Embed Size (px)
Citation preview
THE PREVALENCE OF BACTERIA-SEDIMENT ASSOCIATIONS AND THEIR
EFFECT ON SEDIMENT SETTLING VELOCITIES IN AQUATIC
ENVIRONMENTS OF AGRICULTURAL SASKATCHEWAN AND ALPINE
BRITISH COLUMBIA
A Thesis
Submitted to the Faculty of Graduate Studies and Research
In Partial Fulfillment of the Requirements
For the Degree of
Master of Science
in
Geography
University of Regina
by
David Clem Barrett
Regina, Saskatchewan
January, 2014
Copyright © 2014: D. C. Barrett
UNIVERSITY OF REGINA
FACULTY OF GRADUATE STUDIES AND RESEARCH
SUPERVISORY AND EXAMINING COMMITTEE
David Clem Barrett, candidate for the degree of Master of Science in Geography, has presented a thesis titled, The Prevalence of Bacteria-Sediment Associations and Their Effect on Sediment Settling Velocities in Aquatic Environments of Agricultural Saskatchewan and Alpine British Columbia, in an oral examination held on December 5, 2013. The following committee members have found the thesis acceptable in form and content, and that the candidate demonstrated satisfactory knowledge of the subject material. External Examiner: Dr. Dirk deBoer, University of Saskatchewan
Supervisor: Dr. Kyle R. Hodder, Department of Geography
Committee Member: Dr. Ulrike Hardenbicker, Department of Geography
Committee Member: Dr. Dena McMartin, Faculty of Engineering & Applied Science
Chair of Defense: Dr. Britt Hall, Department of Biology
i
ABSTRACT
Bacteria-sediment associations (BSA) are the natural interactions
between bacteria and sediment particles that influence both bacterial survival
rates and properties of sediment particles. A newly developed method for
quantifying BSA was applied to an agricultural watershed in southeastern
Saskatchewan and an alpine watershed in the southern Coast Mountains of
British Columbia. Both watersheds exhibited spatial and temporal trends of BSA.
Impacts corresponding to land use and anthropogenic activity are also present.
Development of a novel method for determining BSA allowed for samples
taken from remote areas to be assessed. Membranes with 8 μm pore diameter
separated planktonic organisms from sediment-associated organisms. The use
of acridine orange fluorochrome and a fluorescence microscope allowed for all
organisms to be viewed.
Samples from the Pipestone Creek watershed in southeast
Saskatchewan confirmed the presence of BSA within in-field runoff, and in
Pipestone Creek. The percentage of bacteria associated with sediment range
from 2% at in-field locations to a maximum of 70% in Pipestone Creek with
samples from the creek having significantly higher BSA values than those from
in-field locations. Results from the Pipestone Creek watershed show that winter
bale grazing (WBG) of cattle, when compared to fall manure spreading (FMS)
and a control of no manure spreading, leads to a larger export of total bacteria
ii
during spring snowmelt. The application of WBG also leads to an increase in
sediment export and a decrease in mean particle diameter.
In the Lillooet River watershed, BSA were also noted to be present at all
sample locations with the percentage of bacteria associated with sediment
ranging from 1% - 40% depending on time of day and location within the
watershed. Decreased BSA was measured at locations associated with
planktonic bacteria influx via intensive agricultural runoff and wastewater
treatment.
Laboratory experiments using a laser backscatter device reveal an
increase in sediment settling velocity associated with increased bacteria
concentrations. Particle-size distributions of the sediment vary with an increase
in some size classes and a decrease in others. The changes in particle sizes
and settling velocities are consistent with the process of flocculation occurring
within the 23-hour sample duration.
The results from this study provide insight into the impacts of BSA in
contrasting environments through the application of a novel BSA quantification
method. Furthermore, the documentation of the impact of BSA on the settling
velocity of sediments provides a more holistic understanding of varved
sediments and therefore better informs process-inference, leading to more
holistic paleoenvironmental reconstructions.
iii
ACKNOWLEDGEMENTS
Financial support for the research undertaken in this thesis was provided
through the Agriculture and Agrifood Canada Watershed Evaluation of Beneficial
Management Practices (WEBs) program as well as a Natural Sciences and
Engineering Research Council of Canada discovery grant provided to Dr. Kyle
Hodder. Funding was also provided from the Faculty of Graduate Studies and
Research as well as the Department of Geography.
I am forever grateful for the academic mentoring, feedback and coaching
that was received from my supervisor, Dr. Kyle Hodder. The knowledge,
enthusiasm and encouragement provided both in the field and otherwise were
paramount to my success and to the positive experience I had during my
studies. The academic freedom, yet careful guidance, promoted my personal
and professional development and allowed for maintained productivity.
