404.

Lawmakers can begin by looking closely at the numerous local and regional partnerships that have been created over the last 20 years or so. Such partnerships are central to a watershed approach and currently represent the more successful components of the national wetlands policy. Through federal support of these partnerships, regulators can continue moving beyond the old “command-and-control” policies of the 1970s and towards a more locally oriented, incentives-based approach. Also, with 75% of the remaining wetlands in the U.S. privately owned, continued and improved government outreach to property owners about the value of saving wetlands is critical.

Acquisition of wetlands by federal, state, tribal, and local governments is another key to reform. More generous funding for the Land and Water Conservation Fund through passage of a bill similar to the one introduced in Congress in 2000 (the Conservation and Reinvestment Act) offers an opportunity for improved intergovernmental management of a vital public resource. One cannot help but wonder whether the lengthy and expensive


litigation between the Corps of Engineers and Cook County, Illinois, over a mere 17 acres of land was worth it: Perhaps the more cost-effective approach would have been federal acquisition of the land in dispute.

Congress also needs to reassess the No-Net-Loss policy that was formulated by President Bush and continued, at least symbolically by his successors. While this approach may have been reasonable when it was first proposed in 1988, more recent scientific data on wetlands structure and function show that not all wetlands are created equal. As we discuss in our conclusion, some wetlands simply are more critical than are others, at least from a societal point of view. Current restoration efforts also must be reviewed, since these practices are attracting increasing criticism (Kaiser, 2001; Malkoff 1998; Mitsch et al 1998).

Finally, given the huge demographic and economic changes that have taken place over the past thirty years, it is imperative to look ahead at the probable causes of wetlands loss in the coming decades. The US Fish and Wildlife Service reported that urban development, or sprawl, accounted for the greatest percentage – some 30% -- of wetlands loss between 1986 and 1997 (Dahl 1990, 69). Given the pace of metropolitan growth today, especially in the Sunbelt, policymakers must address this. As urban areas expand, so does the need for protecting the ecological services provided by naturally occurring wetlands within their parameters. This means that some areas of the country will require greater attention and resources. Restoration of the Everglades -- reduced by 70% from its original state -- will require national attention and is getting it (Florida DEP 2005). So too does Louisiana – a state containing 40% of the wetlands in the contiguous United States, and accounting for 80% of wetland loss in the lower 48 states. Scientists have argued that preserving Louisiana’s wetlands would have lessened hurricane damage along the coast (Knickerbocker 2005). One observer argues that Louisiana’s wetlands are the “hardest working wetlands in America, a watery world of bayous, marshes, and barrier islands that either produces or transports more than a third of the nation's oil and a quarter of its natural gas, and ranks second only to Alaska in commercial fish landings.” (Bourne 2004). Similiarly, changes in the landscape, including loss of wetlands and native grasses that absorb water, along with development and filling of floodplains, contributed to flooding in the Midwest in the summer of 2008 (Achenbach 2008).

CONCLUSION In her 1947 best-selling book, The Everglades: River of Grass, Marjory Stoneman

Douglas wrote, “There are no other Everglades in the world” (Clarke and Cortner 2002, 186). That same year the Everglades were designated a national park. But at the very moment that Douglas and park officials were promoting understanding about the world’s only Everglades, the federal government was assisting local economic interests in developing Florida for intensive agricultural production and for population growth. Throughout the 1940s and 1950s the Army Corps of Engineers built a system of canals, levees, pumping stations, and drainage ditches that effectively choked off the Everglades’ supply of fresh water from Lake Okeechobee. Gradually, the River of Grass began to die.

Towards the end of the 1980s, alarm over the possible loss of the Everglades prompted the Bush Administration and Congress, acting in cooperation with Florida officials, to embark on a massive restoration project. With the Clinton White House strongly supporting it, the


project gained momentum through the 1990s. In 2000 Congress passed a seminal piece of legislation, The Everglades Restoration Act, which authorized the expenditure of nearly eight billion cost-shared dollars over the next fifty years to be spent on seventy different projects aimed at restoring the flow of clean water to the Everglades. In its breadth, complexity, and vision, the Everglades Restoration Project rivals the regional water projects of the 1930s, projects such as the Tennessee Valley Authority and the Columbia River Basin Plan. The restoration of the Everglades is a goal that continues to enjoy broad political support in 2008, although with a far less federal investment than the environmental community had hoped (Grunwald, 2006).

But will it work? Although a former Chief of Engineers was sanguine about what he characterized as simply a large re-plumbing project on the part of the Army Corps, (Corps 1995) a number of scientists and policy scholars have their doubts. Not only is the science of ecological restoration in its infancy, there continues to be formidable political obstacles to restoring the Everglades. These include Florida’s powerful sugar interests (New York Times 2003). Political scientist William Lowry summarizes many of the problems confronting the Everglades effort in his recent book, Dam Politics: Restoring America’s Rivers:

“This legislation [Everglades Restoration Act of 2000] may have signaled a new era in

the Everglades, but funding arrangements require Florida and the Corps to go back to Congress every two years for authorization of additional projects and every year for implementation money…This resolution may yet work, but to this point the Everglades restoration effort has only produced minor, disjointed changes... During 2002, the National Academy of Sciences and numerous other analysts raised significant doubts as to the ultimate impact of ongoing restoration efforts…” (Lowry 2003). In response to the federal government’s slow progress in the region, the state of Florida

recently moved to accelerate project implementation. With its ACCELER8 the state has “stepped up the pace” on eight restoration projects that include reservoirs, stormwater treatment areas, and wetland restoration (SFWMD 2008). The projects were selected on their potential to show immediate benefits but also where the necessary lands were already in public ownership. They illustrate a general dissatisfaction and impatience with the federal process and according to the South Florida Water Management District, they are designed to serve as the “initial foundation for other comprehensive restoration efforts to follow” (Day 2005). Construction has begun on the Everglades Agricultural Area Reservoir, designed to protect coastal estuaries and reduce the flow of nutrients into the Everglades by capturing and storing agricultural stormwater runoff and freshwater release from Lake Okeechobee (Florida DEP 2005). Environmentalists argue that the projects are all water-supply projects and the environmental components to restoration have essentially been abandoned under the present state-led restoration effort (Economist 2005). Other critiques point to implementation concerns related to uncertainties in technologies and costs, project prioritization, and restoration effectiveness (Sheikh and Carter 2006). Most recently, the state of Florida signed an historic deal with U.S. sugar to purchase over 187,000 acres of farmland just north of Everglades National Park, substantially expanding the land under protection and connecting Lake Okeechobee to the Park (Cave 2008).

It is a sobering thought that we may not be able to save the Everglades, despite spending billions of dollars and applying the most up-to-date science on restoring them. The world then


will have lost one of its most unique ecosystems. So, what is the lesson from the Everglades? It seems obvious. As Barry Commoner argued over a decade ago, it is simpler and in the long run cheaper to prevent something bad from happening than from trying to clean up after the fact (Commoner 1990). Commoner was writing about pollution prevention as opposed to pollution control, but the argument applies equally well to natural ecosystems. To the extent that the national No-Net-Loss policy allows for, or even encourages through trade-offs, the further loss of America’s remaining original wetlands, it is not the best policy. As the Everglades case illustrates, our first priority must be preserving and protecting what is left of what once was a vast network of natural wetlands that helped to cleanse the continent. And establishing that clear policy resides with the Congress when, as it must, revisit the Clean Water Act.

REFERENCES

Achenbach, Joel. Iowa Flooidng Could be an Act of Man. Washington Post (June 19, 2008). Allen A.O. and J.J. Feddema. 1996. Wetland Loss and Substitution by the Section 404 Permit

Program in Southern California, USA. Environmental Management 20(2): 263-274. Alvayay, J. and J.S. Baen. 1990. The Implications of Federal Wetlands Protection Programs

for the Real Estate Appraisal Industry. Journal of Real Estate Research 5(1):153-165. Association of State Wetland Managers. 2006. State Wetland Programs, on-line at

http://www.aswm.org/swp/index.htm, last visited February 25, 2007.Barringer, F. 2006. Reach of the Clean Water Act Is at Issue in 2 Supreme Court Cases This Week. New York Times, February 20, 2006, A8.

Bianucci, L.R. and R.R. Goodenow. 1991. The Impact of Section 404 of the Clean Water Act on Agricultural Land Use. UCLA Journal of Environmental Law and Policy 10:41-65.

Bishop, T. 2006. Special Projects Coordinator, Wildlife Bureau, Iowa Department of Natural Resources. Personal Communication, March 10, 2006.

Bourne, J. K. 2004. Gone with the Water. National Geographic Magazine (October 2004), available at http://magma.nationalgeographic.com/ngm/0410/feature5/?fs=www3. nationalgeographic.com, last accessed March 10, 2007.

Brown, P.H. and C.L. Lant. 1999. The Effect of Wetland Mitigation Banking on the Achievement of No-Net-Loss. Environmental Management 23(3):333-345.

Cave, Damien. 2008. Florida Buying Big Sugar Tract for Everglades. New York Times (June 25, 2008).

Clarke, J.N. and H.J. Cortner. 2002. The State and Nature: Voices Heard, Voices Unheard in America’s Environmental Dialogue. Upper Saddle River, N.J.: Prentice Hall.

Cole, C.A. and D. Shafer. 2002. Section 404 Wetland Mitigation and Permit Success Criteria in Pennsylvania, USA, 1986-1999. Environmental Management 30(4):508-515.

Constanza, R., R. d’Arge, R. de Groot, S. Farber, M. Grasso, B. Hannon, K. Limburg, S. Nameem, R.V. O’Neill, J. Paruelo, R.G. Raskin, P. Sutton, and M.van der Belt. 1997. The Value of the World’s Ecosystem Services and Natural Capital. Nature 387:253-260.

Council on Environmental Quality. 2005. Executive Office of the President, Conserving America’s Wetlands, Implementing the President’s Goal. Washington, DC: Coastal America.


Dahl, T.E. 1990. Wetlands Losses in the United States 1780s to 1980s. Washington, DC: U.S. Department of the Interior, Fish and Wildlife Service.

Dahl, T.E. 2000. Status and Trends of Wetlands in the Conterminous United States, 1986-1997. Washington, DC: United States Department of the Interior, Fish and Wildlife Service.

Dahl, T.E. 2006. Status and Trends of Wetlands in the Conterminous United States, 1998-2004 Washington, DC: U.S. Fish and Wildlife Service.

Dale, L. and A.K. Gerlak. 2007. It’s All in the Numbers: Acreage Tallies and Environmental Program Evaluation, Environmental Management 39: 246-260.

Davis, M.L. 1997. U.S. Army Corps of Engineers, Deputy Assistant Secretary of the Army (Civil Works). Testimony before the U.S. House of Representatives Transportation and Infrastructure Committee, Subcommittee on Water Resources and Environment (April 29, 1997).

Day, M. 2005. Everglades Restoration: Stepping Up the Pace. South Florida Water Management District, ACCELER8 Everglades Now Project (July 18, 2005), available at http://xlr8.sfwamd.gov/portal/page?_pageid=35,224274and_dad=portaland_schema=Portal (last visited July 20, 2005).

The Economist. 2005. Water, bird and man (October 6, 2005), available at http://www.economist.com/world/na/displaystory.cfm?story_id=4484300 (accessed January 30, 2007).

Eilperin, Juliet. 2008. EPA Enforcement is Faulted: Agency Official Cites Narrow Reading of Clean Water Act. Washington Post (July 8, 1008).

Environment Opinion Survey. 1991. Second National Environmental Survey. Washington, DC: Market Strategies, June 1, 1991.

Environmental New Service. 2007. Poll: Iowa Sportsmen Say Global Warming Must Be Curbed. November 2, 2007. Available at http://www.ens-newswire.com/ens/nov2007/2007-11-02-092.asp (last visited February 211, 2008).

Florida Department of Environmental Protection. 2005a. About the Everglades, available at http://www.dep.state.fl.us/secretary/everglades/about.htm (last visited July 20, 2005).

Florida Department of Environmental Protection 2005b. Florida Breaks Ground on First ACCERLER8 Project, Press Release (January 14, 2005), available at http://www.dep.state.fl.us/secretary/news/2005/01/0114_02.htm (last visited March 27, 2007).

Gates, P.W. 1968. History of Public Land Law Development. Washington, DC: Public Land Law Review Commission.

Grunwald, Michael. 2006. The Swamp: The Everglades, Florida, and the Politics of Paradise. New York: Simon and Schuster.

Heimlich, R.E. K.D. Wiebe, R. Claassen, D. Gadsby, and R.M. House. 1998. Wetlands and Agriculture: Private Interests and Public Benefits. Washington, DC: United States Department of Agriculture, Economic Research Service.

Hornyak, M.M. and K.E. Halvorsen. 2003. Wetland Mitigation Compliance in the Western Upper Peninsula of Michigan. Environmental Management, published online: 26 November 2003.

Izzo, D. 2002. Principal Deputy Assistant Secretary of the Army for Civil Works, Statement before the Subcommittee on Energy Policy, Natural Resources and Regulatory Affairs


Committee on House Government Reform, U.S. House of Representatives (September 19, 2002).

Kaiser, J. 2001. Recreated Wetlands No Match for Original. Science 6 (July 2001), 233. Knickerbocker, B. 2005. New Drive to Save Wetlands. Christian Science Monitor (October

19, 2005), available at http://www.csmonitor.com/2005/1019/p03s01-sten.htm, accessed March 27, 2007.

Lemonick, M.D. 1991. War over the Wetlands. Time, (August 26). Lieber, H. 1975. Federalism and Clean Waters: The 1972 Water Pollution Control Act.

Lexington, KY: Lexington Books. Lindblom, C. 1959. The Science of Muddling Through. Public Administration Review 19:79-

88. Lowry, W.R. 2003. Dam Politics: Restoring America’s Rivers. Washington, DC: Georgetown

University Press. Malakoff, D. 1998. Restored wetlands flunk real-world test. Science 280 (April 17):371-372. Meindl, C.F. 2000. Past Perceptions of the Great American Wetlands: Florida’s Everglades

during the Early Twentieth Century. Environmental History 5:378-394. Meyer, W.B. 1994. When Dismal Swamps Become Priceless Wetlands. American Heritage

(May/June). Mitchell, J.G. 1992. Our Disappearing Wetlands. National Geographic 182(4):13-14. Mitsch, W.J. X. Wu, R.W. Nairn, P.E. Weihe, N. Wang, R. Deal, and CE. Boucher. 1998.

Creating and Restoring Wetlands: A whole-ecosystem experiment in self-design. Bioscience 48:1019-1030.

National Research Council. 1995. Wetlands: Characteristics and Boundaries. Washington, DC: National Academy Press.

National Research Council. 2001.Compensating for Wetland Losses Under the Clean Water Act. Washington, DC, National Academy Press.

National Wetlands Coalition at http://www.thenwc.org/home.htm, accessed July 31, 2004. National Wetlands Newsletter. 1993. Congress. Washington, DC: Environmental Law

Institute, January-February:17-19. National Wetlands Newsletter. 2001. The SWANCC Decision. Washington, DC:

Environmental Law Institute, January-February:2. National Wildlife Federation. 1989. Attitudes of the American Public Toward Wetlands and

Waterfowl. Washington, DC: Market Research Division, September 1. “Everglades in Peril,” The New York Times (April 21, 2003), A24.

Pianin, E. 2003a. Administration Establishes New Wetlands Guidelines. Washington Post (January 11):A05.

Pianin, E. 2003b. EPA Scraps Changes to Clean Water Act: Plans Would Have Reduced Protection. Washington Post (December 17):A20.

Pope, C. 1991. That Question of Balance: George Bush and Wetlands Preservation. Sierra (November-December):22-24.

Pressman, J.L. and A. Wildavsky. 1973. Implementation: How Great Expectations in Washington are Dashed in Oakland. Berkeley, CA: University of California Press.

Richey, W. 2006. Supreme Court Splits Over Protecting Wetlands. Christian Science Monitor, June 20, 2006 at http://www.csmonitor.com/2006/0620/p01s01-usju.html?s=t5, accessed March 23, 2007).


Sanger, D.E. and D.M. Halbfinger. 2004. For Earth Day, Bush and Kerry Vie on Environment. New York Times (April 23, 2004).

South Florida Water Management District, ACCELER8 Everglades Now! available at https://my.sfwmd.gov/portal/page?_pageid=1855,2830547,1855_2831083and_dad=portaland_schema=PORTALandnavpage=home, last visited February 11, 2008).

Sheikh, P. and N.T. Carter. 2006. South Florida Ecosystem Restoration and the comprehensive Everglades Restoration Plan. Washington, DC: Congressional Research Service, January 11, 2006.

Steinhart, P. 1993. Mud Wrestling. Sierra (January-February):52-59; 148-150. Sudol, M.F. and R.F. Ambrose. 2002. The US Clean Water Act and Habitat Replacement:

Evaluation of Mitigation Sites in Orange County, California, USA. Environmental Management 30(5):727-734.

U.S. Advisory Committee on Intergovernmental Relations. 1992. Intergovernmental Decisionmaking for Environmental Protection and Public Works. Washington, DC: U.S. Advisory Committee on Intergovernmental Relations.

U.S. Army Corps of Engineers. 1995. Interview with Chief of Engineers. Washington, DC, May.

U.S. Army Corps of Engineers. 2006. Regulatory Program, Regulatory Statistics: FY 2003 at http://www.usace.army.mil/inet/functions/cw/cecwo/reg/2003webcharts.pdf, accessed May 24, 2006.

U.S. Department of Defense and Environmental Protection Agency. 2006. Compensatory Mitigation for Losses of Aquatic Resources. Federal Register, Volume 71, Number 59 (March 28, 2006), available at http://www.epa.gov/owow/wetlands/pdf/MitRule NPRM.pdf, accessed May 25, 2006).

U.S. Environmental Protection Agency. 1999. Facts about Wetlands. Available at http://www.epa.gov/OWOW/wetlands/wetland2.html, accessed July 21, 1999.

U.S. Environmental Protection Agency. 2003. Administration to Reaffirm Commitment to No Net Loss of Wetlands and Address Approach for Protecting Isolated Waters in Light of Supreme Court Ruling on Jurisdictional Issues. Washington, DC, U.S. EPA, January 10, 2003 at http://www.epa.gov/epahome/headline2_011003.htm, accessed January 15, 2003.

U.S. General Accounting Office. 1998. Wetlands Overview: Problems with Acreage Data Persist. Washington, DC: General Accounting Office.

U.S. General Accounting Office. 2004. Waters and Wetlands: Corps of Engineers Needs to Evaluate Its District Office Practices in Determining Jurisdiction. Washington, DC: Government Accountability Office. GAO-04-297.

U.S. General Accounting Office. 2001. Assessments Needed to Determine the Effectiveness of In-Lieu Mitigation. Washington, DC: General Accounting Office. GAO-01-325.

Vileisis, A. 1997. Discovering the Unknown Landscape: A History of America’s Wetlands. Washington, DC: Island Press.

White House Office of Environmental Policy. 1993. Protecting America’s Wetlands: A Fair and Flexible Approach. Washington, DC: White House Office on Environmental Policy, August 24).

Wilen, B.O. 1995. The Folly of the Numbers Game. National Wetlands Newsletter, May-June 1995, 8-10.

Zinn, J.A. and C. Copeland. 1993. Wetlands Issues in the 103rd Congress. Washington, DC, Library of Congress, Congressional Research Service, March 23, 1993.


Chapter 4

DYNAMICS OF COASTAL WETLANDS AND LAND USE CHANGES IN THE WATERSHED:

IMPLICATIONS FOR THE BIODIVERSITY

Miguel Ángel Esteve, M. Francisca Carreño, Francisco Robledano, Julia Martínez-Fernández and Jesús Miñano

Departamento de Ecología e Hidrología, Universidad de Murcia Campus de Espinardo, 30100-Murcia (Spain)

ABSTRACT

The Mediterranean coastal landscapes have suffered significant changes along the last decades due to the agricultural intensification and tourist development. Such changes have modified the water flows and specifically the hydrological regime of wetlands, as has occurred in the Mar Menor (Southeast Spain). The Mar Menor coastal lagoon and associated wetlands present noticeable ecological and biodiversity values. However, the land-use changes in the watershed and the consequent changes in the water and nutrient flows along the period 1980-2005 are threatening the conservation of these wetlands. A dynamic model has been developed to simulate the key environmental and socio-economic factors driving the export of nutrients to the Mar Menor lagoon and associated wetlands, where some eutrophication processes have appeared.

In the present chapter the changes in the vegetal and faunistic assemblages are analysed. Vegetal communities are studied by means of remote sensing techniques, which have provided information about the changes in area and habitat composition of the wetlands along the considered period. This has shown that the habitats more negatively affected by the hydrological changes are those most threatened in the international context and with a highest interest from the point of view of biodiversity conservation. It has also been possible to verify the direct relationships between all these changes at wetlands scale and the agricultural changes at the watershed scale.

Two faunistic communities especially sensitive to these ecosystemic changes have also been studied: i) Wandering beetles and ii) Birds (waterbirds and steppe passerines). Wandering beetles (Coleoptera) were studied with pitfall traps in 1984, 1992 and 2003 and steppe passeriforms with line transects in several years along the period. In both communities evident changes have been observed. Regarding beetles, the most

Miguel Ángel Esteve, M. Francisca Carreño, Francisco Robledano et al. 134

halophilous species have been favoured, some of them especially relevant due to its rarity in the European context. The ratio Carabidae/Tenebrionidae has shown to be a good indicator of the hydrological changes of the wetlands. Waterbirds have shown dramatic changes in their relative abundances within the lagoon, with a long-term decline in the most characteristic original species, increases in generalist piscivores and a recent appearance and rapid growth of the herbivores guild. In the case of steppe passeriforms, this community has been negatively affected, especially some species like Melanocorypha calandra. The family Alaudidae has lost importance to the benefit of the families Turdidae and Fringillidae. These changes can be considered a loss of value in relation with the original passeriform community, since the wetland qualifies as a Specially Protected Area under the EU’s Bird Directive, precisely on the basis of its genuine steppe bird assemblage.

In conclusion the changes at wetlands scale clearly reflect the hydrological modifications at the watershed scale and have significant effects on the most characteristic biodiversity of the wetlands of coastal arid systems.

INTRODUCTION The importance of wetlands is increasingly recognised as systems supporting an specific

and valuable biodiversity and as areas playing a key role in essential ecological functions such as the control of nutrient flows and the removal of diffuse pollution at the landscape scale. This has promoted different protection and conservation strategies which in the context of arid landscapes, as those existing in Southeastern Spain have an especial importance due to the singularity and key role of wetlands in such arid systems. However, conventional protection and conservation strategies usually do not take into account the close dependency of wetlands on the dynamics and management outside the protected area and this may interfere on the protection and conservation goals. Land use and management practices at watershed scale affect the wetlands in many ways and one of the most relevant ones is linked to the water and nutrients flows entering into the wetlands from the watershed. To what extent do wetlands react to land use changes in the watershed?; Can these changes be tracked in different ecological compartments of the wetlands?; Do that changes draw a similar pattern?. We have tried to answer some of such questions studying a complex of wetlands in a arid landscape which have suffered important long-term land use changes: The Mar Menor lagoon and associated wetlands.

The Mar Menor lagoon is a hypersaline Mediterranean coastal lagoon located in Southeast Spain. Ramsar Site since 1994, it is the largest water surface of the western Mediterranean coast (135 km2 surface area and a 580 hm³ volume), and a remarkable biodiversity and scientific resource (EU Bird Specially Protected Area and Barcelona Convention’s Specially Protected Areas of Mediterranean Importance since 2001). It is almost closed by a sand bar 22 km long with a width varying between 100 and 1,200 m, with a very narrow connection with the open sea.

Associated to its internal shore there is a series of coastal wetlands, Marina del Carmolí, Playa de la Hita and Saladar de lo Poyo (figure 1), which are protected at both national and international level (Ramsar site and Site of Community Importance for the Natura-2000 Network) due to their natural and ecological interest. They include several natural habitats of

Dynamics of Coastal Wetlands and Land Use Changes in the Watershed 135

priority and community interest according to the Habitats Directive. They are defined as coastal crypto-wetlands (Vidal-Abarca et al., 2003).

Figure 1. Location of Mar Menor wetlands. PH: Playa de la Hita; MC: Marina del Carmolí; LP: Saladar de Lo Poyo.

The Mar Menor watershed is a 1,270 km² plain slightly inclined towards the lagoon and drained by several ephemeral watercourses (ramblas), which flow into the lagoon after episodic storm rainfall events. The area has a Mediterranean arid climate, with warm winters, an annual mean temperature about 17 ºC, annual mean rainfall of 330 mm and a high inter-annual rainfall variation. Agriculture is the predominant land-use in the Mar Menor watershed.

Urban changes, tourist development and especially the spread of irrigated lands favoured by the opening of the Tagus-Segura water transfer system in 1979, have led to a significant increase in the water and nutrient flows reaching the Mar Menor lagoon-associated wetlands complex. In the wetlands this has caused changes which are relevant at watershed scale, due to the role of the active wetland area in the removal of nutrients from diffuse sources, and also at wetland scale, due to the effects on biodiversity. Understanding the dynamics of change is therefore important for conserving the biodiversity of the wetlands and for the sustainable management of the Mar Menor lagoon and watershed.

Previous studies on Mar Menor wetlands have focused on conservation and management (Robledano et al., 1987, 1991a, 1991b, Robledano and Esteve, 1992; Ortega et al., 1992), on restoration (Robledano, 1995) and on specific topics such as heavy metals in Lo Poyo wetland (Álvarez-Rogel et al., 2002b, 2004) and nutrients (Álvarez-Rogel et al., 2002a, 2006, 2007a).


In the present chapter the changes in the vegetal and faunistic assemblages of Mar Menor wetlands are studied to analyse their responses to the land use and hydrological changes at watershed scale, the implications in terms of biodiversity and conservation value and to assess potential indicators in the analysed assemblages of the long-term watershed changes. It has been studied the vegetal communities in the Mar Menor wetlands and two faunistic communities especially sensitive to the ecosystemic changes: wandering beetles and birds (waterbirds and steppe passerines). These communities have been studied at specific spatial scales and study areas: vegetal communities were studied in three wetlands associated to the Mar Menor shore (Playa de la Hita, Marina del Carmolí and Saladar de Lo Poyo); wandering beetles and steppe passerines were analysed in the Marina del Carmolí wetland whereas waterbirds were surveyed in the whole Mar Menor lagoon.

LAND USE CHANGES AND THEIR EFFECTS ON HABITATS OF COASTAL WETLANDS

Study Area and Methodology

Three Mar Menor wetlands, Playa de la Hita, Marina del Carmolí and Saladar de Lo

Poyo, have been studied to analyse the temporal and spatial changes in the area and in the internal composition of such wetlands from 1984 to 2001, by means of remote sensing, in order to analyse their implications, especially as regards the application of the Habitats Directive (92/43/CEE).

All studied wetlands present salt steppes, salt marshes, reedbeds and sandy areas, although with a different relative importance. Following the typology of the Habitats Directive (92/43/CEE), the sandy areas unit is mostly composed of the following habitats: 1210 “Annual vegetation of drift lines” and 2210 “Fixed beach dunes with Crucianellion maritimae”. The salt steppe unit is 95% composed of the priority habitat 1510 “Mediterranean salt steppes, Limonietalia”, whereas the remaining 5% comprises habitat 1430 (Halo-nitrophilous scrubs (Pegano-Salsoletea) and 92D0 (Southern riparian galleries and thickets). Main species in salt steppe are Lygeum spartum, Suaeda vera, Frankenia corymbosa and Limonium caesium. The salt marsh unit is dominated by habitat 1420 (Mediterranean and thermo-Atlantic halophilous scrubs, Sarcocornetea fruticosi), although there are also small patches of habitat 1410 (Mediterranean salt meadows). Main species in salt marsh are Sarcocornia fruticosa, Arthrocnemum macrostachyum, Halimione portulacoides and Limonium cossonianum. Finally the reedbeds unit is dominated by Phragmites australis. All habitats are designated as being of Community Interest with the exception of habitat 1510 (Mediterranean salt steppes), which is designated as of Priority Interest. Reedbeds are not included in the Habitats Directive.

Remote sensing has been extensively used in wetlands studies and inventories (Noriega and Lozano-García, 2000), in the detection of water bodies and vegetation in wetlands (Toÿrä et al., 2001), in the elaboration of land use-land cover and vegetation maps (Cihlar et al., 1996; Michelson et al, 2000; Hess et al., 2003; Wang y Tenhunen, 2004; Kumar Joshi et al., 2006), including those of wetlands (Wang et al., 2007), in the detection of land use-land cover


changes (Narumalani et al., 2004; Cakir et al., 2006; Mundia and Aniya, 2006) and in the analysis of hydrological and land cover changes in wetlands (McHugh et al., 2007).

Land cover maps for the three wetlands were obtained by photo interpretation of aerial photographs and supervised classification using the widely used maximum likelihood algorithm (Michelson et al., 2000; Richards, 1995). Landsat images sensors TM and ETM+ covering five years of the period 1984 to 2001 (years 1984, 1992, 1995, 1997 and 2001) were used. Each classification was carried out with two images (summer and winter) and the NDVI (Normalised Difference Vegetation Index) for each image. The images were processed with GRASS (Geographic Resources Analysis Support System, http://grass.itc.it) an open GIS system under Linux. The methodology was verified by cross-validation using a stratified random sampling. Overall accuracy (percentage of sampled pixels which are well classified) reaches 85% in Marina del Carmolí and Playa de la Hita and 89% in Saladar de Lo Poyo.

The following land cover classes were identified (Carreño et al., 2008): natural vegetation (three units: salt steppe, salt marsh and reedbeds); agricultural fields, water bodies (mainly portions of the Mar Menor lagoon); bare ground (inactive ponds in saltworks, river beds in ephemeral channels) and infrastructures (urban settlements, roads, rubbish tips). The narrow strips of sandy areas are included in the nearest vegetal units. These maps have allowed the tracking of the changes in the area and in the internal composition of wetlands between 1984 and 2001. Data were analysed under the statistical package R (The R Foundation for Statistical Computing, 2005, http://www.r-project.org).

Field data on 2003 and 2004 of soil moisture and conductivity in 35 sample units of the whole set of Mar Menor wetlands were also available and were used in this work.

Results and Discussion

Playa de la Hita Wetland Figure 2 shows the land cover maps of Playa de la Hita obtained by supervised

classification of Landsat TM and ETM+ images between 1984 and 2001. In Playa de la Hita there was a slight reduction in the area of natural vegetation due to the construction of infrastructures (table 1). Regarding its internal composition, the main changes among habitats are the conversion of 29% of salt steppe into salt marsh and the change of 16% of salt marsh into reedbed.

Table 1. Matrix of land use change in Playa de la Hita wetland 1984-2001

2001

Land use/ land cover (ha)

Salt steppe

Salt marsh

Reebed

Crops

Infra-structures

Water bodies

Total 1984 (ha)

Salt steppe 0.88 1.63 0.00 0.06 3.00 0.00 5.57 Salt marsh 0.13 13.94 3.50 0.06 3.88 0.00 21.51 Reedbed 0.00 0.13 2.31 0.00 0.00 0.25 2.69 Crops 0.13 1.13 0.06 0.00 2.19 0.00 3.51 Infrastructures 0.06 0.00 0.13 0.00 4.75 0.00 4.94 Water bodies 0.00 0.00 1.19 0.00 0.00 8.63 9.82

1984 Total 2001 (ha) 1.20 16.83 7.19 0.12 13.82 8.88


Figure 2. Land cover maps of Playa de la Hita wetland from 1984 to 2001 obtained by supervised classification of Landsat TM and ETM+ images.

Figure 3. Area of salt steppe, salt marsh and reedbed in Playa de la Hita wetland from 1984 to 2001.

Taking into account all changes, salt steppe in 2001 had lost 78% of its area in 1984; salt marsh had also lost area, although in a lesser degree (22%), whereas reedbed had increased by 167%. These trends of change are specially marked since 1995 (figure 3).


In overall, all these changes constitutes a loss of value from the Habitats Directive point of view, since it represents a significant loss of a Priority Interest habitat (salt steppe) and a loss in a Community Interest habitat (salt marsh) to the benefit of reedbed, not included in the Directive.

Saladar de Lo Poyo Wetland

Figure 4 shows the land cover maps of Lo Poyo in 1984 and 2001.

Figure 4. Land cover maps of Saladar de Lo Poyo wetland in 1984 and 2001 obtained by supervised classification of Lansat TM and ETM+ images.

From 1984 to 2001 there is a slight increase in the total area of the Saladar de Lo Poyo wetland due to the increase in reedbed. The two habitats present in Saladar de lo Poyo wetland (salt marsh and reedbed) shift in relative dominance between 1984 and 2001 (table 2): salt marsh occupies in 1984 a 63 % of total habitats area, whereas in 2001 66% of total habitats area is occupied by reedbed. These trends of change, similarly to the case of Playa de la Hita, also point to a loss of value from the Habitats Directive point of view.

Table 2. Matrix of land use change in Saladar de Lo Poyo wetland 1984-2001

2001

Land use/ Land cover

Salt marsh Reedbed

Bare ground Crops Infrastructures

Water bodies

Total 1984 (ha)

Salt marsh 14.19 7.06 11.81 6.69 6.19 0.00 45.94 Reedbed 1.94 11.56 2.25 6.88 3.88 0.00 26.50 Bare ground 7.06 7.31 130.56 20.56 5.19 3.63 174.31 Crops 5.00 28.13 12.06 139.25 26.00 0.00 210.44 Infrastructures 0.38 0.69 5.38 6.44 15.75 0.00 28.63 Water bodies 0.00 0.00 3.38 0.00 0.00 237.50 376.88

1984

Total2001 (ha) 28.56 54.75 165.44 179.81 57.00 377.13

Marina del Carmolí Wetland Figure 5 shows the land cover maps of Marina del Carmolí between 1984 and 2001. Total

wetland area increased by 12% between 1984 and 2001 and the habitat composition suffered a dramatic change. Marina del Carmolí was basically a salt steppe of 237 ha in 1981 (table 3) whereas in 2001 this habitat had lost half of its initial area and salt marsh and reedbed,


practically absent in 1984, occupy a significant extension (82 and 64 ha, respectively). Again, these trends of change imply a loss of value from the Habitats Directive point of view.

Figure 5. Land cover maps in Marina del Carmolí wetland from 1984 to 2001 obtained by supervised classification of Landsat TM and EMT+ images.

The pattern of change of the Marina del Carmolí habitats along time (figure 6) suggest two different periods. Between 1984 and 1995 there is a significant loss of salt steppe to the benefit of salt marsh. Between 1995 and 2001 the area of salt steppe is stabilised whereas there is a reduction of salt marsh associated to a steadily increase in reedbed. Therefore, the conversion of part of the initial salt steppe into reedbed, shown in table 1, is mediated by an intermediate stage as salt marsh.

Table 3. Matrix of land use change in Marina del Carmolí wetland 1984-2001

2001

Land use/ Land cover

Salt steppe

Salt marsh Reedbed

Bare ground Crops

Infrastruc-tures

Water bodies

Total 1984

Salt steppe 91.75 49.79 23.13 10.19 41.94 20.44 0.00 237.22Salt marsh 0.00 2.03 0.00 0.00 0.00 0.00 0.00 2.03 Reedbed 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 Bare ground 18.00 12.25 29.63 10.75 11.06 20.25 0.19 102.13Crops 9.63 17.25 10.44 5.25 38.31 8.50 0.00 89.37 Infrastructures 2.81 0.25 0.94 9.06 6.06 52.31 0.75 72.19 Water bodies 0.00 0.00 0.00 1.31 0.00 5.44 295.44 302.19

1984 Total 2001 (ha) 122.19 81.56 64.13 36.56 97.37 106.94 296.38

The relative changes between salt steppe, salt marsh and reedbed might be explained by

the interaction between the soil moisture and conductivity gradients, as show data taken in 2003 and 2004 in the Rambla del Miedo area, one of the ephemeral channels entering into


Marina del Carmolí (figure 7). The regression model (R2adj.= 0.78; p<0.001) shows that

conductivity presents a quadratic response to soil moisture: at low values of soil moisture, conductivity increases with higher water content until a certain threshold, around 30% of soil moisture, above which conductivity decreases as soil moisture increases.

Figure 6. Area of salt steppe, salt marsh and reedbed in Marina del Carmolí wetland from 1984 to 2001.

Figure 7. Data and regression model of soil moisture respect to conductivity in the Rambla del Miedo area, in Marina del Carmolí.


Figure 8 shows the spatial expression of this complex gradient along a transect crossing Marina del Carmolí in direction NW-SE, with the first and last samples located in the boundaries of the wetland. Samples with lower values of soil moisture, located far from the Miedo and Miranda ephemeral channels, present positive correlation between soil moisture and conductivity, whereas the samples located close to such watercourses, with higher soil moisture, presents a negative correlation between both variables.

Figure 8. Soil moisture and conductivity along a transect crossing the Miedo and Miranda ephemeral channels, in Marina del Carmolí. Location of Miedo (MIE) and Miranda (MIR) ephemeral channels are indicated.

In synthesis, the initial increase in the water flows affecting Marina del Carmolí may have favoured higher soil moisture and therefore higher conductivity, what might have caused the increase in salt marsh at the expense of salt steppe. At a later stage beginning around 1995, greater water inputs would have caused a decrease in conductivity and allowed the extension of reedbed, as shown in figure 6. To this process may have also contributed, although with secondary importance, the inputs of urban wastewater into the ephemeral channels reaching Marina del Carmolí and some channelling works in the watercourses, which favour the spread of reedbed (Carreño et al., 2008).

