Statistical and hydrological modeling of soil and subsoil salt-accumulation caused by tree plantations established above shallow saline groundwater

Introduction

In Hungary there was a great increase in the acreage of afforested areas during the last hundred years, from 1.1 to 1.8 million ha. Most of these are plantations were established on previous grassland/cropland, ca 60%. European Union incentives favor further afforestations, and there are ca 15 000 hectares being afforested on earlier croplands each year. For the coming 30 years some 700 000 ha more afforestation is planned (Andrasevits et al., 2005). Although it has been demonstrated that trees cause subsurface salt accumulation above shallow saline groundwater in areas with negative water balance (Bazykina 2000, Nosetto et al., 2007, 2008), when the decision on afforestation of a particular plot is made there is no allowance made for the salinity of groundwater, therefore a future risk is incorporated in every such case. The area suggested for afforestation by the National Forest Strategy, 2009 covers areas where other agricultural activity is not profitable, and this area is overlapping with the area of saline shallow groundwaters as shown by Map 1. and there is a considerable risk that undesired, deleterious salt accumulation will occur in newly afforested areas.

Map 1. Total soluble salt concentration of groundwater in the Great Hungarian Plain

The model of the study

The hydrological effect of trees is much different from the crops or grasses, being characteristic for the earlier land-use. Due to their deep roots, trees extract water from much deeper layers and there is an increased water absorption and transpiration. As Fig 1 from Jobbágy et Jackson, 2004 shows there is a tendency of salt accumulation under the established tree stand. Where the

conditions are favorable a horizontal groundwater flux will accelerate the rate of salt accumulation.

Figure 1. The scheme of salt accumulation above a shallow groundwater as described by Jobbágy and Jackson, 2004

With our project our intention is to quantify the risk and provide guidelines for the optimal selection of plots and trees in order to reduce the risk of subsurface salt accumulation in Argentine and Hungary with the same methodology and evaluation techniques. The model of the study is shown in Fig 2 of Nosetto et al., 2008. Here we list those factors that affect the underground salt accumulation under tree stands, and which are studied in the project one-by-one. The positive factors on the left of the figure increase the chance of salt accumulation. The project aims at describing the complex interrelation of the listed factors in such a way that the effect of new afforestations could be predicted.

Objectives

The objective of the suggested studies is the detailed investigation of the Hungarian situation through the systematic study of the affecting factors and to give suggestions for the prevention of expected problems. There will be two scales as hinted by Fig 2. At the regional scale throughout the Great Hungarian Plain altogether 243 afforested plots will be visited to collect data on groundwater, soil and vegetation characteristics. These data will be used in a statistical modeling for the prediction of the effect of afforestation on soil, subsoil, groundwater characteristics. At the stand-scale studies carried out at 16 selected plots the effect of age group of trees, groundwater level, groundwater salinity and soil texture on the transpiration of trees will be assessed by detailed temporal monitoring. The suggested research is fundamental research, because the main objective is to find a predictive model of expected salt accumulation. This a novel research because of the comprehensive consideration of all important factors of salt accumulation. Only limited number of factors were studied so far by Nosetto 2007, 2008, Bazykina, 2000 and others, and the spatial variability of soil texture is disregarded. In Hungary it is a very strong limiting factor and deserves consideration as suggested by Várallyay, 2002.

Factors studied

All the modifying factors listed in Fig 2 will be studied in the project. In our previous paper the effect of climatic water balance was described in detail for the Argentine Pampas. Throughout the Great Hungarian Plain there is a great range found in the annual climatic water deficit (difference between potential evaporation and precipitation), ranging from 100 to more than 250 mm (Pécsi et al., 1985) as shown by Fig 3, but the effect of these differences on soil or subsoil salt accumulation have not been evaluated yet.

Fig 3. The annual climatic water deficit in Hungary

The range of groundwater depths (watertable) in the Great Hungarian Plain (GHP) was tabulated by Tóth et al., 2001 and ca 77% of the area was found to lie above groundwater depths between 1 and 4 meter. The same publication showed that in ca 59% of the GHP areas have shallow groundwaters more saline than 1 g/l. Also Tóth et al. (2001) listed the typical subsoil textural sequences in GHP and found that 45% of the area can be characterized as clay, 12% as sand down to the depth of 10 m. Some 30% of the area are characterized by variable layers of silt, sand and clay.