Without the advice, support, equipment and lab space of Dr. Dena
McMartin from the Faculty of Engineering and Applied Science, many aspects of
this research could not have been undertaken and for that I am grateful.
I need also to thank the fellow graduate students in the Department of
Geography whom often provided support and humour. Without the constant
reassurance, positive feedback and health and safety meetings, more of my
marbles would have undoubtedly been lost. To my parents, I am grateful for your
continued encouragement, understanding and support through this stage of my
life.
iv
DEDICATION
This thesis is dedicated to my dog, Tess, who throughout my undergraduate and
graduate programs has helped me to maintain a shred of sanity at the toughest
of times. She has had to join me on the roller coaster that is student life and has
not complained, but rather reminded me to keep my priorities in line. She has
acted as a constant reminder that life needs to be enjoyed – every last tail wag
of it.
“Everyone you will ever meet knows something you don’t”
- Bill Nye
v
TABLE OF CONTENTS
Abstract.................................................................................................................. i Acknowledgements .............................................................................................. iii Dedication ............................................................................................................ iv Table of contents .................................................................................................. v List of Tables ...................................................................................................... vii List of Figures ..................................................................................................... viii List of Abbreviations ............................................................................................. x Co-Authorship ...................................................................................................... xi 1 Introduction ..................................................................................................... 1
1.1 Thesis objectives .............................................................................. 4 1.2 Study sites ........................................................................................ 5
1.2.1 Prairie sites ......................................................................... 6 1.2.2 Alpine site ......................................................................... 13
1.3 Literature Cited ............................................................................... 17 2 Literature Review .......................................................................................... 22
2.1 The Presence of Bacteria in Agricultural and Alpine Environments 22 2.2 Bacterial Properties and Attachment Methods ............................... 24
2.2.1 Adsorbed vs planktonic ..................................................... 27 2.3 Water Quality Implications .............................................................. 33 2.4 The Impact of Bacteria on Sediment Characteristics ...................... 34
2.4.1 The Effect of Sediment on Bacterial Properties ................ 39 2.5 Separating Adsorbed and Free-Phase Bacteria ............................. 40 2.6 Agricultural beneficial management practices ................................ 43
2.6.1 Winter bale grazing ........................................................... 43 2.7 Literature cited ................................................................................ 45
3 The impact of management practices on the prevalence of bacteria-sediment associations in Southeastern Saskatchewan ...................................................... 51
3.1 Abstract .......................................................................................... 51 3.2 Introduction ..................................................................................... 52 3.3 Study site ........................................................................................ 57 3.4 Methods .......................................................................................... 60
3.4.1 Sample Collection ............................................................. 60 3.4.2 Partitioning sediment-associated bacteria from free-phase bacteria ........................................................................................ 61 3.4.3 Enumeration and quantification of total bacteria ............... 62 3.4.4 Measurement of particle size, suspended sediment concentration and water quality .................................................... 63
vi
3.5 Results ........................................................................................... 64 3.6 Discussion ...................................................................................... 77 3.7 Conclusions .................................................................................... 84 3.8 Literature cited ................................................................................ 86
4 The presence of bacteria-sediment associations in an alpine environment and their effect on the settling velocity of sediment ............................................ 93
4.1 Abstract .......................................................................................... 93 4.2 Introduction ..................................................................................... 94 4.3 Study site ........................................................................................ 96 4.4 Methods .......................................................................................... 99
4.4.1 Sample Collection ............................................................. 99 4.4.2 Quantifying BSA ............................................................. 100 4.4.3 Measurement of particle size and SSC ........................... 102 4.4.4 Measuring settling velocities ........................................... 103
4.5 Results ......................................................................................... 104 4.5.1 Lillooet River ................................................................... 104 4.5.2 Settling velocity and particle-size distribution .................. 110
4.6 Discussion .................................................................................... 115 4.6.1 Conclusions .................................................................... 121
4.7 Literature cited .............................................................................. 124 5 Synthesis ..................................................................................................... 131
5.1 Literature Cited ............................................................................. 137 Appendix A – Relationship between particle sizes and BSA............................. 138 Appendix B – SOP’s ......................................................................................... 142
vii
LIST OF TABLES
Table 1.1 Geographical coordinates of locations referenced in this study ........... 9 Table 1.2 Treatments applied to agricultural sites compared in this study. Note
that sites 1 and 7-13 had other BMPs applied, and were not considered in this study. ..................................................................................................... 9