Although conductivity and soil moisture data before 2003 are not available, we have used the 2003-2004 data and the area occupied by each habitat between 1984 and 2001, obtained from the corresponding land cover maps, to estimate the change along time of these variables in Marina del Carmolí. Figure 2 shows the characterisation of the habitats in the wetlands (Playa de la Hita, Saladar de Lo Poyo and Marina del Carmolí) in terms of soil moisture and conductivity, respectively. The salt marsh presents the highest values of conductivity although the ranges of the three habitats partially overlap. On the contrary, salt steppe, salt marsh and reedbed are clearly differentiated along a gradient of increasing soil moisture.

Figure 10 presents a different illustration of the explained process. It has been calculated the weighted average of cuantile 25 of conductivity and soil moisture in Marina del Carmolí for each year, using the value of this parameter in the salt steppe, salt marsh and reedbed in


2003-2004 (figure 9) and the relative proportion of each habitat in Marina del Carmolí from 1984 to 2001, derived from the corresponding land cover maps. Results show a pattern coherent with the complex gradient between soil moisture and conductivity shown in figures 7 and 8. As shown in figure 10, the estimated quantile 25 of soil moisture and conductivity in Marina del Carmolí increases from 1984 to 1995, after which the further increase in soil moisture is accompanied by a decrease in conductivity, period in which the expansion of reedbed takes place.

In synthesis, the increase in agricultural drainages has caused important changes in the wetlands associated to the Mar Menor lagoon, both in total area and internal composition. There is an increase in 35 ha in natural vegetation (salt steppe, salt marsh and reedbeds), which occurred in Marina del Carmolí and Saladar de Lo Poyo, whereas in Playa de la Hita there was a slight reduction in the area of natural vegetation due to the construction of infrastructures.

Figure 9. Ranges of conductivity (left) and soil moisture (right) of habitats in the Mar Menor wetlands. Boxes represent the cuantiles 25, 50 (median) and 75.

Figure 10. Estimated change in conductivity and soil moisture in Marina del Carmolí wetland between 1984 and 2003 (see text for details).


Regarding the internal composition, the habitats with higher requirements of soil moisture (salt marsh and reedbeds) have doubled and quadrupled their area, respectively, while the salt steppe has decreased to a half. The expansion of reedbed associated with shallower soil water conditions, has also been reported in other studies in the Marina del Carmolí wetland (Alvarez-Rogel et al., 2007b).

The net loss of salt steppe is very relevant, since it is the habitat with the highest interest from the Habitats Directive point of view (Priority Interest). Moreover, salt steppe is a rare habitat with a total area in Spain of only 12,976 ha, of which no more than 37% presents a good conservation state (Esteve and Calvo, 2000). Therefore, any reduction in the area of this habitat constitutes a significant loss, especially taking into account that in Murcia province, where the Mar Menor is located, the conservation state of this priority habitat is well over the average in Spain, with 83% of salt steppes in good conservation status (Esteve and Calvo, 2000).

It has been calculated an index to quantify the changes in overall value of wetlands from the Habitats Directive point of view along the period 1984-2001, as weighted average taking into account the area of each habitat and assigning the values 2, 1 and 0 to the Priority habitat (salt steppe), Community interest habitat (salt marsh) and rest, respectively. Figure 11 shows the sustained decrease (reduction in the index around 38%) in the overall value of the wetlands from the Habitats Directive point of view.

Land use changes at watershed scale are the primary factor explaining the described changes in the habitats of Mar Menor wetlands. The expansion of irrigated lands (Martínez et al., 2005; Velasco et al., 2006; Carreño et al., 2008) has doubled the irrigation water volume, causing the consequent increase in the drainage flows, part of which reach the Mar Menor wetlands. This has been confirmed by the rise in water tables in the aquifers of the watershed (ITGE, 1994; García Lázaro, 1995).

Figure 11. Value of the Mar Menor wetlands from the point of view of the Habitats Directive from 1984 to 2001, using the priority index (see text for details).


These changes have increased the levels of groundwater, flooding periods and soil water content in the wetlands, as reported in other field studies (Alvarez-Rogel et al., 2007b).

The role of the agricultural drainage coming from the watershed on the changes in the wetlands is supported by the close relationship between the sum of salt marsh and reedbeds in the wetlands and the area of irrigated lands in the watershed (Martinez and Esteve, 2002). Both variables show a similar pattern (Figure 12), characterised by sigmoid growth, which begins to slow down in 1991 in the case of irrigated lands and five years later in the case of salt marsh plus reedbeds. The regression analysis between both variables (figure 13) reaches a high significance (R2

adj. = 0.917; p = 0.01).

Figure 12. Temporal pattern of irrigated lands in Mar Menor watershed and area of habitats in the Mar Menor wetlands.

Figure 13. Regression model of the area of salt marsh and reedbed in the Mar Menor wetlands respect to the area of irrigated lands in the watershed.


When considering a five year time lag the relationship substantially improves (R2adj. =

0.945; p < 0.001), what might be interpreted as the time lag required by the surface and subsurface drainage flows to reach the wetland and by the habitats to respond to the increase in such water inputs.

It should be noted that the increase in the total wetland area does not appear to be a good indicator of the hydrological changes that have taken place in the watershed, since local conditions, in particular the spread of infrastructures like in Playa de la Hita, may limit the overall expansion of the wetland. On the contrary, the changes in the internal composition of each wetland towards more hygrophilous vegetation seems to be a good indicator of the hydrological changes at watershed scale, since there are no physical limitations to its expansion inside the wetland.

In synthesis, the long-term land use changes in the watershed, in particular the spread of irrigated lands and its effects on the hydrological dynamics, have important effects on the Mar Menor wetlands, of which the area occupied by salt marsh and reedbeds constitutes a good indicator. These changes implies a loss of value from the Habitats Directive point of view. In the next sections we try to analyse whether faunistic assemblages, in particular wandering beetles, steppe passerines and waterbirds, are also affected by the changes in the water regime of the watershed.

CHANGES IN WANDERING BEETLE ASSEMBLAGES

Introduction In this section we analyse if the described changes in the Mar Menor watershed have also

modified the beetles community of the wetlands, in particular in Marina del Carmolí. Carabidae and Tenebrionidae beetles are well known families that have frequently been

used in ecological studies (Dajoz, 2002). Beetles of the former family are known to show measurable responses to environmental disturbance and degradation (Desender et al., 1994; Brandmayr et al.., 2000; Rainio and Niemelä, 2003), thus complying with the definition of an ecological indicator as described by Niemi and McDonald (2004). Beetles are therefore widely used in studies on habitat conservation (Eyre and Luff, 2002). Tillage, pesticide treatments, harvesting and other agricultural practices can cause disturbances (Cole et al., 2002; Holland, 2002; Lövei and Sunderland, 1996; Serrano et al., 2005) and significant changes in beetle assemblages (Belaoussoff et al., 2003). However, little attention has been paid to changes in the assemblages of wetlands adjacent to cultivated areas as a result of fluctuations in water tables related to agricultural management practices. Fuellhaas (2000) demonstrated that a rise in the water table led to the recolonisation of wetlands by carabids with good dispersal capabilities. Similar results were reported by Främbs (1990) in North-European wetlands.

Tenebrionid beetles have also been used as environmental bioindicators (De los Santos, 1983), particularly in arid environments such as those commonly found in the Mediterranean Basin. The loss of these species in arid Mediterranean systems has been related to processes of environmental degradation (Cartagena and Galante, 2002). Some species with a wide ecological tolerance are able to colonise peripheral wetland habitats with varying degrees of


soil moisture (Bujalance et al., 1987; Giménez and Esteve, 1994). The combined study of beetles of both families may therefore be useful to assess changes caused by variations in the water table and soil salinity. To this end, De los Santos (1983) proposed an index based on the Carabidae/Tenebrionidae ratio to evaluate the effects of climatic factors on beetle assemblages in Mediterranean areas. In this work we have studied the Carabidae and Tenebrionidae families to analyse their response to changes in the water and salinity conditions of the wetlands.

Study Area and Methodology In year 2003 the wandering beetles assemblages in ten sites located in the set of Mar

Menor wetlands and peripheral areas were sampled with pitfall traps. Four sites, located in the Marina del Carmolí, were also sampled in 1984 with the same methodology, what allows a long-term survey in this wetland and an assessment of the changes in the beetles community along this period.

Three sampling sites were located in the inner parts of the wetland (W1, W2 and W3), while another site (P1) was located in peripheral areas close to the wetland (figure 14). This last site showed no marked vegetation changes during the study period (1984-2003) and so no significant changes in beetle assemblages were expected in this peripheral site.

The survey also included data on soil moisture and conductivity in 2003 in the four sampling sites. Table 4 summarises the plant communities, soil moisture and conductivity in the sampled years and sites.

In each year the sampling period lasted six months, from 1 April to 30 September since, according to Esteve (1987) and Giménez (1999), the period of maximum activity of most of these species in wetland areas corresponds to spring and summer.

Figure 14. Location of the sampling sites in Marina del Carmolí wetland. W1, W2, W3: wetland sites; P1: peripheral site.


Table 4. Details of the sampling sites of the Marina del Carmolí wetland (SE Spain). W= Sites deeply located in the wetland; P= peripheral site

Site 1984 2003 Soil moisture 2003 (%) Soil conductivity 2003 (dS/m) W-1 salt marsh salt marsh 29.78 7.28 W-2 salt steppe reedbeds 38.22 5.16 W-3 cultivated salt marsh 24.13 6.88 P-1 sands sands 3.15 0.16

This has also been found to be the case in other beetle studies in arid systems (Yaacobi et

al., 2007). Previous sampling campaigns in the area showed that any loss of information with regards to composition and abundance of Tenebrionidae and Carabidae in the Mar Menor wetlands is negligible. The traps consisted of plastic bottles of 1 litre capacity and 10 cm mouth diameter, filled with a saturated solution of NaCl. In each sampling site there were two rows of five traps each. Traps were emptied every 45 days (i.e., four times per year).

Beetles were identified to species level. The inventories and abundance data were used to calculate the species abundance per site and year. Population abundance is described through a capture-effort index consisting of the number of captures of each species per 10 active pitfalls during the six months sampling period. We calculated the logarithm of Carabidae/Tenebrionidae ratio and also the normalised Carabidae/Tenebrionidae ratio, which takes values between –1 and +1 (Pardo et al., in press), depending on the relative predominance of tenebrionidae or carabidae, respectively.

Wandering beetles were classified as halobionts, halophiles and other preferences, according to their adaptation to and tolerance of salt and water in the soil (Serrano et al., 2002). Briefly, halobionts are defined as species linked to salty soils that are rarely found in other environments. Halophiles are defined as those able to live in sites with a widely ranging salt content and thus must have efficient physiological mechanisms to maintain their water balance. The third group is made up of species with heterogeneous preferences (psammophiles, xerophiles, ripicoles) that are usually found at the periphery of saline habitats. This classification allows the response of communities to changes in the soil water content and salinity to be characterised.

Results and Discussion At all wetland sites there was an increase in the abundance of carabids and a decrease in

tenebrionids (Figure 15), resulting in an increase in the normalised C/T ratio between 1984 and 2003 (Table 5). Peripheral site remained unchanged. The peak value corresponded to W2 in 2003, which also had the highest soil moisture.

This long-term change might be explained by the soil moisture changes caused by the modifications in the water regime at watershed scale, as discussed in the precedent section. Figure 16 shows the close relationship between soil moisture and logarithm of Carabidae/Tenebrionidae ratio using data from the ten sampling sites of the set of Mar Menor wetlands and peripheral areas.


Figure 15. Abundance of carabids and tenebrionids in the wetland sites of Marina del Carmolí (W1, W2 and W3) in 1984 and 2003.

Table 5. Normalised ratio of carabidae respect to tenebrionidae (C/T) in each site of

Marina del Carmolí in 1984 and 2003. P1: peripheral site; W1, W2, W3: wetland sites

Sampling site 1984 2003 P1 -0.9479 -0.9743 W1 -0.8148 0.801 W2 -0.9953 0.99446 W3 -0.9605 0.66145


Figure 16. Logarithm of the Carabidae/Tenebrionidae ratio and soil moisture in the ten sampling sites of the set of Mar Menor wetlands and peripheral areas.

The effects of agricultural management practices on carabid assemblages through changes in the soil moisture content have been noted by Eyre et al., (1986). On the contrary, tenebrionid beetles were not favoured by an increase in soil moisture, as these beetles are better adapted to more arid environments and high soil moisture seems to be a limiting factor (Bujalance et al., 1987; De los Santos, 1983; De los Santos, 2002). Thus, a decrease in abundance of tenebrionid beetles could be a good descriptor of changes in the hydrological conditions of the wetland. Main tenebrionid species were Tentyria laevis, Pimelia baetica and Zophosis punctata.

Along with the change in the Carabidae/Tenebrionidae ratio there is a change in the composition of carabids. The assemblage of 1984 was dominated by xerophiles such as Orthomus barbarus, Dixus sphaerocephalus and Ditomus tricuspidatus, whereas halobionts remained only in specific sites (W1). On the contrary, the assemblage of 2003 was dominated by halobionts and halophile species, which seem to be favoured by the increase in the groundwater level, flooding period and soil moisture in the wetland sites in 2003, along with associated increased soil surface salinity (Figure 17).

The predominance of this type of species has been observed in littoral areas of saline lagoons (Rueda, 1990) and in small-scale gradients linked to the micro-relief of coastal dunes Georges, 1999), which allow the presence of halophilous species in the intermediate levels susceptible to flooding. The halobiontic species were classified by Rueda (1990) as halophilic species adapted to floods and show medium to low body size, functional wings and high dispersal power. According to Den Boer (1987), these characteristics are typical of species inhabiting unstable habitats. In contrast, species that make up the 1984 assemblages were typical of more stable habitats.

Halobionts and halophiles dominated the assemblage in 2003 with species such as Megacephala euphratica, Scarites procerus eurytus, Cylindera paludosa, Pogonus chalceus and Anisodactylus virens. These last three species were absent in 1984. Megacephala euphratica and Scarites procerus eurytus (Figure 18) are the most singular carabid species


from a biogeographical point of view, as good examples of the Turano Mediterranean distribution pattern (Serrano et al., 2002).

Figure 17. Proportion of halobionts, halophiles and groups with other environmental preferences in the sampling sites of Marina del Carmolí in 1984 and 2003.

Figure 18. Left: Scarites procerus eurytus; Right: Megacephala euphratica.


These two species expand their distribution and /or multiply by twenty their abundances respect to 1984 (Table 6), as a response to the increase in soil moisture. Figure 19 shows the response model (R2

adj.=0.93) of Megacephala euphratica to soil moisture and conductivity. The usual available habitat for Megacephala varies from more xerophitic salt marshes with 20% soil moisture and 18 dS/m conductivity to more humid salt marshes with 35% soil moisture and 7 dS/m conductivity.

In the peripheral site the situation was more stable, as xerophiles predominated from 1984 to 2003 and tenebrionid beetles, particularly core species such as Tentyria laevis, Zophosis punctata and Gonocephalum rusticum (Giménez, 1999), increased in abundance.

In synthesis, these results suggest that long-term trends in the groundwater table and soil moisture conditions in the wetland are well reflected in the changes in species composition at different taxonomic scales: proportion between families (Carabidae and Tenebrionidae) and the proportion between biological types within the dominant family under wetter conditions (halobionts and halophiles respect to groups with other environmental preferences).

Table 6. Number of sampling sites with presence of Megacephala euphratica and Scarites procerus eurytus and average abundance of individuals per effort unit (10 pitfall traps)

in Marina del Carmolí

1984 2003 NS AA NS AV

Megacephala euphratica 2 0.75 2 14.80 Scarites procerus eurytes 1 0.25 3 6.50

NS: Number of sampling sites; AV: Average abundance of individuals per effort unit.

Figure 19. Response model of Megacephala euphratica to soil moisture and conductivity.


CHANGES IN BIRD POPULATIONS AND COMMUNITIES

Introduction Surveys of bird populations and communities have been carried out in the Mar Menor

lagoon and its associated wetlands since the early 70’s of the past century. Previous pioneer studies (Guirao, 1859; Zamorano, 1932) recognise this as a rich area in terms of ornithological diversity. Monitoring effort, however, has been quite heterogeneous and still in recent years varies greatly in space and time. Long time-series of biological data are scarce, with the exception of wintering waterbird census made in the framework of the International Waterbird Census (IWC) scheme, and more recently complemented by breeding waterbird population surveys. Tipically terrestrial or palustrine bird communities (e.g. steppe and reedbed passerines) have been studied much less intensively, and good long-term records are scarce or totally lacking. In this part of the chapter we focus on the two communities with the best series of data, when searching for biological indicators of landscape and environmental change in the wetland complex: steppe birds and waterbirds. As mentioned above, the latter have been surveyed through January census, more or less continuously since 1972 (Martínez et al., 2005; Robledano et al., 2008). For the former, we have been able to reconstruct a series of data extending from 1984 to 2008, through the compilation of several surveys undergone in the same wetland area (Marina del Carmolí) by the own authors and other researchers (Hernández, 1995; Torralva et al., 2003; Robledano et al., 2006).

Study Area and Methodology

Steppe Passeriform Birds The area surveyed is the Marina del Carmolí wetland, described elsewhere in this

chapter. Census of passerines along line-transects in the salt steppe and salt marsh, were used to track changes in the terrestrial bird communities of the wetland. We used IKA values (birds/km) recorded in six years of transect sampling (spread along a period of 24 years): 1984, 1989, 1995, 1997, 2003 and 2008. Dryland fields were sampled only in 1984. Seasonal averages (summer: April-September; winter: October-March) have been calculated when more than one transect census was available.

Line transects had a length of 0.5-1 km depending on the year, with the outer limit of the counting strip usually set at 40 m, on both sides of the transect. When a narrower sampling strip was used, the results have been corrected to refer to similar surface areas. Data for 2003 (Torralva et al., 2003), given in birds per hectare, have also been made comparable by referring them to an equal surface of that covered by the transects. On the other hand, by expressing the results of all surveys in birds/10 ha, we were also able to compare local densities with those recorded elsewhere, and to track their temporal variation as an index of change in ornithological value.

Sampling was carried out by one or two observers who followed a pre-established route at a more or less constant speed (approximately 2 km/h), always during the first four hours of daylight. The Marina del Carmolí (salt steppe and salt marsh) keeps a network of ancient paths, closed to vehicles and scarcely used by pedestrians, but distinguishable enough to


facilitate the bird surveys. We believe that the fragmentation or edge effects caused by these paths are minimal due to their partial colonisation by vegetation. Although we cannot determine the exact sampling places during all the study period, we are confident (and most researchers involved in sampling have confirmed so) that the routes followed are representative of the average conditions of the terrestrial phase of the Carmolí wetland. Part of the variation, however, can be attributable to observer or methodological bias, and the results have to be interpreted cautiously.

Sampling effort ranged between a single seasonal survey per year (1984) to 10-11 surveys per season (1995), and on average approached bi-monthly surveys (5,3-5,6 surveys/season). The final series consists approximately of a survey every five years. We used several indices of abundance to relate changes in passerine populations and taxocenoses to habitat and landscape variables. We summed the IKAs of the most important families in terms of abundance and habitat specificity. We also computed total community abundance, species richness and diversity (Shannon-Wiener index). In order to detect changes in the conservation value of avifauna, we also computed indices of conservation status, adapted from de procedure used by Pons et al. (2003) and ranked the species according to their inclusion in Birds in Europe SPEC cathegories (Birdlife International, 2004), IUCN Spanish Red Data Book (Madroño et al., 2004) and Annex I of the EU Bird’s Directive. These values were multiplied by the abundance index (IKA) logarithmically transformed (Pons et al., 2003; Paquet et al., 2006). Separate analyses were performed on summer and winter community abundance matrices.

Abundance, richness, diversity and conservation status indices were related through simple linear regression with four predictor variables describing changes in the structure of vegetation and landscape characteristics, in turn related with the land-use changes in the watershed and the consequent changes in the water flows into the Marina del Carmolí wetland. These variables were the area of salt steppe, salt marsh, reedbed and crops in a 3 km2 square window containing the wetland, as described earlier in this chapter. Mean IKA of passerine species and families and the community structural indices for 1984, 89, 95, 97 and 2003 were regressed on habitat variables for 1984, 92, 95, 97 and 2001 (data not available thereafter). Although data points are not exactly coincident, we consider them representative of consecutive stages in the wetland ecosystem with an approximately 5 year spacing.

Waterbirds

The study area for waterbirds is the whole Mar Menor water mass (figure 1), a 135 km2 coastal lagoon located in the southeast of the Iberian Peninsula, with a mean depth of 4 m. We have used the results of the January waterbird censuses co-ordinated occasionally by the Regional Environmental Authority and most frequently by amateur environmental organisations (especially before 1995, and from 2000 onwards), the results of which have been compiled by Hernández and Robledano (1991), Martínez et al., (2005) and Hernández et al. (2006). The Mar Menor lagoon has the best coverage among the wetlands of Murcia Region, with 25 censuses between 1972 and 2005 (no census available for 1974, 1976, 1980-82, 1998, 2000-01 and 2003). Birds are censused through a standardised boat route that has remained approximately constant between years. Numbers counted of each species are pooled and reported together for the whole lagoon. We focus on the five most abundant waterbird species (Great Crested Grebe Podiceps cristatus, Black-necked Grebe Podiceps nigricollis, Great Cormorant Phalacrocorax carbo, Red Breasted Merganser Mergus serrator and


Common Coot Fulica atra), also characterised by a strong feeding association to the lagoon. With the exception of Phalacrocorax carbo, which can fly daily to carry out part of its feeding activity in nearby water surfaces (irrigation ponds and the sea), the individuals of these species spend all or a great proportion of their time budget inside the lagoon, and their feeding occurs typically in the water column and benthos. Individually or at the family level, they represent different foraging strategies, from specialist piscivores to generalist herbivores. Annual abundance (total number of each species in the January census) was converted into biomass using constant weight values obtained from the literature (Cramp, 1977). The percent contribution of each species, familiy or guild (for the Coot these were coincident) to total waterbird biomass was used to illustrate long-term changes in the trophic structure of the community. We interpreted these changes on the basis of the estimated nitrogen load reaching the lagoon the previous years, and two biotic variables: total adult jellyfish present in the lagoon, based on visual estimates and annual totals captured by boats committed to their removal, and the annual catch of the two main fish species exploited by the local fleet (Atherina boyeri and Engraulis encrasicholus), as a surrogate index of fish productivity.

Results and Discussion

Response of the Steppe Bird Communities to Habitat Changes in the Carmolí Wetland Our compilation of terrestrial bird surveys shows dramatic changes in abundance and

community structure, with important implications for the conservation value and function of this habitat for birds. In summer, total community abundance declined during the 24 year period of study (R2

adj.=-0.6; n=6; p<0.05), but species richness and Shannon diversity index increased. Diversity values, in any event, were initially and on average lower that those reported in the literature for the same or similar areas (Hernández, 1995). Alaudidae was the only family showing an overall significant decline with year (R2

adj.=-0.81; p <0.01), while there was an increase in the proportional abundance of Fringillidae, Turdidae and Sylviidae (Figure 20).

In winter, a similar pattern emerges for most families and structural indices, except that total community abundance appears more stable (or slightly increasing), and the overall contribution of other families, mainly Motacillidae and Emberizidae, is greater (as corresponds to a richer community). Shannon H’ and richness significantly increase with year (R2

adj.=0.97 and 0.72; p<0.001 and 0.05, respectively). The overall decline of Alaudidae is less pronounced than in summer yet also significative (R2adj.=-0.77; p<0.05), and there is a remarkable transient increase of Fringillidae in 1995-97.

Using data coming from so many different sources, however, raises the problem of bias induced by observer, date, sampling effort, and other sources of variation that can be confounded with true changes in population or community parameters (Devictor et al., 2007). But it is quite improbable that all or most of the changes detected simply reflect these methodological factors. Strip transects are considered adequate to track within-site changes in the relative abundance of wintering grassland birds, and comparable to area searches in producing density estimates (Roberts and Schnell, 2006). The general concordance of summer and winter patterns, in the species and taxocenoses affected, and in the magnitude of the changes, suggests that sampling bias is not masking the main picture of bird community changes in any season.


Figure 20. Changes in the relative abundance of the main families (expressed as % of total IKA), total community abundance and structural parameters of the passerine bird community (left, summer; right, winter).

The habitat variables most related with bird variation (Table 7) also suggest that summer and winter communities are affected by similar structural changes, driven by the pressures exerted on the steppe area by irrigation at the watershed scale. Many species did not respond individually to habitat variables, but representative species and families that can be considered habitat specialists did so. Figures 21 and 22 show some examples.


Table 7. Response of some representative species and families to habitat variables

LINEAR REGRESSION (sign, adjusted R2 and significance level) BIRD VARIABLES Summer community

STEP

SALT

REED

CULT

CALA Melanocorypha calandra 0.99*** -0.76* -0.72* COGU Galerida cristata 0.72* -0.69* CUTO Sylvia conspicillata 0.76* TACO Saxicola torquata 0.94** TECO Calandrella brachydactlyla 0.80* TRIG Miliaria calandra 0.76* VERD Chloris chloris 0.61(p=0.07) ALAU Total Alaudidae 0.76* -0.5 (p=0.11) TURD Total Turdidae -0.61(p=0.08) 0.75* FRIN Total Fringillidae -0.49 (p=0.11) 0.85* TOTC Total community 0.75* -0.73* RDBO Spanish Red Book Index 0.70* Winter community ALRE Lanius excubitor/meridionalis -0.93** 0.79* BUIT Cisticola juncidis 0.67 (p=0.05) CALA Melanocorypha calandra 0.63 (p=0.06) -0.44 (p=0.13) COLT Phoenicurus ochruros 0.68 (p=0.05) ESPA Emberiza schoeniclus 0.46 (p=0.11) ESTO Sturnus unicolor 0.46 (p=0.12) GOCO Passer domesticus 0.38 (p=0.15) JILG Carduelis carduelis 0.46 (p=0.11) MICO Turdus merula 0.46 (p=0.12) MOCO Phylloscopus collybita 0.46 (p=0,12) 0.46 (p=0.12) PETI Erithacus rubecula 0.46 (p=0.12) PECH Luscinia svecica 0.46 (p=0,12) PICO Fringilla coelebs 0.46 (p=0.12) TACO Saxicola torquata 0.70* TECO Calandrella brachydactlyla 0.59 (p=0.07) -0.58 (p=0,07) ZOCO Turdus philomelos 0.46 (p=0.12) ALAU Total Alaudidae 0.66 (p=0.05) 0.48 (p=0.11) TURD Total Turdidae -0.94** 0.88* HSHA Shannon H' -0.75** 0.52 (p=0.10) BDIR Birds Directive Index 0.84* -0.69* DESCRIPTOR VARIABLES PEARSON CORRELATION COEFFICIENTS STEP Area of salt steppe SALT Area of salt marsh -0.92* REED Area of reedbed NS NS CULT Area of crops NS NS NS

Correlations among variables are also shown. Marginally significative but biologically meaningful relationships are also included. Bird variables are mean IKAs of species or families except for structural or conservation indices. Significance levels: p<0,05 = *; p<0,01 = **; p<0,001 = ***


Figure 21. Response of some representative species and families of the summer community to habitat variables. Melanocorypha calandra, Galerida cristata, Calandrella rufescens and total Alaudidae are plotted against the area of salt steppe; Calandrella brachydactyla against irrigated lands and total Fringillidae against the area of salt marsh.

The indicator value of bird species, families and indices, regarding changes induced by agricultural influences, is illustrated by an early negative response of Alaudidae, particularly of Calandra Lark (Melanocorypha calandra) and Crested Lark (Galerida cristata) in summer,


and M. calandra and Short-toed Lark Calandrella brachydactyla in winter, to the loss of salt steppe in the habitat mosaic. Some species characterise the intermediate phases of change (by mid 1990’s), when the loss of steppe habitat is stabilised and the development of salt marsh reaches its maximum, as described in the previous section). This is the case of Fringillidae in both seasons. Sylviidae, although favoured in general terms by the development of salt marsh, show different responses depending on the species, with Spectacled Warbler (Sylvia conspicillata) peaking in the early stages of change (1989), Dartford Warbler (S. undata) during the middle ones (1994-97), and Sardinian Warbler (S. melanocephala) towards the late ones (2003), probably illustrating species-specific habitat preferences. At the end of the period, wintering species of partial palustrine character like Reed Bunting (Emberiza schoeniclus), as well as some Turdidae tend to increase in an apparent response to the partial invasion of salt marsh by reedbeds. In other wetlands, these species characterise the transitional (mixed) habitats located between the reed belt of lagoons and the salt marsh vegetation of immediate terrestrial areas (Peiró, 2006). The Lesser Short-toed Lark (Calandrella rufescens) shows also a pattern of intermediate increase in numbers (peaking between 1989-97). Its apparent indifference to any habitat variable can be due to the particular preferences of this species in terms of coverage and vertical structure of the vegetation. C. rufescens is common in coastal steppe areas with a good representation of salt marsh and low vegetation cover (Cramp, 1988; Tellería et al., 1999; Torralva et al., 2003). Hernández (1995) added to its preferences a weaker association with perennial grasses (e. g. Lygeum spartum, a dominant species in the salt steppe), and stronger with intermediate heights of these two types of vegetation (salt steppe and salt marsh).

The positive response of Calandrella brachydactyla (in summer) to the area of irrigated crops is somewhat unexpected since this species is usually favoured by dryland, unploughed agricultural habitats (De Juana and Suárez, 2004). In our study area it was more abundant in the peripheral dryland fields surveyed in 1984 (24 birds/km in summer and 14 in winter) that in the Carmolí wetland in any year (8 and 1,78 birds/km as maximum values, respectively, in the same season). An explanation for this is the particular nature of irrigated agriculture in this area. Due to the low quality of water, the consequent salinisation of soils, and the use of short crop cycles, the fields surrounding the wetland stay uncultivated for long periods, which favours the presence of many typical steppe species. In Murcia Region C. brachydactyla is a typical species of uncultivated agricultural fields (Hernández, 1995). In summer these fields seem to be playing a similar positive effect to that attributed by Brotons et al. (2005) to improved pastures around French steppes. The abundance of Corn Bunting (Miliaria calandra), a species of European concern (Birdlife International, 2004), is also positively related with the area of crops around the wetland. The recent breeding of Collared Pratincole (Glareola pratincola) in these areas, made possible through agreements between landowners and conservation organisations, is an example of the ecological potential of these areas, that should be also a priority conservation target. In winter C. brachydactyla seems to depend more on the salt steppe habitat.

It is not possible nor realistic, owever, to try to explain most variation in all species or groups solely on the basis of local habitat changes. External factors, illustrated by population trends at higher geographical scales, can also explain the numerical changes of some bird species, particularly those showing cyclical phases of increase and decline in abundance. Unfortunately, the Spanish bird monitoring programs that could illustrate these patterns (Del Moral et al., 2008) cover only the most recent part of our study period.


Figure 22. Response of some representative species of the winter community to habitat variables. Alaudidae species and total abundance are represented against the area of salt steppe and total Turdidae against the area of salt marsh.

Among the conservation indices, those based on SPEC and on the Spanish Red Data Book show an overall decrease in the summer community (although with a partial recovery in recent years), and fluctuant trends with even some gain at the end of the period. The Birds Directive index decreases both in summer and winter (Figure 23). The latter trend is the only significant one (R2

adj.=-0.58; p<0.05). In both seasons, the decline in the Birds Directive index parallels that of the Annex I species present, namely Melanocorypha calandra and Calandrella brachydactyla in summer, and these same species plus Sylvia undata in winter. Apart from these, the general loss of conservation value in summer is due to the decline in species of open spaces and mosaic habitats like Miliaria calandra and Lanius senator (SPEC3). Among “winner” species adding conservation value we can remark Emberiza schoeniclus and Carduelis cannabina in winter. In overall, besides increasing species richness and diversity, the change towards less steppic conditions represents a small gain in conservation value. The species that benefit from this change are usually the less representative of the original conditions and often marginal to the region.

Although high values of species richness and diversity are often set as general targets in protected area management, the use of more specific conservation indices allow a best evaluation of performance against particular objectives. Our indices describe different trajectories depending on season and on the legal or scientific value involved in the


calculation. The Birds Directive based index is the only one that shows a marked decline both in summer and winter, while those based on SPEC and IUCN Red Book categories exhibit ups and downs, or even improve along the period of habitat change. But, considering the legal status of the study area, the greater concern should be put on the former. The Carmolí wetland has been declared as a Bird SPA (Special Protection Area) under 79/409 EC Directive, and the loss in conservation value for Annex I species implies that this function has not been achieved.

Among other criteria, this SPA was declared in 2001 on the basis of one steppe passerine bird (Calandrella rufescens) exceeding the population threshold established at the European level (Viada, 1998). The effectiveness of protection could be questioned on the basis of its recent population decline. Our estimates of summer population density for this species are 20 birds/10 hectares at the start of the study, rising to 44 and 47.5 birds/10 ha in 1989 and 1997, which probably represent the highest densities of the Iberian Peninsula (Hernández and Pela, 1987; Suárez et al., 2002; Samprieto and Pelayo, 2003).

Figure 23. Changes in the indices used to assess the conservation value of the study area (upper graph, summer; lower graph, winter).


This transient positive effect can be attributed to the expansion of salt marsh vegetation which stabilises by 1997. In the long term, however, its abundance is lowered to values between 5.5-8.75 birds/10 ha, still high but nearly an order of magnitude lower than the peak values (on which the evaluation against the SPA criteria is based).

Dynamics of Waterbird Communities in the Mar Menor Lagoon

Of the five species studied, all but the Red Breasted Merganser have increased in numbers along most of the period of study. The greatest fluctuations and highest counts correspond to the Black-necked Grebe. The trend of the Great Crested Grebe is parallel to that of its congeneric species, but less fluctuant. Cormorants seem to have grown more steadily. Coot display also a marked increase, but starting later than these three piscivorous species. Figure 24 summarises the general progression of waterbird biomass and the relative contribution of each species. Despite the gaps at the start and the end of the period, there are recognisable periods characterised by the dominance of particular species. The Red Breasted Merganser is the dominant piscivorous during most of the decade of 1970’s. The Great Cormorant dominates most years since then, representing ca. 50% of the biomass except between 1988 and 1996, when other piscivores, including the two Grebes, dominate. Herbivores (Coot) join the community in 1992 and their contribution rises to more than 30% in 2005. The biomass of Red Breasted Merganser does not change markedly in the long term, although its relative contribution decreases gradually (from more than 90% to less than 15%). In terms of feeding guild numbers, the composition of the waterbird community has changed from a 100% of piscivores until 1991, to nearly 50% of herbivores in 2005. When expressed as biomass, the contribution of herbivores is slightly lower (under 40%).

The development of intensive agriculture and residential tourism during the last three decades have become the main pressures driving environmental change in the Mar Menor Lagoon. Total waterbird biomass has increased more than 4-fold during this period, which supports the increase in carrying capacity as a plausible explanation (Ysebaert, 2000; Van Eerden et al., 2005; Roomen et al., 2006). The positive response of bird biomass to estimated nutrient loads into the Mar Menor lagoon (Martinez et al., 2005) is in accordance with this explanation, as does the lack of a general relationship with biogeographical population trends (F. Robledano et al., in prep.). Local trends of bird numbers differ also markedly from those observed in other nearby sites or regions, in some species with opposite rather than divergent trends. This is the case of the Coot, which declined sharply as a breeder in the Valencia region during the period 1984-2004, and especially from 1991 onwards in the winter census, falling from more than 14,000 birds to less than 2,000 (Gómez et al., 2006; compare this with the 10-fold increase in the Mar Menor between 1992 and 2005). The carrying capacity of the lagoon for this phytophagous species seems to have risen during the second half of this period, opposite to what happened in a number of apparently more suitable wetlands in the Valencia region. This is not surprising since, in their preferred sections within the Mar Menor lagoon, Coot attains higher densities than in brackish coastal wetlands of southeastern Spain with dense Ruppia meadows (Robledano et al., 2008).

Nutrient load appears as the first determinant of waterbird biomass in three out of five species, with the dynamics of a fourth one (Great Cormorant) primarily governed by external factors (see e.g. Van Eerden and Gregersen, 1995; Van Bommel et al., 2003) but also favoured by local eutrophication. On the other hand, Red Breasted Merganser seems to have been indifferent (tolerant) to nutrient loading during most of the study period, although


negatively affected in the long term (Martínez et al., 2005). The trends of Great Cormorant and Red Breasted Merganser in the Mar Menor match best those of their Spanish populations (Martínez et al., 2005), and consequently they could be considered as poorer indicators of lagoon condition.

Figure 24. Contribution of the five species studied to the total biomass (kg) of the waterbird community of the Mar Menor Lagoon (only years with data for all the species altogether are shown).

The question remains about through which mechanisms eutrophication improves feeding conditions for waterbirds. Some authors (e.g., Nienhuis, 1992) argue against a direct relationship between eutrophication and higher secondary production. But at a general scale, the changes in waterbird numbers have been commonly related to the fertilisation of coastal waters, with increases and declines interpreted on the basis of processes of nutrient addition and removal (see references in Robledano et al., 2008). Nutrient loads cause increased nutrient concentration that stimulates phytoplankton production and thereby increases chlorophyll concentration and light attenuation (Cloern, 2001). The effect of nutrient loads on fish biomass is also usually positive (Van Rijn and Van Eerden, 2003; Pérez-Ruzafa et al., 2007).