At present the most common tree species used for afforestation in the GHP are English oak, (Quercus robur L.), black locust, (Robinia pseudoacacia L.) and various poplars, most often (Populus alba L., P. ×canadensis Moench). These trees are characterized by different growth rate, rooting system and rooting depth according to Crow, 2005. The order of salt tolerance between the species is hypothesised to be Q robur > P species > R pseudoacacia but all three species have some tolerance according to UMES, 2004.

A further factor affecting the salt accumulation is the stand age. This factor determines rooting depth, growth rate and related water uptake and results in increasing salt accumulation underground by time (Bazykina,2000).

Methods used in the project

During three years of field data collection 243 afforested plots will be visited and there groundwater, subsoil, soil and tree growth parameters will be collected. The studied 243 plots will represent the most important combinations of the factors which affect the subsoil salt accumulation as shown in Fig 2. The number of study sites is received as 35 based on the following combinations: three mentioned species [Q robur, R pseudoacacia, P species] X common texture sequences [clayey, sandy, variable silty] X common groundwater depth [1-2, 2-4, 4-8 m] X groundwater salinity [1-2, 2-5, 5-10 g/l] X stand age [10-20, 30-40, 50-60 year]. There will be two versions of the equations: one numerical based on the measured data and one using categories of widely available information sources, such as AGROTOPO database (Várallyay and Molnár, 1989), existing groundwater depth and salinity maps by Tóth et al., 2001. At each plot four boreholes will be made (two inside, two outside of the plot in order to provide replicates) in a transect delineated based on the heterogeneous growth of vegetation.The effect of existing soil and subsoil salinity on the tree growth characteristics will be analysed in those studied plots, which are covered by a series of available ASTER satellite images (Fig 4.) with spring and autumn recordings.The novelty of this study plan is the vertical and horizontal integration of three different approaches plus the parallel Argentine studies. This way the undisclosed constants and regression coefficients in the equations formulated for the regional scale can be explained based on hydrological measurements carried out at stand scale. The final evaluation of the two databases will be carried out together by the two national teams. The databases of the two national groups will have the same structure and each element of study at both spatial scales will be compatible.

Fig 4. The cover of available ASTER satellite images

Expected results

The results of the study will show the effect of contrasting vegetation types on salt accumulation. Not evident interactions between tree salt tolerance, soil and subsoil texture sequences and groundwater depth and salinity will be described by the numerical equations between salt accumulation and affecting factors. The results will give very clear answer to the long-time speculated effect of afforestations on groundwater depth in the Great Hungarian Plain. The extent of progressive salt accumulation under older stands will help to evaluate the phenomenon of tree dieback, being characteristic in salt-affected regions. The data collected on the temporal fluctuation of groundwater will give a very important physiological indicator and will help the future selection of promising tree genotypes for the GHP. With the data collected an open database will be created about areas lying above shallow groundwater. The scientific importance of the database is manifold: it provides a basis for the analysis of land use effects on basic soil and subsoil parameters, first of all salt accumulation. Besides it will contain basic biometric data on the tree stand and groundwater depth, which are basic for environmental monitoring. This way the project will provide useful, often highly demanded data for spatial and temporal monitoring both on soils and subsoils, adding new data to the database of Research Institute for Soil Science

and Agricultural Chemistry of the Hungarian Academy of Sciences (RISSAC-TAKI) and Hungarian State Geological Institute (HSGI-MÁFI) as well. The practical importance of the results is straightforward, because with the use of the equations describing salt accumulation (dependent variable) with the categories of affecting factors collected from easily available databases (independent variable) the planning of new afforestations is greatly facilitated, and this way the undesirable process of salt accumulation in soils and subsoils will be prevented.