Table 2.1 Methods used in past studies for determining BSA ratios .................. 30 Table 2.2 A comparison of prior study results evaluating sediment adsorbed
fractions of E. coli and Enterococci, and the methods used to determine sediment association. ................................................................................. 32
Table 3.1 The agricultural practices undertaken at each of the field locations ... 59 Table 3.2 A comparison of BSA ratios at three sample locations within the
Pipestone Creek main stem on three sample days. Missing samples due to analysis errors resulting in unusable data. .................................................. 66
viii
LIST OF FIGURES
Figure 1.1 (A) Location of the agricultural field sites within the Pipestone Creek Watershed of Southeastern Saskatchewan (PFRA, 2001) (B) Location of study areas within the Canadian prairies .................................................... 11
Figure 1.2 Daily discharge for the Pipestone creek, upstream of the Moosomin reservoir in 2012 (Image source: Water Survey of Canada; station 05NE003; http://www.wateroffice.ec.gc.ca/) ............................................... 12
Figure 1.3 (A) The Lillooet River valley in the Coast Mountains of British Columbia (B) Sample locations along Lillooet River (C) Sample locations between Lillooet Glacier and Silt Lake (BC TRIM shapefiles) ..................... 15
Figure 1.4 (A) Average daily discharge of Lillooet River from 2000-2010 (B) Annual daily discharge for Lillooet River, near Pemberton, British Columbia (Water Survey of Canada) .......................................................................... 16
Figure 2.1. An ESEM image of a floc, with arrow labelling a bacterium. The largest objects visible in the image are diatoms. Source: Figure 3B in Droppo (2001; pg 1556). ............................................................................. 38
Figure 3.1 BSA from all in-field sampling locations and the Pipestone Creek during 2012 and 2013. Red line indicates mean, black line indicates median, box is 25-75th percentile, whiskers represent 90th and 10th
67 percentile, black
dots are outliers. ......................................................................................... Figure 3.2 The relationship between BSA and volume concentration within the
Pipestone Creek watershed, at the edge-of-field scale for three different particle sizes. Linear fit curve applied; n=20. .............................................. 69
Figure 3.3 The relationship between BSA and volume concentration within Pipestone Creek for three different particle sizes. Linear fit curve applied; n=7. ............................................................................................................ 70
Figure 3.4 BSA and local precipitation from an in-field location during a storm runoff event in summer, 2012 ..................................................................... 71
Figure 3.5 Bacteria-sediment associations measured in winter bale grazing, manure spreading and control sites during spring snowmelt. Box plot parameters same as Figure 3.1. ................................................................. 72
Figure 3.6 Concentrations of organisms per unit volume at winter bale grazing, manure spreading and control sites on during spring snowmelt. Box plot parameters same as Figure 3.1. ................................................................. 73
Figure 3.7 The impact of WBG, FMS and control sites on total concentrations of particles during the 2013 snowmelt. Box plot parameters same as Figure 3.1............................................................................................................... 75
Figure 3.8 The impact of WBG, FMS and control sites on volume moment mean calculations during the 2013 snowmelt. Box plot parameters same as Figure 3.1............................................................................................................... 76
Figure 4.1 Temporal changes in bacteria-sediment associations at the Lillooet River location over a 72-hour sampling period in early summer, 2012. .... 106
ix
Figure 4.2 A transect along the Lillooet River showing calculated BSA and total organisms per litre. Dashed line is a linear regression curve showing an increase (R2 107=0.51) in organisms in a downstream direction. ....................
Figure 4.3 Suspended sediment concentrations of the Lillooet River, proximal to Lillooet Glacier, measured over a 40 hour period. Outliers were a result of sampler intake being submerged in riverbed sediment. ........................... 108
Figure 4.4 The relationship between BSA and particle concentration along Lillooet River, at three different particle size bins ..................................... 109
Figure 4.5 Difference in particle-size distribution between mixture with 40 ml of bacterial solution and one to which no bacteria were added. Positive values denote a (higher/lower) concentration in the run containing bacteria. Sample numbers represent logarithmically spaced sample intervals over 23-hour sample period. .......................................................................................... 112
Figure 4.6 Difference in particle-size distribution between mixture with 10 ml of bacterial solution and one to which no bacteria were added. Postive values denote a (higher/lower) concentration in the run containing bacteria. Sample numbers represent logarithmically spaced sample intervals over 23-hour sample period. .......................................................................................... 113
Figure 4.7 A comparison of settling velocities without bacteria, and at 6 Polyseed dilutions. Also shown is the calculated Stokes' settling velocity for quartz minerals. ................................................................................................... 114
Figure A. 1 Relationship between volume concentration of 32 particle size bins
and associated R2 139 values from samples in the Lillooet River ................... Figure A. 2 Relationship between volume concentration of 32 particle size bins
and associated R2
140 values from samples at in-field locations in the Pipestone
Creek watershed ...................................................................................... Figure A. 3 Figure A. 4 Relationship between volume concentration of 32 particle
size bins and associated R2
141 values from samples taken in Pipestone Creek
................................................................................................................. Figure A. 5 Materials required for separation of sediment-associated bacteria
and planktonic bacteria in the field ........................................................... 143 Figure A. 6 Materials required for laboratory analysis of samples ................... 145
x
LIST OF ABBREVIATIONS
BSA
WBG
FMS
SV
PSD
LISST
VMM
Bacteria-sediment associations
Winter bale grazing
Fall manure spreading
Settling velocity
Particle-size distribution
Laser In-Situ Scattering and Transmissometer (Sequoia Scientific;
LISST-100X device)
Volume moment mean diameter
xi
CO-AUTHORSHIP
This dissertation is organized with two discrete papers intended for peer-
review publication. Due to this structure, there may be some redundancy in the
methods and introduction sections. Every effort has been made to minimise
repetition without affecting the integrity of the independent chapters.