Of course, factors other than agricultural nitrogen load can also explain changes in waterbird abundance. Phosphorus loads from urban sewage can have an important role in the eutrophication process. The general increase in productivity can also respond to climatic factors. Changes in the salinity of the lagoon can also account for some changes in waterbird populations. Long term variation in several other geomorphologic and hydrographic parameters, through its influence on fish community composition and abundance (PÉREZ-RUZAFA et al., 2007), can also influence waterbird numbers. Weather (outside and in the wintering places) and disturbance are factors that can also affect waterbird populations (Davidson and Rothwell, 1993; Rönkä et al., 2005; O’Connell et al., 2007) and hence census results. It is also possible that variations in the winter pattern of occurrence of the species explain apparent interannual changes (Roomen et al., 2006).


Aggregated variables (familiy, guild or total waterbird biomass) seem to respond more markedly than individual species’ abundances to gross nutrient enrichment. An aggregated approach can also be applied to the temporal sequence of years, trying to identify phases of relative stability and sharp changes in the lagoon ecological status. Within the general increase in waterbird biomass that starts by mid 1980’s, we can distinguish four main phases plus a short one of rapid change (Figure 25): Phase 1 and 2 correspond to a period of more or less homogeneous nutrient loading, the first part of which (1972-79) still shows high fish catches that dramatically drop at its end (probably due to overfishing); the second phase (1980-87) shows a first positive response of generalist piscivores, mainly Great Cormorant, which dominates the community with Red Breasted Merganser. It follows a period of gradually increasing nutrient loads extending approximately from 1988 to 1995 (Phase 3), during which the relative biomass of Podicipedidae (vs that of Cormorant plus Merganser) increases markedly. This is followed by a short phase (1996-98) with low fish catches, higher nutrient loads and an incipient jellyfish development (Phase 4), coincident with a fall in Podicipedidae biomass. From 1999 onwards, there is some recovery of Podicipedidae but their relative contribution to total biomass is lowered due to the large increase of Cormorant and the also increasing share of herbivores (Coot). This last period (Phase 5) is characterised by further increases in nutrient loads, but also by a greater abundance of jellyfish that can exert some control on eutrophication. In fact there is an apparent recovery of fish catches that can be related to such control, which in turn could be responsible for the recovery of piscivores after 1998. Towards the end of the study period, the continued increase in nutrient loads and the decline of jellyfish numbers could illustrate a new shift, in this case towards conditions more favourable for herbivores. Eutrophication causes a marked deterioration of seagrass and macroalgal communities, a situation apparently not beneficial for herbivores (Noordhuis et al., 2002), but it also can promote a shift towards increased abundance of opportunistic macroalgae (Krause-Jensen et al., 2008), a source of food for generalist herbivores like Coot (Perrow et al., 1997). The proliferation of such algae is a phenomenon already observed in some stretches of the lagoon shoreline.

Surprisingly, four out of five waterbird species increase in numbers along a period of fisheries decline, and only one of them is typically non-piscivorous (Coot). However, it is possible that the piscivorous waterbirds select different prey types or size classes than those commercially exploited (Liordos and Goutner, 2007), which in the long term seem negatively affected by eutrophication. But the decrease in fish yield can also be a consequence of overfishing, leading to a dominance of small fish in the community due to the selective extraction of larger ones (Van Rijn and Van Eerden, 2003). This larger biomass of small fish is known to create good feeding conditions for species like the Great Crested Grebe and the Great Cormorant (Gwiazda, 1997; Smit et al., 1997; Van Rijn and Van Eerden, 2003). Other species can survive entirely on prey other than fish, like Black-necked Grebe, a typical invertebrate-feeder (Jehl, 2001).

Interaction among variables (F. Robledano et al., in prep.) suggests that some biotic components might have effects on the relationship between waterbirds and trophic variables. Particularly, jellyfish seem to modify the response of waterbirds to nutrient enrichment. Jellyfish populations are the main top-down agent in the Mar Menor lagoon, controlling the effects of the eutrophication process through two mechanisms: the direct use of nutrients by endosymbiotic zooxanthella, and the direct predation on plankton (Pérez Ruzafa and Aragón, 2003; Rodríguez et al., 2005).


Figure 25. Simplified representation of the main phases identified by the response of terrestrial passerines and waterbirds to habitat and environmental variables. Species or families are represented by areas proportional to their contribution to total community abundance (passerines) or biomass (waterbirds), and environmental variables by curves or areas. Values are averages of summer and winter IKAs (passerines) and of January census for each period (waterbirds). Taxa represented are (from bottom to top): Waterbirds: Coot, Great Cormorant, Podicipedidae and Red Breasted Merganser; Passerines: Alaudidae, Sylviidae, Turdidae, Fringillidae and others.

Jellyfish play a role of sink for organic matter, by taking up and storing nutrients, and typically disrupt the trophic web at the level of secondary consumers. By feeding on organisms that serve as prey of large consumers like large crustaceans and fish larvae, they also reduce their feeding opportunities (Gili and Pagès, 2005). Increasing anthropogenic nutrient loads favours jellyfish, acting synergistically with overfishing to result in a degradation of the ecosystem, characterised by harmful algae and jellyfish blooms (Vasas et al., 2007).

Summarising, long-term negative effects are only apparent for the Red Breasted Merganser, which seems to prefer low to moderately eutrophic conditions. The two Podicipedidae increase in moderately eutrophic conditions and keep their contribution to waterbird biomass along most of the study period, except during the short phase of higher


enrichment when jellyfish are still unable to alleviate eutrophication (Pérez Ruzafa et al., 2002). However, we think that further nutrient addition, particularly in the absence of this top-down control (jellyfish numbers have been lowered to a minimum during the last removal campaigns), can lead the system to conditions less favourable for most piscivores. The feeding activity of great crested grebes in eutrophic waters has been shown to be limited by factors like underwater visibility (Van Eerden et al., 2003). The Great Cormorant, also, has been proposed as an indicator of waters of intermediate turbidity, above (or below) which, it feeds less efficiently (Van Rijn and Van Eerden, 2003). Present favourable conditions for herbivores can also deteriorate in the long term if the system shifts to a phytoplankton-dominated state (Noordhuis et al., 2002; García-Pintado et al., 2007).

CONCLUSION We can conclude, answering the questions starting this chapter, that wetlands have

effectively reacted in a strong manner to the land use changes in the watershed and associated alterations of the water and nutrient flows. Moreover, the changes can be tracked in different biological communities, in particular the vegetation assemblages, the carabid and tenebrionid communities, steppe passerines and waterbirds. It has been shown that the effects on steppe passerines are mediated by the changes in the vegetation assemblages, as shown by the regression models between the steppe passerines and the area occupied by each type of habitat along time. Some trends constitute a common pattern of change in all studied communities, especially the reduction of the elements more specifically linked to arid characteristics, like the area of salt steppe habitat, the abundance of tenebrionids and steppe passerine birds such as Alaudidae.

In each assemblage it has been identified some good indicators, not defined at the species level, of the observed long-term changes. Regarding the vegetation assemblages, the overall increase in the total wetland area seems to be a poor indicator of the increase in water input at watershed scale, whereas the increase in the hygrophilous vegetation (salt marsh plud reedbeds), observed overall and in each particular wetland, constitutes a good indicator of such water changes. A very significant relationship was found between the temporal trend of the area of irrigated lands in the watershed with a five years time lag and the area of salt marsh plus reedbeds in the wetlands. In the case of wandering beetles, long-term trends in the groundwater table and soil moisture conditions in the wetland are well reflected in the changes in species composition at different taxonomic scales: proportion between families (Carabidae and Tenebrionidae, as expressed by the Carabidae/Tenebrionidae ratio) and the proportion between biological types within the dominant family under wetter conditions (halobionts and halophiles respect to groups with other environmental preferences). In particular, the C/T ratio, which does not require identification of taxons to the species level, only to the family level, seems to be a very good indicator of the soil moisture to explain both spatial patterns and long-term changes. The value of terrestrial birds as indicators of land use changes is illustrated by their integrative, multi-stage response to the habitat changes. The main phases, defined by the relative proportions of habitats, are associated with numerical changes in species, families and assemblages. In our case the decrease in abundance of Alaudidae in the wetland seems to be a good indicator of the land use changes in the


watershed. Specific Alaudidae species seems to explain in more detail the nature of such changes. Waterbird lagoonal species, although having lower conservation scores (none of the species studied qualify for SPA designation), emerge as an equally valuable tool in the surveillance of agricultural (and urban) nutrient impacts on the lagoon. The warning role of waterbirds about changes in nutrient status in the eutrophication process of the lagoon (suggested as early as mid 1980’s), could also have been played by steppe passerines with regard to habitat changes in the peripheral wetlands (evident by mid 1990’s, and dramatic at the start of this century), had their populations been monitored more continuously. Multispecies assessments of protected areas are a valuable tool, not only for its own management, but for the investigation of human impacts on biodiversity at a wide array of scales (Devictor et al., 2007).

Regarding the biodiversity and conservation value, different trends have been identified. In the case of the vegetation communities, the habitats favoured by the increase in irrigated lands and the water changes in the watershed are those less valuable from the Habitats Directive perspective. The habitat showing the highest increase is reedbeds, which is not included in the Directive, followed by the salt marsh, designated as Community Interest, at the expense of a substantial reduction in salt steppe, of Priority Interest due to its singularity and level of threat. This is relevant since it is a rare habitat in Spain and the conservation status of salt steppe in Murcia is well over the average in Spain. In the case of wandering beetles, the changes in the water regime have benefited the most singular carabid species from a biogeographical point of view, Megacephala euphratica and Scarites procerus eurytus, good examples of the Turano Mediterranean distribution pattern. In the case of steppe passerines, the Birds Directive based index is the only one that shows a marked decline both in summer and winter, while those based on SPEC and IUCN Red Book categories exhibit ups and downs, or even improve along the period of habitat change. However, the Carmolí wetland has been declared as a Bird SPA (Special Protection Area) under the Birds Directive, hence, the loss in conservation value of Marina del Carmolí according to the Bird Directive based index implies that the protection function for this natural site has not been achieved.

The original landscape and habitat setting of a hypersaline, oligotrophic lake fringed by steppe habitats, has been gradually modified towards the present scenario of a Mediterranean-like water mass with incipient eutrophication symptoms, surrounded by expanding salt marsh and reedbed communities. The Mar Menor wetlands, in particular Marina del Carmolí, exemplifies the deterioration of a steppe area through hydrological changes in the watershed, without direct habitat reclamation. The need to manage agricultural impacts at the watershed scale emerges here as a critical biodiversity issue. Ecological changes induced by excess drainage water and nutrient loading into the Mar Menor lagoon and its peripheral wetlands have modified its biota, causing a loss of ecological integrity and conservation value. The observed trends, which will probably continue in the future unless action is taken in the whole area of influence, question the effectiveness of the protection measures given to the area and highlight the urgent need to apply management measures outside the protected area for an effective conservation the Mar Menor lagoon and associated wetlands.


ACKNOWLEDGEMENTS The research has been partially supported by the European project DITTY. Development

of Information Technology Tools for the Management of European Southern lagoons under the influence of river-basin runoff (EVK3-CT-2002-00084), support which is acknowledged. The research has also been partially supported by the scientific project Estado ecológico de los humedales mediterráneos semiáridos: propuesta de indicadores para su evaluación (CGL 2006-08134), funded by the Ministry of Science and Education, whose support is also acknowledged. We also thank Asociación de Naturalistas del Sureste (ANSE) for coordinating, and many anonymous ornithologists for carrying out waterbird censuses in the Mar Menor during the years without public support to this scheme. Vicente Hernández Gil kindly provided the results of passerine bird surveys carried out within his Ph.D. research in the Marina del Carmolí. Passerine data for 1995-97 were gathered as a part of the wetland monitoring scheme financed by Project LIFE/1973/93/11-10: “Conservación de humedales y otros ecosistemas característicos de zonas áridas”

REFERENCES

Álvarez-Rogel, J., Jiménez-Cárceles, F.J., Barbosa. O. A., Egea, C., Ortiz, R., Alarcón, A., Delgado, M. J. and Martínez, F. (2002a). Preliminary data on the role of a coastal salt marsh in the retention of nitrates and phosphates leached from intensive agricultural zones in the Campo de Cartagena (Murcia. SE Spain). In: A. Faz Cano, R. Ortiz Silla and A. R. Mermut (Eds.), Final Proc. Inter. Symp. on Sustainable Use and Management of Soils in Arid and Semiarid Regions (469-472). Cartagena.

Álvarez-Rogel, J., Jiménez-Cárceles, F.J., Barbosa, O. A., Egea, C., Ortiz, R., Alarcón, A., Delgado, M. J. and Martínez, F. (2002b). Quantification of Zn and Pb in the top soil and aerial biomass in a salt marsh polluted by metal mine wastes in SE Spain. In: A. Faz Cano, R. Ortiz Silla and A. R. Mermut (Eds.), Final Proc. Inter. Symp. on Sustainable Use and Management of Soils in Arid and Semiarid Regions (473-476). Cartagena.

Alvarez-Rogel, J., Jiménez-Cárceles, F.J. and Egea-Nicolás, C. (2006). Phosphorus and nitrogen content in the water of a coastal wetland in the Mar Menor lagoon (SE Spain): relationships with effluents from urban and agricultural areas. Water, Air, and Soil polluted, 173, 21-38. DOI: 10.1007/s11270-006-9020-6.

Alvarez-Rogel, J., Jiménez-Cárceles, F.J. and Egea-Nicolás, C. (2007a). Phosphorus retention in a coastal salt marsh in SE Spain. The Science of Total Environment, 378, 71-74.

Alvarez-Rogel. J., Jiménez-Cárceles. F.J., Roca. M.J. and Ortiz. R. (2007b). Changes in soils and vegetation in a Mediterranean coastal salt marsh impacted by human activities. Estuarine. Coastal and Shelf Science, 73, 510-526.

Alvarez-Rogel, J., Ramos Aparicio, J., Delgado Iniesta, M. J. and Arnaldos Lozano, R. (2004). Metals in soils and above-ground biomass of plants from a salt marsh polluted by mine wastes in the coast of the Mar Menor lagoon. SE Spain. Fresenius Envirnmental Bulletin, 13, 274-278.


Belaoussoff, S., Kevan, P. G., Murphy, S. and Swanton, C. (2003). Assessing tillage disturbance on assemblages of ground beetles (Coleoptera: Carabidae) by using a range of ecological indices. Biodiversity Conservation, 12, 851- 882.

BirdLife International. (2004). Birds in Europe. Population Estimates, Trends and Conservation Status. Cambridge, Birdlife Conservation Series, 12.

Brandmayr, P., Lövei, G. L., Zetto-Brandmayr, T., Casale, A. and Vigna Taglianti. A. (Eds.). (2000). Natural history and applied ecology of carabid beetles. Sofia. Pensoft.

Brotons, L., Wolff, A., Paulus, G. and Martin, J. (2005). Effect of adjacent agricultural habitat on the distribution of passerines in natural grasslands. Biological Conservation, 124 (3), 407-414.

Bujalance, J. L., Cárdenas, A. M. and Bach, C. (1987). The relationship between local and regional diversity. Ecology, 78, 70-80.

Cakir, H. I., Khorram, S. and Nelson, S. A. C., (2006). Correspondence analysis for detecting land cover change. Remote Sensing of Environment, 102, 306-317.

Carreño, M.F., Esteve, M.A., Martinez, J., Palazón, J.A. and Pardo, M.T. (2008). Habitat changes in coastal wetlands associated to hydrological changes in the watershed. Estuarine. Coastal and Shelf Science, 77, 475-483.

Cartagena. M.C. and Galante. E. (2002). Loss of Iberian island tenebrionid beetles and conservation management recommendations. Journal of Insect Conservation, 6, 73–81.

Cihlar, J., Ly, H. and Xiao, Q. (1996). Land Cover Classification with AVHRR Multichannel Composites in Northern Environments. Remote Sensing of Environment, 58, 36-51.

Cole, L. J., McCracken, D. I., Dennis, P., Downie, I. S., Griffin, A. L., Foster, G. N., Murphy, K. J. and Waterhouse, T. (2002). Relationships between agricultural management and ecological groups of ground beetles (Coleoptera: Carabidae) on Scottish farmland. Agriculture, Ecosystems and Environment, 93, 323-336.

Cloern, J.E. (2001). Our evolving conceptual model of the coastal eutrophication problem. Mar. Ecol. Prog. Ser., 210, 223–253.

Cramp, S. (Ed.). (1977). The birds of the western Palearctic. Vol. 1: Ostrich to Ducks. London, Oxford University Press.

Cramp, S. (Ed.). (1988). The birds of the western Palearctic. Vol. 5: Tyrant Flycatchers to Thrushes. London, Oxford University Press.

Dajoz, R. (2002). Les Coléoptères Carabidés et Ténébrionidés. Ecologie et Biologie. Paris, Tec et Doc Lavoisier.

Davidson, N. and Rothwell, P. (Eds.). (1993). Disturbance to waterfowl on estuaries. Wader Study Group Bulletin, 68 (Special Issue).

De Juana, E. and Suárez, F. (2004). Terrera Común Calandrella brachydactyla. In: A. Madroño, C. González and J. C. Atienza (Eds.), Libro Rojo de las Aves de España. Madrid, Dirección General para la Biodiversidad-SEO/Birdlife.

De los Santos, A. (1983). Biología y ecología de dos comunidades mediterráneas de coleópteros de superficie. PhD Thesis. Murcia, University of Murcia.

De los Santos, A., de Nicolás, J.P. and Ferrer, F. (2002). Habitat selection and assemblage structure of darkling beetles (Col. Tenebrionidae) along environmental gradients on the island of Tenerife (Canary Islands). Journal of Arid Environments, 52, 63-85.

Del Moral, J.C., Escandell, V., Molina, B., Bermejo, A. and Palomino, D. (Eds.). (2008). Programas de seguimiento de SEO/Birdlife en 2006. Madrid, SEO/Birdlife.


Den Boer, P. J. (1987). On the turnover of carabid populations in changing environments. Acta Phytopathology et Entomologica Hungarita, 22, 71-83.

Desender, K., Dufrêne, M. and Maelfait, J. P. (1994). Long term dynamics of carabid beetles in Belgium: a preliminary analysis on the influence of changing climate and land use by means of a database covering more than a century. In: K. Desender et al. (Eds.), Carabid beetles: ecology and evolution. Dordrecht, Kluwer Academic Publishers.

Devictor, V., Godet, L., Julliard, R., Couvet, D. and Jiguet, F. (2007). Can common species benefit from protected areas? Biological Conservation, 139(1-2), 29-36.

Esteve, M. A. (1987). Evaluación ecológica. comunidades animales y ordenación del territorio: aplicación al área del Mar Menor (SE España). PhD Thesis. Murcia, University of Murcia.

Esteve, M. A. and Calvo, J. F. (2000). Conservación de la naturaleza y biodiversidad en la Región de Murcia. In J. F. Calvo, M. A. Esteve and F. López Bermúdez (Coords.), Biodiversidad. Contribución a su conocimiento y conservación en la Región de Murcia. (193-214). Murcia, Servicio de Publicaciones de la Universidad de Murcia.

Eyre, M. D., Rushton, S. P., Luff, M. L., Ball, S. G., Foster. G. and Topping. C. J. (1986). The use of invertebrate community data in environmental assessment. Agricultural Environment Research Group, University of Newcastle upon Tyne.

Eyre, M. D., Luff, M. L. (2002). The use of ground beetles (Coleoptera: Carabidae) in conservation assessments of exposed riverine sediment habitats in Scotland and northern England. Journal of Insect Conservation, 6, 25–38.

Främbs, H. (1990). Changes in carabid beetle populations on a regenerating. excavated peat bog in North-west Germany. In: N. E. Stork (Ed.). The role of ground beetles in ecological and environmental studies (157-169). Intercept, Andover.

Fuellhaas, U. (2000). Restoration of degraded fen grassland-effects of long-term inundation and water logging on ground beetle populations (Coleoptera. Carabidae). In: P., Brandmayr, G. L. Lövei, T. Zetto Brandmayr, A. Casale and A. Vigna Taglianti (Eds.), Natural History and Applied Ecology of Carabid Beetles (289-294). Sofía, Pensoft.

García Lázaro, U. (1995). Estudio de posibilidades de instalación de plantas desalinizadoras con aguas subterráneas de los acuíferos del Campo de Cartagena (Murcia). Federación de Cooperativas Agrarias de la Región de Murcia. Murcia.

García Pintado J., Martínez Mena, M., Barberá G. G., Albadalejo J. and Castillo V. (2007). Anthropogenic nutrient sources and loads from a Mediterranean catchment into a coastal lagoon: Mar Menor, Spain. Science of the Total Environment, 373, 220–239.

Georges, A. (1999). Ground beetles (Coleoptera : Carabidae) of old littoral grasslands - Species assemblages and indicator species. Annales de la Societe Entomologique de France, 35 (SUPPL.), 255-258.

Gili, J. M. and Pagès, F. (2004). Jellyfish blooms. Boll. Soc. Hist. Nat. Balears, 48, 16-22. http://www.recercaenaccio.cat/agaur_reac/AppJava/resources/documents/070712-gili-pags-2005-44237.pdf (accessed on 5/05/2006).

Giménez, A. and Esteve. M. A. (1994). Estructura espacial y ambiental de las comunidades de tenebriónidos de la comarca del Mar Menor (SE de España). Studia. Oecologica, 10-11, 409-419.

Giménez, A. (1999). Patrones, modelos y procesos en las comunidades de tenebriónidos de ecosistemas mediterráneos semiáridos. PhD Thesis. Murcia, University of Murcia.


Gómez, J. A., Dies, J. I. and Vilalta, M. (Eds.). (2006). Las aves acuáticas de la Comunitat Valenciana. Censos y evolución de las poblaciones (1984-2004). Generalitat Valenciana, Consellería de Territori i Habitatge, Valencia.

Guirao, A. (1859). Catálogo Metódico de las Aves observadas en una gran parte de la provincia de Murcia. Boletín de la Real Academia de Ciencias de Madrid, 4, 1-50.

Gwiazda, R. (1997). Foraging ecology of the Great Crested Grebe (Podiceps cristatus L.) at a mesotrophic-eutrophic reservoir. Hydrobiologia, 353: 39–43.

Hernández, F. and Pela, E. (1987). Sobre comunidades de aves esteparias en planicies del valle medio del Ebro. I. Int. Simp. Aves Esteparias, León (Spain), 379–393.

Hernández, A. J., Fernández-Caro A. and Ibáñez J. M. (2006). Censo Invernal 2005 de acuáticas de la Región de Murcia. El Naturalista Digital, Asociación de Naturalistas del Sureste, Murcia. www.asociacionanse.org/naturalista-digital (accessed on 28/02/2008).

Hernández, V. (1995). Ecología de las estepas de la Región de Murcia. Estructura y dinámica de sus comunidades orníticas. Cuadernos de Ecología y Medio Ambiente, 4. Universidad de Murcia.

Hernández, V. and Robledano, F. (1991). Censos invernales de aves acuáticas en la Región de Murcia, SE de España (1972-1990). Anales de Biología (Biología Animal), 17(6), 71-83.

Hess, L., Melack, J. M., Novo, E. M. L. M., Barbosa, C. C. F. and Gastil, M. (2003). Dual-season mapping of wetland inundation and vegetation for the central Amazon basin. Remote Sensing of Environment, 87, 404-428.

Holland, J. M. (Ed.). (2002). The agroecology of carabid beetles. Andover, Intercept. ITGE. (1994). Las aguas subterráneas del Campo de Cartagena (Murcia). Secretaría General

de la Energía y Recursos Minerales. Murcia. Jehl, J. R. Jr. (2001). The abundance of the Earead (Black-necked) Grebe as a recent

phenomenon. Waterbirds, 24(2), 245-49. Krause-Jensen, D., Sagert, S., Schubert, H and Boström, C. (2008). Empirical relationships

linking distribution and abundance of marine vegetation to eutrophication. Ecological Indicators, 8, 515-529.

Kumar Joshi, P. K., Roy, P. S., Singh, S., Agrawal, S. and Yadav, D. (2006). Vegetation cover mapping in India using multi-temporal IRS Wide Field Sensor (WiFS) data. Remote Sensing of Environment, 103, 190-202.

Liordos, V. and Goutner, V. (2007). Spatial Patterns of Winter Diet of the Great Cormorant in Coastal Wetlands of Greece. Waterbirds, 30(1), 103-111.

Lövei, G. L. and Sunderland, K. D. (1996). Ecology and behaviour of ground beetles (Coleoptera: Carabidae). Annual Review of Entomology, 41, 231-256.

Madroño, A., González, C. and Atienza, J.C. (Eds.). (2004). Libro Rojo de las Aves de España. Madrid, Dirección General para la Biodiversidad-SEO/Birdlife.

Martínez, J. and Esteve, M. A. (2002). El regadío del Campo de Cartagena y la exportación de fertilizantes agrícolas al Mar Menor. In J. Martínez and M. A. Esteve (Eds.), Agua, regadío y sostenibilidad en el Sudeste ibérico (133-150). Bilbao, Bakeaz.

Martínez-Fernández, J., Esteve-Selma, M. A., Robledano-Aymerich, F., Pardo-Sáez. M. T. and Carreño-Fructuoso, M. F. (2005). Aquatic birds as bioindicators of trophic changes and ecosystem deterioration in the Mar Menor lagoon (SE Spain). Hidrobiología, 550, 221-235.

McHugh, O. V., McHugh, A. N., Eloundou-Enyegue, P. M. and Steenhuis, T.S. (2007). Integrated qualitative assessment of wetland hydrological and land cover changes in a


data scarce dry Ethiopian highland watershed. Land Degradation and Development (in press). www.interscience.wiley.com DOI:10.1002/ldr.803

Michelson, D. B., Liljeberg, B. M. and Pilesjö, P. (2000). Comparison of Algorithms for Classifying Swedish Landcover Using Landsat TM and ERS_1 SAR Data. Remote Sensing of Environment, 71, 1-15.

Mundia, C. N. and Aniya, M. (2006). Dynamics of landuse/cover changes and degradation of Nairobi City, Kenya. Land Degradation and Development, 17, 97-108.

Nienhuis, P. H. (1992). Eutrophication, water management, and the functioning of Dutch estuaries and coastal lagoons. Estuaries, 15(4), 538-548.

Niemi, G. J. and McDonald, M. E. (2004). Applications of ecological indicators. Annual Review of Ecology. Evolution and Systematics, 35, 89-111.

Noordhuis, R., Van der Molen, D. T. and Van der Berg, M. S. (2002). Response of herbivorous water-birds to the return of Chara in Lake Veluwemeer, The Netherlands. Aquatic Botany, 72, 349-367.

Noriega, J. and Lozano-Garcia, D.F., (2000). Spatial filtering of radar data (RADARSAT) for wetlands (Brackish marshes) classification. Remote Sensing of Environment, 73, 143-151.

Narumalani, S., Mishra, D. R. and Rothwell, R. G. (2004). Change detection and landscape metrics for inferring anthropogenic processes in the greater EFMO area. Remote sensing of Environment, 91, 478-489.

O'Connell, M. J., Ward, R. M., Onoufriou, C., Winfield, I. J., Harris, G., Jones, R., Yallop, M. L. and Brown, A. F. (2007). Integrating multi-scale data to model the relationship between food resources, waterbird distribution and human activities in freshwater systems: preliminary findings and potential uses. Ibis, 149 (suppl. 1), 65–72.

Ortega, M., Gomez, R., Suarez, M. L., Vidal-Abarca, M. R., Calvo, J. F., Palazon, J. A. and Ramirez-Diaz, L. (1992). Hydrochemical characterization and typification of ponds in a semi-arid region of Southeastern Spain. In R.D. Robarts and M.L. Bothwell (Eds.), Aquatic ecosystems in semi-arid regions: Implications for resource managemen,. NHRI Symposium Series 7, Environment Canada (117-126). Saskatchewan, Saskatoon.

Paquet, J.-Y.; Vandevyvre, X.; Delahaye, L. and Rondeux, J. (2006). Bird assemblages in a mixed woodland–farmland landscape: The conservation value of silviculture-dependant open areas in plantation forest. Forest Ecology and Management, 227(1-2), 59-70.

Pardo, M. T., Esteve, M .A., Giménez, A., Martínez-Fernández, J., Carreño, M. F., Serrano, J. and Miñano, J. Assessment of the hydrological alterations on wandering beetle assemblages (coleoptera: Carabidae and Tenebrionidae) in coastal wetlands of arid mediterranean systems). Journal of Arid Environments (in press).

Peiró, I. G. (2006). Estudios ornitológicos aplicados a la gestión del hábitat de los Passeriformes del carrizal en el Parque Natural de El Fondo (Alicante, SE de España). Tesis Doctoral, Murcia, Universidad de Murcia.

Pérez Ruzafa, A. and Aragón, R. (2003). Implicaciones de la gestión y el uso de las aguas subterráneas en el funcionamiento de la red trófica de una laguna costera. In: J.M. Fornés and R. Llamas (Eds.) Conflictos entre el desarrollo de las aguas subterráneas y la conservación de los humedales: litoral mediterráneo (215-245). Madrid. Ediciones Mundi-Prensa.

Pérez Ruzafa, A., Gilabert, J., Gutiérrez, J.M., Fernández, A.I., Marcos, C. and Sabah, S. (2002). Evidence of a planktonic food web response to changes in nutrient input dynamics in the Mar Menor coastal lagoon, Spain. Hidrobiología, 475/476, 359-369.


Pérez Ruzafa, A., Mompeán, M.C. and Marcos, C. (2007). Hydrographic, geomorphologic and fish assemblage relationships in coastal lagoons. Hydrobiologia, 577(1), 107-125.

Perrow, M. R., Schutten, J. H., Howes, J. R., Holzer, T., Madgwick, F. J. and Jowitt, A. J. D. (1997). Interactions between coot (Fulica atra) and submerged macrophytes: the role of birds in the restoration process. Hydrobiologia, 342-343, 241-255.

Pons, P., Lambert, B., Rigolot, E., and Prodon, R. (2003). The effects of grassland management using fire on habitat occupancy and conservation of birds in a mosaic landscape. Biodiversity and Conservation, 12(9), 1843-1860.

Rainio, J. and Niemelä, J. (2003). Ground beetles (Coleoptera: Carabidae) as bioindicators. Biodiversity and Conservation, 12, 487-506.

Richards, J. A. (1995). Remote Sensing Digital Image Analysis. An Introduction. (2nd edition). Berlin, Springer-Verlag.

Roberts, J. P. and Schnell, G. D. (2006). Comparison of survey methods for wintering grassland birds. J.Field Ornithol., 77 (1), 46-60.

Robledano, F. (1995). Bases para la conservación de humedales en el área del Mar Menor (Murcia. SE de España): criterios para la regeneración y restauración. In: C. Montes, G. Oliver, F. Molina and J. Cobos (Eds.), Bases Ecológicas para la restauración de Humedales en la Cuenca Mediterránea (251-262). Sevilla, Consejería de Medio Ambiente. Junta de Andalucía.

Robledano, F., Calvo, J. F., Esteve, M. A., Mas, J., Palazon, J. A. and Ramirez-Diaz, L. (1987). Tipología, conservación y gestión de las zonas húmedas del sureste español. Limnética, 3(2), 311-320.

Robledano, F., Calvo, J. F., Esteve, M. A., Mas, J., Palazon, J. A., Suarez, M. L., Torres, A., Vidal-Abarca, M. R. and Ramirez-Diaz, L. (1991a). Estudios ecológicos de los humedales costeros del sudeste español. I. Inventario y tipificación. Anales de Biología, 17(6), 153-163.

Robledano, F., Calvo, J. F., Esteve, M. A., Mas, J., Palazon, J. A., Suarez, M. L., Torres, A., Vidal-Abarca, M. R., Ramirez-Diaz, L. (1991b). Estudios ecológicos de los humedales costeros del sudeste español. II. Evolución histórica, situación actual y perspectivas de conservación. Anales de Biología, 17 (6), 165-176.

Robledano, F. and Esteve, M. A. (1992). Conservation and management of wetlands in Murcia Region (SE Spain). The Mar Menor Lagoon. In: M. Finlayson, T. Hollis and T. Davis (Eds.), Managing Mediterranean Wetlands and their birds for the year 2000 and beyond (240-242). Slimbridge, IWRB Special Publication.

Robledano, F., Esteve, M.A., Hernández Gil, V., Blanco Gago, J.C., Pardo, M.T., Martínez, J. and Carreño, M.F. (2006). Cambios en la actividad agrícola y respuesta de las comunidades de passeriformes en los humedales del Mar Menor (Murcia, SE España). Congreso Ibérico de Ornitología. Elche (Alicante, Spain).

Robledano, F., Pagán, I. and Calvo, J. F. (2008). Waterbirds and nutrient enrichment in Mar Menor Lagoon, a shallow coastal lake in southeast Spain. Lakes and Reservoirs: Research and Management, 13, 23–35.

Rodríguez, S., Martínez, J., Carreño, M. F., Pardo, M. T., Miñano, J., Esteve, M. A., Mas, J. and Jiménez, F. (2005). Report on Lagoon Modelling in Mar Menor Site. Univ. Murcia. http://www.dittyproject.org/Public%5CAnnexes%5CUMU%5Creport19.pdf.


Rönkä, M.T.H., Saari, C.L.V., Lehikoinen, E.A., Suomela, J. and Häkkilä, K. (2005). Environmental changes and population trends of breeding waterfowl in northern Baltic Sea. Annales Zoologici Fennici, 42, 587-602.

Roomen, M. van, Koffijberg, K., Noordhuis, R. and Soldaat, L. (2006). Long-term waterbird monitoring in The Netherlands: a tool for policy and management. In: G.C. Boere, C.A. Galbraith and D.A. Stroud (Eds.), Waterbirds around the world (463-470). Edinburgh, UK, The Stationery Office.

Rueda, F. (1990). Biología y ecología de los carábidos de las orillas de lagunas salinas españolas. PhD Thesis. Madrid, Universidad Autónoma de Madrid.

Sampietro, F.J. and Pelayo, E. (2003). Terrera marismeña Calandrella rufescens. In: R. Martí and J.C. del Moral (Eds.), Atlas de las Aves reproductoras de España (370-371). Madrid, Dirección General de Conservación de la Naturaleza-Sociedad Española de Ornitología.

Serrano, J., Lencina, J. L., Andujar, A., Ruano, L. and Andujar, C. (2002). The carabid fauna of the saltmarsh of Cordovilla (Albacete. SE Spain) (Insecta. Coleoptera). In: J. Szyszko, P. J. Boer and T. den Bauer (eds.), How to protect or what we know about carabid beetles (213-224). Warsaw, Warsaw Agricultural University Press.

Serrano, J., Ruiz, C., Andujar, C. and Lencina, J. L. (2005). Land use and ground beetle assemblages in the National Park of Cabañeros. Central Spain (Coleoptera: Carabidae). In: G. Lövei and S. Toft (Eds.), European Carabidology (Proceedings of the 11th European Carabidologists`s Meeting) DIAS Report Plant Production. No. 14. (275-289). Danish Institute Agricultural Sciences.

Smit, H., Van der Velde, G., Smits, R. and Coops, H. (1997). Ecosystem responses in the Rhine-Meuse Delta Turing two decades alter enclosure and stops towards estuary restoration. Estuaries, 20(3), 504-520.

Suárez, F., Garza, V. and Morales, M.B. (2002). Habitat use of two sibling species, the Short-Toed Calandrella brachydactyla and the Lesser Short-Toed C. rufescens Larks, in mainland Spain. Ardeola, 49(2), 259-272.

Tellería, J. L., Asensio, B., Díaz, M. (1999). Aves ibéricas. Volumen II. Passeriformes. Madrid, José Manuel Reyero.

Torralva, M., Vargas, M., Jiménez, R., Arcas, J.M., García, A., Robles, M. and Caballero, A. (2003). Análisis de la viabilidad de la comunidad ornitológica. In: Estudio de la Fauna del Mar Menor y directrices para su manejo y conservación. Vol. II. Technical Document. Consejería de Agricultura, Agua y Medio Ambiente. Murcia, Comunidad Autónoma de la Región de Murcia.

Töyrä, J., Pietroniro, A. and Martz, L.W. (2001). Multisensor hydrologic assessment of a freshwater wetland. Remote Sensing of Environment, 75, 162-173.

Van Bommel, S., Röling, N.G., van Wieren, S.E. and Gossow, H. (2003). Social causes of the cormorant revival in the Netherlands. Cormorant Research Bulletin, 5, 16-25.

Van Eerden, M.R. and Gregersen, J. (1995). Long-term changes in the northwest population of Cormorants Phalacrocorax carbo sinensis. Ardea, 83 (1), 61-79.

Van Eerden, M. R. and Piersma, T. and Lindeboom, R. (2003). Competitive food exploitation of smelt Osmerus eperlanus by great crested grebes Podiceps cristatus and perch Perca fluviatilis at Lake Ijsselmeer, The Netherlands. Oecologia, 93, 463-474.

Van Eerden, M. R., Drent, R. H., Stahl, J. and Bakker, J. P. (2005). Connecting seas: western Palearctic continental flyway for water birds in the perspective of changing land use and climate. Global Change Biology, 11, 894-908.


Van Rijn, S. and Van Eerden M. R. (2003). Cormorants in the Ijsselmeer area: competidor or indicator? Cormorant Research Bulletin, 5, 31-32.

Vasas, V., Lancelot, C., Rousseau, V. and Jordán, F. (2007). Eutrophication and overfishing in temperate nearshore pelagic food webs: a network perspective. Marine Ecology Progress Series, 336, 1-14.