Preliminary Research

The basic series of the geology of the GHP was developed by Hungarian State Geological Institute in the seventies-eighties of the last century by the participation of László Kuti. He, together with RISSAC staff members has assessed the hydrogeological factors of soil salinization in Tóth et al., 2001. The current tendencies of soil salt concentratin in relation with the groundwater level have been described by Tóth et al., 2006.

During the preceding years the collaboration with the Argentine team resulted in two papers on the situation in a few oak stands inside Hungary, which is described in Nosetto et al., 2007 and another on the regional situation in Argentina in Nosetto et al., 2008. These were the bases of one PhD thesis and several other publications.The competence of the RISSAC team in the use of field techniques for the characterization of soil salt accumulation with geoelectric instruments was shown by the publications Douaik et al., 2007, Tóth et al., 2006. The suggested augering techniqe was used by Tóth and Várallyay, 2001 for the description of the relationship of soil and subsoil salt accumulation with groundwater depth and salinity. Workplan

1st year:

Development of the joint stand preselection methodology, database structure, instrumentation and field protocol for the project by the two national teams.

Selection of the studied afforested plots based on the combination of affecting factors as presented in publications of Tóth et al., 2001, the series of Geological Atlas of Hungary and the register of forests at the National Forest Authority. Each site and the corresponding boreholes and transects will be selected with the visual aid of generally available satellite and aerial records of Google Earth and MEPAR (Hungarian Agricultural Plot Identification System).

From the 243 plots selected for regional-scale study 16 will be selected for stand-scale study based representativity.

Already during first year and each following year the local data collection of groundwater changes together with accompanying soil salinity, moisture and temperature plus weather data will be collected at one fourth of the plots selected for it during more than six months preceding leaf fall. The reason for salinity and moisture monitoring is the need to study the effect of changing salinity on transpiration. Based on the temporal monitoring of detailed groundwater level fluctuations and the chemical composition of groundwater and soil and subsoil layers and other affecting factors the transpiration of the stands will be estimated and compared to estimates received on the basis of chloride balance calculations.

2nd year:

Regional-scale data collection at one third of the plots during September and October. At each selected plot the elevation above sea level will be determined for the boreholes. There will be two boreholes outside the studied plot in grassland/cropland on two opposite sides of the afforested plot at distances not shorter than 50 m from the edge of the afforested plot. There will be two boreholes inside the plot at distances not shorter than 50 m from the edge and each other. Along the transects, the bulk electrical conductivity of the soil will be measured with an electromagnetic induction probe and calibration samples will be collected. From the boreholes the texture sequence will be determined by field geological assessment. Also from the each 20 cm layer laboratory texture analysis will be performed for sand, silt and clay fractions. From the same layers soil salinity, alkalinity and sodicity and chloride content will be determined. The depth of groundwater found and risen after 30 minutes will be determined. A groundwater sample will be collected and analysed in the laboratory for electrical conductivity, Na, Cl, carbonates and bicarbonates. Along the transect the tree trunk circumference, tree height will be determined in order to estimate the timber volume. In the transect segments outside the tree stand the biomass will be measured by cutting the aboveground plant mass and subsequent drying.

Continuation of stand-scale data collection at another fourth of the plots selected for this research.

Biomass estimation of plots covered by the satellite image limits will be carried out by the ASTER satellite images.

3rd year:

Continuation of regional-scale data collection at another third of the plots. Continuation of stand-scale data collection at another fourth of the plots selected for this

research. Biomass estimation of plots covered by the ASTER satellite images.

4th year:

Continuation of regional-scale data collection at another third of the plots. Continuation of stand-scale data collection at another fourth of the plots selected for this

research. Biomass estimation of plots covered by the ASTER satellite images. Statistical analyses of data will be carried out for the collected data. At regional scale for each tree species the salt accumulation of stands versus nearby

grasslands/croplands will be predicted by the studied independent variables shown by Fig 2, climatic water balance, clay percent, groundwater salinity, groundwater depth, stand age. The correlation between timber volume and soil salt accumulation will be studied at this regional scale also inside the area covered by ASTER satellite images the tree biomass will be estimated by Normalized Differential Vegetation Index and its correlation with soil and subsoil salt accumulation will be calculated. At stand scale the analysis of groundwater fluctuations and the chloride curves of underground layers will provide two independent estimates of tree transpiration as described by Nosetto et al., 2007.