This dissertation is based on two manuscripts:
Chapter 3: Barrett, D.C., Hodder, K.R.
The impact of management practices on the prevalence of
bacteria-sediment associations in Southeastern Saskatchewan
Chapter 4: Barrett, D.C., Hodder, K.R.
The presence of bacteria-sediment associations in an alpine
environment and their effect on the settling velocity of sediment
For both of the manuscripts, the author of this dissertation, DB, was
responsible for the design and implementation of the study, data analysis and
manuscript authoring. KH provided ongoing feedback of experimental design
and manuscript preparation.
1
1 INTRODUCTION
Understanding current geomorphic processes and anthropogenic impact
on said processes is key to understanding both current environmental changes
as well as past conditions. Interpreting paleoenvironmental conditions through
the use of proxies such as laminated lake sediments relies on our understanding
of the process that create these proxies and their inter-relationships in a
process-network. As such, to better understand processes of the past, we must
better understand contemporary processes.
The current level of anthropogenic influence is more ubiquitous than ever
before. Humans exert a significant influence on current environmental conditions
and geomorphic processes, prompting some researchers to describe humans as
the premier geomorphic agent sculpting the landscape (Hooke 2000). Water
quality is a key environmental parameter undergoing change across the globe.
Through the application of intensive agriculture and the concentration of human
waste, contamination of waterways is a growing challenge. The release of
bacteria into hydrologic systems often leads to ecological impacts and ultimately
can impact human health.
To better understand the movement and impact of bacteria in waterways,
water quality modelling systems are often employed. Water quality models have
been validated for watershed parameters such as runoff and discharge
(Kushwaha and Jain 2013), nutrient loading (Santhi et al. 2001, Jha et al. 2007)
and sediment loads (Kirsch et al. 2002) to name a few. Models focusing
2
specifically on microbial water quality fall into one of three categories (Jamieson
et al. 2004): 1) loading models focussing on landscape bacterial concentrations
2) models simulating persistence of bacteria in receiving waters and 3) models
that attempt to combine both loading and in-stream bacterial processes.
Although bacteria-sediment interactions are known to occur and
composite particles are know to be present in a wide variety of fluvial and
lacustrine environments (Liss et al. 1996, Droppo et al. 1997, Woodward et al.
2002, Droppo and Ongley 2004, Hodder and Gilbert 2007, Hodder 2009), their
impacts on geomorphological processes and overall water quality are usually not
well understood (Jamieson et al. 2004, Viles 2012). Often regarded as too small
to have geomorphic impact (Viles et al. 2008), bacteria are ubiquitous across the
Earth (Pikuta et al. 2007) and may play a larger role in surface processes than is
currently recorded in the literature.
The presence of bacteria within flocculated particles has been well
documented in a multitude of different environments (Zabawa 1978, Ongley et
al. 1992, Liss et al. 1996, Droppo 2001, Li and Yang 2007, Droppo et al. 2009).
Contemporary research quantifying bacteria-sediment associations (BSA) has
been primarily limited to laboratory or highly controlled environments where
access to electricity and analysis equipment is immediate (Schillinger and
Gannon 1985, Krometis et al. 2009, Soupir et al. 2010). Current methods used in
the literature vary widely and cannot be applied in remote environments where
sample analysis may be delayed by days, or even weeks. The two main
procedures for separating planktonic bacteria from sediment-associated bacteria
3
are fractional filtration (Schillinger and Gannon 1985, Mahler et al. 2000, Soupir
et al. 2008) and centrifugation (Characklis et al. 2005, Fries et al. 2006). The
variability in analysis methods provides significant challenges when comparing
results between studies.