Velasco. J., Lloret. L., Millán. A. Barahona. J. Abellán. P. and Sánchez-Fernández. D. (2006). Nutrient and particulate inputs into the Mar Menor lagoon (SE Spain) from an intensive agricultural watershed. Water. Air and Soil Pollution, 176, 37-56.

Viada, C. (Ed.). (1998). Areas Importantes para las Aves en España (2nd edition). Monografía nº 5. Madrid, SEO/Birdlife.

Vidal-Abarca. M. R., Esteve Selma. M. A. and Suárez Alonso. M. L. (2003). Los Humedales de la Región de Murcia: Humedales y Ramblas de la Región de Murcia. Murcia, Consejería de Agricultura. Agua y Medio Ambiente.

Wang. Q. and Tenhunen. J. D. (2004). Vegetation mapping with multitemporal NDVI in North Eastern China Transect (NECT). International Journal of Applied Earth Observation and Geoinformation, 6, 17-31.

Wang. C., Menenti. M., Stoll. M. P., Belluco. E. and Marani. M. (2007). Mapping mixed vegetation communities in salt marshes using airbone spectral data. Remote Sensing of Environment, 107, 559-570.

Yaacobi. G., Ziv. Y. and Rosenzweig. M.L. (2007). Effects of interactive scale-dependent variables on beetle diversity patterns in a semi-arid agricultural landscape. Landscape Ecology, 22, 687-703.

Ysebaert, T., Meininger, P.L., Meire, P., Devos, K., Berrevoets, C.M., Strucker, R.C.W. and Kuijken, E. (2000). Waterbird communities along the estuarine salinity gradient of the Schelde estuary, NW-Europe. Biodiversity and Conservation, 9 (9), 1275-1296.

Zamorano Ruiz, M. (1932). Notas sobre algunas especies de los antiguos órdenes de Palmípedas y Zancudas, que frecuentan el Mar Menor (Murcia). Cartagena, Libreria Melero.


Chapter 5

PATHOGEN REMOVAL IN CONSTRUCTED WETLANDS

Kela P. Weber and Raymond L. Legge∗ Department of Chemical Engineering, University of Waterloo,

Waterloo, ON, Canada, N2L 3G1

ABSTRACT

Conventional secondary and tertiary wastewater treatment methods include activated sludge, trickling filters, slow sand filtration, chlorination, ozonation and UV radiation. Chlorination being the most widely used pathogen disinfection method is presently under scrutiny as chlorination can produce carcinogenic trihalomethanes when natural organic matter is present in the wastewater. Constructed wetlands (CWs) have proven to be an effective treatment alternative for the removal and inactivation of pathogens in wastewaters. Constructed wetlands have low principle and operating costs and are fairly simple to design and implement, making them an attractive wastewater treatment alternative when compared to conventional secondary or tertiary treatment processes. Constructed wetlands designed for pathogen treatment are most often preceded by filtration or sedimentation. Pathogen removal efficiencies upwards of 99.99% have been reported by multiple authors employing many different constructed wetland designs. Constructed wetland design tends to be based largely on rule of thumb sizing, as the specific mechanisms and fundamental variables involved in pathogen removal are only vaguely understood. Suggested mechanisms of pathogen treatment in CWs include but are not restricted to sedimentation, natural die-off, temperature, oxidation, predation, unfavourable water chemistry, biofilm interaction, mechanical filtration, exposure to biocides and UV radiation. Pathogen removal has been shown to correlate well with hydraulic retention time. Use of first order decay kinetics is the preferred method to describe and predict pathogen removal in CWs. A severe lack of attention has been given to the comparative quantification of the specific mechanisms contributing to pathogen treatment in constructed wetlands. Small-scale controllable constructed wetland systems

∗ Corresponding author: Prof. Raymond L. Legge, Department of Chemical Engineering, University of Waterloo,

200 University Avenue W. Waterloo, ON N2L 3G1 Canada. Tel.: + 1 519 888 4567 ext. 36728.. Fax: + 1 519 746 4979. E-mail: [email protected]

Kela P. Weber and Raymond L. Legge 178

are identified as systems which can be used in conducting well-designed controlled experiments where fundamental mechanisms and variables involved in pathogen removal can be comparatively quantified. It is proposed that if the fundamental mechanisms and variables affecting pathogen removal in constructed wetlands are better understood and quantified the large performance variations reported for similarly designed treatment wetland systems can be better explained, engineered and controlled.

NOMENCLATURE AND ABBREVIATIONS A wetland area [m2] C bacterial concentration [#/m3] C* background conc. [#/100mL] C0 initial conc. [#/100mL] CW constructed wetland DO dissolved oxygen EPS extracellular polymer substances FC fecal coliform FLB fluorescently labeled bacteria FLM fluorescently labeled micro-spheres FS fecal streptococcus FWS free water surface h water depth [m] HLR hydraulic loading rate [mass/time] HRT hydraulic retention time HSSF horizontal sub-surface flow IRC iron oxide coated k area based, first order rate constant k/q Da (Damkőhler number) k1 first-order, zero background aerial rate constant [m/d] kv1 volume-based, first-order decay rate [1/d] NOM natural organic matter R death rate [#/day] SC somatic coliphage SMZ surfactant modified zeolite TC total coliform THM trihalomethane TOC total organic carbon UV ultra violet V wetland water volume [m3] VF vertical flow ε volume fraction of water τ nominal detention time [days]

Pathogen Removal in Constructed Wetlands 179

1. INTRODUCTION

1.1. Pathogens Pathogen removal, pathogen treatment, pathogen disinfection, and pathogen inactivation

are all terms used to describe a similar water quality objective. That objective is to either destroy pathogenic organisms or eliminate the ability of the pathogens found in water to reproduce or create adverse health effects in humans upon entry into the human body. This objective can be met in a variety of ways producing the same effect. Pathogens can be completely removed from the water, destroyed or lysed within the water, or be effectively inactivated by eliminating the ability for the microorganism to reproduce or perform life sustaining functions. Pathogens are quite often treated, within water bodies, by a variety of these methods through one treatment technique or combination of techniques. To avoid confusion the term “pathogen treatment” will be used within this chapter to describe any number of these treatment methods. Whenever possible the specific mechanism will be discussed and the general term “pathogen treatment” will be used when the specific pathogen treatment mechanism (i.e. death, removal or inactivation) is unknown.

A pathogen is a biological agent that causes disease or illness in its host. Parasites in contrast are organisms that live in symbiosis with their host although some parasites are also considered pathogens. Water-born pathogens can be broken up into 5 categories: viruses, bacteria, protozoa, fungi and helminths. This chapter will focus on pathogens of bacterial, viral and protozoan nature, as treatment of these water born pathogen types is often more difficult.

For treatment purposes, viruses can be considered submicroscopic (20-200 nm), non living particles of genetic material that are enclosed in a sheath. Viruses cannot reproduce or divide on their own but infect and reproduce at the expense of a host organism. There are over 100 virus types in human feces, with as low as 1 organism infective dose for some viruses. Some of the most common water born viruses are enteroviruses including poliovirus (polio), coxsackievirus (meningitis and colds), and echovirus (meningitis and colds) (Kadlec and Knight, 1996).

Bacteria are prokaryotic microorganisms with sizes ranging from 0.1-5 μm. Bacteria are commonly found in feces with normal populations around 1011 organisms/g (Leclerc et al., 1977). Most bacteria found in humans live in symbiosis with their host although some are known human pathogens. Bacteria when engaged in their normal germination cycle are quite easy to remove or inactivate in waster-water. Some bacteria when faced with adverse environmental conditions can form a sporozoite or spore. Spores can be considered in a dormant state with a thick outer layer to their cell wall which allows them to withstand harsh environmental conditions until such time when the spore can “reactivate”, and again participate in the normal germination cycle. Spores are much more difficult to inactivate or kill than regularly germinating bacteria. Some of the most common bacteria pathogens are Salmonella spp. and E. coli.

Protozoans range in size from 4-10 μm and are known to feed on other pathogens such as bacteria. Pathogenic protozoans include Entamoeba histolytica and Giardia lamblia, both causing diarrhoea in their host. The most common pathogenic water born protozoan belongs to the genus Cryptosporidium (Slifko et al., 2000; Morsy et al., 2007).


Cryptosporidium is an intracellular coccidian protozoan parasite that infects the gastrointestinal tract of a wide range of vertebrate hosts including humans (Xiao et al., 2004; Morsy et al., 2007). One reason Cryptosporidium is effective at evading pathogen treatment and infecting hosts is its ability to form oocysts. Oocysts when compared to a regularly occurring protozoan can be roughly related to the relationship between bacterial spores and regularly germinating bacteria as previously discussed. The clinical signs of cryptosporidiosis in humans are mainly diarrhoea, dehydration, malabsorption, weight loss and/or wasting. Infection is self-limiting, but chronic infections may establish, particularly in young and immunosuppressed individuals (Fayer et al., 2000, Morsy et al., 2007).

1.2. Pathogen Quantification Methods In order to fully characterize the microbial population within a wastewater complete

enumeration and identification of all microorganisms is needed. However, this is often technically, economically and temporally infeasible. Therefore, to quantify pathogenic microbial populations, methods for quantifying groups of bacteria or enumerating specific indicator organisms are utilized. Indicator organisms need to be easy to monitor and also need to correlate with pathogenic populations. No indicator organism or group of organisms is a perfect indicator as the characteristics of different pathogens can be quite different. For example viruses, bacteria and protozoa all behave quite differently within CWs and therefore will be removed at different rates and correlate with different indicator organisms to different degrees.

The two most popular and commonly used indicator tests are total coliforms (TC) and fecal coliforms (FC). The total coliform test enumerates gram negative, facultatively anaerobic, rod shaped bacteria that do not form spores (Kadlec and Knight, 1996). The enumeration is done with a lactose substrate over a 48 hr period at 35ºC. Fecal coliform are a subset of total coliform, testing for a more specific set of bacteria originating from a mammalian source. The fecal coliform test is performed with a lactose substrate over a 24 hr period at 44.5ºC.

One limitation to keep in mind with TC is that many of the bacterial species being enumerated are commonly found in surface waters and therefore TC includes many bacterial species not originating from human or other animal fecal sources. Therefore, TC is not a perfect indicator for human fecal contamination. FC are largely fecally derived species, but also include free coliform groups (Klebsiella spp. primarily) and bacteria from other warm blooded animals like birds and mammals (Kadlec and Knight, 1996). Although FC is a better indicator than TC in testing for human fecally derived species, it is not a perfect indicator either.

Two popular indicator organisms are E. coli and fecal Streptococcus (FS). E. coli is a bacteria belonging to the coliform group. It is often used to indicate human fecal contamination. E. coli constitutes about 20-30% of TC found in wastewater (Dufour 1977; Kadlec and Knight, 1996). It can also be argued that E. coli is not a good fecal contamination diagnostic because E. coli also originates in other warm blooded animals. As well, if E. coli has been shown to correlate so well with TC one could simply use the TC test. FS can be found in the feces of humans and warm blooded animals such as birds and mammals (Kadlec and Knight, 1996). FS are often found in fecal contaminated waters but are not observed to


multiply in natural or polluted waters. FS also have a long lifespan compared to fecal coliforms and are therefore used as a second indicator. FS is said to be a better indicator for longer living viruses (Clausen et al., 1977; Kadlec and Knight, 1996). Some species of Streptococcus such as Streptococcus faecalis and S. durans are considered to be diagnostic of human fecal contamination. FC/FS ratios can also be used to provide indications of a contaminant source. FC/FS ratio for nonhuman waste is typically less than 0.7, whereas the FC/FS ratio for human waste is usually around 4.0 (Clausen et al., 1977, Kadlec and Knight, 1996). It should be mentioned that die-off affects and therefore for diagnostic purposes are only applicable up to ~24 hrs after waste discharge (Vymazal, 2005).

It is generally accepted that FC is not a good indicator for viruses because viruses are more resistant to chlorination and environmental deactivation (Kraus, 1977; Gersberg et al., 1987). By this same logic, FC will not be a good indicator for bacterial spore removal. Bacteriophages (viruses that infect bacteria) such as Coliphage MS-2 have been used as viral indicators in wetland treatment systems (Gersberg et al., 1987; Kadlec and Knight, 1996). Positive aspects of using MS-2 include ease of enumeration, MS-2 is nearly the same size as the enteroviruses, and MS-2 is more resistant to UV, heat and disinfection than most enteric viruses. Therefore MS-2 is a fairly representative, conservative viral pathogen indicator.

1.3. Conventional Pathogen Treatment Methods Conventional methods of pathogen removal are both effective and reliable. Activated

sludge, trickling filter and slow sand filtration are the most commonly cited methods. All of these treatment methods, when properly used, are very effective at removing many different types of pathogens with typical removal efficiencies around 99-99.99% (Vymazal, 2005). Most of these processes still employ a tertiary or polishing step for further pathogen treatment.

Chlorination, ozone and UV disinfection are three of the most cited and commonly used tertiary treatment (or polishing) pathogen removal methods. Chlorination has been the most popular method of disinfection as it is simple to implement, relatively inexpensive, and requires minimum effort to add on to an existing system. Chlorination can also provide a residual chlorine concentration within the distribution system. Chlorination has been shown to be effective at treating both bacteria and viruses, however less effective at treating protozoa oocysts such as Cryptosporidium parvum (LeChevallier and Au, 2004).

Although chlorination is a popular method of disinfection, a growing concern over chlorination has developed in recent years for a variety of reasons. Chlorine residuals can be harmful to a variety of microorganisms and fish in receiving waters, and free chlorine when in contact with a significant amount of natural organic matter can form trihalomethanes (THM’s) or other organo-chlorine compounds which are known to be carcinogenic (LeChevallier and Au, 2004). The possible formation of harmful by-products is the main reason for the development of other tertiary (or polishing) conventional-type pathogen treatment methods such as UV and ozonation.

Although UV and ozonation have been developed to compete with chlorination, several technical limitations still exist such as fluence measurement and difficulties with turbidity for UV. As well, neither technology is able to ensure continued disinfection once the water enters the distribution system and, in comparison to chlorination, both technologies require an


increase in energy usage. Cost and maintenance have limited the use of UV and ozonation (Kadlec and Knight, 1996).

Low energy non chemical methods of pathogen treatment have been used when receiving waters are non potable or disinfection requirements relaxed. Wetlands have shown to reduce pathogen counts to varying degrees (Vymazal, 2005). Constructed Wetlands have low principle and operating costs making them popular in remote areas and developing countries. The wetland concept has become an attractive cost-effective wastewater treatment alternative compared to conventional or tertiary treatment processes (Morsy, 2007).

1.4. Constructed Wetlands The natural ability of wetland ecosystems to improve water quality has been recognized

for more than 25 years (Knight et al., 1999). The main idea behind a constructed wetland is to try and reproduce a wetland ecosystem with the expectation that the contaminated water will be treated as it passes through the wetland system. Constructed wetlands have been used to treat a number of different contaminated waters including organic farm waste (Cronk, 1996), food processing waste (Burgoon et al., 1999), human wastewater (Decamp and Warren, 2000), and acid mine drainage (Mitsch and Wise, 1998).

There are three general types of constructed wetland systems; free surface water (FSW), horizontal subsurface flow (HSSF), and vertical flow (VF) constructed wetlands. FSW CWs are built by first digging a trench with a slight incline slightly (~1º) from inlet to outlet to allow gravity movement of the water. This trench is then lined with an impermeable polymer, or low a permeability soil such as densely packed clay. This lined trench is then filled with the desired bed media. In some geographical locations the native soil may have a permeability low enough such that no liner is required. In this case the trench can simply be dug and filled with the desired bed media. Common media include peat, gravel, sand, soil and compost. FWS constructed wetlands contain standing water on the surface of the treatment system and can be fed from either below ground or above ground (Figure1).

HSSF CWs are similar in design to FSW CWs however are fed from below ground and are designed so the maximum flow rate will not allow surface water formation on top of the HSSF CW (Figure 2).

Figure 1. Free surface water constructed wetland.


Figure 2. Horizontal subsurface flow constructed wetland.

VF CWs are similar in design to HSSF CWs however are not built with an incline as the water is fed into the top of the wetland system. A large impermeable enclosure is built and filled with a substrate, water is then fed into the top of the VF CW. Water is usually fed into VF wetlands intermittently allowing for bed aeration and to deter surface water formation (Figure 3). All three types of CW systems may contain different emerging macrophytes (large aquatic plants rooted to the bed media). In some cases, usually for study purposes, macrophytes are not included in the CW design. The most commonly utilized macrophytes in CW designs are Phragmites australis (common reed) and Typha species (cattail or bulrush).

Figure 3. Vertical flow constructed wetland.


Wetland design is based mainly on configuration, size and flow records from previously employed systems. Wetlands are land (area) intensive technologies making implementation infeasible in populated areas. Although CWs have proven to be an effective treatment technology for a variety of contaminants, systems will fail due to temperature changes, hydraulic issues (clogging/short circuiting), and other undetermined reasons. Although there are more than 650 full-scale constructed wetland systems in place throughout the world (Diemont, 2006), they are for the most part consistently treated as a black box. The specific mechanisms and the underlying fundamental variables affecting the functionality of treatment wetlands have been given little attention. This fact can be mainly attributed to the lack of feasible testing methods and associated systems (Stottmeister et al., 2003) which has hampered the confidence and support for the technology. There has long been a fundamental inability to make a priori predictions in terms of the effectiveness of a proposed wetland for a given flow and input water chemistry (Wieder et al., 1989).

2. PATHOGEN REMOVAL IN CONSTRUCTED WETLANDS Pathogen removal is most often required when dealing with domestic wastewater. In

some instances ground water can exhibit high levels of pathogenic counts but pathogen removal in CWs is usually discussed in the context of water with a domestic waste origin. CWs have recently been implemented to treat many different types of municipal wastewaters (Desena, 1999) with examples that include effective secondary and tertiary applications for the removal of pathogens such as fecal coliforms, Cryptosporidium, and Giardia (Gerba et al., 1999; Neralla et al., 2000) and parasites such as Ascaris suum, Toxacara vitullorum and Hymenolepis diminuta (Stott et al., 1999).

Vymazal (2005) conducted an extensive worldwide survey of CWs treating enteric bacteria and grouped 60 CW systems into FWS, HSSF, VF and hybrid systems, with hybrid systems representing the use of more than one type of CW system in series. The most common configurations of hybrid systems were FWS to HSSF and VF to HSSF systems. A relatively even distribution of configuration type was found with slightly more HSSF systems and slightly fewer VF systems being used. Although 60 systems represent a large number of systems, in comparison to the number of conventional treatment systems in use around the world, it is rather insignificant. From this data it was also seen that the majority of CWs used to treat pathogens can be found in Europe and the Americas.

2.1. Constructed Wetland Design CW systems are complicated in their chemistry, hydraulics, and distribution of specific

removal mechanisms, therefore so simplified relationships and models for pathogen removal have been sought. Pathogen removal in CW systems has been shown to correlate well with hydraulic retention time (HRT). Since CW size is proportional to HRT, pathogen removal in CWs has been modelled based on different size aspects of the CW systems.

Hydraulics, kinetics and other design criteria such as minimum depths, maximum superficial velocities, minimum HRTs and wetland slopes all need to be accounted for when


designing a CW system. For general considerations regarding wetland design see Kadlec and Knight (1996). Wetland systems are most often modelled as plug flow reactors incorporating both dispersion and kinetics into the model (Cronk, 1996). Hydraulics are quantified through hydraulic conductivity, porosity, and in some cases, specific values for dispersion. Kinetics are almost always accounted for by using 1st order rate constants. Although regression and Monod type equations are used for other types of water treatment, pathogen removal is almost exclusively modelled using 1st order kinetics. Other design criteria are based mostly on past experience and include but are not limited to a maximum superficial velocity of 8.6m/d to avoid damage to plant root systems (Sauter and Leonard, 1997), a minimum specific surface area of 1m2 for tertiary treatment of municipal wastewater in vertical flow systems (Schoenerklee et al., 1997), and a slope of 1% for proper water movement.

2.2. Kinetic Models Volume based first order decay models have been used to describe pathogen treatment in

lagoons and wetland systems (Kadlec and Knight, 1996).

CAhkVCkR vv )(11 ε== (Equation 1) where:

R = death rate [#/day] kv1 = volume-based, first-order decay rate [1/d] V = wetland water volume [m3] C = bacterial concentration [#/m3] ε = volume fraction of water A = wetland area [m2] h = water depth [m] Another, more popular model, used to describe pathogen treatment in CW systems is the

area based 1st order decay model (Kadlec and Knight, 1996).

ACkR 1= (Equation 2) where:

k1 = first-order, zero background aerial rate constant [m/d] Equation 2 is one of the most widely used models to describe pathogen removal in CWs.

Although it is simple to use, and for individual purposes can often accurately describe pathogen removal in any one CW system, the use of this equation has made the comparison of different CW systems difficult. There are many latent variables not accounted for when reporting a single aerial based first order rate constant. One of the most obvious latent variables is wetland depth. As wetland depth can affect oxygen concentrations and therefore specific biological populations and reactions, it seems that Equation 1 which makes use of a volume based first order rate constant would, to a certain degree, better describes pathogen


treatment in CW systems. Treatment performance continues to be most often recorded and reported based on an aerial first order rate constant.

The k-C* model is another popularly used equation to describe pathogen treatment in CW systems (Kadlec and Knight, 1996). The k-C* model incorporates background concentration (C*) into the model to try and better represent the conditions within the CW system.

*)( CCkAR −= (Equation 3)

)exp()exp(*)(*)(

0

τvkqkCCCC

−=−=−−

(Equation 4)

where:

k = area based, first order rate constant C* = background conc. [#/100mL] C0 = initial conc. [#/100mL] τ = nominal detention time [days] k/q = Da (Damkőhler number) Common background concentrations are found in the range of 10 to 500 units/100mL for

fecal coliforms, with typical rate constants in the area of 75 m/yr for FWS CWs and 95 m/yr for HSSF CWs (Kadlec and Knight, 1996). Although the k-C* model is an improvement of Equation 2, removal is still most often described using an aerial based rate constant which, as discussed above, yields complications when comparing different systems.

2.3. Constructed Wetland Design and Implementation Constructed wetland design tends to be based largely on rule of thumb sizing rather than

insightful engineering. From a scientific and engineering perspective a better understanding of the fundamentals of CWs is required so that CWs can then be designed and optimized for better performance.

As previously mentioned there are many drawbacks to using aerial based 1st order decay kinetics and sizing a wetland based solely on area and inlet concentrations. Using these methods it is assumed that effluent concentrations are directly related to inlet concentrations and flow, and that removal efficiency will be based on wetland area. However, as seen with the wide range of removal efficiencies for similarly designed CWs, performance cannot be accurately described based solely on these variables and doing so can lead to erroneous conclusions (Kadlec, 1997; Werker et al., 2000). Simple system changes such as a change in water levels will affect the concentrations of contaminants within the system (Kadlec, 1997; Neralla et al., 2000). As well, seasonal cycles affecting inlet flows and concentrations, solar radiation, temperature (air, soil and water), precipitation, evapotranspiration and biomass in wetlands, will all affect system performance (Kadlec, 1999).

Methods involving averaging certain variables over longer time periods have been suggested (Tanner et al., 1998). With this idea in mind it is also suggested that models used


for design should not use time periods shorter than 3 detention times to account for variable retention times (Kadlec, 1997).

Other factors to be considered in CW design include but are not limited to local climate, site topography, site geology, loadings, local drainage areas, land availability, cost, size and extent of receiving water body and water quality objectives (Shutes et al., 1999). Designing and implementing CW systems is not an easy task and an understanding of total CW system dynamics is lacking, as a result most designs turn to a conservative-based design using the parameters based on past experience.

2.4. Removal Efficiencies Vymazal (2005) presented removal efficiencies and 1st order aerial rates recorded for

different in-use CW systems for 4 different indicator organisms. Removal efficiencies ranging from 65% to 99% were observed with the highest removal rates observed for hybrid systems, followed by HSSF and finally FWS systems.

Pathogen treatment in wetland systems relies on a host of mechanisms within the CW system including sedimentation, natural die-off, temperature, oxidation, predation, unfavourable water chemistry, adhesion to biofilm, mechanical filtration, exposure to biocides and UV radiation (Kadlec and Knight, 1996; Vymazal, 2005; Cronk, 1996; Gerba et al., 2000). With these mechanisms in mind, some of the most prevalent latent variables not described within a simple first order aerial based rate constant are substrate type, plant type, microbial activity within the CW system, microbial ecology within the CW system, biofilm interactions, temperature, incoming water quality characteristics, detection variability and, as previously discussed, wetland depth. Many other variables could be identified, however this short list has been restricted to provide the most prevalent and obvious.

Currently, the design of CWs as a treatment technology is based largely on area and configuration. Most published research consists of case studies reporting percent removal or a 1st order k value. The possible mechanisms of bacterial removal are, to large extent, not discussed. Due to the lack of systematic analyses, the specific removal processes and the fate of potential pathogenic bacteria in constructed wetlands is not yet well understood (Vacca, 2005).

2.5. Pre-Treatment Pre-treatment has been cited as a requirement when using CW systems to treat domestic

wastewater with a large amount of particulate matter (Anderson et al., 1996; Cronk, 1996; Kern and Idler, 1999; Perfler et al., 1999; Peterson, 1998; Sauter and Leonard, 1997; Williams et al., 1999). Suspended solids can temporarily or permanently clog CW systems (Cronk, 1996; Schoenerklee et al., 1997). Clogged systems are subject to short circuiting, creating unwanted flow patterns and reduced contact time between the contaminant and the wetland substrate (Tanner et al., 1998).

Sedimnetation is cited as the most common method of pre-treatment for CW systems (Anderson et al., 1996). Sedimentation can be performed through the use of lagoons, equalization basins, ponds, settling tanks or septic tanks (Frostman, 1996; Philippi et al.,


1999; Sauter and Leonard, 1997). Mechanical aeration can also be used as a pre-treatment process (Cronk, 1996); however utilizing this technology can significantly increase the cost of the treatment system in question. Another form of pre-treatment is to use a hybrid “FWS to HSSF system”. Sending wastewater first through a FWS CW can effectively reduce the number of particles found within the water while not observing the effects of clogging (Rochfort et al., 1997). Use of a pre-treatment system can also help in equalizing flows over long periods of time, reducing the negative effects of low or high flows (Frostman, 1996).

2.6. Design Limitations

2.6.1. Hydraulic Challenges Channelling or short-circuiting has been shown to impact treatment performance (Rash

and Liehr, 1999; Sauter and Leonard, 1997; Scholes et al., 1998). As discussed previously, pre-treatment can greatly reduce the chances of clogging and channelling. The presence of plant roots has been shown to direct water below the root zone in wetland systems (Rash and Liehr, 1999), creating a vertical stratification. In some cases there is evidence that plants may cause short circuiting, however this may be limited to start-up periods (Frostman, 1996). Wetland width can also have an effect on channelling as edge effects have been shown to influence CW hydraulics (Tanner et al., 1998). Lastly, the position of effluent removal has been shown to have an effect on channelling. Collecting effluent from the bottom of the CW system has been shown to result in vertical stratification (McNevin et al., 2000).

2.6.2. Water Level Maintenance

Water levels need to be maintained in wetland systems to ensure plant health, consistent microbial activity, and consistent microbial ecology within the CW system. Variable or seasonal flows need to be equalized to ensure a CW system does not dry out or overflow (Anderson et al., 1996). The need for continuously flowing water to avoid pipe freezing during winter, coupled with unexpected snow melts, can cause system overflows in cold weather regions (Wittgren and Maehlum, 1997). Dry periods in arid geographical regions can effectively dry out CW systems through lack of inlet flow and increased evapotranspiration. As previously discussed use of a pre-treatment system can help to maintain equalized flows throughout short and long time periods. One simple way to help ensure a CW system does not dry-out due to reduced inlet flow is to place the effluent pipe at a height that ensures that the minimum necessary water level is maintained. This however may not be possible if treating water with a high level of particulate matter as this may result in plugging.

2.6.3. Cold Climates

Cold climates present another challenge for CW design. Decreased microbial activity, plant dormancy, and freezing of the water column can all occur due to cold temperature operation (Werker et al., 2002). The simplest was to alleviate these issues is to find ways to insulate the CW system. HSSF systems are often used in cold temperature climates to eliminate free water freezing and make use of warmer ground temperatures in the winter (Dusel and Pawlewski, 1997; Revitt et al., 1997). Vegetation can be used to help produce an insulating layer of mulch on top of the CW system (Smith et al., 1997). Emerging vegetation


can also entrap snow, further insulating the system (Wittgren and Maehlum, 1997). It has also been suggested that CW systems could be made deeper, to avoid surface freezing. This may have a disadvantage in cases where higher levels of dissolved oxygen are required as installing deeper CW system can hinder treatment performance. It should be noted that colder temperatures can in some cases improve pathogen removal in CW wetlands and therefore cold climate design only becomes an issue in extreme cases where freezing may occur.

3. MECHANISMS OF PATHOGEN TREATMENT IN CONSTRUCTED WETLANDS

3.1. Mechanistic and Factor Specific Studies

Through superior fundamental studies and insightful characterisation and study of full

scale wetlands, the fundamental variables and mechanisms involved in water treatment can be identified to provide a better understanding of these systems. With a better understanding enhanced design, implementation and monitoring of in-use systems will ensue, allowing for improved performance for CWs of all sizes and configurations.

Suggested mechanisms of pathogen treatment in CWs include but are not restricted to sedimentation, natural die-off, temperature, oxidation, predation, unfavourable water chemistry, biofilm interaction, mechanical filtration, exposure to biocides and UV radiation (Kadlec and Knight, 1996; Vymazal, 2005; Borisko et al., 2000; Cronk, 1996; Gerba et al., 2000). Table 1 lists these mechanisms with some suggested design parameters which may have an effect on each respective mechanism. Also included in this table are estimated times for each mechanism to have an effect on pathogen treatment in CW systems.

Studies examining pathogen removal in CWs have been limited mainly to case studies and full-scale system research. In most cases a simple first order rate constant is reported or an overall removal efficiency given.

Table 1. Summary of pathogen treatment mechanisms in constructed wetlands

Removal Mechanism Design Factor Effective Time

Sedimentation bed media type, configuration minutes-years Natural die off hydraulic retention time days-weeks Temperature location, plant presence, microbial activity days Oxidation plant presence, configuration hours-days Predation microbial ecology and activity minutes-days Unfavourable water chemistry bed media type minutes-days Adhesion to biofilm microbial ecology and activity minutes-hours Mechanical filtration bed media type, configuration minutes-hours Exposure to biocides plant type minutes-hours UV radiation configuration seconds-minutes


There have been a small number of studies looking at specifically selected mechanisms. The following discussion will focus on mechanistic and factor specific studies incorporating CW systems as either a secondary or tertiary pathogen removal treatment system in either field or laboratory operation. For a detailed discussion of the mechanisms of pathogen removal in a more generally defined porous media see Stevik et al. (2004).

3.1.1. Sedimentation

Karim et al. (2004) found Giardia cysts and Cryptosporidium oocysts to have a 1-3 order of magnitude higher concentration in the sediment of FWS CWs as compared to the water column, suggesting sedimentation to be a factor in pathogen removal. This same study found sedimentation to be size related with larger particles having a higher sedimentation rate. Karim et al. (2004) also found that wetland sediments prolonged the survival of bacteria but not pathogens, suggesting that pathogens may be removed or inactivated within wetland sediment where other bacteria are not.

3.1.2. Natural Die-off

Natural die-off has been cited as a major mechanism of pathogen treatment in CWs (Green et al., 1997; Curds, 1992; Decamp and Warren, 1998; Wand et al., 2007) and is accordingly cited to be correlated with HRT (Kadlec and Knight, 1996). Wand et al. (2007) found natural die off to be a relevant treatment mechanism in their study. Karim et al. (2004) were able to quantify natural die-off rates in conjunction with a study into sedimentation. Die-off rates were found to be greater for bacteria and coliphage in the water column as compared to the sediment. The die-off rates of fecal coliforms in the water and sediment were 0.256 log10 day-1 and 0.151 log10 day-1, respectively. Die-off rates of Salmonella typhimurium in the water and sediment were 0.345 log10 day-1 and 0.312 log10 day-1, respectively. The die-off rates of naturally occurring coliphage in the water column and sediment were 0.397 log10 day-1 and 0.107 log10 day-1, respectively. Giardia die-off rates were greater in the sediment than the water column with die-off rates of 0.37 log10 day-1 and 0.029 log10 day-1, respectively. Of all the microorganisms studied, coliphage survived the longest amount of time in the sediment and the least amount of time in the water column. In contrast Giardia survived longest amount of time in the water column and the least amount of time in the sediment (Karim et al., 2004).

3.1.3. Temperature

Human pathogens function most efficiently around internal body temperatures (~37°C), it is therefore assumed that lower temperatures will either inactivate pathogens or force them into a spore or oocysts state. Inactivation helps treat pathogens in the water although it is increasingly difficult to treat spores or oocysts. Increasing the temperature will help increase the activity of non pathogenic organisms such as grazing protozoa, possibly increasing pathogen ingestion rates by grazing protozoa, however increased pathogen activity is possible as well. There is therefore some debate as to whether an increase in temperature increases pathogen removal rates. Temperature dependence or lack thereof most likely depends on what mechanisms are dominant within specific CW systems.

Ulrich et al. (2005) found that in summer vs. winter operation, higher wastewater temperatures improved pathogen removal performance by about 1 log unit if other factors


such as microorganism concentrations and hydraulic loading did not influence the annual variation pattern. Quinonez-Diaz et al. (2001) showed that temperature appeared to play a significant role in the reduction of enteric bacteria and viruses in gravel bed CWs through correlation coefficients ranging from 0.35 to 0.61. Pundsack et al. (2001) found that wetlands, sand filters and peat filters all performed better during summer months. Perkins and Hunter (2000) found that applying straw to the top of unplanted areas of CW systems increased bacterial removal efficiency. Although not discussed by Perkins and Hunter (2000), straw will act as an insulation blanket possibly increasing bed temperature which may partly account for higher removal efficiencies when straw was applied to unplanted areas. Rivera et al. (1995) discussed that CW systems in temperate areas showed seasonal variations in removal with higher removals during the summer months. Williams et al. (1995) showed a correlation in BOD and FC in their wetland units and observed that BOD removal was higher in the summer.

Karathanasis et al. (2003) found vegetated systems performed best during warmer months and that unplanted systems performed best during the winter. Jillson et al. (2001) reported an average removal of FC during summer periods of 96.7%, and an average winter removal of 99%. If the original start-up period is excluded, summer removal was shown to be 98.5%. This cautionary statement made by Jillson et al. (2001) is well taken as many studies do not identify if start-up periods are included in the quantification of the mechanistic effects.

Zdragas et al. (2002) showed coliform population levels to be strongly inactivated by solar radiation at low temperatures (<10°C), but inactivated to a lesser degree at higher temperatures. This demonstrates that pathogen treatment mechanisms are synergistic and cannot be studied or quantified independently. Zdragas et al. (2002) used multiple linear regression to attain surface plots showing the effects of cross terms. This study was well performed using the data available and these analysis techniques should be used as a template for further studies.

3.1.4. Oxidation

Enteric bacteria are either facultative or obligate anaerobes and thus the presence of oxygen creates unfavourable conditions for these organisms (Vymazal, 2005). Plants are known to exude O2 through their roots into the rhizosphere space of CW systems (Kadlec and Knight, 1996). As a result, plant effects and DO are quite often assumed to be directly correlated. Green et al. (1997) measured DO concentrations in selected wetland systems and found that greater dissolved oxygen (DO) allowed for increased E. coli removal.

3.1.5. Predation

Predation or grazing by protozoan ciliates and flagellates is often cited as a mechanism for pathogen removal in CWs (Stott et al., 1997; Kadlec and Knight, 1996). Morsy (2007) found an average Cryptosporidium oocyst predation rate by wetland ciliates of 10 oocysts cell-1 h-1 at high oocyst concentrations. Wand et al. (2007) suggested predation, most likely by protozoa in general, to be the dominant removal mechanism of bacteria in CWs. Decamp et al. (1999) recorded mean E. coli grazing rates using fluorescently labelled bacteria (FLB) for Paramecium spp. (1.85 FLB/cell/min), which was the largest ciliate used in the study, followed by oxytrichids (1.104 FLB/cell/min), Halteria (0.648 FLB/cell/min) and scuticociliates (0.433 FLB/cell/min), the smallest ciliates used in the study. Decamp et al.


(1999) also found that at the measured HRT within their studied wetland, grazing could account completely for the E. coli removal observed. Rice et al. (1998) determined bacterial prey concentrations of 104/mL would be needed to sustain a CW population of Bdellovibrio, pointing out that the complex ecological populations within CW systems require a balance of food and nutrients to consistently perform to expectations. Any periods where prey is lacking will adversely affect grazing protozoa within the wetland system. Therefore, not having consistent FC counts in the inflow may impact CW performance and should be considered in predictive models and design.

3.1.6. Unfavourable Water Chemistry

No studies were found regarding the effects of unfavourable water chemistry in CW systems on pathogen treatment. However, the role of pH, ionic strength and NOM concentration has been discussed elsewhere with respect to porous media filters (Stevik et al., 2004).