Work flow between Hungarian partners, covering the personal responsabilities

Selection of study sites

Preparation of basic hydrogeological spatial database on the combinations of hydrogeological factors: L Kuti

Selection of relevant afforestations: I Csiha and Z Gribovszki Interpretation of salt tolerance of species: I Csiha Field geological augering and description: L Kuti Field characterization of stands: A Szabó Field installation of groundwater monitoring stations: Z Gribovszki Sample analysis: A Szabó Statistical modeling: T Tóth Hydrological modeling: Z Gribovszki Remote sensing analysis: A Szabó

History of collaboration between the Hungarian and Argentine teams

The two teams collaborated in a Hungarian-Argentine Intergovernmental project 2004-2005 “Vegetation and salt/water dynamics in flat landscapes: Comparative andcomplementary studies in Pampa Deprimida”. The principal investigators of the project were Tibor Tóth from the Hungarian side and Esteban Jobbágy from the Argentine side. The results of the collaboration are published in the papers Nosetto et al., 2007 on the salt accumulation in a Hungarian study site and in Nosetto, 2008 on the analysis of the situation in the Pampa Deprimida, and also his PhD thesis: “Conversion of grasslands into forests: the impact on the water and salt dynamics. Universidad de Buenos Aires, 2007.

Most important interactions between Argentine and Hungarian research groups during the project

Collaboration in the study site selection and parallel instrumentation at mutual level Joint development of the structure of field database and field protocol. Supervision of site selection at mutual level. Supervision of soil property selection and analytical methods by Hungarian team. Supervision of remote sensing analysis by Argentine team. Supervision of hydrological instrumentation by Hungarian team. Supervision of ecohydrological interpretation by Argentine team. Supervision of statistical interpretation by Hungarian team.

Required infrastructure

The maps required were developed by the Research Institute for Soil Science and Agricultural Chemistry of the Hungarian Academy of Sciences and Hungarian State Geological Institute and are available. The location of possible afforested stands will be determined based on the advise given and database accessed by the Station for Afforestation of Salt-Affected Areas of the Forestry Research Institute.

The augerings will be carried out by the staff of Hungarian State Geological Institute. The field devices necessary for the field bulk electrical conductivity measurements

(EM38 2K, EMRC-120) are available in RISSAC. The devices necessary for temporal monitoring of groundwater depth and meteorological

elements will be purchased with the contribution of the project funding. The devices used for timber volume estimation are available at the Station for

Afforestation of Salt-Affected Areas of the Forestry Research Institute. The GIS software ArcView is available at RISSAC. The ASTER satellite images are

available at RISSAC together with the ERDAS processing software.

The necessary desktop computers, GPS and leveling instruments are available at RISSAC.

A field notebook will be purchased with the funds provided by the project. A four-wheel vehicle for access and transport is available at RISSAC.

Personnel of the project

At RISSAC the following personnel is available for the project Tibor Tóth principal investigator András Szabó PhD student Newly employed PhD student will be payed with project funding.

At the Hungarian State Geological Institute the following personnel is available for the project László Kuti, head of the Department of Environmental Geology

At the Station for Afforestation of Salt-Affected Areas of the Forestry Research Institute the following personnel is available

Imre Csiha, station director

At the University of Western Hungary the following personnel is available Zoltán Gribovszki, associate professor

References:

Andrasevits, Z., Buzás, Gy., Schiberna E. 2005. Current afforestation practice and expected trends on family farms in West Hungary. Journal of Central European Agriculture. 5:297-302.http://www.agr.hr/jcea/issues/jcea5-4/pdf/jcea54-8.pdf

Bazykina, G. S. 2000. Ecological Assessment of Meadow-Chestnut Soils of the Solonetzic Complex Ameliorated by Means of Afforestation in Nonirrigated Conditions in the Northern Caspian Region. Pochvovedenie. No.11. 1340-1348.