To develop more accurate water quality models, BSA needs to be
quantified in a range of different environments. This will allow for a better
understanding of the factors influencing BSA and the range of values observed
in contrasting watersheds. Along the same lines, it is important to understand
how land use affects the presence and persistence of BSA. Doing so will allow
for more accurate water quality models and consequently more informed policy
decisions. Since freshwater is a valued and threatened resource in the Canadian
Prairies (Schindler 2006), making informed policy decisions is imperative.
Sediment movement within watersheds affects water quality (Young et al.
1989) as well as geomorphic processes (Gao 2008). When flocculated particles
are formed, their settling velocity will be greater than that of the individual
particles that comprise the floc (Droppo et al. 1997, Mikkelsen and Pejrup 2001,
Hodder 2009). As bacteria are noted to associate with sediment particles to form
flocs, it can be surmised that the presence of bacteria will alter the settling
velocity of sediment particles. By examining Stokes’ Law for calculating the
settling velocity of particles (Equation 1.1), it becomes evident that particle
radius is the major factor affecting settling velocity. The formation of flocs and
resulting changes in particle size will therefore have an effect on the settling
4
velocity of particles. To date, the effect of bacteria on sediment settling velocities
has yet to be measured.
Equation 1.1 Stokes' law as arranged to calculate settling velocity (vs). ρp is density of particle; ρf
In glacier-fed lakes laminated sediments are often present and are
frequently used as a paleoenvironmental proxy (Hodder et al. 2007). The
presence of laminated sediments requires sediment to settle at a rate that allows
it to reach the lake bottom. Hodder (2009) found that flocculation was a key
processes allowing for sediment deposition in a lacustrine environment located
in an alpine watershed. Understanding the environmental factors influencing the
creation of flocs and as such, laminated sediments, allows for more holistic
paleoenvironmental reconstructions to undertaken.
is density of fluid; g is acceleration due to gravity; r is particle radius; μ is viscosity of the fluid medium
𝑣𝑠 = 29
(𝜌𝑝 − 𝜌𝑓)𝜇
𝑔𝑟2
1.1 Thesis objectives
The objectives of this thesis are as follows:
1) To determine a practical method for evaluating the presence of bacteria-
sediment associations in remote environments
2) To measure the presence of bacteria-sediment associations in an agricultural
environment and an alpine environment
3) To compare the values obtained thus measuring the impact of specific land
management practices in both locations.
5
4) To measure the impact that bacteria have on the settling velocities of
sediment in freshwater environments in contrast with their constituent
particles.
1.2 Study sites
To compare the presence of BSA in different environments, two
contrasting watersheds were chosen:
a) a predominantly agricultural, lowland watershed in southeastern
Saskatchewan
b) an alpine watershed in southwestern British Columbia.
These two watersheds were chosen in an attempt to capture two extremes for
BSA analysis. The agricultural watershed provided a highly modified,
anthropogenically dominated environment whereas the alpine watershed
provided a less modified area near Lillooet Glacier and a gradient of
anthropogenic impact in the downstream direction. Documenting the range of
BSA ratios present in the two contrasting watersheds will help parameterize
water quality models that attempt to include sediment-associated bacteria.
Validation of water quality models occurs through specific case studies in a
range of environments (Jamieson et al. 2004). Applying and validating models in
different environments is a key aspect to fine-tuning those models (Kirsch et al.
2002, Jha et al. 2007). The initial development of water quality models requires
some background work to be done examining the range of values that may be
present in a natural environment (Bai and Lung 2005). By examining contrasting
6
environments, the range of expected values can be developed and used for
model development. Furthermore, the spatial gradient of anthropogenic
presence can provide insights into how values may change based on land use.
The testing and development of the novel method for measuring BSA
was undertaken in the agricultural watershed and then applied to the alpine
watershed. The Lillooet River watershed has a spatial gradient of anthropogenic
presence. Areas proximal to Lillooet Glacier have minimal direct anthropogenic
influence as they are remote and undeveloped. In contrast, agricultural land use
becomes predominant in the lower region of the Lillooet River valley, proximal to
Lillooet Lake. The gradient of anthropogenic influence, from minimal at Lillooet
Glacier, to heavily modified at Lillooet Lake, provided insights into the impacts of
different land uses on BSA ratios. The expectation is that the results from these
watersheds will be end-members of BSA in the environment. It is expected that
other similar watersheds that are driven by snowmelt should have BSA values
somewhere between those measured in this study depending on the supply of
sediment and bacteria within the watershed.