3.1.7. Biofilm Interaction

The activity of microbial communities, particularly those associated with biofilms, is considered to play a key role in the removal, mineralisation and transformation of particulate and dissolved contaminants in treatment systems (Stott and Tanner, 2005). The biofilm surrounding the bed media and the biofilm found both in the larger bed media area and in the rhizosphere region can impact pathogen treatment by facilitating the retention of pathogens through attachment and by harbouring grazing protozoa in and on the surface of the biofilm region. The rhizosphere region has been shown to contain elevated levels of O2 exuded from the plant roots (Batty et al., 2000), allowing for a contained aerobic microenvironment in HSSF wetland systems. Rhizosphere biofilms have also been shown to enhance the development of bacterial populations with antibiotic activity (e.g., Pseudomonas) (Broadbent et al., 1971).

Richardson and Rusch (2005) observed greater filtration efficiency of CW bed media over time and attributed this observation to a suggested initial acclimation phase when inadequate solids accumulation and biofilm growth occurred on the bed media. Stott and Tanner (2005) showed that removal of fluorescently labelled micro-spheres (FLM) was higher in the presence of thicker biofilms for a range of particle sizes. Removals of 79-81% (0.1 mm FLM), 92-96% (1 mm FLM), and up to 98% (4.5 mm FLM) were observed when thicker (autotrophic) biofilms were present in FWS CW microcosm systems. Lower removals of 43% (0.1 mm FLM), 59% (1 mm FLM) and 84% (4.5 mm FLM) occurred in microcosms containing thinner heterotrophic biofilms.

Vacca et al. (2005) showed that the development of differing microbial populations occur within CW biofilm regions depending on soil type and whether or not the CW system contains vegetation. Weber et al. 2008 found plant presence/absence to affect the type of bacterial populations found within the interstitial water of mesocosm CWs. Larsen and Greenway (2004) also found different microbial communities in planted and unplanted gravel bed CWs. The same study showed that biofilm coverage based on extracellular polymer substances (EPS) was greater on 5 mm gravel as compared to 20 mm gravel. However, on a mass basis biofilm development was not significantly different suggesting that if the same measurements had been taken further along in the development of the CW systems perhaps both types of gravel media would have shown full biofilm coverage.


3.1.8. Mechanical Filtration Mechanical filtration has been well described as a stand alone conventional treatment

technology (Stevik et al., 2004). Mechanical filtration in terms of attachment (or what is sometimes referred to as “adsorption”) of cells to filter bed solids has been cited as a mechanism involved in pathogen treatment in CWs (Williams et al., 1995; Wand et al., 2007). Pundsack et al. (2001) found peat media-based CWs to remove a larger amount of Salmonella than similarly designed sand media CWs. Schulze-Makuch (2003) found iron oxide coated (IRC) sand did not significantly adsorb MS-2 or E. coli, where surfactant modified zeolite (SMZ) adsorbed 99% and 100% for MS-2 and E. coli, respectively. Garcia et al. (2003) showed a microbial inactivation ratio to range between 0.1 and 2.7 log-units for FC and from 0.5 to 1.7 log-units for somatic coliphage (SC) in beds with coarse granular material (5–25 mm). This same inactivation ratio ranged between 0.7 and 3.4 log-units for FC and from 0.9 to 2.6 log-units for SC in the bed with finer material (2–13 mm) showing that media size can also have an effect on attachment of pathogens to bed media. Rivera et al. (1995) showed bacterial removal to be more effective in soil beds when compared to gravel beds, however also showed protozoan removal to be greater in gravel beds as compared to soil beds.

Wand et al. (2007) found an average of attachment of 8.0*106 cells g-1sand in VF CW beds. This value was much lower than the excepted. Theoretical attachment, assuming a monolayer of E. coli cells of 1x3 μm dimensions (Brock et al., 1994) and a specific surface area of 41cm2 g-1sand, was 8.3*108 cells g-1 (2 orders of magnitude greater than the observed attachment); therefore it is suggested that theoretical attachment is not maintained or perhaps never reached in CW systems. Wand et al. (2007) added that adsorption does not play a dominant role in the removal process of bacteria because bacteria were ineffectively attached or “adsorbed” to the sand matrix. However, it should also be noted that hindering pathogens in any way possible gives grazing protozoa a better chance to feed, thus increasing pathogen removal. Again no one mechanism works solely alone but each mechanism is linked in some way to another.

3.1.9. Exposure to Biocides

It has been shown that root excretions of certain aquatic macrophytes including Scirpus lacustris and Phragmites australis kill fecal indicators and pathogenic bacteria (Seidel, 1976; Gopal and Goel, 1993; Neori et al., 2000; Vymazal, 2005). In a study by Soto et al. (1999) it was found that bactericidal excretions within the biofilm microenvironment could be a possible mechanism for TC and FC removal in planted gravel bed CWs.

3.1.10. UV Light

Solar radiation has been cited as a major mechanism in pathogen inactivation. Mid-UV (UV-B; 290–320 nm) and near-UV (UV-A; 320–400 nm) have been identified as having lethal and detrimental effects on coliform populations (Jagger, 1983; Zdragas et al., 2002). Wavelengths up to 700 nm have also been shown to have detrimental effects on coliform populations (Nasim and James, 1978; Jagger 1983; Zdragas et al., 2002). Non-lethal effects of solar radiation such as growth delay, growth inhibition, inhibition of induced enzyme synthesis, reduced active transport, and mutagenesis have been cited (Maki, 1993; Zdragas et al., 2002). Viruses have also been shown to be inactivated by solar radiation, but to a lesser degree than bacteria (Bitton 1980; Zdragas et al., 2002).


Although UV inactivation has been studied in other treatment systems only one study observing UV inactivation in CWs was identified. Zdragas et al. (2002) showed solar radiation has a lethal effect on coliforms, and this effect was greatest at lower temperatures (<10°C) as compared to higher temperatures.

3.1.11. Hydraulic Retention Time and Hydraulic Loading Rate

It is generally accepted that pathogen treatment in CWs is primarily influenced by hydraulic loading rate (HLR) (Brix et al., 2003; Hill and Sobsey, 2001; Tanner et al., 1995), and hydraulic residence time (HRT) (Vymazal, 2005; Tanner et al., 1995). Garcia et al. (2003) showed the microbial inactivation ratio to increase as the HRT increased until a saturation value was reached (3 days for the system studied). Garcia et al. (2003) also showed that the value of the microbial inactivation ratio at the saturation level depended on the granular medium contained in the bed. Green et al., 1997 showed CW pathogen removal efficiency to decrease during wet weather, possibly due to decreased HRT in the CW system and decreased loading rate of pathogens.

3.1.12. Plants

Brix (1997) suggested that macrophytes (aquatic plants) stabilize the surface of constructed wetlands, provide physical filtration through the root systems, insulate the surface against frost during winter conditions, and render huge surface areas for microbial attachment and growth. The root-substrate complexes and associated biofilm may also have the capacity for filtration and adsorption of pathogens (Morsy, 2007). On the other hand, shading by vegetation could reduce exposure from UV light and prevent heating of the wastewater by sunlight thus decreasing the rate of inactivation of microorganisms in FWS CW systems (Morsy, 2007).

Vymazal (2005) suggested that planted CW systems are more effective in removal of bacteria as compared to unplanted beds or unplanted ponds, and that this phenomenon is primarily due to two factors: (1) presence of oxygen in the water column of FWS CWs or rhizosphere region within HSSF CWs and (2) presence of plant exudates with antimicrobial properties. Hench et al. (2003) found increased DO and increased reduction of fecal coliform, enterococcus, Salmonella, Shigella, Yersinia, and coliphage populations in planted wetlands when compared to unplanted CW systems. Quinonez-Diaz et al. (2001) concluded that an observed increase in removal rates of coliforms and viruses (coliphages and enteroviruses) in planted CWs, when compared to unplanted CWs, was due to the presence of plant roots. Gersberg et al. (1987) also found planted CWs, when compared to unplanted CWs, to have greater removal efficiencies for viruses (bacteriophages) and concluded that this increase in removal was due to adsorption and filtration in the root system. Hill and Sobsey (2001) found the presence of vegetation in a HSSF wetland improved the removal of Salmonella, fecal coliforms and E. coli when compared to a non-planted HSSF CWs. Soto et al. (1999) also found planted gravel bed CW systems to be more effective at removing TC and FC. Rivera et al. (1995) found that planted CW systems were more effective at removing TC and FC in soil media CWs, however no significant difference between plant types [P.australis (common reed) and typha (cattail or bulrush)] was observed. This same study also showed plants to have no effect on TC or FC removal in gravel beds. Lastly Rivera et al. (1995) showed plants to significantly affect protozoan removal in both soil and gravel bed CWs. Although these


studies outline the improvements in pathogen treatment due to the presence of plants, there are also a number of studies showing no effect or a negative effect on pathogen removal, when plants are included in CW design.

Keffala and Ghrabi (2005) found that plants did not improve FC removal in their CW systems. In fact the planted CWs performed slightly worse when compared to unplanted systems. Karathanasis et al. (2003) found no significant difference in the average yearly removal of fecal bacteria between planted and unplanted gravel bed CWs. Tanner et al. (1995) also found no significant difference in FC removal in gravel bed CWs.

Wand et al. (2007) at first found no difference between removal rates of bacteria in unplanted and planted CWs, however when the effects of transpiration on concentration were accounted for, a much higher removal efficiency could be concluded for the planted CWs. This point is well taken, as transpiration is something not considered in many studies.

Although the effects of plants on removal efficiencies has sometimes been shown to be negligible, the effect plants can have on overall system health and stability cannot to be understated. As previously mentioned plants can help serve to insulate CW systems during cold periods, provide a source of oxygen and carbon to the system through root excretions, and have been recently been shown to perhaps stabilize the bacterial communities within the CW systems during disturbances. Weber et al. (2008) showed the interstitial bacterial community within mesocosm systems containing Phragmites australis could better tolerate and recover from a system perturbation, suggesting that planted CW systems may be more robust and able to better handle system disturbances, enhancing system stability and long term treatment performance.

4. EXPERIMENTAL RESEARCH Case studies involving large scale in-use wetland systems give practical information for

design and implementation. Pilot studies involving both the treatment of contaminated waters in conjunction with comparison of wetland design orientations aim at understanding the effect of different design aspects on treatment performance. However, when working with in-use constructed wetlands there is a lack of practical replication and environmental control, and extended response times hinder true comparative or quantitative studies. Most constructed wetlands have been installed for non-research purposes (Wieder, 1990); therefore, most advances in the field can be attributed to large compilations of case study data. True experimental design and subsequent statistical analysis is not achievable for most constructed wetland research endeavors. Many case study reports do not take into account numerous factors which can vary between wetland systems over the study period due to the naturally divergent nature of many uncontrolled variables in wetland systems. These factors include but are not limited to biological activity, bacterial community changes, hydraulic dispersivity, porosity, and evapotranspiration effects. The resulting lack of understanding towards the specific mechanisms and the underlying fundamental variables affecting the functionality of treatment wetlands has resulted in a wide variation in reported performance values for similarly designed systems (Kadlec and Knight, 1996), with little explanation as to the reasons for this variation.


Several challenges are apparent when designing experiments aimed at the study of pathogen removal mechanisms. 1) Large scale in-use wetland systems offer little control over environmental variables. 2) Pathogen removal mechanisms do not act independently, rather certain removal mechanisms may be correlated or act synergistically. 3) Large-scale in-use treatment wetland monitoring data often does not contain useful information regarding the action of specific pathogen removal mechanisms. 4) Appropriate statistical analysis is often unattainable using in-use wetland monitoring data.

Experimental designs yielding quantitative and comparative information are needed to understand the interdependent nature of pathogen removal mechanisms. The four difficulties identified can be overcome through the use of small laboratory scale wetland systems, viable and non viable experimental tracers, well designed experiments, and the use of analysis of variance (anova) and general linear models for subsequent experimental analysis.

4.1. Laboratory Scale Experimental Systems Although laboratory scale wetland systems do not completely recreate or represent full

scale in-use wetland systems they do offer a number of advantages when investigating fundamental variables or mechanisms affecting treatment performance. Fundamental investigations are often not directed at accurately estimating removal coefficients in large scale systems, but rather at gaining an understanding of the quantitative comparison of certain variables or mechanisms on overall removal. In other words, trying to understand which mechanisms or variables have a significant impact on removal. As well, what factors may influence these significant variables and mechanisms, and how do these mechanisms or variables relate to each other. Although overall removal rates will not be representative in small scale systems, these systems offer advantages not available in large scale-systems.

Laboratory scale treatment wetland systems are often referred to as mesocosms or microcosms. Examples of small scale mesocosm system studies include those of (Kapplemeyer et al., 2001; Prado, 2004; McHenry and Werker, 2005; Werker et al., 2007; Weber et al., 2008). Configurations, flow rates, feed entry points, and recycle rates can all be varied to achieve different internal system conditions. As the internal conditions of in-use treatment systems are varied, so can the experimental conditions investigated with laboratory scale mesocosm systems. Figure 4 shows an example of one such mesocosm system used for laboratory scale research.

The constructed wetland mesocosm approach was developed as a quantitative method to gain insight into wetland performance, and ultimately use this insight to assist in the design and treatment optimization of constructed wetlands by applying principles of tracers, reactor theory, modeling, and enzymology, while undertaking experiments to critically assess factors that could influence performance and reliability of treatment wetlands (Werker et al., 2004). Through system replication, environmental control and shorter response-times, wetland mesocosms allow for the implementation of factorial designed experiments and subsequent quantitative and comparative statistical analysis. The mesocosm approach has been described as a powerful way to test research hypotheses using quantitative experimentation (Perrin et al., 1992). Several mesocosm studies have recently been used in undertaking a quantitative approach to the study of constructed wetland systems (Kappelmeyer et al., 2001; Prado, 2004; McHenry and Werker, 2005; Werker et al., 2007; Weber et al., 2008).


d

c

f

a

e

b

g

A B

d

c

f

a

e

b

g

d

c

f

a

e

b

g

A B

Figure 4. Mesocosm schematic. In (A) water (a) is fed into the mesocosm (b) and allowed to percolate through the pea gravel bed to be collected at the bottom (c). An atmosphere exposed port serves as an injection (d) and sampling (e) point. Drainage ports are located near the top to prevent overfilling (f), and near the bottom (g) for mesocosm drainage (Weber, 2006).

4.2. Pathogen Tracers

To help decipher between biotic and abiotic removal mechanisms the use of both active

and inactive pathogen tracers proves useful. As previously discussed wastewater often contains are large mixture of different pathogens therefore indicator organisms are often used to study the general trend of pathogen removal in wetland systems. Bacterial pathogens are the most commonly occurring, and therefore most commonly studied pathogens in the laboratory. E. coli is the most commonly used pathogen tracer as it is easy to work with, relatively non-infectious, rapidly replicates, can be cultured on media, and is well documented and understood in the literature. Fluorescent micro-spheres are also often used in pathogen removal studies. The use of both E. coli and similarly sized micro-spheres allows for a comparison of removal rates for both a biotic pathogen and a surrogate abiotic molecule (microsphere) (Werker et al., 2007).

When studying the removal of viruses the bacteriophage MS-2 is most commonly employed. As previously discussed, positive aspects of using MS-2 include ease of enumeration, MS-2 is nearly the same size as enteroviruses, and MS-2 is more resistant to


UV, heat and disinfection than most enteric viruses making MS-2 a fairly representative, conservative viral pathogen indicator.

4.3. Study of Biotic Pathogen Removal Factors It is generally accepted that constructed wetlands contain a biological regime associated

with the wetland substrate (Wynn and Liehr, 2001). The role of the biological regime, and the related mechanisms associated with contaminant treatment, have been largely overlooked in favor of using hydrodynamic and simple first order removal rate models to describe water treatment in constructed wetlands. The role and influence which the biological regime has on specific aspects within constructed wetlands has been given little to no attention.

The abiotic pathogen removal mechanisms in wetland systems can be said to be similar to those in sand or gravel filters and are well documented and quantified (Stevik et al., 2004), but the biotic removal mechanisms in wetland systems are what make them unique and more effective. Both the planted regions and the internally developed bacterial community within wetlands can be considered as biotic or “living” components of the wetland system. These biotic components are more difficult to study and quantify leading to a gap in our knowledge surrounding the associated removal mechanisms and it is likely. Of further note is the likely interdependence of abiotic and biotic removal mechanisms. For example, physical pathogen removal filtration rates will be affected by biofilm type and size. Biofilm type and size will be based on the overall activity of the microbial population with the wetland system and the different bacterial species populating the media.

Different communities will theoretically create different microenvironments including different biofilm environments. As previously discussed the existence of, and general integrity and make-up of these biofilms, can have an effect on pathogen removal in wetlands (Broadbent et al., 1971; Richardson and Rusch, 2005; Vacca et al., 2005; Larsen and Greenway, 2004). The different bacterial communities in wetlands play a vital role in water treatment performance and ecosystem health (Parkinson and Coleman, 1991; Aelion and Bradely, 1991). Although the bacterial community in the wetland has been recognized as having a large influence on water treatment performance, little attention has been give to understanding the microbial ecology in wetlands.

The interaction between plants and the bacterial communities found in the wetland substrate is another major factor affecting bacterial community dynamics in wetland systems. Aquatic plants, such as Phragmites australis (the common reed), have the ability to transfer oxygen from their aerial tissues and release it into their rhizosphere (Karathanasis and Johnson 2003; Batty et al., 2000). Plant root systems also provide mechanical support and perform many roles including the synthesis, accumulation, and secretion of compounds (Flores et al., 1999). The chemicals secreted into the surrounding rhizosphere by roots are referred to as root exudates. Plants have been shown to exude 5-21% of all photosynthetically fixed carbon into the surrounding rhizosphere as root exudates (Walker et al., 2003; Marschner, 1995). Through this exudation, roots can often influence the microbial community structure within the surrounding rhizosphere (Walker et al., 2003; Nardi et al., 2000). Recent work has gone so far as to take advantage of this bacteria-rhizosphere interaction to help promote plant growth using engineered bacteria (Reed et al., 2005).


Biotic factors can be difficult to quantify in wetland systems. Two factors which can be studied with relatively little difficulty are microbial activity and microbial community diversity. The activities of different bacterial species and the overall community structure affect treatment performance of constructed wetland systems. By gaining better insight into bacterial community activity and diversity, improvements to existing wetland models should be possible.

4.3.1. Microbial Activity

One of the most commonly utilized microbial activity measures is the 5 day biochemical oxygen demand BOD5. Other more easily applied methods include carbon utilization measures (Weber et al., 2008; Tietz et al., 2008), and microbial-related enzymatic activity measures based on the conversion of FDA to FL (Schnürer and Rosswall, 1982; McHenry and Werker 2005; Weber et al., 2008).

4.3.2. Microbial Community Assessment

Another important factor in wetland systems which has recently received attention is the bacterial community structure (Vacca et al., 2005; Hallberg and Johnson, 2005; Weber et al., 2008). Both the genetic diversity and functional adaptation of bacterial communities in wetland systems allow for improved long term treatment performance (Kadlec and Knight, 1996). Genetic diversity can give an idea of the number and distribution of species within a community while the study of the functional response or diversity, takes a more holistic approach, and yields an idea of the overall community response or function without cataloging specific species. A brief look at some of the more commonly used bacterial community assessment methods follows.

Non-Molecular Methods

Light Microscopy

Traditional non-molecular methods include microscopy and culture based identification. Microscopy has advantages due to its ease of use and quick assessments, making microscopy a convenient and dependable method when monitoring communities of fixed or similar species distribution. Light microscopy allows for qualitative-heavy identification based on morphology however it is not possible to distinguish between living and dead organisms (Madigan et al., 2002). Other drawbacks include the need for specialized expertise and the fact that different organisms share similar morphology (Ferris et al., 1996; Duineveld et al., 2001).

Traditional Plating

Culture-based identification can be used to identify some organisms. By using previously developed expertise, species identification can be accomplished through sequential plating with different nutrient sources (Cullimore, 2000). These methods require a large amount of time, resources, and expertise. As well, many organisms may not be cultureable under plating conditions (Amman et al., 1995); it has been suggested that direct or plate count techniques account for a meager 0.1-20% of the original population (Muyzer et al., 1993).


BIOLOG™ Plates Community level physiological profiling (CLPP) is an approach used to characterize

microbial community function based on sole carbon source utilization patterns (CSUPs). CLPP can be used as an indicator of the metabolic characteristics and overall stability of a specific microbial community over time. Recent work in the area of microbial soil ecology has utilized BIOLOG™ plates as a tool for CLPP. BIOLOG™ plates consist of 96 wells, each well containing a different carbon source and a redox dye indicator, tetrazolium violet, which changes colour in response to carbon utilization. Garland and Mills (1991) were the first to use BIOLOG™ plates for characterizing heterotrophic soil bacteria communities through principle component analysis (PCA). A number of subsequent studies are discussed in Konopka et al., (1997). Most recently Weber et al., (2007, 2008) successfully used BIOLOG™ ECO plates to profile the interstitial bacteria in constructed wetland mesocosms. In contrast to the original BIOLOG™ plates with 96 different carbon sources, ECO plates by the same manufacturer are based on 31 different carbon sources with built-in triplicates allowing for better replication. CLPP has advantages over both classic cell culturing techniques and molecular level RNA/DNA amplification techniques as these other techniques are time consuming and require specialized expertise (Garland, 1997).

Limitations pertaining to the CLPP approach using BIOLOGTM ECO plates have been discussed in the literature (Garland, 1997; Knopka et al., 1998; Preston-Mafham et al., 2002). Limitations and pitfalls pertaining to data analysis have also been recently described (Weber et al., 2007). Some of the most pertinent limitations include the bias in the technique toward rapidly growing bacteria, the need to ensure similar sample sizes, the need to reduce time between sampling and inoculation of the BIOLOGTM microplates, and difficulties with meaningful data analysis and interpretation.

Molecular Methods

With the advent of the polymerase chain reaction (PCR) and a growing library of genetic information on bacterial species, molecular identification methods have become increasingly popular in the field of bacterial community analysis. Although many of the molecular techniques developed for community analysis could be used to study the bacterial community in wetlands, only a small number of molecular-based community studies have been performed for wetland systems (Vacca et al., 2005; Hallberg and Johnson, 2005).

Denaturing Gradient Gel Electrophoresis (DGGE)

DGGE separates PCR amplified bacterial community rDNA gene segments by electrophoresis on a denaturing gradient gel. PCR is a method of DNA or RNA amplification using a heat stable polymerase, an excess of nucleotide bases (dNTPs), and an excess of 2 (20-base pair or smaller) primers (Tozeren and Byers, 2005). These 2 primers match highly conserved regions on the DNA, most commonly in the gene encoding bacterial 16s ribosomal RNA (rRNA). The primers are selected with some a priori knowledge of what bacterial species may be expected in the samples. PCR has been shown to have a number of limitations including contaminants present in the samples, suboptimal reaction conditions, lack of primer specificity, and differential annealing (Suzuki and Giovannoni, 1996).

To perform a DGGE analysis, PCR amplified rDNA segments are run on a gel containing urea and formamide. The urea and formamide denatures the rDNA as it runs on the gel, completely restricting its movement in the gel at a specific location based on the original


sequence. Therefore different proportions of nucleotides in the original species sequence will allow rDNA to move a characteristic distance unique to the original sequence (species). Each band present on the gel is then said to be representative of a specific bacterial species in the original sample (Nadarajah, 2007).

Terminal-Restriction Fragment Length Polymorphism (TRFLP)

Either or both of the primers used in the PCR amplification is labeled with a fluorescent marker. These primers are again most often targeted at the 16s rRNA encoding region of the DNA, which has a highly conserved length (500-200 bp). A restriction enzyme is then added which cleaves the rDNA at a specific site dependant on the species sequence. The lengths of the fluorescently labeled fragment can then be determined using capillary electrophoresis (Dunbar et al., 2001). Two electropherograms are then generated with two sets of colour peaks, usually blue for TRFs created from the 5’ end and green created from the 3’ end (Osborn et al., 2000; Nadarajah, 2007). The resulting fingerprint gives a measure of phylogenetic diversity (Liu et al., 1997). Biases for TRFLP originate from the same limitations discussed for PCR.

Fluorescent In-Situ Hybridization (FISH)

A library of fluorescently labeled probes is designed based on the expected 16s ribosomal RNA sequences of the bacterial species in the sample (MacDonald and Brozel, 2000). These single stranded probes bind with the denatured DNA of the respective species within the samples. Fluorescence is then detected using fluorescence microscopy. FISH allows the visualization of the spatial distribution of organisms in a sample (Karp, 1999). FISH limitations include problems with cell permeability, target site accessibility, and a possible lack of a priori knowledge regarding species within a sample.

4.4. Experimental Design When trying to ascertain the quantitative contributions of different mechanisms, designs,

or operating variables on system performance a well designed experiment is of crucial importance. Many experimental regimes are based on single factor experiments where one variable is varied while all other controllable variables are kept constant. This method allows for the quantification of the effects a single independent variable has on a dependant variable such as a system removal rate. However, in wetland systems many variables which may at first seem to be independent may in fact be interdependent or act synergistically. Therefore is important to allow for the quantification of both single factor and synergistic variable effects on system performance.

One such experimental design method which allows for the quantification of both single factor and the interdependent effects of variables on the final dependant variable (system performance) is the factorial design. An xn factorial design is performed at x number of levels for n number of variables. For example, an experimental design could be performed at 2 levels looking at the effect of plant presence and bed media type on overall pathogen removal performance in mesocosm systems. The 2 levels for the plant presence could be -1 for no plants present, and +1 for plants present. The two levels for the bed media could be -1 for sand, and +1 for gravel. This experimental regime would require 4 different mesocosm


setups, and would require duplication for statistical analysis purposes giving a total of 8 experiments. Subsequent analysis is analogous to performing a multiple linear regression. Statistical analysis of the results for a 22 factorial design experimental regime yields an equation of the form:

Y = α0 + α1x1 + α2x2 + α3x1x2 where: Y = dependent variable (removal performance)

α = coefficient x = independent variable

This equation once attained is not for use in predicting removal performance in the

studied systems, rather it is simply used to quantitatively compare the magnitudinal effect of each studied independent variable (xi) on the dependent variable (Y). The size of the |α| signifies the magnitude of that variable’s effect on the dependent variable. The α terms can be positive or negative depending on the effect each independent variable has on the dependent variable. One of the large benefits of using a factorial design is the final “cross” term that is attained. The α3x1x2 in the final general linear equation signifies any synergistic or “cross” effects the two independent variables may have on the dependent variable (Y). Quantifying this cross term can significantly increase the fit of data and should be considered in wetland system experiments due to the synergistic nature of the variables in a wetland system. Although quantifying the cross term does not give information as to the nature of the synergistic effect it does allow research to be directed in the proper direction. For example, if a cross term is found to be quite large in comparison to the single factor terms perhaps more consideration in design and further research into the nature of this synergistic effect is needed.

Many statistical packages offer extensive analysis options for factorial experiments. Model fit evaluation and error propagation analysis are both needed to evaluate the validity of the general linear model. Data transformations may be needed for proper statistical analysis. For further details regarding factorial experiments and experimental design see Montgomery (2001). Other statistical methods such as multivariate analysis have been shown to be useful in evaluating microbial ecology in CW systems (Weber et al., 2007) and could be similarly utilized in looking at other large CW data sets.

Through the use of small scale systems, active and inactive tracers, proper experimental design and statistical analysis, investigations regarding the fundamental mechanisms and variables affecting overall pathogen removal performance can be performed. Biotic variables such as microbial activity and microbial community structure should also be evaluated as these variables may have a large influence on removal performance. Including these variables in future studies could perhaps help to better characterize and explain the large performance variations reported for similarly designed wetland systems.

CONCLUSIONS AND RECOMMENDATIONS Constructed wetlands have proven to be an effective treatment alternative for the removal

and inactivation of pathogens in wastewaters. Constructed wetlands designed for pathogen


treatment are most often preceded by filtration or sedimentation. Pathogen removal efficiencies upwards of 99.99% have been reported by multiple authors employing many different constructed wetland designs.

Vymazal (2005) showed hybrid CW systems to perform better than HSSF systems which in turn performed better than FSW systems. Constructed wetland design tends to be based largely on rule of thumb sizing, as the specific mechanisms and fundamental variables involved in pathogen removal are only vaguely understood. Suggested mechanisms of pathogen treatment in CWs include but are not restricted to sedimentation, natural die-off, temperature, oxidation, predation, unfavourable water chemistry, biofilm interaction, mechanical filtration, exposure to biocides and UV radiation. Pathogen removal has been shown to correlate well with hydraulic retention time. First order decay kinetics are, almost exclusively, used to model and predict pathogen removal in constructed wetlands.

Of the research reviewed for this chapter, regarding specific mechanisms of pathogen treatment in CWs, only a small number studies arrived at quantitative conclusions. Studies such as Zdragas et al. (2002) utilizing multiple linear regression to evaluate synergistic effects between mechanisms should be used as an example of strong experimental design and analysis.

Of all the pathogen treatment mechanisms proposed only the effects of temperature and plants are under debate. Multiple studies showed colder temperatures to improve pathogen removal efficiency; however a similar number of studies also showed higher temperatures to improve pathogen removal efficiency. Temperature effects depend on the type of pathogen under consideration, the bed media and what mechanisms are dominant within the specific CW system. Plants were shown to improve, show no improvement, and in one case decrease, removal efficiencies within CW systems. The effect of plants on specific pathogen removal rates depended on the pathogen under consideration, the bed media, and what mechanisms were dominant within the specific CW system. This again highlights the synergistic nature and interdependence of pathogen removal mechanisms.

CW systems are complex ecosystems where any number of chemical, biological and physical transformations can be taking place in seemingly random or ordered fashion. There is a severe lack of attention given to the comparative quantification of the specific mechanisms of pathogen treatment in constructed wetlands. Small scale controllable constructed wetland systems should be more widely used to conduct well-designed controlled experiments, where mechanisms of pathogen removal are comparatively quantified using statistical analysis techniques such as multiple linear regression and multivariate analysis. Only once the quantitative role and interdependence of pathogen treatment mechanisms are understood, can insightful engineering design and treatment optimization be performed.

REFERENCES

Aelion C.M., Bradley P.M. (1991) Aerobic biodegradation potential of subsurface microorganisms from a jet fuel-contaminated aquifer. Applied and Environmental Microbiology. 57, 57-63.


Amman, R.I, Ludwig, W., Schleifer, K.H. (1995) Phylogenetic identification and in-situ detection of individual microbial cells without cultivation. Microbial Reviews. 59:1, 143-169.

Anderson, B.C., Watt, W.E., Marsalek, J., Crowder, A.A. (1996) Integrated urban stormwater quality management: field investigations at a best management facility. Canadian Water Resources Journal. 21, 165-182.

Batty, L. Baker, A. Wheeler, B., Curtis, C. (2000) The effect of pH and plaque on the uptake of Cu and Mn in Phragmites australis. Annals of Botany. 26, 647-653.

Bitton, G. 1980. Adsorption of viruses to surfaces: Technological and ecological implications. pages 331–374 in G. Bitton and K.C. Marshall (eds.) Adsorption of microorganisms to surfaces. John Wiley and Sons, New York.

Borisko, J.P., Slawson, R. M., and Warner, B.G. (2000) An examination of Escherichia coli removal by an alternative wetland-based wastewater treatment system: Preliminary findings, Canadian Society for Civil Engineering 2000 Conference Proceedings, London, Ontario, 523-529.

Brix, H. (1997) Do macrophytes play a role in constructed treatment wetlands?. Water Science and Technology. 35, 11-17.

Brix, H., Arias, C.A., Cabello, A., Johansen, N.H. (2003) Removal of indicator bacteria from municipal wastewater in an experimental two-stage vertical flow constructed wetland system. Water Science and Technology. 48:5, 35-41.

Broadbent, P., Bakker, K.F., Waterworth, Y. (1971) Bacteria and actinomycetes antagonistic to fungal root pathogens in Australian soils. Aust. J. Biol. Sci. 24, 925–944.

Brock, Th., Madigan, M.T., Martinko, J.M., Parker, J. (1994) Biology of Microorganisms. Prentice-Hall International, Inc., Englewood Cliffs, NJ.

Burgoon, P. S., Kadlec, R. H., Henderson, M. (1999) Treatment of potato processing wastewater with engineered natural systems. Water Science and Technology. 40, 211-215.

Clausen, E.M., Green, B.L., Litsky, W. (1977) Fecal Streptococci: Indicators of Pollution. pages 247-264 in A.W. Hoadley and B.J. Dutka (eds.) Bacterial Indicators/Health Hazards Associated with Water. American Society for Testing and Materials. Philadelphia, PA.

Cronk, J. K. (1996) Constructed wetlands to treat wastewater from dairy and swine operations: A review. Agriculture. 58, 97-114.

Cullimore, R.D. (2000) Practical atlas for bacterial identification, CRC Press, Boca Raton, FL, USA.

Curds, C.R. (1992) Protozoa and the water industry. Cambridge University Press, Cambridge, UK.

Decamp, O., Warren, A. (1998) Bacterivory in ciliates isolated from constructed wetlands (reed beds) used for wastewater treatment. Water Research. 32:7, 1989-1996.

Desena, M. (1999) Constructed wetlands provide cost-effective treatment for Florida town. Water, Environment and Technology. 11, 38-39.

Diemont, A.W. (2006) Mosquito larvae density and pollutant removal in tropical wetland treatment systems in Honduras. Environment International. 32, 332 – 341.

Dufour, A.P. (1977) Escherichia coli. the fecal coliform. pages 48-58 in A.W. Hoadley and B.J. Dutka (Eds.) Bacterial Indicators/Health Hazards Associated with Water. American Society for Testing and Materials. Philadelphia, PA.


Duineveld, B.M., Kowalchuk, G.A., Keijzer, A., van Elsas, J., van Veen, J. (2001) Analysis of bacterial communities in the rhizosphere of Chrysanthemum via Denaturing Gradient Gel Electrophoresis of PCR-amplified 16S rRNA as well as DNA fragments coding for 16S rRNA. Applied and Environmental Microbiology. 67:1, 190-197.

Dunbar, J., Ticknor, L.O., Kuske, C.R. (2001) Phylogenetic specificity and reproducibility and new method for analysis of terminal restriction fragment profiles of 16s rRNA genes from bacterial communities. Applied and Environmental Microbiology. 67:1, 190-197.

Dusel, C.J., Pawlewski, C.W. (1997) Constructed wetlands offer flexibility. Land and Water. 41, 27-28.

Fayer, R., Morgan, U. M., Upton, S. J. (2000) Epidemiology of Cryptosporidium: Transmission, detection, and identification. International Journal for Parasitology. 30, 1305–1322.

Ferris, M.J., Muyzer, G., Ward, D.M. (1996) Denaturing gradient gel electrophoresis profiles of 16S rRNA-defined populations inhabiting hot spring microbial mat community. Applied and Environmental Microbiology. 61:2, 340-346.

Flores, H.E., Vivanco, J.M., Loyola-Vargas, V.M. (1999) “Radicle” biochemistry: the biology of root-specific metabolism. Trends Plant Sci. 4, 220–226.

Frostman, T.M. (1996) Constructed wetlands for water quality improvement. Land and Water. 40, 14-17.

Garcia, J., Vivar, J., Aromir, M., Mujeriego,R. (2003) Role of hydraulic retention time and granular medium in microbial removal in tertiary treatment reed beds. Water Research. 37:11, 2645-2653.

Garland, J.L. (1997) Analysis and interpretation of community-level physiological profiles in microbial ecology. FEMS Microbiol. Ecol. 24, 289-300.

Garland, J.L., Mills, A.L. (1991) Classification and characterization of heterotrophic microbial communities on the basis of patterns of community-level sole-carbon-source utilization. Applied and Environmental Microbiology. 57, 2351-2359.

Gerba, C.P., Nokes, R., Karpisack, M. (2000) Reduction of enteric microorganisms through small scale subsurface flow constructed wetlands, Technology Expo and International Symposium on Small Drinking Water and Wastewater Systems, Phoenix, AZ (USA).

Gerba, C.P., Thurston, J.A., Falabi, J.A., Watt, P.M., Karpiscak, M.M. (1999) Optimization of artificial wetland design for removal of indicator microorganisms and pathogenic protozoa. Water Science and Technology. 40, 363-368.

Gersberg, R.M., Brenner, R., Lyon, S.R., Elkins, B.V. (1987) Survival of bacteria and viruses in municipal wastewaters applied to artificial wetlands. in K.R. Reddy and W.H. Smith (Eds.) , Aquatic plants for waste treatment and resource recovery. Orlando: Magnolia Publishing.

Gopal, B., Goel, U. (1993) Competition and allelopathy in aquatic plant communities. Botanical Review. 59, 155–210.

Green, M.B., Griffin, P., Seabridge, J.K., Dhobie, D. (1997) Removal of bacteria in subsurface flow wetlands. Water Science and Technology. 35:5, 109-116.

Hallberg, K.B., Johnson, D.B. (2005) Microbiology of a wetland ecosystem constructed to remediate mine drainage from a heavy metal mine. Science of the Total Environment. 338, 53-66.


Hench, K.R., Bissonnette, G.K., Sexstone, A.J., Coleman, J.G., Garbutt, K., Skousen, J.G. (2003) Fate of physical, chemical, and microbial contaminants in domestic wastewater following treatment by small constructed wetlands. Water Research. 37:4, 921-927.

Hill, V.R., Sobsey, M.D. (2001) Removal of Salmonella and microbial indicators in constructed wetlands treating swine wastewater. Water Science and Technology. 44:11-12, 215-222.

Jagger, J. (1983) Physiological effects of near ultraviolet radiation on bacteria. Photochemical and Photobiological Reviews. 7:1–75.

Jillson, S.J., Dahab, M.F., Woldt, W.E., Surampalli, R.Y. (2001) Pathogen and pathogen indicator removal characteristics in treatment wetlands systems. Practice Periodical of Hazardous, Toxic, and Radioactive Waste Management. 5:3, 153-160.