Crow, P. 2005. The influence of soils and species on tree rooting depth. Forestry Commission Information Note. November 2005www.forestry.gov.uk/pdf/fcin078.pdf/$file/fcin078.pdf

Douaik, A., M. Van Meirvenne, T. Tóth . (2007.) Statistical methods for evaluating soil salinity spatial and temporal variability. Soil Science Society of America Journal. 71:1629-1635.http://www.taki.iif.hu/english/soilsci/toth/abstr/235DMT2007FULL.pdf

National Forest Strategy, accessed 2009.Ministry of Agriculture and Regional Development. www.fvm.hu/main.php?folderID=2336

Nosetto, M D., E. G. Jobbágy, T Tóth and C. M. Di Bella (2007.) The effects of tree establishment on water and salt dynamics in naturally salt-affected grasslands. Oecologia. 152: 695-705. http://www.taki.iif.hu/english/soilsci/toth/abstr/230NJTB2007FULL.pdf

Nosetto M.D., Jobbágy M. G., Tóth T., Jackson R. B. (2008.) Regional patterns and controls of ecosystem salinization with grassland afforestation along a rainfall gradient. Global Biogeochemival Cycles. 22 doi:10.1029/2007GB003000

http://www.taki.iif.hu/english/soilsci/toth/abstr/NJTJ2008.pdf

Pécsi, M. 1985. National Atlas of Hungary.Kartográfia. Budapest.

Tóth T., Kovács, D., Marth, P., 2006 Mesoregional variability of the salt content of the genetic horizons of salt-affected soils as shown by the Hungarian Soil Monitoring System. 2006 Talajvédelem Különszám Talajtani vándorgyűlés Sopron, pp 192-201. (in Hungarian)

Tóth, T., and L. Kuti. (1999.) Geological factors affecting the salinization of the Nyirolapos Sample Area (Hortobagy, Hungary). I. General geological characterization, calcite concentration and pH values of subsurface layers. (in Hungarian). Agrokémia és Talajtan. 48: 431-444. http://www.taki.iif.hu/english/soilsci/toth/abstr/TK99FULL_5-6.PDF

Tóth, T., and Gy. Várallyay (2001.) Variability of soil conditions in relationship with salinization. (in Hungarian). Agrokémia és Talajtan. 50:19-34. http://www.taki.iif.hu/english/soilsci/toth/abstr/TV2001FULL.pdfTóth, T., L. Kuti, S. Kabos, and L. Pasztor L. (2001.) Use of digitalized hydrogeological maps for evaluation of salt-affected soils of large areas. Arid Land Research and Management. 15: 329-346.http://www.taki.iif.hu/english/soilsci/toth/abstr/TKKP2001FULL.pdf

Tóth T., L. Kuti, and U. Fügedi (2003.) Monthly studies at Zab-szék saline lake. Temporal changes of lake water, groundwater, soil and vegetation. (in Hungarian) Természetvédelmi Közlemények. 10:191-206. http://www.taki.iif.hu/english/soilsci/toth/abstr/TKF2003FULL.pdf

Tóth, T., A. Ristolainen, V. Nagy, D. Kovács, and Cs. Farkas. (2006.) Measurement of soil electrical properties for the characterization of the conditions of food chain element transport in soils. Part II. Classification of management units. Cereal Research Communications. 34: (No.1) 163-166. http://www.taki.iif.hu/english/soilsci/toth/abstr/TRNKF2006FULL.pdf

UMES (University of Minnesota Extension Service). 2004. "Protecting Trees & Shrubs Against Winter Damage", Advocate, Winter 2004http://on-line-seminars.com/pb/wp_3a6c2c92.html

Várallyay Gy., Molnár E., (1989). The agro-topographical map of Hungary (1:100,000 scale). Hungarian Cartographical Studies. 14th World Conference of ICA-ACI, Budapest: pp. 221-225.

Várallyay, Gy. 2002. Environmental stresses induced by salinity/alkalinity in the Carpathian Basin (Central Europe). Agrokémia és Talajtan. 51:233-242.