1.2.1 Prairie sites
The agricultural sites examined were located in the southeastern portion
of the Province of Saskatchewan. The two farms are titled Farm B and Farm M
(Table 1.1). Both sites are located within the Pipestone Creek watershed and are
part of the Canadian Prairies and more specifically the prairie pothole region.
Although the geographical area of the Prairies has multiple definitions, in this
7
paper the term is used to refer to those areas in Western Canada that are
classified as grasslands (Raddatz 1998) and dominated by agricultural land use.
The Pipestone Creek watershed drains a gross area of 3 684 km2
Figure 1.1
, most
of that area in Saskatchewan ( ). Within the Pipestone Creek
watershed there is a major reservoir, Moosomin Lake. According to PFRA
(2001), the majority of land in the Pipestone Creek watershed is used for
agricultural purposes with percentage land use as follows: crop production
(53%), perennial forage (31%) and grassland (8%); the remaining land cover
classifications in the watershed are ‘shrubs’, ‘trees’, ‘water bodies’, ‘wetlands’
and ‘other lands’. The watershed is located in the prairie pothole region of
Saskatchewan (Johnson et al. 2005), and is characterised by the presence of
small depressional wetlands formed due to glacial action and then the fill-and-
spill nature of the pothole topography (Shook and Pomeroy 2011). Seventy
percent of the watershed is considered to be hydrologically non-contributing to
outflow from the watershed (PFRA, 2001).
The Pipestone Creek watershed was chosen due to its widespread
agricultural use. The specific research sites were chosen because they had a
variety of beneficial management practices (BMPs) being applied. The
management practice of winter bale-grazing was applied at two sites, fall
manure spreading was applied at two sites and there were two control sites with
no manure spreading (Table 1.2). The process of winter bale grazing is one in
which cattle are allowed to feed and excrete directly on pasture land during
winter months (Jungnitsch et al. 2011, Kelln et al. 2012). This is in contrast to
8
wintering cattle at a feedlot and distributing the manure over fields as a fertilizer
in the fall.
Hydrological data has been collected at two locations within the
watershed by Environment Canada. The station upstream of Moosomin
Reservoir (Table 1.1; Water Survey of Canada gauging station 05NE003) has
recorded since 1960 and the station at the edge of the reservoir (Table 1.1;
Environment Canada gauging station 05NE002) has recorded reservoir levels
since 1955.
Discharge in Pipestone Creek is driven by precipitation and does not
occur during the winter months due to freeze up (Figure 1.2). Although
Pipestone Creek continues to flow during the summer, the discharge decreases
drastically after snowmelt. After snowmelt concludes, the flows at the edge of
field scale are ephemeral. During 2010 and 2011, total annual precipitation at
the field sites was 549 mm and 436.3, respectively. The closest Environment
Canada climate station in Broadview (Table 1.1; station id: 4010879) recorded a
mean annual precipitation of 412.6 mm from 1971-2000 of which 28% arrives as
snow. Since the in-field flows are ephemeral, not all summer precipitation
generates runoff. The amount of summer precipitation recorded at the field sites
that triggered the automatic samplers was 186 mm in 2010 and 171 mm in 2011.
During both spring snowmelt and summer precipitation, runoff transports
sediment within the watershed. The primary sources of sediment in Prairie
watersheds are considered to be agricultural practices (de Boer 1997).
9
Table 1.1 Geographical coordinates of locations referenced in this study
Site Location
In-field Farm M 50.24°N 102.22°W Farm B 50.27°N 102.18° W
Pipestone Creek
Grab sample 1 50.18°N 102.27°W Grab sample 2 50.15°N 101.87°W Grab sample 3 50.08°N 101.78°W
Water Survey of Canada stations
Pipestone Creek near Moosomin Reservoir - 05NE003
50.15°N 101.83°W
Moosomin Reservoir - 05NE002 50.05°N 101.68°W
Environment Canada climate station Broadview - 4010879 50.37°N 102.57°W
Table 1.2 Treatments applied to agricultural sites compared in this study. Note that sites 1 and 7-13 had other BMPs applied, and were not considered in this study.
Farms Site # Treatment Farm M 2
3 4 5 6
Winter bale grazing Winter bale grazing
Perennial pasture - control Spreading of manure Spreading of manure
Farm B 14 Perennial pasture – control
10
The sediment delivery from agricultural practices is, however, decreasing (de
Boer 1997), and it is hypothesized that this decrease in sediment delivery is due
to the implementation of soil conservation BMP’s (de Boer 1997).