Kadlec, R.H. (1997) Deterministic and stochastic aspects of constructed wetland performance and design. Water Science and Technology. 35, 149-156.

Kadlec, R.H. (1999) Chemical, physical and biological cycles in treatment wetlands. Water Science and Technology. 40, 37-44.

Kadlec, R.H., Knight, R.L. (1996) Treatment Wetlands, Lewis Publishers, Boca Raton, FL, USA.

Kapplemeyer, U., Wiessner, A., Kuschk, P., Kastner, M. (2001) Planted fixed bed reactor (PFR) – Eine universelle testenheit fur bewachsene bodenfilter. Chemie Ingenieur Technik. 73, 1472-1477.

Karathanasis, A., Johnson, C. (2003) Metal removal by three aquatic plants in an acid mine drainage wetland. Mine Water and the Environment. 22, 22-30.

Karathanasis, A.D., Potter, C.L., Coyne, M.S. (2003) Vegetation effects on fecal bacteria, BOD, and suspended solid removal in constructed wetlands treating domestic wastewater. Ecological Engineering. 20:2, 157-169.

Karim, M.R., Manshadi, F.D., Karpiscak, M.M., Gerba, C.P. (2004) The persistence and removal of enteric pathogens in constructed wetlands. Water Research. 38:7, 1831-1837.

Karp, G. (1999) Cell and Molecular Biology, 2nd edition. Wiley, NY, USA. Keffala, C., Ghrabi, A. (2005) Nitrogen and bacterial removal in constructed wetlands

treating domestic waste water. Desalination. 185:1-3, 383-389. Kern, J., Idler, C. (1999) Treatment of domestic and agricultural wastewater by reed bed

systems. Ecological Engineering. 12, 13-25. Knight, R., Kadlec, R., Ohlendorf, H. (1999) The use of treatment wetlands for petroleum

industry effluents. Env. Sci. Technol. 33, 973-980. Knopka, A., Oliver, L., Turco, R.F. (1998) The Use of Carbon Substrate Utilization Patterns

in Environmental and Ecological Microbiology. Microbial Ecol. 35, 103-115. Kraus, M.P. (1977) Bacterial Indicators and Potential Health Hazards of Aquatic Viruses.

pages 196-217 in A.W. Hoadley and B.J. Dutka (eds.) Bacterial Indicators/Health Hazards Associated with Water. American Society for Testing and Materials. Philadelphia, PA.

Larsen, E., Greenway, M. (2004) Quantification of biofilms in a sub-surface flow wetland and their role in nutrient removal. Water Science and Technology. 49:11-12, 115-122.

LeChevalier, M.W., Au, K. (2004) Water Treatment and Pathogen Control. IWA Publishing. London, United Kingdom.

Leclerc, H., Mossel D.A., Trinel P.A., Gavini, F. (1977) Microbiological Monitoring – A New Test for Fecal Contamination. pages 23-36 in A.W. Hoadley and B.J. Dutka (eds.),


Bacterial Indicators/Health Hazards Associated with Water. American Society for Testing and Materials. Philadelphia, PA.

Liu, W.T., Marsh, T.L., Cheng, H., Forney, L.J. (1997) Characterization of microbial diversity by determining terminal restriction fragment length polymorphism of genes encoding 16s rRNA. Applied and Environmental Microbiology. 63, 4516-4522.

MacDonald, R., Brozel, V.S. (2000) Community analysis of bacterial biofilms in simulated recirculating cooling-water system by fluorescent in situ hybridization with rRNA targeted oligonucleotide probes. Water Research. 34:9, 2439-2446.

Madigan, M.T., Martinko, J.M., Parker, J. (2002) Brock Biology of Microorganisms, 10th edition. Prentice-Hall, NJ, USA.

Maki, J. (1993) The air–water interface as an extreme environment. in T. Ford (ed.) Aquatic microbiology. Blackwell Scientific, Cambridge, MA.

Marschner, H. (1995) Mineral Nutrition of Higher Plants, (2nd edition). Academic Press, London, England.

McHenry J, Werker A. (2005) In-situ monitoring of microbial biomass in wetland mesocosms. Water Sci Technol. 51:9, 233–41.

McNevin, D., Harrison, M., King, A., David, K., Mitchell, C. (2000) Towards an integrated performance model for subsurface flow constructed wetlands. Journal of Environmental Science and Health- Part A - Toxic Hazardous. 35, 1415-1431.

Mitsch, W.J., Wise, K.M. (1998). Water quality, fate of metals, and predictive model validation of a constructed wetland treating acid mine drainage, Water Research, 32: 1888-1900.

Montgomery, D.C., (2001) Design and Analysis of Experiments 5th Edition. John Wiley and Sons, Inc., New York, New York.

Morsy, E.A., Al-Herrawy, A.Z., Ali, M.A. (2007) Assessment of cryptosporidium removal from domestic wastewater via constructed wetland systems. Water Air Soil Pollut. 179:207–215.

Muyzer, G., De Waal, E.C., Utterlinden, A.G. (1993) Profiling complex microbial populations by denaturing gradient gel electrophoresis analysis or polymerase chain reaction-amplified genese coding for 16s rRNA. Applied and Environmental Microbiology. 59:3, 695-700.

Nadarajah, N. (2007) Activated sludge bacterial community composition responses to temperature fluctuations. PhD Thesis, Graduate Department of Chemical Engineering and Applied Chemistry, University of Toronto, ON, Canada.

Nardi S., Concheri G., Pizzeghello D., Sturaro A., Rella R., Parvoli G. (2000) Soil organic matter mobilization by root exudates. Chemosphere. 5, 653-658.

Nasim, A., James, A.P. (1978) Life under conditions of high irradiation. in D. J. Kushner (ed.) Microbial life in extreme environments. Academic Press, London.

Neori, A., Reddy, K.R., Cızkova-Koncalova, H., Agami, M. (2000) Bioactive chemicals and biological-biochemical activities and their functions in rhizospheres of wetland plants. Botanical Review. 66, 351–378.

Neralla, S., Weaver, R.W., Lesikar, B.J., Persyn, R.A. (2000). Improvement of domestic wastewater quality by subsurface flow constructed wetlands, Bioresource Technology. 75, 19-25.


Osborn, A.M., Moore, E.R.B., Timmins, K.N. (2000) An evaluation of terminal-restriction fragment length polymorphism (T-RFLP) analysis for the study of microbial community structure and dynamics. Environmental Microbiology. 2:1, 39-50.

Parkinson D., Coleman D.C. (1991) Microbial communities, activity, and biomass. Agriculture, Ecosystems and Environment. 34, 3-33.

Perfler, R., Laber, J., Langergraber, G., and Haberl, R. (1999) Constructed wetlands for rehabilitation and reuse of surface waters in tropical and subtropical areas - First results from small-scale plots using vertical flow beds. Water Science and Technology. 40, 155-162.

Perkins, J., Hunter, C. (2000) Removal of enteric bacteria in a surface flow constructed wetland in Yorkshire, England. Water Research. 34:6, 1941-1947.

Perrin, C.J., Wilkes, B., Richardson, J.S. (1992) Testing stream ecosystem responses to additions of treated acid mine drainage: a mesocosm approach. Environmental Toxicology and Chemistry. 11, 1513-1525.

Peterson, H. G. (1998). Use of constructed wetlands to process agricultural wastewater. Canadian Journal of Plant Science/Revue Canadienne de Phytotechnie. 78, 199-210.

Philippi, L. S., da Costa, R. H. R., Sezerino, P. H. (1999) Domestic effluent treatment through integrated system of septic tank and root zone. Water Science and Technology. 40, 125-131.

Prado, A.R. (2004) Application of wetlands for the treatment of chromium-containing wastewater. PhD Thesis, Department of Civil Engineering, University of Waterloo, ON, Canada.

Preston-Mafham, J., Boddy, L., Randerson, P.F. (2002) Analysis of microbial community functional diversity using sole-carbon-source utilisation profiles - a critique. FEMS Microbiology Ecology. 42, 1-14.

Pundsack, J., Axler, R., Hicks, R., Henneck, J., Nordman, D., McCarthy, B. (2001) Seasonal pathogen removal by alternative on-site wastewater treatment systems. Water Environment Research. 73:2, 204-212.

Quinonez-Diaz, M.D.J., Karpiscak, M.M., Ellman, E.D., Gerba, C.P. (2001) Removal of pathogenic and indicator microorganisms by a constructed wetland receiving untreated domestic wastewater. Journal of Environmental Science and Health - Part A Toxic/Hazardous Substances and Environmental Engineering. 36:7, 1311-1320.

Rash, J. K. Liehr, S. K. (1999) Flow pattern analysis of constructed wetlands treating landfill leachate. Water Science and Technology. 40, 309-315.

Reed, M.L.E., Warner, B.G., Glick, B.R. (2005) Plant growth-promoting bacteria facilitate the growth of the common reed Phragmites australis in the presence of copper or polycyclic aromatic hydrocarbons. Current Microbiology. 51, 425-429.

Revitt, D. M., Shutes, R. B. E., Llewellyn, N. R., Worrall, P. (1997) Experimental reedbed systems for the treatment of airport runoff. Water Science and Technology. 36, 385-390.

Rice, T.D., Williams, H.N., Turng, B.F. (1998) Susceptibility of bacteria in estuarine environments to autochthoneous bdellovibrios. Microb. Ecol. 35, 256–264.

Richardson, S.D., Rusch, K.A. (2005) Fecal coliform removal within a marshland upwelling system consisting of scatlake soils. Journal of Environmental Engineering. 131:1, 60-70.

Rivera, F., Warren, A., Ramirez, E., Decamp, O., Bonilla, P., Gallegos, E., Calderon, A., Sanchez, J.T. (1995) Removal of pathogens from wastewaters by the root zone method (RZM). Water Science and Technology. 32:3, 211-218.


Rochfort, Q. J., Anderson, B. C., Crowder, A. A., Marsalek, J., Watt, W. E. (1997). Field-scale studies of subsurface flow constructed wetlands for stormwater quality enhancement. Water Quality Research Journal of Canada. 32, 101-107.

Sauter, G. Leonard, K. (1997) Wetland design methods for residential wastewater treatment. Journal of the American Water Resources Association. 33, 155-162.

Schnürer, J., Rosswall, T. (1982) Fluorescein diacetate hydrolysis as a measure of total microbial activity in soil and litter. Applied and Environmental Microbiology. 43, 1256-1261.

Schoenerklee, M., Koch, F., Perfler, R., Haberl, R., Laber, J. (1997) Tertiary treatment in a vertical flow reed bed system - A full scale pilot plant for 200-600 people. Water Science and Technology. 35, 223-230.

Scholes, L., Shutes, R. B. E., Revitt, D. M., Forshaw, M., Purchase, D. (1998) The treatment of metals in urban runoff by constructed wetlands. Science of the Total Environment. 214, 211-219.

Schulze-Makuch, D., Bowman, R.S., Pillai, S.D., Guan, H. (2003) Field evaluation of the effectiveness of surfactant modified zeolite and iron-oxide-coated sand for removing viruses and bacteria from ground water. Ground Water Monitoring and Remediation. 23:4, 68-74.

Seidel, K. Macrophytes and water purification. (1976) in Biological control of water pollution, Tourbier, J.; Pierson, R.W., Jr., Eds., University of Pennsylvania Press: Philadelphia, PA, 109–121.

Shutes, R. B. E., Revitt, D. M., Lagerberg, I. M., Barraud, V. C. E. (1999) The design of vegetative constructed wetlands for the treatment of highway runoff. Science of the Total Environment. 235, 189-197.

Slifko, T. R., Smith, H. V., Rose, J. B. (2000) Emerging parasite zoonoses associated with water and food. International Journal for Parasitology. 30, 1379–1393.

Smith, I. D., Bis, G. N., Lemon, E. R., and Rozema, L. R. (1997) A thermal analysis of a sub-surface, vertical flow constructed wetland. Water Science and Technology. 35, 55-62.

Soto, F., Garcia, M., de Luis, E., Becares, E. (1999) Role of Scirpus lacustris in bacterial and nutrient removal from wastewater. Water Science and Technology. 40:3, 241-247.

Stevik, T.K., Aa, K., Ausland, G., Hanssen, J.F. (2004) Retention and removal of pathogenic bacteria in wastewater percolatingthroug h porous media: a review. Water Research. 38, 1355-1367.

Stott, R., Jenkins, T., Bahgat, M., Shalaby, I. (1999) Capacity of constructed wetlands to remove parasite eggs from wastewaters in Egypt. Water Science and Technology. 40, 117-123.

Stott, R., Jenkins, T., Shabana, M., May, E. (1997) Survey of the microbial quality of wastewaters in Ismailia, Egypt and the implications for wastewater reuse. Water Science and Technology. 35:11-12, 211-217.

Stott, R., Tanner, C.C. (2005) Influence of biofilm on removal of surrogate faecal microbes in a constructed wetland and maturation pond. Water Science and Technology. 51:9, 315-322.

Stottmeister, U., Wießner, A., Kuschk, P., Kappelmeyer, U., Kastner, M., Bederski, O., Muller, R.A., Moormann, H. (2003) Effects of plants and microorganisms in constructed wetlands for wastewater treatment. Biotechnology Advances. 22, 93- 117.


Suzuki, M.T., Giovannoni, S.J. (1996) Bias caused by template annealing in the amplification of mixtures of 16s rRNA genes by PCR. Applied and Environmental Microbiology. 62:2, 625-630.

Tanner, C. C., Sukias, J. P., Upsdell, M. P. (1998) Relationships between loading rates and pollutant removal during maturation of gravel-bed constructed wetlands. Journal of Environmental Quality. 27, 448-458.

Tanner, C.C., Clayton, J.S., Upsdell, M.P. (1995) Effect of loading rate and planting on treatment of dairy farm wastewaters in constructed wetlands - I. Removal of oxygen demand, suspended solids and faecal coliforms. Water Research. 29:1, 17-26.

Tietz, A., Langergraber, G., Watzinger, A., Haberl, R., Kirschner, A.K.T. (2008) Bacterial carbon utilization in vertical subsurface flow constructed wetlands. Water Research, 42, 1622-1634.

Tozeren, A., Byers, S.W. (2004) New Biology for engineers and computer scientists. Pearson Education Inc., NJ, USA.

Ulrich, H., Klaus, D., Irmgard, F., Annette, H., Juan, L.P., Regine, S. (2005) Microbiological investigations for sanitary assessment of wastewater treated in constructed wetlands. Water Research. 39:20, 4849-4858.

Vacca, G., Wand, H., Nikolausz, M., Kuschk, P., Stner, M. (2005) Effect of plants and filter materials on bacteria removal in pilot-scale constructed wetlands. Water Research. 39:7, 1361-1373.

Vymazal, J. (2005) Removal of enteric bacteria in constructed treatment wetlands with emergent macrophytes: A review. Journal of Environmental Science and Health - Part A Toxic/Hazardous Substances and Environmental Engineering. 40:6-7, 1355-1367.

Walker, T.S., Bais, H.P., Grotewold, E., Vivance, J.M., (2003) Root exudation and rhizosphere biology. Plant Physiology. 132, 44-51.

Wand, H., Vacca, G., Kuschk, P., Kruger, M., Kastner, M. (2007) Removal of bacteria by filtration in planted and non-planted sand columns. Water Research. 41:1, 159-167.

Weber, K.P. (2006) Investigation of the mechanisms and fundamental variables affecting acid mine drainage treatment in wetland mesocosms. M.A.Sc. Thesis, Department of Chemical Engineering, University of Waterloo, ON, Canada.

Weber, K.P., Gehder, M., Legge, R.L. (2008) Assessment of the changes in the microbial community in response to acid mine drainage exposure. Water Research. 42:1-2, 180-188.

Weber, K.P., Grove, J.A., Gehder, M., Anderson, W.A., Legge, R.L. (2007) Data transformations in the analysis of community-level substrate utilisation data from microplates. J. Microbiol. Methods. 69, 461-469.

Werker, A. G., Legge, R. L., Warner, B. G. (2000) Understanding wetland wastewater treatment variability. Canadian Society for Civil Engineering 2000 Conference Proceedings, London, Ontario. 538-544.

Werker, A., Dougherty J., McHenry. J., Van Loon, W., Legge, R., Warner, B. (2004) Mesocosms applied for design, optimisation, and benchmarking of wetland wastewater treatment. 9th International Conference on Wetland Systems, Avignon, France.

Werker, A.G., Dougherty, J.M., McHenry, J.L., Van Loon, W.A. (2002) Treatment variability for wetland wastewater treatment design in cold climates. Ecological Engineering. 19, 1–11.


Werker, A.G., Van Loon, W., Legge, R.L. (2007) Tracers for investigating pathogen fate and removal mechanisms in mesocosms. Science of the Total Environment. 380, 188-195.

Wieder, R.K. (1989) A survey of constructed wetlands for acid coal mine drainage treatment in the eastern United States. Wetlands. 9, 299-315.

Wieder, R.K., Linton, M.N., Heston, K.P. (1990) Laboratory mesocosm studies of Fe, Al, Mn, Ca, and Mg dynamics in wetlands exposed to synthetic acid coal mine drainage. Water, Air and Soil Pollution. 51, 181-196.

Williams, J. B., Zambrano, D., Ford, M. G., May, E., Butler, J. E. (1999) Constructed wetlands for wastewater treatment in Colombia. Water Science and Technology. 40, 217-223.

Williams, J. Bahgat, M., May, E., Ford, M., Butler, J. (1995) Mineralisation and pathogen removal in gravel bed hydroponic constructed wetlands for wastewater treatment. Water Science and Technology. 32:3, 49-58.

Wittgren, H. B. Maehlum, T. (1997) Wastewater treatment wetlands in cold climates. Water Science and Technology. 35, 45-53.

Wynn, T.M., Liehr, S.K. (2001) Development of a constructed subsurface-flow wetland simulation model. Ecological Engineering. 16, 519–536.

Xiao, L., Fayer, R., Ryan, U., Upton, S. J. (2004) Cryptosporidium taxonomy: Recent advances and implications for public health. Clinical Microbiology Reviews. 17, 72–97.

Zdragas, A., Zalidis, G.C., Takavakoglou, V., Katsavouni, S., Anastasiadis, E.T., Eskridge, K., Panoras, A. (2002) The effect of environmental conditions on the ability of a constructed wetland to disinfect municipal wastewaters. Environmental Management. 29:4, 510-515.


Chapter 6

THE ROLE OF HARVEST AND PLANT DECOMPOSITION IN CONSTRUCTED WETLANDS

Juan A. Álvarez1 and Eloy Bécares2 1Environmental Institute, University of León, La Serna, 56, 24007, León, Spain

2Department of Ecology, Genetic and Microbiology, Faculty of Biology, University of León, 24071, León, Spain

ABSTRACT

Upon decomposition, at the end of the summer and during the autumn, wetland vegetation releases organic carbon into the wetland system. A part of this organic matter remains in the wetland, and is degraded at different rates during the rest of the year. Therefore, litter decomposition has important consequences on constructed wetlands because it is related to the autochthonous production of organic matter, clogging rates in surface-flow wetlands, and terrestrialization in free-water surface wetlands.

The effect of harvest was studied in two free-water surface-flow wetlands. Both wetlands were planted with Typha latifolia with one of the wetlands harvested. On the other hand, decomposition rates of Typha latifolia were quantified during both winter and summer in the non-harvested surface constructed wetland using the litter bag technique. Nutrient concentrations were always lower in the effluent of the harvested wetland, indicating nitrogen and phosphorus release by decomposition of vegetation, in the non-harvested system. In addition, harvesting reduced the effluent TSS and BOD concentrations by 37.3% and 49.2%, respectively, when compared to the non-harvested wetland in spring. Seasonal background concentrations (C*) in the wetlands, increased from winter to spring and decreased again in summer. Organic load and nutrients produced per gram of Typha were evaluated by using in-situ Typha degradation experiments. Taking into account the experiments of litter bag technique, no significant differences were found in both variables among the different mesh sizes, with the exception of the control bags in winter. Meso or macrofauna did not play any role in plant decomposition. Decomposition rates were significantly different between winter and summer when considering each mesh size separately. Decomposition rates from adjusted exponential models ranged from 0.0014 to 0.0026 d-1 in winter (5ºC), and from 0.0043 to 0.0052 d-1 in summer (20ºC). Typha decomposition rates were compared with others macrophytes. From these decomposition rates, it is estimated that 31% of the

Juan A. Álvarez and Eloy Bécares 214

initial mass of plant detritus would remain in the system after one year. Based on the research conducted during several experiments, harvesting can be recommended as an operational and management strategy in warm climates and diluted wastewater conditions.

ABBREVIATIONS AFDW ash free dry weight, BOD biological oxygen demand, CW constructed wetland DO dissolved oxygen, DW dry weight, FWS free water surface, SSF subsurface flow, SD standard deviation, T Temperature, T latifolia: Typha latifolia. TCOD, SCOD total and soluble chemical oxygen demand, TKN total Kjeldahl nitrogen, TN total nitrogen TOC total organic carbon, TP t otal phosphorus, TSS, VSS total and volatile suspended solids.

1. INTRODUCTION

1.1. Role of Vegetation on Constructed Wetland Performance Over the last two decades, constructed wetland systems have been increasingly used to

treat domestic sewage, industrial wastewater, and agricultural runoff because they have particular mechanism and function in treatment of wastewater. Wetland treatment process has been gaining international interests and applications due to its low-energy cost, low maintenance and operational cost, and high removal capacity. The advantages of constructed wetland also include correspondingly low investment, multiversity and pertinence of treatment systems, efficient purification capability and especial environment function. They are especially suitable for the rural areas and small cities. Constructed wetland systems are an ecological technique by using natural purifying functions of soil, vegetation and microorganisms. Because of the presence of surface water, the systems are classified as free-water surface (FWS) systems, in which, a small depth of water (approximately 25-30 cm) is maintained on the top of the gravel bed, or subsurface flow (SSF) when water is a few centimetres below the gravel bed level. According to flow type the systems are classified as horizontal or vertical flow systems. The combination of plant species and flow types has produced different technologies. The most common combine FWS with horizontal flow or SSF with horizontal or vertical flow. Detailed descriptions on these technologies can be found

The Role of Harvest and Plant Decomposition in Constructed Wetlands 215

in several books and manuals (Hammer, 1989; Moshiri, 1993; Reed et al., 1995; Kadlec and Knight, 1996, Vymazal et al,1998; Young et al. 1998; Kadlec et al., 2000). On the other hand, nowadays, one of the great problems of wetland operation is the gravel bed clogging, this phenomenon can be avoided or delayed using a proper pre-treatment as can be anaerobic treatment (Álvarez et al., 2008).

Originally, the basis for employing constructed wetland for wastewater treatment is the ability of water plants to translocate oxygen to their roots, and the surrounding water (wastewater, in case of treatment wetlands) environment. Although a number of other pollutants removal processes have been identified, the wetlands plants play a major role in the occurrence of most of these processes (Sundaravadivel et al., 2001). Within the water column, the stems and leaves of the wetland plants significantly increase surface areas for biofilm development. Plant tissues, more over are colonised by photosynthetic algae as well as by bacteria and protozoa. Likewise, the roots and rhizomes that are buried in the wetland substrate provide for attached for grow microorganisms (Brix, 1997).

Besides, aquatic plants have important functions in the ecological restoration of a polluted water body. With these functions, the water body returned to ecologic balance, and biological diversity was improved greatly. The community of aquatic plants can be divided into three kinds by their life type: floating plant community, submersed-plant community, and emergent-plant community. In the recent decades, a large number of studies on constructed wetlands for treating contaminated waters have produced data on pollutant reductions, and the influence of aquatic plants on organic matter and nutrient removal efficiency in constructed wetlands is often presented in the literature (Coleman et al., 2001; He and Mankin, 2002; Yuan et al., 2004; Zhang and Hong, 2006; Akratos and Tsihrintzis, 2007; De Feo, 2007; Vymazal, 2007). Summarizing many studies for constructed wetlands, wetland plants play many important roles in constructed wetlands (table 1) (Brix, 1997; Sundaravadivel et al., 2001; Scholz and Lee, 2005; Zhang and Hong, 2006). The most two important functions include:

1. Uptaking the nutrients, absorb and accumulate heavy metal and poisonous substances

from wastewater, but the absorbency and functionary mode of removing nutrients and poisonous substances by wetland plants are different.

2. Transferring oxygen to rhizophere for the growth, reproduction and decompositions of microorganisms, and playing important roles in the simultaneous nitrification and denitrification.

1.1.1. Wetland Plant Species: Requirements

A wide variety of aquatic plants can be used in constructed wetland systems designed for wastewater treatment. Depending on the type of plants, the systems can include submerged plants (limnophytes, e.g. Myriophyllum spp., Ceratophyllum spp.), emergents (helophytes, Typha spp., Scirpus spp., Phragmites spp.) or floating plants (pleustophytes, Lemna spp., Eichhornia spp., Salvinia spp.). Commonly, however, constructed wetlands are planned as marsh-type wetlands and are planted with emergent macrophytes (rooted plants that anchor to the substrate media) that are adapted to water-dominated environment. Frequently used macrophytes species are cattails (Typha spp.), reeds (Phragmites spp.), bulrushes (Scirpus spp.); rushes (Juncus spp.), and sedges (Carex spp.). The general requirements of plants suitable for use in constructed wetland wastewater treatment systems include (Tanner, 1996):


1. Ecological acceptability, that is, no significant weed or disease risks or danger to the ecological or genetic integrity of surrounding natural ecosystems;

2. Tolerance of local climatic conditions, pests and diseases; 3. Tolerance of pollutants and hypertrophic water-logged conditions; 4. Ready propagation, and rapid establishment, spread and growth; and 5. High pollutant removal capacity, either through direct assimilation or storage, or

indirectly by enhancement of microbial transformations. Specific requirements will vary depending on the functional role of wetland plants in the

treatment systems. This will be related to the type of wetland design and its mode of operation (continuous or batch), loading rate, and wastewater characteristics. Other ancillary objectives (such as ecological, aesthetic, recreational, and economic) of wetland developments may also affect the choice of the plants.

Ansola et al. (1993) did not find differences among plant species (Scirpus lacustris, Typha angustipholia, Iris pseudacorus, Phragmites australis) for all variables studied with the exception of phosphates for Iris pseudacorus, which presented significantly higher removals compared to Typha and Phragmites. In other experiment (Ansola 1994), compared three-way combinations of plants (Scirpus-Iris-Phragmites, Typha-Iris-Scirpus, Typha-Scirpus-Phragmites) growing in separate tanks. Results showed no differences among combinations of plants for nutrients or organic matter. Concerning faecal indicators and pathogen removal, López and Bécares (1993) found that Scirpus lacustris had higher bacteria removal rates than the aforementioned species, as shown by most faecal indicators used.

Table 1. Major roles of macrophytes in constructed wetlands

Wetland plant part Role

Aerial plant tissues - Light attenuation: reduced growth of phytoplankton - Influence on microclimate: insulation during winter - Reduced wind velocity: reduced risk of resuspension of solids - Aesthetic appearance - Nutrient storage

Plant tissue in water - Filtering effect: filter out large debris - Reduced current velocity: increased rate of sedimentation, reduced

risk of resuspension - Surface area for attached microorganism - Excretion of photosynthetic oxygen: increased aerobic degradation - Nutrient uptake

Roots and rhizomes - Stabilising the sediment surface: less soil erosion - Prevents the medium for clogging in vertical flow systems - Release of oxygen increase organic degradation and nitrification - Nutrient uptake - Secretion of antibiotics for detoxification of root zone: pathogen

removal


Coleman et al. (2001) indicated that unplanted systems provided significant wastewater treatment, but planted systems further improved many treatment efficiencies. Typha significantly out-performed Juncus and Scirpus both in growth and in effluent quality improvement. There was also some evidence that the species mixture out-performed species monocultures. Typha was the superior competitor in mixtures.

Akratos and Tsihrintzis (2007) concluded that Typha planted systems achieved higher organic matter and nutrient removal efficiency than Phragmites planted systems.

1.1.2. Oxygen Supply by Plants

Oxygen can be considered to cycle within wetlands. Oxygen enters via inflows or by diffusion at the water surface when the surface is turbulent (for example, due to wind mixing). Oxygen is also produced within the water column during photosynthesis. It is well documented that aquatic macrophytes release oxygen from roots which influences the biogeochemical cycles in the sediments due to the effects on the redox status of the sediments (Barko et al., 1991; Sorrel and Boon, 1992). Qualitatively this is visualized by reddish color associated with oxidized forms of iron on the surface of the roots.

Dissolved oxygen is a key factor affecting the simultaneous nitrification and denitrification process and higher dissolved oxygen in the wetland system can improve the simultaneous nitrification and denitrification. According to some microenvironmental studies regarding plant, oxygen release in a wastewater environment are important to understand the principles of constructed wetlands for wastewater treatment. Many studies insighted into root-induced microenvironments and would be helpful for the quantification of the total amount of oxygen contributed by plants in constructed wetlands (read above). Amount of oxygen released by plant roots has been quantified based on the radial oxygen loss (ROL) exhibited by the roots, the number and the length of active lateral roots, and the field plant density (Zhan and Hong, 2006).

However, there are also some controversial claims about the magnitude of oxygen supply by plants to the wastewater in constructed wetland systems. Armstrong and Armstrong (1988) found little oxygen actually escaping from root zone, and the small amounts of this surplus oxygen vary with the plant species. Lawson (1985) calculated that up to 4.3 g/m2d of oxygen flux is possible from the roots of Phragmithes spp. Brix and Schierup (1990) found that of a total oxygen influx of 5.86 g/m2d to a wetland planted with Phragmites, 3.76 g come from the atmosphere directly to the water column. The remaining 2.1 g oxygen influx was through the plant tissue, of which 2.08 g gets transferred to the rhizome system for root respiration. Only 0.02 g of oxygen that was in excess of plant requirement was leaked by the roots into the water environment. Gries et al. (1990) measured root oxygen release to be in the range of 1 to 2 g/m2d.

1.1.3. Physical Effects Caused by Vegetation

The physical presence of vegetation in wetlands distributes and reduces the current velocities of the water, which creates better conditions for sedimentation of suspended solids. Light attenuation by the wetland plants hinders the production of algae in water below the vegetation cover. The vegetation cover in a wetland can be regarded as a thick biofilm located between the atmosphere and the wetland soil or water surface in which significant gradients in different environmental parameters occur. Wind velocities are reduced near the soil or water surface compared to the velocities above the vegetation, which reduces resuspension of


settled material and thereby improves the removal of suspended solids by sedimentation. In temperate areas, the plant cover provides insulation during winter and helps keep the substrate free of frost (Smith et al., 1997).

1.1.4. Role of Vegetation on Wetland Operation: Hydraulic Conductivity

The contribution of macrophytes to the overall treatment performance is assumed to vary to a large extent depending on the wetland design and operation practices. Macrophytes are likely to affect considerably the removal of pollutants in horizontal sub-surface constructed treatment wetland, while their role is minor in pollutant removal for periodically loaded vertical-flow wetlands (Brix, 1997). Nevertheless, the secondary role of macrophytes concerning oxygen transport, clogging prevention and provision of an energy source for micro-organism can influence positively the treatment performance of wetlands (Cooper, 1996).

In systems with subsurface water flow, substrate hydraulic conductivity is an important design parameter. Wastewater interacts directly with the rhizosphere, and roots have additional functions apart from being the physical support for the biofilm. Nevertheless, in FWS systems, the hydraulic conductivity of the gravel bed and therefore the role of the rhizosphere is negligible (Kadlec and Knight, 1996). The main role of macrophytes is to provide additional surface for the development of a biofilm on the submerged parts of plants. Comparing planted and unplanted tanks, Ansola et al. (1993, 1995) proved that planted tanks were significantly different from control plots with regards to DBO, COD and total phosphorus, but no differences were found between planted and unplanted tanks with regards to nitrogen forms (ammonia, nitrates, organic nitrogen) nor phosphates. Differences between FWS and SSF in the type of substances removed (nutrients in SSF, organic matter in FWS) are evidence that plants have a passive role in the superficial-flow systems, acting as a physical support for bacterial growth on their submerged leaves and stems.

On the other hand, the influence of roots and rhizomes on the hydraulic conductivity of SSF wetlands depends on the type of substrate used. In SSF systems, wastewater flow is largely intended to be below the surface through channels created by living and dead roots as well as through the pore space of the substrate medium. As roots and rhizomes grow, they disturb and loosen the substrate medium. Furthermore, when the roots and rhizomes die and decay, they may leave behind tubular pores and channel, which can improve the hydraulic conductivity of the substrate medium. This may be largely true with bed medium based in gravel substrate. On the contrary, the hydraulic conductivity of soil-based systems often decreases (Marsteiner et al., 1996). Data on hydraulic conductivity in soil-based reed beds in Austria, Denmark, and in the UK also do not support the increase in hydraulic conductivity due to wetland plants in soil-based systems (Conely et al., 1991; Haberl and Perfler, 1990).

1.1.5. Role of Vegetation on Wetland Treatment Efficiency

The removal of organic and inorganic matter and bacteria from wastewater carried out by macrophytes has been explained through several mechanisms, such as sedimentation, mechanical filtration or nutrient assimilation by plants (Bécares, 2006). Their roots may also serve as substrates for attached bacteria degrading and taking up nutrients and organic carbon (Brix, 1995). The latter process is favoured by oxygen release into the rhizosphere (Gersberg et al., 1986), and by plant exudates (Stengel, 1985). Despite this evidence, there is still controversy about the mechanisms of functioning of macrophytes for wastewater treatment in


constructed wetlands. Some researchers have found wastewater treatment is improved in the presence of macrophytes (Rogers et al. 1991; Farahbakhshazad et al., 1995), while other studies have not detected significant differences in treatment results between planted and unplanted systems (Tanner et al., 1995). Nevertheless, comparisons between studies are difficult because they utilize diverse aquatic plant species, wastewaters and flows.

1.1.5.1. Organic Matter and Suspended Solid Removal

Plants do not take up organic matter but foster the growth of organisms in the rhizosphere, which could help on such task (Brix 1994). Bécares (2006) indicated that removal rates for biological oxygen demand (BOD) and suspended solids (SS) were slightly higher in summer than winter but no statistically significant differences were found when comparing winter/summer or planted/unplanted conditions. Absence of differences between tanks showed by Bécares (2006) could be due to two reasons: First, solids (organic matter) accumulate in the first third of the system, mostly at the head of the tanks (Tanner and Sukias 1995). Therefore, efficiency is potentially independent of plant presence. Second, BOD loads were very low, below 3 g/m2d. Tanner et al. (1995) found this value to be the limit for detecting significant plant effects on BOD removal.

Yuan et al. (2004) and DeFeo (2007) studied vegetation effect on organic matter removal on vertical flow wetland. Yuan et al (2004) showed the following COD removal: 80.5%, 75.5%, 70.5% and 61.4% in wetlands with Acorus gramineus, Juncus effuses, Iris japonica, and without plants, respectively. So, vegetation role on COD removal is significant and it is correlated with the ability of absorbing organic substances. DeFeo (2007) indicated that vertical wetlands planted with Phragmites removed more COD (1.9-8.2%) than the unvegetated wetlands due to the isulating effect of the vegetation that created better temperature condition for the development of the biological processes inside the wetlands.

In case of horizontal flow wetlands, Akratos and Tsihrintzis (2007) showed significant differences on BOD removal between wetlands unplanted (85.7%) and planted with Typha (88.3%), but the differences were not significant between wetlands unplanted (85.7%) and planted with Phragmites (84.6%). The explanation of this phenomenon may be that the main mechanism responsible for organic matter removal is the microbial activity of aerobic and anaerobic bacteria (Greenway and Woolley, 1999; Steer et al., 2002; Vymazal, 2002). The higher removal efficiencies in organic matter by the Typha unit, is probably due to the fact that cattails have a more vigorous root system (Huang et al., 2000).

1.1.5.2. Nutrient Removal

Nutrient requirement for growth of wetland macrophytes, mainly the nitrogen and phosphorus, are taken up primarily through their root systems. Marginal uptake occurs also through immersed stems and leaves from the surrounding water (Gumbricht, 1993). Thus, the vegetation may be helpful in removal of nutrients from wastewater. Shaver and Mellio (1984) have shown that nutrient uptake is maximum during the initial period of establishment of the plants in constructed wetlands, and the efficiency tends to decrease as available nutrient input rises, that is, when the nutrient loading rates increases, the uptake of nutrient by plants decreases. Tanner (1996) found that during the first 2-year period of operation, plant uptake of nutrients could only account for 6 to 10% for nitrogen, and 6 to 13% of phosphorus removal from wetlands.


Nitrogen Removal The processes that affect removal and retention of nitrogen during wastewater treatment

in constructed wetlands are manifold and include NH3 volatilization, nitrification, denitrification, nitrogen fixation, plant and microbial uptake, mineralization (ammonification), nitrate reduction to ammonium (nitrate-ammonification), anaerobic ammonia oxidation (ANAMMOX), fragmentation, sorption, desorption, burial, and leaching (Kadlec and Knight, 1996; Vymazal, 2002; Mayo and Mutamba, 2004). However, only few processes ultimately remove total nitrogen from the wastewater while most processes just convert nitrogen to its various forms. Removal of total nitrogen in studied types of constructed wetlands varied between 40 and 55% with removed load ranging between 250 and 630 g N/m2yr depending on CW type and inflow loading (Vymazal, 2007). However, the processes responsible for the removal differ in magnitude among systems. Single-stage constructed wetlands cannot achieve high removal of total nitrogen due to their inability to provide both aerobic and anaerobic conditions at the same time. Vertical flow constructed wetlands remove successfully ammonia but very limited denitrification takes place in these systems. On the other hand, horizontal-flow constructed wetlands provide good conditions for denitrification but the ability of these system to nitrify ammonia is very limited. Therefore, various types of constructed wetlands may be combined with each other in order to exploit the specific advantages of the individual systems (Vymazal, 2007).