Aside from in-field collection, samples were also obtained along
Pipestone Creek at three different locations; Pipestone 1, Pipestone 2 and
Pipestone 3 (Table 1.1). These locations were chosen due to their easy access
and are roughly equidistant apart between the field sites and Moosomin
Reservoir. The majority of the agricultural component of this thesis was
undertaken in conjunction with the Agriculture and Agri-Food Canada (AAFC)
Watershed Evaluation of Best Management (WEBs) research project. This study
uses a subset of sites from the larger WEBs project.
11
Figure 1.1 (A) Location of the agricultural field sites within the Pipestone Creek Watershed of Southeastern Saskatchewan (PFRA, 2001) (B) Location of study areas within the Canadian prairies
A
B
12
Figure 1.2 Daily discharge for the Pipestone creek, upstream of the Moosomin reservoir in 2012 (Image source: Water Survey of Canada; station 05NE003; http://www.wateroffice.ec.gc.ca/)
13
1.2.2 Alpine site
Lillooet Glacier and Lillooet River were chosen as the alpine sampling
location due to the presence of a large proglacial meltwater channel and glacier-
fed lake, allowing for a source-to-sink-to-sink sampling transect (Figure 1.3).
Lillooet Glacier is located in the southwestern portion of British Columbia, in the
Coast Mountains. Although other locations were considered, this location was
chosen for safe accessibility. The location was also chosen due to its lack of
development upstream combined with spatial gradient of anthropogenic
presence, increasing in the downstream direction. Also, the region has had
numerous prior geomorphological studies, upon which this study builds,
including the presence of flocs (Hodder and Gilbert 2007, Hodder 2009),
paleoenvironmental reconstruction (Gilbert 1975, Desloges and Gilbert 1994,
Gilbert et al. 2006) and sediment transfer (Jordan and Slaymaker 1991, Friele et
al. 2005, Simpson et al. 2006, Schiefer and Gilbert 2008, Guthrie et al. 2012).
The presence of well-defined laminated sediments in Lillooet Lake has been
noted by numerous authors (Gilbert 1975, Desloges and Gilbert 1994). To better
understand these laminated sediments, we must document the processes that
influence their formation. This thesis expands the understanding of the Earth
surface processes occurring within the watershed that have influenced the
presence and characteristics of laminated sediments in Lillooet Lake.
The majority of samples at the alpine site were taken from the Lillooet
River upstream of the braided channel (Figure 1.3; B). Other samples were
taken proximal to Lillooet glacier (Figure 1.3; A), Silt Lake (Figure 1.3; D&E) and
14
downstream at multiple locations in Lillooet River (Figure 1.3; F-O). Sample
locations A-E were chosen due to their ease of access to the main river channel.
Samples locations F-O were obtained roughly 10 km apart from the east bank of
Lillooet River where it runs near to the Lillooet forest service road. Samples G
and O were taken in an attempt to capture changes occurring due to the
confluence of Meager Creek and the outfall of the Lillooet wastewater treatment
plant respectively.
The Lillooet River flows year-round with the annual peak occurring during
glacial melt at approximately ordinal day 180 (Figure 1.4). Most samples were
taken upstream of the Lillooet-Green River confluence, thus the high SSC in
Green River (Gilbert 1975) is of little consequence to this study. The primary
sources of sediment within the watershed are from glacial erosion (Desloges and
Gilbert 1994), and volcanic debris originating from the Mount Meager volcanic
complex (Guthrie et al. 2012) – both in the headwater reaches of the watershed.
15
Figure 1.3 (A) The Lillooet River valley in the Coast Mountains of British Columbia (B) Sample locations along Lillooet River (C) Sample locations between Lillooet Glacier and Silt Lake (BC TRIM shapefiles)
B
A
C
16
Ordinal Day
Figure 1.4 (A) Average daily discharge of Lillooet River from 2000-2010 (B) Annual daily discharge for Lillooet River, near Pemberton, British Columbia (Water Survey of Canada)
A
B
17
1.3 Literature Cited Bai, S., Lung, W.-S. 2005. Modeling sediment impact on the transport of fecal
bacteria. Water Research, 39: 5232–5240. Characklis, G.W., Dilts, M.J., Simmins, O.D., III, Likirdopulos, C.A., Krometis, L.-
A.H., Sobsey, M.D. 2005. Microbial partitioning to settleable particles in stormwater. Water Research, 39: 1773–1782.
de Boer, D.H. 1997. Changing contributions of suspended sediment sources in
small basins resulting from European settlement on the Canadian prairies. Earth Surface Processes and Landforms, 22: 623–639.