A strong positive correlation was found between ammonia surface loading and removal rate (Soto et al. 1999, 1999b). Removal rate was significantly higher (p<0.05) in planted tanks and also higher in summer than winter. Planted tanks removed 50% more ammonia than unplanted under winter conditions (0.145 g/m2d.), and about 63% more in summer conditions (0.297 g/m2d). This means that plant activity in summer was only responsible for 13 % of the ammonia uptake (i.e. 0.152 g/m2d), while other mechanisms unrelated to plant uptake but linked to plant presence, were responsible for the rest of the removal observed (Bécares, 2006).

Nitrification occurred in both planted and unplanted tanks. Bécares (2006) indicated that unplanted tanks were marginally more effective than those with plants in the removal of nitrate (differences were not statistically significant). The ratio C:N (as BOD5:TN) was 2.42 in winter, and 3.55 during summer, high enough for denitrification to occur (Radtke, 1995). Nitrogen accumulation by plants was calculated in about 30% of the TN removed into the planted system, including the N accumulated in submerged parts, while the remaining 70% was likely removed by the microbial pool and lost through denitrification (Soto et al. 1999).

Another studies in horizontal wetlands (Akratos and Tsihrintzis, 2007) showed that vegetation type affects nitrogen removal more than organic matter removal, something that was also observed by He and Mankin (2002). Akratos and Tsihrintzis (2007) obtained a TKN removal efficiency of 66.8, 54,8 and 34.2% in Typha planted wetland, Phragmites planted wetland and no planted wetland, respectively. So, It is showed the important role of vegetation on nitrogen removal. In case of vertical flow wetlands, Yuan et al. (2004) and DeFeo (2007) reached data in the same way, the constructed wetlands with macrophytes had a higher nitrogen removal efficiency than unplanted wetlands.

Phosphorous Removal

Phosphorus removal in constructed wetlands is a result of bacteria removal, plant uptake, adsorption by the porous media and precipitation, where phosphorus reacts with the porous


media and with minerals such as ferric oxyhydroxide and carbonate (Kadlec and Knight, 1996; Yang et al., 2001). Bacteria removal and plant uptake are responsible for P-PO4

-3

removal, while precipitation and adsorption are responsible for the removal of all phosphorus forms (Kadlec and Knight, 1996). Removal of total phosphorus varied between 40 and 60% in all types of constructed wetlands with removed load ranging between 45 and 75 gN/m2yr depending on CW type and inflow loading (Vymazal, 2007).

It is generally accepted that the mechanisms for phosphorus removal are more related to gravel surface processes, such as physical adsorption and chemical precipitation by Ca or Fe (see, for example, Richardson, 1985; Faulkner and Richardson, 1989), than to biological processes, such as uptake by plants and microorganisms. However, and similarly to that found by Faulkner and Richardson (1989), Bécares (2006) showed that the removal rate of reactive phosphorus and the removal efficiency were higher in planted than unplanted tanks. According to this, macrophytes were directly involved in the removal of 29% of TP in winter and 47.3% in summer (i.e. 0.024 and 0.057 g/m2d removed, respectively). These differences were statistically significant in summer.

In the same way that nitrogen removal, studies carried out in horizontal (Akratos and Tsihrintzis, 2007) and vertical wetlands (DeFeo, 2007) showed that the vegetation system reached higher phosphorous removal efficiency that the unvegetated systems.

1.1.5.3. Bacteria Removal

Similarly to previous researches (Gersberg et al. 1990, Rivera et al. 1995, Tanner et al. 1995, Loveridge et al. 1995), from differences between planted and unplanted systems, Bécares (2006) concluded that macrophytes also play an active role in the removal of microorganisms from wastewater. Rooted biofilms provide a better substrate than gravel surfaces for microbial activity (Loveridge et al. 1995). Constructed wetlands are more efficient than conventional systems in the removal of bacteria but generally less efficient than stabilization ponds (García and Bécares 1997). The comparison of planted and unplanted systems generally showed higher rates of bacteria removal in the presence of plants, although results were highly variable and were dependent on plant type, hydraulic design and wastewater characteristics (Hammer 1989, Tanner et al. 1995).

Several mechanisms have been proposed, and on occasion demonstrated, to be responsible for bacteria removal. Oxygen production and bacterial activity in the rhizosphere (Brix 1987, 1997), sedimentation, filtration and adsorption (Gersberg et al. 1989, Williams et al. 1995) are commonly cited mechanisms in the literature. Decamp and Warren (1998) and Rivera et al. (1995) have pointed bacterivory as a key mechanism in constructed wetlands, as microfaunal densities and predatory activity were higher in the presence of plants in their experiments. Excretion of antibacterial compounds by plants is another controversial mechanism frequently cited in the literature, but not yet clearly proved. Plants modify the soil microenvironment and probably release substances, which enhance the development of specialized bacterial on its rhizosphere (Hatano et al., 1993). Commonly cited papers, such as those by Seidel (1955, 1976), Gopal and Goel (1993), Ottová et al. (1997) and others have not been able to prove a direct effect of plants on bacteria inhibition but have found that plant presence is related to higher bacteria reductions. There is evidence showing that some plants produce secondary metabolites with antibacterial properties (e.g. Dellagreca et al. 2001), although their role in wastewater treatment has still to be proved.


Studies comparing the role of Scirpus lacustris in planted and un-planted subsurface flow tanks (García et al., 1999, Soto et al., 2000) showed that planted tanks were more efficient at removing microbes (up to 99.9%) than unplanted tanks. There were statistical differences between planted and control conditions (p<0.01) for total coliforms, faecal streptococci and total heterotrophs, and removal rates were higher in summer than winter. Antibacterial activity potentially exerted by plants was not detected by using filtered effluents from planted tanks as dilution water for total bacterial growth in the influent (García et al., 2004). Predation, as a potential mechanism for bacteria removal was also evaluated in these systems (García et al. 2004). Decamp et al. (1999) found a higher ciliate abundance and predatory activity in planted than unplanted gravel wetlands. Bécares (2006) showed that the abundances of ciliates and flagellates, the only bacterivorous organisms found, were much lower than reported in other similar systems (Decamp et al. 1999, Panswar and Chavalparit 1997). Differences in abundance in planted and unplanted tank were statistically significant for flagellates but not for ciliates. The presence of plants increased protozoan abundance, which could be another reason for the higher bacteria removal in planted systems (García et al. 2004).

1.1.6. Organic and Antibiotic Excretion by Plants

Root systems of wetland plants also release substances other than oxygen. Early experiments of the Max Planck Institute in Germany showed that the bulrush Schoenoplectus released antibiotics from its roots. It is also known that a range of submerged macrophytes release compounds that affect the growth of other species (Seidel, 1955). However, the role of these compounds in wetland treatment processes has not yet been experimentally verified. Plants also release a wide range of organic compounds through their roots (Rovira, 1969; Barber and Martin, 1976). Reported values of these organic compounds are in the range of 5 to 25% of the photosynthetically fixed carbon (Brix, 1997). The organic carbon so excreted may act as a carbon source for denitrification process, and hence enhance the nutrient removal process in constructed wetlands (Platzer, 1996).

1.2. Wetland Vegetation Harvest and Decomposition As it was indicated in section 1.1.5 of this chapter, macrophytes play a fundamental role

in nutrient and organic matter removal and processing in constructed wetlands (Kadlec and Knight, 1996, Mbuligwe, 2004). In general, wetland vegetation grows and emerges during spring and summer, but at the end of summer, it starts to decay and it is degraded during the rest of the year, making an additional contribution of organic matter and nutrients to the system. In constructed wetlands, this autochthonous input of organic matter is added to allochthonous input from the influent wastewater. Several studies have dealt with plant decomposition and their degradation rates in constructed wetlands (e.g. Vymazal, 1995; Kadlec and Knight, 1996; Álvarez and Bécares, 2006), but very few papers study the fate of the organic matter produced by plant decomposition (autochthonous matter) and its relation to allochthonous organic matter (Wetzel, 2001). In this way, Pinney et al. (2000) showed that from 100% of plant carbon in constructed wetlands, 5-8% was in the form of dissolved organic carbon; 45-60% was accumulated as total biomass and 35-60% was used either in bacteria growth or it was degraded to CO2 and CH4.


Under certain conditions, such as in systems treating diluted wastewater or in highly productive areas, the autochthonous input can contribute significantly to the background concentration of organic matter in the wetland. Under these circumstances, harvesting is an appropriate operation for improving system efficiency (Yang et al., 2001; Karathanasis et al., 2003). On the other hand, macrophyte decomposition increases the gravel clogging in SSF systems and the sediment layer in FWS systems, leading to the terrestrialization of the wetland (Kirschner et al., 2001). Decomposition of aquatic macrophytes may also regulate the recycling of nutrients in ecosystems and thus it influences the long-term carbon storage of the ecosystem (Xie et al., 2004). For this reason, harvesting is an exploitation technique that it is still subject to debate.

The performance of constructed wetlands may be improved by harvesting the plants, thus additional BOD and TSS inputs from decaying biomass litter will be reduced (Karathanasis et al., 2003). With respect to nutrients, it has been suggested that FWS systems should be harvested frequently to enhance nutrient removal (Yang et al., 2001). Vymazal (2007) indicated that removal of both nitrogen and phosphorus via harvesting of aboveground biomass of emergent vegetation is low but it could be substantial for lightly loaded systems (100-200 gN/m2yr and 10-20 gP/m2yr). Systems with free-floating plants may achieve higher removal of nitrogen via harvesting due to multiple harvesting schedule (Vymazal, 2007). Nevertheless, other studies have proved that harvesting produces a negligible effect on nutrient removal from the system; it is labour intensive and it may not be always economically or ecologically feasible (Kim and Geary, 2001; Wang and Mitsch, 2000).

Plant matter decay begins before plant parts fall into the water, and litter starts degrading in the aerial standing dead phase (Kuehn et al., 1999; Kuehn and Suberkropp, 1998). The decomposition of aquatic macrophytes in water can be divided into three distinct phases: an initial rapid loss due to leaching, a period of microbial decomposition and colonisation, followed by mechanical and invertebrate fragmentation (Megonigal et al., 1996; Webster and Benfield, 1986). The length of each phase depends on many environmental factors, although its variability is associated with the food quality of the detritus (Anderson and Shedell, 1979), which is related to the initial C/N ratio, hardness of the plant material and availability of nutrients from the environment (Menéndez et al., 2001; Lee and Bukaveckas, 2002)

In general, aquatic invertebrate communities enhance leaf litter breakdown and decay (Cummins, 1988; Menéndez et al., 2001; Menéndez et al., 2003). On the other hand, the hydrologic regime and nutrient availability also contribute to the decomposition of plant detritus (Webster and Benfield, 1986; Chauvet, 1997). Biochemical properties and decomposition rates of the litter have been shown to vary greatly among wetlands in diverse physiographic settings (Mitsch and Gosselink, 1993). Decomposition rates also depend on physical and chemical water conditions, like water temperature, hydrologic regime (period of inundation), and various other water quality attributes, including pH and redox conditions (Hobbie, 1996; Murphy et al., 1998).

In this chapter, a study has been carried out with the objective of testing the effect of vegetation on the organic matter dynamics of a surface flow wetland comparing a harvested versus non-harvested wetland. Organic matter and nutrient production of Typha latifolia was also calculated. Besides, it was analysed the decomposition rates of Typha latifolia in a constructed wetland in winter and summer conditions and evaluated the consequences of litter accumulation to the system performance.


2. MATERIAL AND METHODS

2.1. Site Description and Harvest Study Two free water surface wetlands with an area of 44 m2 each, and with 30 cm of 6-8 mm

diameter gravel and 40 cm of water depth were planted in 2000 with Typha latifolia (3008 gDW/m2 plant density in summer 2004). In autumn (October 2004), one of the wetlands was harvested by cutting plants by hand 2 cm above the water level to preserve the hydraulics and the effect of stem biofilm in the harvested wetland. These two wetlands were situated in the municipal wastewater plant of Cubillas de los Oteros, a small village of 150 inhabitants situated in León (North-west Spain). This treatment facility consisted of a septic tank as pre-treatment followed by a stabilization pond. The effluent from the stabilization pond entered in two surface flow wetlands and finally in a subsurface flow wetland.

The operation period (272 days) was divided into three seasons, winter (1st December 2004 to 8th March 2005, days 1 to 98), spring (8th March 2005 to 14th June 2005, days 98 to 196) and summer (14th June 2005 to 29th August 2005, days 196 to 272). Samples of the pond effluent (wetland influents) and the effluents of each wetland were taken weekly. Analysis of total and volatile suspended solids, (TSS, VSS), total and soluble chemical oxygen demand (TCOD, SCOD), biological oxygen demand (BOD), total Kjeldahl nitrogen (TKN), ammonium-nitrogen (NH4), total phosphorus (TP), total organic carbon (TOC) and chlorophyll-a were carried out following Standard Methods (1995). In addition, measurements of pH, temperature (T), and dissolved oxygen concentration (DO) were also taken.

To maintain an average organic loading rate below 12 gBOD/m2d in the superficial wetlands, according to facility design data, and due to seasonal changes in the organic matter concentration of the wastewater, the flow was held constant by an influent pump timer. The rate was maintained at about 11.0 m3/d until day 79 of the study and changed to 8.6 m3/d from day 79 until the end of the study; the excess flow was not treated in the plant. The average theoretical hydraulic retention time (HRT) in each wetland was 17.6 h. Evaporation was calculated in summer and it was considered negligible as it represented less than 10% of the lowest flow applied to the system.

Taking into account the operation data of this study, background concentrations (C*) of TSS, BOD, TKN and TP were calculated for each seasonal period as well as for the whole study period according to Kadlec and Knight (1996). The observed C* concentrations were evaluated by subtracting the harvested wetland concentrations from the non-harvested wetland concentrations. Calculated C* were compared with the observed background concentrations in this study.

2.2. Typha Degradation Tanks Typha degradation was measured in 10 L plastic tanks. Tanks were filled with 9.1 litres

of non-chlorinated tap water and 250 g of dried senescent Typha from the same wetland (100 g of leaf and 150 g of stems, following leaf:stem weight ratio in plants). Three tanks were


placed inside the non-harvested wetland. The tanks were opened to the atmosphere but covered with the lid half-twisted to avoid the addition of rainwater or insects.

The experiment was carried out in winter (starting December 2004) and in summer (starting June 2005) and it lasted about 90 days. Water samples of 800 mL were collected from each tank after 1, 3, 8, 22, 58, and 87 days to analyse TSS, VSS, TCOD, SCOD, BOD, TKN, TP, and TOC. The same volume of clean water was added to the tanks on each sample day and the dilution effect in the organic matter data was taken into account. One of the tanks remained in the wetland until day 262. Samples from this tank were also measured at day 176 and 262. The physico-chemical characteristics of Typha degradation trials (pH, DO and T) were measured in the tanks in both winter and summer experiments.

2.3. Decomposition Experiments Dry sections of Typha from the harvested wetland were separated into stems and leaves,

and cut into 10 cm long pieces. Stems and leaves were air-dried to constant weight at 65ºC and then nitrogen percentage and litter weight were measured. The litter bags were tubes made of PVC (15 cm long; 10 cm diameter) closed at both ends with mesh of varying size (25 mm, 100 mm, 1 mm or 5 mm). Control tubes without mesh were also used, allowing water free flow. Each litter tube contained 2 g of leaves and 3 g of stems, (5 g DW) imitating the proportions previously measured in the Typha. Three tubes of each mesh size were retrieved after 1, 3, 8, 20, 58 and 90 days. In December 2004, 18 tubes of each mesh size (i.e. 90 in total) were submerged along the non-harvested wetland placing a stone inside each tube to maintain them under water at all time. The three replicate tubes of each mesh size were placed at the beginning, middle and end of the wetland. A summer experiment was performed in the same way from 6 June 2005 to 1 September 2005 to evaluate temperature effects in Typha decomposition by comparing results obtained in winter and in summer.

2.3.1. Sample Processing and Analysis

At each sampling time, tubes were collected and dry weight of the detritus remaining was measured by drying at 65ºC to constant weight (48 h) and weighing. A weight fraction of the detritus was ground and burnt at 550ºC for 12 h to measure ash content and calculate ash-free dry weight (AFDW). Nitrogen percentage was assessed by the Kjeldahl method (Standard Methods, 1995). In situ measurements of pH, temperature and dissolved oxygen concentration of the wetland water were also taken at each sampling time.

2.4. Data Analysis Statistical analyses were carried out using Statistica (StatSoft, 1995). Pair-wise

comparisons (ANOVA) were performed using Tukey HSD post-hoc tests (HSD, honest significant difference), using vegetation and seasonal effect as covariates. ANOVAs were also used to evaluate all wetland inflow/outflow water quality differences.

Decomposition rates were estimated using a linear model (Howard and Howard, 1974), in which the rate follows the equation X=-kt, where X is the loss of material as proportion of the initial mass, t is the time (in days) after the initial exposure and k is the decomposition


constant. An exponential model (Olson, 1963) was also used, in which decomposition rate follows the equation X= e-kt. Analysis of covariance (ANCOVA) of the percentage of AFDW remaining after each sampling time, with time as a covariate, was used to detect differences in decomposition rates between treatments (summer-winter and mesh sizes). Subsequent pair-wise comparisons were performed using Tukey HSD post-hoc tests (HSD, honestly significant difference).

3. RESULTS OF HARVEST STUDY

3.1. Harvested and Non-Harvested Wetland Differences Influent pH was about 8.8 when the entire operation period was taken into account

(Figure 1). Effluent pH decreased to 8.5 and 8.0 in the harvested and non-harvested wetlands, respectively. Temperature and dissolved oxygen values were lower in the non-harvested wetland in all seasons.

Figure 1. Physico-chemical parameters of wetland influent and effluents (mean ± SD) during winter, spring, and summer. * indicates significant difference between the harvested and non-harvested wetland effluent (ANOVA, p<0.05).

There were significant differences in these variables between the harvested and non-harvested wetland effluent when the entire operation period was considered as a whole (pH: p<0.003; DO: p<0.02 and T: p<0.0004). Nevertheless, when each season was taken into


account separately, statistical differences were only observed for pH and T during winter (p<0.02 and p<0.005 for pH and T, respectively). TCOD, BOD and nutrient concentrations (TKN, NH4, and TP) were lower in the harvested wetland effluent than in the non-harvested wetland effluent during the entire operation period (Figure 2).

No significant differences between the harvested and non-harvested wetland effluent were found for any variable (Figure 2) when each season was considered separately. Nevertheless, statistical differences in BOD concentration were found between the harvested and non-harvested wetland effluent when the entire operation period was taken into account (p<0.005), BOD concentration was therefore significantly higher in the non-harvested wetland.

Figure 2. Average organic matter and nutrient values of wetland influent and effluents (mean ± SD) in winter, spring, and summer.


As can be observed in Figure 2, the higher mean differences in TSS, TCOD, and BOD concentrations were in spring. TSS and BOD effluent concentrations decreased 37.3% (from 67 to 42 mg/L) and 49.2% (from 63 to 32 mg/L) in the harvested wetland with respect to the non-harvested wetland, respectively. In addition, TSS reduction caused a TCOD decrease of 26.2% in the harvested wetland, since that SCOD concentration was similar in both effluents.

With respect to nutrients, significant differences for TKN, NH4, and TP were found when the entire study period was taken into account (p<0.006, p<0.01 and p<0.02, for TKN, NH4, and TP, respectively), although no significant differences were found for each season separately (Figure 2). TKN, NH4, and TP concentrations were always higher in the non-harvested wetland effluent, suggesting a nitrogen and phosphorus release due to plant decomposition. On the other hand, the sunlight in the harvested wetland could promote periphyton productivity, which could lead to an increase in nutrient uptake within the harvested wetland (Grimshaw et al., 1997), i.e., the harvested wetland removed more nutrients.

Ratios between COD, BOD, and TOC were calculated in influent and effluents to evaluate the variation in biodegradable organic matter (Table 2). No significant differences between effluent ratios were found during the entire operation period. The BOD/TCOD ratio of the effluents was much higher than those found in active sludge (usually less than 0.25, Henze et al., 1997) or in other wetland effluents (0.12-0.18, Baptista et al., 2003). This indicates the high biodegradability of the effluents from these wetlands.

Table 2. Ratios of the parameters analysed in influent and both effluents of the systems

studied. Comparisons among seasons (p values, ANOVA test) for all ratios are presented in the last row. Comparisons (p-values, t-test) between the harvested and non-

harvested effluent are presented in the last column

Ratios Influent Non-harvested effluent

Harvested effluent

Effluent comparisons

BOD/TCOD 0.44 0.57 0.46 p=0.067 BOD/SCOD 1.82 1.75 1.75 p=0.423 SCOD/TCOD 0.39 0.42 0.41 p=0.537 TCOD/TOC 13.44 11.80 11.20 p=0.593 BOD/TOC 6.42 6.57 4.84 p=0.063 Seasonal comparisons p< 0.03 p=0.66 p=0.154

3.2. Organic Matter Released by Wetland Vegetation Background concentrations (C*) were calculated for each seasonal period as well as for

the whole study period following Kadlec and Knight (1996) methodology. These calculated C* were compared with observed background concentrations (Table 2). Differences between effluents can be considered to be mostly due to the presence of vegetation, therefore the observed C* concentrations were calculated by subtracting the harvested wetland concentrations from the non-harvested wetland concentrations. The effect of vegetation was


more important in spring when concentration differences between the non-harvested and harvested wetland effluent were higher (Table 2). Comparison of the observed C* and the calculated C* indicated that there was an additional supply of organic matter and nutrients from wetland vegetation. Observed TSS concentrations were 1.7 times higher than the calculated TSS in spring, and observed BOD concentrations were 2.5, 4.5, and 2.2 times higher than the calculated BOD in winter, summer, and spring, respectively.

Figure 3. Box plots of organic matter and nutrient production in Typha degradation trials. Mean (mg/d); Box: Mean ± SE (Standard error); Whisker: Mean ± SD (Standard deviation). s: summer; w: winter. Note that the X-axis scale is not linear.

3.3. Typha Degradation Tanks

Organic matter and nutrients released by Typha during the incubation periods are

presented in Figure 3.


The main Typha degradation occurred during the first 3 days of incubation. Typha released organic matter in a continuous but minimal way from day 58 to the end of the experiment on day 262 of incubation. A higher production of TSS, TCOD, SCOD, and TP was generally observed in summer, with exception of BOD and TKN.

Average pH values during the degradation trials were 6.9 and 6.1; average dissolved oxygen values were 3.8 and 1.2 mg/L and average temperature values were 6.6 and 19.8ºC in winter and summer, respectively. The amount of organic matter and nutrients released by Typha is presented in Table 4.

Temperature affected Typha decomposition, since Typha removal was doubled in summer (26.9%) compared to in winter (13.8%) after 92 days of incubation. Typha removal after 262 days was slightly higher than after 92 days. This indicates that vegetation degradation mainly occurs during the first days of incubation and continually decreases during the next two months (Figure 3). Plant degradation after that time is almost negligible and 67.5% of the plant biomass will remain in the system after one year. BOD production per gram of plant was similar in summer and winter but TSS and TCOD production were 6.1 and 1.6 times higher in summer than in winter, respectively.

Table 3. Seasonal background concentrations in the wetlands studied. Observed C*: Non-harvested wetland effluent minus harvested wetland effluent. Calculated C*: as

indicated by Kadlec and Knight (1996). NA: Not applicable. a Calculated C* for TP was constant for all operation periods

Observed C* (mg/L) Calculated C* (mg/L) TSS BOD TKN TP TSS BOD TKN TPa Winter 9.1 13.7 2.5 0.6 13.1 5.7 2.42 0.05 Spring 25.3 30.8 2.7 0.8 15.0 6.9 NA 0.05 Summer -7.3 15.0 2.4 0.9 16.4 6.8 3 0.05 Total period 10.0 17.9 2.5 0.7 14.1 6.0 NA 0.05

Table 4. Typha removal and organic matter production in degradation trials using tanks

with freshwater and an initial biomass of 250 g of Typha (Ty) during 92 and 262 incubation days

Organic matter and nutrient production

Incubation days

Typha remained (g)

Typha removal (%)

mgTSS /gTy

mgBOD /gTy

mgTCOD /gTy

mgTKN /gTy

mgTP /gTy

Winter 92 215.5 13.8 11.9 44.0 70.6 2.3 0.9 Summer 92 182.8 26.9 72.7 41.4 109.5 4.7 1.5

Total period 262 168.8 32.5 33.7 72.6 97.9 3.5 0.9

Additionally, nutrient matter released by vegetation was also affected by temperature,

because TKN and TP production was approximately double in summer than in winter. A


better estimation of the organic matter and nutrient production by Typha in a whole year is obtained from the long-term degradation tank after 262 days. Considering the initial 250 g introduced in the tank, 169 g still remained after that period. Experimental conditions maintained in the tanks permitted the calculation of gross organic matter and nutrient production from Typha, an estimation that will be considered later in this chapter.

4. RESULTS OF DECOMPOSITION EXPERIMENTS Results from decomposition experiments are further detailed elsewhere (Alvarez and

Bécares 2006). Leaching was substantial during the first day following immersion, especially in control mesh tubes. Leaching weight loss was 15% and 12% in winter and summer, respectively. Differences in weight loss (% AFDW remaining) among the mesh sizes in winter showed that weight loss in the control tubes was higher and significantly different from that in all the meshed tubes (Tukey HSD test, p<0.03). No statistically significant differences were found among mesh sizes in winter. In the summer experiment, weight losses were not significantly different among the different mesh sizes. Control weight loss was not significantly different from that observed in meshed tubes, either. On the other hand, there was a clear seasonal difference in the timing of plant degradation. In winter, decomposition was not statistically significant until almost 60 days of incubation had passed. Weight loss at day 58 was different to that at day 1 (Tukey HSD test, p<0.05), and the weight at day 90 was significantly different to that at all previous sampling days (Tukey HSD test, p<0.0009), except to that at day 58. In summer, patterns were different and decomposition was significant at day 20 (Tukey HSD test, p<0.0009), decomposition was substantial from day 20 to day 58, and negligible between day 58 and day 90.

Decomposition rates were more than double in summer than in winter. There were significant differences between summer and winter for each of the mesh sizes (Tukey HSD tests, p<0.0002). However, mesh size overall had no significant effect on decomposition rates, as no differences were found among mesh sizes when data for each season were analysed separately.

Nitrogen percentage increased during the incubation period in all mesh sizes both in winter and in summer, these increases being significantly higher in summer than in winter (Tukey HSD test, p<0.02). Differences between summer and winter were greater in the 5 mm tubes and in controls, while no seasonal differences in N content were found for the rest of mesh sizes.

5. DISCUSSION

5.1. The Influence of Plants on the Wetland Environment Vegetation influenced pH and T variables of the wetland in winter and it influenced pH,

T and OD parameters during all operation periods. Presumably, higher organic matter in the non-harvested wetland was responsible for a higher oxidation activity, causing lower pH and lower dissolved oxygen in the vegetated wetland. Vegetation also maintained a higher


thermal inertia in comparison with the harvested wetland. Plants helped to maintain the lower temperatures reached during night (frequently below 0ºC during winter), which resulted in lower temperatures at the sampling time (noon) within the non-harvested wetland compared to the harvested wetland. Absence of plants in the harvested wetland allowed for higher temperatures at noon during higher light intensities in comparison with the vegetated wetland. This aspect confirms the important thermal inertia of vegetation (Brix, 1994), but in an opposite way.

5.2. Effect of Vegetation on the Performance of Constructed Wetlands The performance of cattail wetlands may be improved by harvesting the vegetation at the

end of the growing season, reducing in this way additional BOD and TSS inputs from decaying biomass litter. Karathanasis et al. (2003) observed that the decline in BOD removal efficiency in cattail systems was mainly in spring, probably due to the high biomass production during the previous season. This caused an increase in the effluent BOD in spring when temperature contributes to higher degradation of the decayed plants. Baptista et al. (2003) studied two laboratory-scale wetlands, with and without vegetation. In this case, they observed that carbon removal was higher in the unplanted system than in the planted system (63% versus 51%). In this work, BOD removal in the harvested and non-harvested wetland was 10.2% and –24.9%, respectively, for the entire operation period, and harvesting improved the removal of biodegradable organic matter in our conditions. The performance of Typha systems may be enhanced if most of the above ground biomass is harvested at the end of the growing season rather than allowed to return and decay within the wetland, which increases the organic and nutrient load (Karathanasis et al., 2003).

Concerning nutrient removal, Toet et al. (2005) and Vymazal (2007) showed that plant harvesting was a useful technique for N and P removal in constructed wetlands with N and P mass loading rates lower than approximately 100-200 gN/m2yr and 10-30 gP/m2yr. Yang et al. (2001) cited that the effect of plant litter was also a significant mechanism which affects nutrient removal in surface flow wetland systems without harvesting. Since the litter releases nutrients within the system, it was suggested that FWS systems should be harvested frequently to enhance their removal of nutrients. These conclusions are corroborated by our work where TKN, NH4, and TP concentrations were always lower in the harvested wetland effluent, indicating a nitrogen and phosphorus release by vegetation decomposition. Besides, there could be more nutrient removal in the harvested wetland due to increased light levels (Grimshaw et al., 1997).

5.3. Consequences of Vegetation Degradation Plants start their degradation process late in the summer, when they are still green, by

releasing dissolved substances (Kuehn and Suberkropp, 1998). The dry stems of Typha plants tend to remain above water level during winter and early spring and their degradation will be very slow. In spring, the stems of plants are immersed in the water resulting in a faster degradation with the increased temperature, and this produces a release of organic matter to the system. Part of this organic matter will be metabolised into the system and another part


will be released with the effluent. In summer, most of the easily biodegradable organic matter has already been degraded but a high amount of dead vegetation will remain in the system in the subsequent years. These decomposition rates will be very slow and the organic matter released cannot be practically measured.

Typha degradation tank experiments showed that after 262 days (winter, spring, and summer) 67.5% of the Typha initial mass still remained in the experimental tanks, whereas it was 73% and 86% after 92 days in winter and summer, respectively. In plant decomposition experiments, it was observed that the percentage of Typha remaining after 90 days of decomposition in the control litter-bag was 63.3%, the same percentage in winter and summer. This means that plant degradation in natural conditions was 10% to 20% higher than in the controlled conditions of the Typha decomposition tanks. These differences were due to the role of bacteria and microorganisms in natural conditions, while the meso and macrofauna had a negligible role in the studied wetland. The percentage of remaining Typha latifolia after 90 days in the constructed wetland of this study was in the range of previous studies in natural wetlands (Kirschner et al., 2001; Taylor and Middleton, 2004).

Laboratory experiments on Typha spp. degradation at different HRT (from 1.6 to 7.4 days) (Pinney et al., 2000) showed that at higher HRT, Typha degradation released organic carbon to the system. Although the dissolved organic carbon released by the plant is low, the high amount of vegetation in a wetland could have an important effect on the BOD of the final effluent (as described above). Another consideration to take into account in a wetland is the role of plankton inside the system. Luyiga and Kiwanuka (2003) observed that plankton increased dissolved oxygen levels and pH, improving BOD, and NH4 removal. They also observed that BOD removal was mainly carried out in the initial portion of the wetland (first eighth) whereas an increase in BOD as consequence of plant degradation and plankton activity was observed in the last part of the wetland. In our system, removal efficiency differences between harvested and non-harvested wetland were higher in spring than winter and summer, this being caused by the increase in vegetation decomposition during spring as temperatures increased.

5.4. Physical and Chemical Factors in Leaf Processing Leaching is the predominant process in the degradation of wetland plant parts during the

initial days following submersion, although leaching rate is dependent on plant species and on the natural breakdown of detritus before submersion (Barlocher, 1997; Menéndez et al., 2003). In this work, 15% and 12% of total weight losses in winter and summer, respectively were due to leaching during the first three days of incubation. Álvarez et al. (2001) found that leaching accounted for 5-9% of mass loss after 5 days depending on the treatment (250 mm and 5 mm mesh size, respectively) when looking at the decomposition of Juncus maritimus. Bedford (2005) found leaching accounted for a 15% of losses in Phragmites leaves. Webster and Benfield, (1986) found that a maximum of 25% of the initial dry weight of leaves may be lost due to leaching in the first 24 hours, depending on water temperature, water turbulence, and plant species.

Different mesh sizes relate to the biological degradation and fragmentation of plant matter by different saprophytic communities (Bradford et al., 2002). Here, we assumed that the 25 µm and 100 µm mesh sizes would only allow the colonization of pure microbial


communities (bacteria, fungi, protozoa and some metazoans) to the exclusion of other organisms, while the 1 mm mesh size would restrict the entrance of meso-fauna but allow the entrance of small metazoans (large rotifers, small nematoda), and the 5 mm mesh size would allow the entrance of meso-fauna (e.g. worms, slugs, chironomids). Therefore, differences among mesh sizes could give clues as to the role of the different saprophytic communities in the biological degradation of organic matter in the studied wetland. In our study, the absence of differences in decomposition rate among mesh sizes shows that, with the exception of microorganisms (bacteria and fungi), the role of other, larger organisms was insignificant in the degradation process of Typha in the studied wetland, both in winter and in summer. Physical and chemical factors, such as water temperature, and the mechanical breakdown of litter particles are mechanisms apparently driving for the most part the degradation of plant matter in our wetland. These factors are also probably responsible for the higher degradation rates observed in control tubes in winter, when there were significant differences with the meshed tubes. Coinciding with other authors (e.g. Braioni et al., 2001), the effect of physico-chemical factors was apparently more important than the effects of biological processes in the processing of plant matter in our wetland, at least during the first 20 days of the experiment.

5.5. Decomposition Rates Typha latifolia decomposition rates (kd) in different studies were compared in natural and

constructed wetlands (Table 5). Decomposition rates of other plants used in constructed wetlands were also reviewed and compared to Typha spp. values (Table 6). Main conditions of the experiments affecting decomposition rates were also included for comparison and correlated with decomposition data. Typha decomposition rates varied across sites operating under similar incubation conditions. Water physico-chemical features, hydrologic regime, micro- and macro-faunal diversity and nutrient availability are relevant factors affecting decomposition rates. Regression analysis showed that only temperature was marginally albeit significantly correlated to decomposition rate, kd (temperature=0.0011*exp (0.0820*kd)), r2= 0.298, p<0.005). Kadlec and Knight (1996) found Typha spp. litter decomposition rates in wetland ecosystems in the range from 0.0013 to 0.0031 d-1. Alvarez an Bécares (2006) found a mean summer value of kd of 0.0043 d-1 (see Table 5), similar to the mean estimate in the revision carried out by Nelson et al. (1990) and expanded by Vymazal (1995) (Typha spp. litter decomposition rates ranging from 0.001 to 0.0104 d-1, mean value of 0.0045 d-1), and by Chimney and Pietro (2006) (Typha spp. litter decomposition rate was 0.0072 d-1).

Typha decomposition rates have been shown to be lower than those in other aquatic macrophyte species. For instance, Álvarez et al. (2001) estimated Juncus spp decomposition rates in values ranging from 0.010 to 0.020 d-1, using 5 mm and 250 mm litter bag mesh sizes at temperatures of 15-25ºC in two lakes in the Doñana National Park (Southern Spain). Decomposition rates of Phragmites australis were similar or slightly greater than the rates for Typha latifolia rates obtained by Alvarez and Bécares (2006). Menéndez et al. (2005) estimated Phragmites australis decomposition rates at 0.0034 and 0.0042 d-1 using 0.1 and 1 mm litter bag mesh sizes, respectively, and at temperatures of 9-15ºC. Fonnesu (2005) reported Phragmites australis decomposition rates using 0.5 mm litter bag mesh sizes of 0.0106 d-1 and 0.0184 d-1, at 10º and 14ºC, respectively. Kirschner et al. (2001) estimated higher degradation rates for Phragmites australis leaves (0.0013 d-1) than for Typha latifolia


(0.00094 d-1). Respect to Carex spp., Kim et al. (2005) reported kd values of 0.0013 d-1 at 9.5ºC and 1 mm mesh size. In the case of Scirpus spp., Chimney and Prieto (2006) reviewed a kd value of 0.0029. In general, it was observed a great range of values as consequence of the influence of different parameters: mesh size, temperature and immersion period. The effect of this last parameter can be observed (Table 6) by the low kd estimated by Moore et al. (2007).

Table 5. A comparison of Typha decay rates (kd, exponential model) in the literature

(natural wetlands) and in this study (constructed wetland) under main conditions found in incubation experiments

kd (d-1) T (ºC) pH Mesh size (mm)

Incubation days Author

0.0098 16.7 7.5 5 140 Jaques and Pinto (1997) 0.0031 18.4 7.5 5 140 Jaques and Pinto (1997) 0.0009 12.5 - 1 840 Kirschner et al. (2001) 0.0100 16.9 7.9 1.3 60 Lee and Bukaveckas (2002) 0.0024 8 5.3 1 150 Taylor and Middleton (2004) 0.0078 26.1 - 3 146 Chimney and Pietro (2006) 0.0044 24.5 - 3 360 Chimney and Pietro (2006) 0.0005 6.9 - 2 1642 Moore et al. (2007) 0.0003 7.8 - 2 1934 Moore et al. (2007) 0.0020 4.9 8.2 1 90 Alvarez and Bécares (2006) 0.0043 19.9 7.4 1 90 Alvarez and Bécares (2006)

Table 6. A comparison of different plant decay rates (kd, exponential model) in the

literature (under main conditions found in incubation experiments)

Plant kd (d-1) T (ºC) pH Mesh size (mm)

Incubation days Author

0.0013 9.5 7.63 1 268 Kim et al. (2005)

0.0030 - - - - Chimney and Prieto (2006) (revision)

Carex spp.

0.0006 6.9 - 2 1642 Moore et al. (2007)

Scirpus spp. 0.0029 - - - - Chimney and Prieto (2006) (revision)

0.0021 - - - - Chimney and Prieto (2006) (revision)

0.0010 - - - 268 Kuehn et al. (2000) 0.0185 20 10.5 5 370 Alvarez et al. (2001)

Juncus spp.

0.0110 20 10.5 5 370 Alvarez et al. (2001) 0.0013 12.5 - 1 840 Kirschner et al. (2001) 0.0106 10 - 0.5 30 Fonnesu et al. (2004) 0.0184 14 - 0.5 30 Fonnesu et al. (2004) 0.0334 21.5 8.76 5 192 Bayo et al. (2005) 0.0171 20.9 8.53 5 192 Bayo et al. (2005)

Phragmites spp.

0.0034 15 7.7 0.1 102 Menéndez (2005) 0.0042 15 7.7 1 102 Menéndez (2005)

Compared to terrestrial plants, the aquatic emergent Typha shows degradation rates as

low as those found for tree leaves. For instance, Menéndez et al. (2003) reports rates for


Populus alba and Populus nigra from incubation in 100 mm and 5 mm litter bag mesh sizes at 10-20ºC of 0.003-0.038 d-1 and 0.006-0.037 d-1, respectively. Thus, Typha and Phragmites decomposition rates are generally lower than those of other aquatic macrophytes, such as hydrophytes (Kirschner et al., 2001), and of the same magnitude as those of tree leaves. Vegetation quality partially explains differences in decomposition rate found within and between plant species, so that leaves of different chemical composition or mechanical toughness decompose at different rates. The results above suggest Typha latifolia is one of toughest and most resistant emergent macrophyte in wetlands. The low Typha decay rate means detritus will remain in the system for a long time and could clog or deepen sediment layers in the wetland, an aspect that we will consider later.

5.6. Nitrogen Dynamics It has been shown that litter materials with a low nitrogen content decompose at a slower

rate and this has led some authors to contend that the C:N ratio of litter is a primary determinant of variation in decomposition rates (Enríquez et al., 1993; Lee and Bukaveckas, 2002). Aber and Melillo (1980) found that the dynamics of leaf decomposition could be described by an inverse linear relationship between the percentage of original mass remaining and N concentration in the residual material. In our study, winter rates do not fit well with this model [r2=0.18 (25 mm); 0.61 (100 mm); 0.50 (1 mm); 0.95 (5 mm) and 0.30 (control)], although summer rates fit much better [r2: 0.93 (25 mm); 0.87 (100 mm); 0.99 (1 mm); 0.70 (5 mm) and 0.81 (control)].

Summer rates could be related to the low initial N content of Typha leaves and high demand during litter decomposition (Menéndez et al., 2003). This hypothesis is supported by the observation that the C:N ratio of the detritus of Typha leaves decreased during the incubation period (data not shown). On the other hand, the increase in the percentage of nitrogen in the detritus, together with the decrease in dissolved oxygen concentrations and higher temperatures in summer, suggest an increase occurs in degradation due to microorganisms in the detritus leaves. This could explain why no significant differences were found in summer between control and meshed litter tubes.

5.7. Terrestrialization in the Studied Wetland Degradation rates and percentage remaining of vegetation biomass in the system after one

year are both good indicators of sedimentation rate, and therefore of the rate of terrestrialization in natural and constructed wetlands. The %AFDW remaining of Typha latifolia after 90 days in the constructed wetland in this study were in the range found in previous studies in natural wetlands. Taylor and Middleton (2004) obtained a value of 80% AFDW remaining after 150 days of incubation of Typha latifolia plant parts in litter bags with a 1 mm mesh size in a natural wetland at 8ºC. Lee and Buckaveckas (2002) found values for Typha of 42,4% and 28% AFDW remaining in autumn (17ºC) and in spring (21ºC), respectively, using a 1.3 mm mesh litter bag incubating 60 days in wetlands impacted by agricultural activities.


Kirschner et al. (2001) studied the terrestrialization process of an oxbow lake on the river Danube (Viena). Their experiments showed that about 45% of leaves and 80% of stems remained in the system after 27 months of degradation. The annual Typha production was 950 g/m2 and their equation of Typha degradation in a 1 mm mesh litter bag at 12.5ºC was: %AFDW (Typha)= 99.1*e-0.00094t. Using this equation, the %AFDW remaining after one year would be 70.32 %, which means 668 g/m2 of Typha accumulating in that wetland after one year. In our constructed wetland, the annual Typha production was 3007.7±1120.2 g/m2 (mean ± SD) The equation of Typha degradation in the winter control tubes is: %AFDW(Typha)= 81.14*e-0.0026t. Considering one whole year (365 days), the %AFDW remaining would be 31.14%. From Typha biomass production, the remaining Typha biomass accumulating in the system after one year would be 932 g/m2. From the summer (20ºC) equation for a 1 mm mesh litter tube, the remaining Typha accumulating after one year would be 547 g/m2. Under the assumption of a uniform distribution of the plants over the whole wetland, a dry to fresh mass mean ratio of 0.145 and an average bulk density of 1.097 g/cm3 of the sediment (Kirschner et al. 2001), the thickness of the sediment layer would increase from 0.3 to 0.6 cm per year. Considering the common 25-30 cm depth of the water layer in this type of constructed wetlands, it is clear that the time expected for the complete terrestrialization of the wetland would exceed the theoretical operation time of the plant.

CONCLUSION The results indicated in this chapter confirm previous studies (e.g. Soto et al., 1999) in

that, when treating diluted wastewater (TCOD <200mg/L), vegetation has a significant role in pollution removal, and in the supply of autochthonous organic matter to the system. The present study demonstrates that vegetation harvesting has a significant effect on the pH, dissolved oxygen, and temperature inside the wetland. Moreover, in comparison with the non-harvested wetland, absence of vegetation in the harvested wetland allowed the reduction of TSS and BOD effluent concentration in spring by 37.3% and 49.2%, respectively. Harvesting can be a recommended operation and management strategy under the climatic and wastewater conditions of the study. (i.e., Mediterranean climate and diluted wastewater).

Typha latifolia degradation rates estimated using the litter bag technique was mainly affected by temperature. The effects of microorganisms (bacteria, fungi) were much more important than the meso- and macro-fauna, whose role was negligible under our experimental conditions. Estimated decomposition rates were 0.0014-0.0026 d-1 and 0.0043-0.0052 d-1 in winter and summer, respectively. These estimates are comparable to other rates given in the literature, which estimate the overall annual degradation rate for Typha at 0,0044 d-1. Compared to other plant species, Typha latifolia is one of toughest, most resistant emergent macrophytes in wetlands. Typha spp. harvesting is recommended in constructed wetlands treating diluted wastewater or in highly productive areas, because detritus will remain in the system a long time, increasing background organic matter levels in the system. Nevertheless, the effect of vegetation necromass on the burial or terrestrialization process of the wetland is negligible when considering the operation time of a CW.


REFERENCES

Aber, J.D., and Melillo, J.M., (1980). Litter decomposition: measuring state of decay and percent transfer into forest soils. Canadian Journal Botany, 58, 416-421.

Akratos, C.S., and Tsihrintzis, V.A. (2007). Effect of temperature, HRT, vegetation and porous media on removal efficiency of pilot-scale horizontal subsurface flow constructed wetlands. Ecological Engineering 29 (2), pp. 173-191.

Álvarez, S., Rico, E., Guerrero, M.C., and Montes, C., (2001). Decomposition of Juncus maritimus in two shallow lakes of Doñana national park. International Review Hydrobiology, 86 (4-5), 541-554.

Álvarez, J.A., Ruíz, I., and Soto, M. (2008). Anaerobic digesters as a pretreatment for constructed wetlands. Ecological Engineering, doi:10.1016/jecoleng.2008.02.001.

Álvarez, J.A., and Bécares, E., (2006). Seasonal decomposition of Typha latifolia in a free-water surface constructed wetland. Ecological Engineering , 28, 99-105.

Anderson, N.H., and Sedell, J.R. (1979). Detritus processing by macroinvertebrates in stream ecosystems. Annual Review Entomology, 24, 351-357.

Ansola, G., De Luis, E. and Fernández, C. (1993). Evaluation of the treatment capacity of an emergent macrophyte experimental plant for low density populations in the initial working stages. Proceed. 2nd. Internat. Conf. Small Wastewater Treatment Plants. Trondheim, Norway: 367-374.

Ansola, G. (1994). Sistemas de experimentación con macrófitos como mecanismo de depuración de la contaminación acuática y evaluación de los procesos de transformación en ambientes rurales. Ph.D. Thesis. University of León, León, Spain

Ansola, G., Fernández, C and De Luis, E. (1995). Removal of organic matter and nutrients from urban wastewater by using experimental emergent aquatic macrophyte system. Ecological Engineering, 5: 13-19.

Armstrong, J., and Armstrong, W. (1988). Phragmites australis — a preliminary study of soil-oxidising sites and internal gas transport pathways. New Phytology, 108, 372–382.

Barber, D. A., and Martin, J. K. (1976). The release of organic substances by cereal roots into soil. New Phytology, 76, 69–80.

Baptista, J.D.C., Donenelly, T., Raine, D., and Davenport, R.J. (2003). Microbial mechanisms of carbon removal subsurface flow wetlands. Water Science Technololgy. 48, 127-134.

Barko, J. W, Gunnison, D, and Carpenter, S. R. (1991). Sediment interactions with submerged macrophyte growth and community dynamics. Aquatic Botany, 41, 41–65.

Bärlocher, F., (1997). Pitfalls of traditional techniques when studying decomposition of vascular plant remains in aquatic habitants. Limnetica, 13, 1-11.

Bayo, M.M., Casas, J.J., and Cruz-Pizarro, L. (2005). Decomposition of submerged Phragmites australius leaf litter in two highly eutrophic Mediterranean coastal lagoons: relative contribution of microbial respiration and macroinvertebrate feeding. Arch. Hydrobiology, 163 (3), 349-367.

Bécares, E. (2006). Limnology of natural systems for wastewater treatment. Ten years of experiences at the experimental field for low-cost sanitation in Mansilla de las Mulas (León, Spain). Limnetica, 25 (1-2), pp. 143-154.

Bedford, A. P., (2005). Decomposition of Phragmites australis litter in seasonally flooded and exposed areas of a managed reedbed. Wetlands, 25 (3), 713-720.


Bradford, M.A., Tordoff, G.M., Eggers, T., Jones, T.H., and Newington, J.E. (2002). Microbiotic, fauna, and mesh size interactions in litter decomposition. Oikos, 99, 317-323.

Braioni, M.G., Gumiero, B., and Salmoiraghi, G., (2001). Leaf bags and natural leaf packs: Two approaches to evaluate river functional characteristics. International Review Hydrobiology, 86 (4-5), 439-451.

Brix, H. and Schierup, H. H. (1990). Soil oxygenation in constructed reed beds in Constructed wetlands in water pollution control. In: Cooper, P. F., and Findlater B C (eds.), Pergamon Press, Oxford, UK, pp 53–65.

Brix, H. (1994). Use of constructed wetlands in water pollution control: Historical development, present status and future perspectives. Water Science Technology, 30 (8), 209-223.

Brix, H. (1995). Use of subsurface flow constructed wetlands for wastewater treatment - an overview. In: Natural and constructed wetlands for wastewater treatment and reuse - Experiences, goals and limits. R. Ramadori, L. Cingolani and L. Cameroni (eds.). Perugia; pp 103-111.

Brix, H. (1997). Do macrophytes play a role in constructed treatment wetlands?. Water Science Technology, 35 (5), 11-17.

Chauvet, E. (1997). Leaf litter decomposition in large rivers: the case of the river Garonne. Limnetica, 13, 65-70.

Chimney, M.J., and Pietro, K.C. (2006). Decomposition of macrophyte litter in a subtropical constructed wetland in south Florida (USA). Ecological Engineering, 27 (4), 301-321.

Coleman, J., Hench, K., Garbutt, K., Sexstone, A., Bissonnette, G., and Skousen, J. (2001). Treatment of domestic wastewater by three plant species in constructed wetlands. Water, Air, and Soil Pollution, 128 (3-4), 283-295.

Conely, L. M., Dick ,R. I., and Lion, L. W. (1991) An assessment of the root zone method of wastewater treatment Research Journal WPCF, 63 (6):239–247.

Cooper, P.F., Job, G.D., Green, M.B., and Shutes, R.B.E. (1996). Reed Beds and Constructed Wetlands for Wastewater Treatment WRc , Swindon, UK.

Cummins, K.W., 1988. The study of stream ecosystems: a functional view. In: Concepts of ecosystem Ecology. Pomeroy, L.R., and Alberts, J. (eds),. Springer-Verlag, New York, USA, pp 247.

Decamp, O. and Warren, A. (1998). Bacterivory in ciliates isolated from constructed wetlands (reed beds) used for wastewater treatment. Water Research, 32, 1989-1996.

Decamp, O., Warren, A., and Sánchez, R. (1999). The role of ciliated protozoa in subsurface flow wetlands and their potential bioindicators. Water Science Technology, 40, 91-98.

De Feo, G. (2007). Performance of vegetated and non-vegetated vertical flow reed beds in the treatment of diluted leachate. Journal of Environmental Science and Health - Part A Toxic/Hazardous Substances and Environmental Engineering, 42 (7), 1013-1020.

Dellagreca, M., Fiorentini, A., Isidori, M., and Zarrelli, A. (2001). Toxicity evaluation of natural and synthetic phenantrenes in aquatic systems. Environmental Toxicololy Chemical, 20, 1824-1830.

Enríquez, S., Duarte, C.M., and Sand-Jensen, K. (1993). Patterns in decomposition rates among photosynthetic organisms: the importance of detritus C:N:P content. Oecologica, 94, 457-471.


Farahbakhshazad, N., Morrison, G.M., Larsson, A., and Weisner, S.B.E. (1995). Effects of grain size on nutrient removal from wastewater in small-scale planted macrophyte system. In: Natural and Constructed wetlands for Wastewater Treatments and Reuse. Ramadori, R., Cingolani and L. Cameroni, L. (eds.). Perugia, pp 143-150.

Faulkner, S. P., and Richardson, C. J. (1989). Physical and chemical characteristics of freshwater wetland soil. In: Constructed wetlands for wastewater treatment. Municipal, industrial and agricultural. Hammer, D.A. (eds.), Lewis Publishers, Chelsea, UK, pp 41-73.

Fonnesu, A., Pinna, M., and Basset, A. (2004). Spatial and temporal variations of detritus breakdown rates in the river Flumendosa basin (Sardinia, Italy). International Review Hydrobiology, 89 (5-6), 443-452.

García, M., and Bécares, E. (1997). Bacterial removal in three pilot-scale wastewater treatment systems for rural areas. Water Science Technology, 35, 197-200.

García, M., Bécares, E., Soto, F., and De Luis, E.. (1999). Papel de los macrófitos en el tratamiento de las aguas residuales. I: Eliminación de bacterias. Tecnología del Agua, 185, 168-72.

García, M., Bécares, E., and Soto, F. (2004). Are bacterial removal efficiencies enhanced by plants? An experimental study using Scirpus lacustris. Proceed. 9th Internat. Conf. Constructed Wetlands. Avignon, France, 4-12.

Gersberg, R.M., Elkins, B.V., Lyon, S.R., and Goldman, C.R. (1986). Role of aquatic plants in wastewater treatment by artificial wetlands. Water Research, 20, 363:368.

Gersberg, R.M., Gearheart, R.A., and Ives, M. (1989). Pathogen removal in constructed wetlands. In: Constructed wetlands for wastewater treatment: Municipal, Industrial and agricultural. Hammer, D.A. (eds.), Lewis Publ. Chelsea, UK, pp 431-445.

Gersberg, R.M., Lyon, S.R., Brenner, R., and Elkins, B.V. (1990). Integrated wastewater treatment using artificial wetlands: A gravel marsh case study. In: Constructed wetlands for wastewater treatment. Hammer, D.A. (eds.), Lewis Publications 145-152.

Gimshaw, H.J., Wetzel, R.G., Brandenburg, M., Segerblom, K., Wenkert, L.J., Marsh, G.A., Charnetzky, W., Haky, J.E., and Carraher, C. (1997). Shading of periphyton communities by wetland emergent macrophytes: Decoupling of algal photosynthesis from microbial nutrient retention. Arch. Hydrobiology, 139, 17-27.

Gopal, B., and Goel, U. (1993). Competition and allelopathy in aquatic plant communities. Botanical Review, 59, 155-210.

Gries, C., Kappen, L., and Losch, R. (1990). Mechanisms of flood tolerance in reed Phragmitis australis, New Phytology, 114, 589–593.

Greenway, M., and Woolley, A., (1999). Constructed wetlands in Queensland: perfomance efficiency and nutrient bioaccumulation. Ecological Engineering, 12, 39–55.

Gumbricht, T. (1993) Nutrient removal processes in freshwater submerged macrophyte systems. Ecological Engineering, 2, 1–30.

Haberl, R., and Perfler, R. (1990). Seven years of research work and experience with wastewater treatment by reed bed system in Constructed wetlands in water pollution control. In: Cooper, P. F., and Findlater, B. C. (eds), Pergamon Press, Oxford, UK, pp 205–214.

Hammer, D. A. (1989). Constructed wetlands for wastewater treatment, municipal, industrial, and agricultural. Lewis Publishers, Chelsea, UK, pp 831.


Henze, M., Harremöes P., La Cour Jansen, J., and Arvin, E. (1997). Wastewater treatment: biological and chemical processes. Heildelberg: Springer.

Hatano, K., Trettin, C.C., House, C.H., and Wollum, A.G. (1993). Microbial populations and decomposition activity in three subsurface flow constructed wetlands. In: Constructed wetlands for water quality improvement. Moshiri, G. A. (eds.), Lewis Publications, pp 541-548.

He, Q., and Mankin, K.R. (2002). Performance variations of COD and nitrogen removal by vegetated submerged bed wetlands. Journal of the American Water Resources Association, 38 (6), 1679-1689.

Hobbie, S.E. (1996). Temperature and plant species control over litter decomposition in Alaskan tundra. Ecolologica. Monographs, 66, 503-522.

Howard, C.P., and Howard, D.M. (1974). Microbial decomposition of tree and shrub leaf litter. I. Weight loss and chemical composition of decomposition litter. Oikos, 25, 341-352.

Huang, J., Reneau, R., and Hageborn, C. (2000). Nitrogen removal in constructed wetlands employed to treat domestic wastewater. Water Research, 34, 2582–2588.

Jaques, N., and Pinto, P (1997). Seasonal differences in the decomposition of Typha angustifolia leaves in a mediterranean river. Limnetica, 13 (2), 19-23.

Kadlec, R.H., and Knight, R.L. (1996). Treatment Wetlands. CRC press, Boca Raton, USA, pp 151.

Kadlec, R. H., Knight, R.L., Vymazal, J., Brix, H., Cooper, P., and Haberl, R. (2000). Constructed wetlands for pollution control. Scientific and Technical Report 8. IWA Publishing, London, UK, pp 156.

Karathanasis, A.D., Potter, C.L., and Coyne, M.S. (2003). Vegetation effects on fecal bacteria, BOD, and suspended solid removal in constructed wetlands treating domestic wastewater. Ecological Engineering, 20, 157-165.

Kim, S.Y., and Geary, P.M. (2001). The impact of biomass harvesting on phosphorus uptake by wetland plants. Water Science Technology,44, 61-67.

Kim, J.G., and Rejmánková, E. (2005). Decomposition of macrophytes and dynamics of enzyme activities in subalpine marshes in Lake Tahoe basin, U.S.A. Plant and Soil, 266 (1-2), pp. 303-313.

Kirschner, A.K.T., Riegl, B., and Velimirov, B. (2001). Degradation of emergent and submerged macrophytes in an oxbow lake of an embanked backwater system: Implications for the terrestrialization process. International Review Hydrobiology, 86 (4-5), 555-571.

Kuehn, K.A., and Suberkropp, K. (1998). Decomposition of standing litter of the freswater emergent macrophyte Juncus effusus. Freshwater Biology, 40, 717-727.

Kuehn, K.A., Gessner, M.O., Wetzel, R.G., and Suberkropp, K. (1999). Decomposition and CO2 evolution from standing litter of the emergent macrophyte Erianthus giganteus. Microbial Ecology, 38, 50-57.

Kuehn, K.A., Lemke, M.J., Suberkropp, K., and Wetzel, R.G. (2000). Microbial biomass and production associated with decaying leaf litter of the emergent macrophyte Juncus effusus. Limnology and Oceanography, 45 (4), 862-870.

Lawson, G.J. (1985). Cultivating reeds (Phragmites australis) for root zone treatment of sewage. Water Research Centre, Cumbria, UK.


Lee, A.A., and Bukaveckas, P.A. (2002). Surface water nutrient concentrations and litter decomposition rates in wetlands impacted by agriculture and mining activities. Aquatic Botany, 74, 273-285.

López, M.J., and Bécares, E. (1993). Primeros datos sobre la eliminación de bacterias en sistemas experimentales de depuración por macrófitos. Proceed. III Congreso de Ingeniería Ambiental. Vol II, pp 196-200.

Loveridge, R.F., Willians, J.M., Hughes, C.M., El-shatoury, S.,Mitchell, J., May, E., and Butler, J.E. (1995). Changes in biofilm composition an the roots of Phragmites australis in gravel-based constructed wetlands. Proceed. Natural and constructed wetlands for wastewater treatment and reuse - Experiences, goals and limits. R. Ramadori, L. Cingolani, L. Cameroni (eds.), Perugia, pp 171-780.

Luyiga, S., and Kiwanuka, S. (2003). Plankton composition, distribution and significance in a tropical integrated constructed treatment wetland in Uganda. Water Science Technology, 48, 241-249.

Marsteiner, E.L., Collins, A.G., Theis, T.L, and Young, T.C. (1996) The influence of macrophytes on subsurface flow wetland hydraulics. Proceed. of the 5th international conference on wetland systems for water pollution control, University of Bodenkultur Wien, Vienna, Austria, pp. 1–7.

Mbuligwe, S.E. (2004). Comparative effectiveness of engineered wetland systems in the treatment of anaerobically pre-treated domestic wastewater. Ecolologycal Engineering, 23, 269.

Megonigal, J.P., Faulkner, S.P., and Patrick, W.H. (1996). The microbial activity season in southeastern hydric soils. Soil Science Society American Journal, 60, 1263-1266.

Menéndez, M, Martínez, M., Hernández, O., and Comín F. (2001). Comparasion of leaf decomposition in two mediterranean rivers: a large eutrophic river and an oligotrophic stream (S Catalonia, NE Spain)”. International Review Hydrobiology, 86 (4-5), 475-486.

Menéndez, M, Hernández, O., and Comín F. (2003). Seasonal comparasions of leaf processing rates in two mediterranean rivers with different nutrient availability. Hydrobiology, 495, 159-169.

Menéndez, M. (2005). Decomposition of the common reed Phragmites australis in a Mediterranean stream pond. Arch. Hidrobiology, 163 (1), 101-115.

Mitsch, W.J., and Gosselink, J.G. (1993). Wetlands. Van Nostrand Reinhold 2nd ed, New York,, USA.

Moore, T.R., Bubier, J.L., and Bledzki, L. (2007). Litter decomposition in temperate peatland ecosystems: The effect of substrate and site. Ecosystems, 10 (6), 949-963.

Moshiri, G.A. (1993). Constructed wetlands for water quality improvement. Lewis Publ. Chelsea, UK, pp 632.

Murphy, K.L., Klopatek, J.M., and Klopatek, C.C. (1998). The effects of litter quality and climate on decomposition along an elevational gradient. Ecological Applications, 8, 1061-1071.

Nelson J.W., Kadlec J.A., and Murkin H.R. (1990). Seasonal comparisons of weight loss of two types of Typha glauca Godr. Leaf litter. Aquatic Botany, 37, 299-314.

Olson, J.S. (1963). Energy storage and the balance of producers and decomposers in ecological systems. Ecological Engineering, 44, 322-330.

Ottova, V., Balcarova, J. and Vymazal, J. (1997). Microbial characteristics of constructed Wetlands. Water Science Techonoly, 30, 117-124.


Panswar, T., and Chavalparit, O. (1987). Water quality and occurrences of protozoa and metazoa in two constructed wetlands treating different wastewaters in thlailand. Water Science Techonology, 36 (12), 183-188.

Platzer, C. (1996). Enhanced nitrogen elimination in subsurface flow artificial wetlands – a multistage concept. Proceed. of the 5th international conference on wetland systems for water pollution control, University of Bodenkultur, Vienna, Austria, pp11–17.

Pinney, M.L., Westerhoff, P.K., and Baker, L. (2000). Transformations in dissolved organic carbon through constructed wetlands. Water Research, 34, 1897.

Reed S.R., Crites, R.W. and Middlecrooks, J.E. (1995). Natural systems for waste management and treatment. McGraw-Hill, New York, USA, pp 270.

Richardson, C.J. (1985). Mechanisms controlling phosphorous retention capacity in freshwater wetlands. Science, 228, 1424.

Rivera, F., Warren, A, Ramirez, E., Decamp, O., Bonilla, P., Gallegos, E., Calderón, A., and Sánchez, J.T. (1995). Removal of pathogens from wastewaters by the root zone method (R.Z.M.). Water Science Techonology., 32 (3): 211-218.

Rogers, K.H., Breen, P.F., and Chick, A.J. (1991). Nitrogen removal in experimental wetland treatment systems: evidence for the role of aquatic plants. Journal Water Pollution Control Federation, 63: 934-941.

Rovira A.D. (1969). Plant root exudates. Botanical Review, 35, 35–57. Taylor, J., and Middleton, B.A. (2004). Comparison of litter decomposition in a natural

versus coal-slurry pond reclaimed as a wetland. Land Degradation Development 15, 439-446.

Scholz, M., and Lee, B.H. (2005). Constructed wetlands: A review. International Journal of Environmental Studies, 62 (4), 421-447.

Seidel, K. (1955). Die flechtbinse Schoenoplectus lacustris, Okologie, Morphologie und Entwirkung Bedeutung. PhD. Thesis Dissertation. Max Plank Institute, Germany.

Shaver, G.R., and Mellio, J.M. (1984). Nutrient budget of marsh plants: efficiency concepts and relation to availability. Ecology, 65, 1491–1510.

Smith, I. D., Bis, G.N., Lemon E.R., and Rozema, L.R. (1997). A thermal analysis of a sub-surface, vertical flow constructed wetland. Water Science Technology, 35 (5): 55–62.

Sorrel, B.K., and Boon, P.I. (1992). Biogeochemistry of billabond sediments — seasonal variations in methane production. Freshwater Biology, 27, 435–445.

Soto, F., Garcia, M., De Luis, E., and Bécares, E. (1999). Role of Scirpus lacustris in bacterial removal from wastewater. Water Science Techonology, 40, 241-247.

Soto, F., Bécares, García, M.., and De Luis, E. (1999b). Papel de los macrófitos en el tratamiento de las aguas residuales. II: Eliminación de nutrientes. Tecnología del Agua, 185, 64-67.

Soto, F., García, M., and Bécares, E. (2000). Seasonal differences in removal efficiencies using S. lacustris for wastewater treatment. Proceed. of 7th Internat. Conf. Wetland Systems Water Pollution Control.

Standard Methods for the Examination of Water and Wastewater, 1995. 19th. Edition, American Public Health Association/American Water Works Association/ Water Environment Federation, Washington D.C., USA.

Statsoft, Inc. 1995. Statistica 6.0, Tulsa (OK), USA


Stengel, E. (1985). Perspektiven der nitratelimination in küstlichen Feuchtgebieten. Grundlagen und praxis naturnaher Klärverfahren. Sammelband des symposiums vom. Liebenburg. Verlagsgruppe Witzenhausen, 173.

Steer, D., Fraser, L., Boddy, J., and Seibert, B. (2002). Efficiency of small constructed wetlands for subsurface treatment of single-family domestic effluent. Ecol. Eng. 18, 429–440.

Sundaravadivel, M., Vigneswaran, S. 2001. Constructed wetlands for wastewater treatment. Critical Reviews in Environmental Science and Technology 31 (4), pp. 351-409

Tanner C C (1996) Plants for constructed wetland treatment systems: a comparison of the growth and nutrient uptake of eight emergent species. Ecological Engineering 7(1): 59–83.

Tanner, C.C., and Sukias, J.P. (1995). Accumulation of organic solids in gravel-bed constructed wetlands. Water Science Techonology, 32, 229-239.

Tanner, C.C., Clayton, J.S. and Upsdell, M.P. (1995). Effect of loading rate and planting on treatment of dairy farm wastewaters in constructed wetlands-I. Removal of oxigen demand, suspended solid and faecal coliforms. Water Research, 29 (1), 17-26.

Taylor, J., and Middleton, B.A. (2004). Comparison of litter decomposition in a natural versus coal-slurry pond reclaimed as a wetland. Land Degradation Development, 15, 439.

Toet, S., Bouwan, M., Cevaal, A., and Verhoeven, J.T.A. (2005). Nutrient removal through autumn harvest of Phragmites australis and Typha latifolia shoots in relation to nutrient loading in a wetland system used for polishing sewage treatment plant effluent. J. Environmental Science. Health, Part A: Toxic/Hazardous Subs. Environmental. Engineering, 40, 1133.

Vymazal, J., (1995). Algae and element cycling in wetlands. Lewis publishers, Boca Raton. Vimazal, J., Brix, H., Cooper, P.F., Green, M.B., and Haberl, R. (1998). Constructed

wetlands for wastewater treatment in Europe. Backhuys Pub., Leiden. UK, pp 264. Vymazal, J. (2002). The use of sub-surface constructed wetlands for wastewater treatment in

the Czech Republic: 10 years experience. Ecological Engineering, 18, 633–646. Vymazal, J. (2007). Removal of nutrients in various types of constructed wetlands. Science of

the Total Environment, 380 (1-3), pp. 48-65. Wang, N., and Mitsch, W.J. (2000). An detailed ecosystem model of phosphorus dynamics in

created riparian wetlands. Ecological Modelling, 126, 101. Webster, J R., and Benfield, E.F. (1986). Vascular plant breakdown in freshwater ecosystems.

Annual Review Ecology Systematics, 17, 567-594. Wetzel, R.G. (1990). Land-water interfaces: metabolic and limnological regulators. Verh.

Inernational Limnology, 24, 6-24. Wetzel, R.G. (2001). Limnology. Academic Press, San Diego, USA. Williams, J., Bahgat, M., May, E., Ford, M., and Butler, J. (1995). Mineralisation and

pathogen removal in gravel bed hydroponic constructed wetlands for wastewater treatment. Water Science Technology, 32, 49-58.

Xie, Y., Dan, Y., and Ren, B. (2004). Effects of nitrogen and phosphorus availability on the decomposition of aquatic plants. Aquatic Botany, 80, 29-37.

Yang, L., Chang, H., and Huang M. (2001). Nutrient removal in gravel- and soil-based wetland microcosms with and without vegetation. Ecological Engineering, 18, 91.


Young, R., White, G., Brown, M., Burton, J., and Atkins, B. (1998). The constructed wetlands manual. Department of Land and Water Conservation, New South Wales, Australia.

Yuan, D., Ren, Q., Gao, S., Zhang, H., Yin, D., and Wang, L. (2004). Purification efficiency of several wetland macrophytes on COD and nitrogen removal from domestic sewage. Chinese Journal of Applied Ecology, 15 (12), 2337-2341.

Zhang, H.-G., and Hong, J.-M. (2006). Functions of plants of constructed wetlands. Wetland Science, 4 (2), 146-154.


Chapter 7

NUTRITION AND TOXICITY OF INORGANIC SUBSTANCES FROM WASTEWATER IN CONSTRUCTED

WETLANDS

Zhenhua Zhang1, Zed Rengel1 and Kathy Meney2 1Soil Science and Plant Nutrition, School of Earth and Geographic Sciences,

The University of Western Australia, 35 Stirling Highway, Crawley WA 6009, Australia

2Syrinx Environmental Pty Ltd, 12 Monger St., Perth WA 6000, Australia

ABSTRACT

The use of constructed wetlands for purification of wastewater has received increasing attention around the world. A variety of wetland plant species (including ornamental ones) as either a monoculture or species mixes are used in constructed wetlands. Plants play an extremely important role in removing pollutants from wastewater. Although there is considerable information on plant productivity, biomass and nutrient dynamics in natural and fertilized wetlands, most studies on constructed wetlands for treatment of wastewaters have only addressed general aspects of plant growth and nutrient accumulation. Nutrition and toxicity of inorganic substances such as nitrogen, sulphur, salts and metals in wastewater on wetland plants has not been fully investigated and their interactive effects and environmental cycling in constructed wetlands remain poorly understood.

Nitrogen nutrition is the most important factor influencing plant performance in constructed wetlands, but higher NH4-N may become toxic to wetland plants. Sulphur is an essential nutrient for plant growth, but under waterlogged conditions sulphate is reduced to hydrogen sulphide that is highly toxic to wetland plants. Many metals in wastewater are essential micronutrients for wetland plants, but become toxic if their concentration exceeds a specific critical point. A proper amount of salts is essential for plant growth, but high concentrations of salts, particularly sodium chloride in wastewater have harmful effects on plant growth.

Wetland plant species have differential capacity to take up nutrients, different preference for nitrogen forms and have evolved various adaptive mechanisms protecting

Zhenhua Zhang, Zed Rengel and Kathy Meney 248

them against toxicity of inorganic substances. Given that plants are an integral part of constructed wetlands, the selection of suitable species, improvement of cultivations and determination of factors affecting growth are needed to produce healthy and effective wetland ecosystems. Understanding biogeochemical cycling in wetlands as well as nutrition and toxicity of inorganic substances from wastewater on plant development and function may help reduce performance variability and enhance pollutant removal in constructed wetlands.

INTRODUCTION During the past decades, the world has experienced growing water stress in terms of both

water scarcity and quality deterioration, which prompted the governments to enforce tough legislations on wastewater discharge and encourage efficient use of water resources, including widespread promotion of water reuse practices.

Treatment technology for wastewater encompasses a vast variety of options. Among them, constructed wetlands are considered a sustainable, low-investment and low-maintenance technology that can complement or replace conventional wastewater treatments. The use of constructed wetlands to treat wastewater from agricultural, mining, municipal and industrial sources etc has undergone dramatic development since the 1990s (Kadlec and Knight, 1996; Sundaravadivel and Vigneswaran, 2001; Scholz and Lee, 2005).

Wastewater quality varies widely among municipal, industrial, agricultural and stormwater categories. Different wastewater sources have unique mixtures of potential pollutants so that even a single wastewater source category such as municipal wastewater or urban runoff may vary depending on local, site-specific circumstance (Kadlec and Knight, 1996). Therefore, constructed wetlands receiving wastewaters are exposed to a wide range of pollutants with varying loads. Contaminants include major elements such as nitrogen (N) and phosphorus (P), mineral oils, pathogens, trace contaminants such as pesticides, heavy metals, radionuclides, and emerging pollutants such as brominated flame retardants, oestrogenic compounds, etc. The contaminants are transformed and distributed in multiple abiotic and biotic compartments, and may cause diverse ecological and ecotoxicological effects (Tack et al., 2007). However, the fate and behaviour of the wide variety of contaminants that enter constructed wetlands are not fully understood.

Wetland plants are often central to wastewater treatment in constructed wetlands (Scholz and Lee, 2005), in addition to the other design factors such as hydraulics, choice of substrate, etc. Macrophytes are assumed to be the main biological component of wetlands. They not only assimilate pollutants directly into their tissues from wastewater and substrates, but also act as catalysts for purification reactions by increasing the environmental diversity in the rhizosphere, and promoting a variety of chemical and biological reactions that enhance pollutant removal (Jenssen et al., 1993; Brix, 1997). However, in some cases the vegetation has either failed completely or proved difficult to establish in wetlands (Batty and Younger, 2004). The reasons for these problems are not thoroughly understood.

Wetland plants require optimum environmental conditions in each phase of their life cycles, including germination and initial plant growth, seasonal growth and senescence and decay. Wetland plants tolerate a wide range of water quality, but do have limits outside which they can not survive (Kadlec, 1999). The toxic inorganic substances such as ammonium,

Nutrition and Toxicity of Inorganic Substances from Wastewater… 249

sulphides, salts and metals in wastewater may inhibit nutrient uptake and plant growth if present in concentrations exceeding those that wetland plants can tolerate.

The function of vegetation in constructed wetlands is a dynamic one, and requires a better understanding of plant nutrient requirements as well as plant tolerance to pollutants to optimise performance in terms of substance removal in constructed wetlands. The objective of this review is to increase the understanding of nutritional value as well as toxicity of inorganic substances in wastewater to wetland plant growth in constructed wetlands. Even though the plant system has a large bearing on microbial activity in constructed wetlands, there may be direct toxicities of inorganic substances to microbes, thus harming the wetland functions. Further limitations to growth of

Science

Wetlands ecology, conservation and restoration

Text of Wetlands ecology, conservation and restoration