Desloges, J.R., Gilbert, R. 1994. Sediment source and hydroclimatic inferences
from glacial lake sediments: the postglacial sedimentary record of Lillooet Lake, British Columbia. Journal of Hydrology, 159: 375–393.
Droppo, I.G. 2001. Rethinking what constitutes suspended sediment.
Hydrological Processes, 15: 1551–1564. Droppo, I.G., Leppard, G.G., Flannigan, D.T., Liss, S.N. 1997. The freshwater
floc: A functional relationship of water and organic and inorganic floc constituents affecting suspended sediment properties. Water, Air, and Soil Pollution, 99: 43–54.
Droppo, I.G., Liss, S.N., Williams, D., Nelson, T., Jaskot, C., Trapp, B. 2009.
Dynamic existence of waterborne pathogens within river sediment compartments. Implications for water quality regulatory affairs. Environmental Science & Technology, 43: 1737–1743.
Droppo, I.G., Ongley, E.D. 2004. Flocculation of suspended sediment in rivers of
southearstern Canada. Water Resources, 28: 1799–1809. Friele, P.A., Clague, J.J., Simpson, K., Stasiuk, M. 2005. Impact of a Quaternary
volcano on Holocene sedimentation in Lillooet River valley, British Columbia. Sedimentary Geology, 176: 305–322.
Fries, J.S., Characklis, G.W., Noble, R.T. 2006. Attachment of fecal indicator
bacteria to particles in the Neuse River Estuary, N.C. Journal of Environmental Engineering, 132: 1338–1345.
Gao, P. 2008. Understanding Watershed suspended sediment transport.
Progress in Physical Geography, 32: 243–263.
18
Gilbert, R. 1975. Sedimentation in Lillooet Lake, British Columbia. Canadian Journal of Earth Sciences, 12: 1697–1711.
Gilbert, R., Crookshanks, S., Hodder, K., Spagnol, J., Stull, R. 2006. The record
of an extreme flood in the sediments of montane Lillooet Lake, British Columbia: implications for paleoenvironmental assessment. Journal of Paleolimnology, 35: 737–745.
Guthrie, R.H., Friele, P., Allstadt, K., Roberts, N., Evans, S.G., Delaney, K.B.,
Roche, D., Clague, J.J., Jakob, M. 2012. The 6 August 2010 Mount Meager rock slide-debris flow, Coast Mountains, British Columbia: characteristics, dynamics, and implications for hazard and risk assessment. Natural Hazards and Earth System Science, 12: 1277–1294.
Hodder, K.R. 2009. Flocculation: a key process in the sediment flux of a large,
glacier-fed lake. Earth Surface Processes and Landforms, 24: 1151–1163. Hodder, K.R., Gilbert, R. 2007. Evidence for flocculation in glacier-fed Lillooet
Lake, British Columbia. Water Research, 41: 2748–2762. Hodder, K.R., Gilbert, R., Desloges, J.R. 2007. Glaciolacustrine varved sediment
as an alpine hydroclimatic proxy. Journal of Paleolimnology, 38: 365–394. Hooke, R.L. 2000. On the history of humans as geomorphic agents. Geology,
28: 843–846. Jamieson, R., Gordon, R., Joy, D., Lee, H. 2004. Assessing microbial pollution of
rural surface waters: A review of current watershed scale modeling approaches. Agricultural Water Management, 70: 1–17.
Jha, M.K., Gassman, P.W., Arnold, J.G. 2007. Water Quality Modeling for the
Raccoon River Watershed Using SWAT. Transactions of the ASABE, 50: 479–493.
Johnson, W.C., Millett, B.V., Gilmanov, T., Voldseth, R.A., Guntenspergen, G.R.,
Naugle, D.E. 2005. Vulnerability of Northern Prairie Wetlands to Climate Change. BioScience, 55: 863–872.
Jordan, P., Slaymaker, O. 1991. Holocene sediment production in Lillooet River
Basin, British Columbia: a sediment budget approach. Géographie physique et Quaternaire, 45: 45–57.
19
Jungnitsch, P.F., Schoenau, J.J., Lardner, H.A., Jefferson, P.G. 2011. Winter feeding beef cattle on the western Canadian prairies: Impacts on soil nitrogen and phosphorus cycling and forage growth. Agriculture, Ecosystems & Environment, 141: 143–152.
Kelln, B., Lardner, H., Schoenau, J., King, T. 2012. Effects of beef cow winter
feeding systems, pen manure and compost on soil nitrogen and phosphorous amounts and distribution, soil density, and crop biomass. Nutrient Cycling in Agroecosystems, 92: 183–194.
Kirsch, K., Kirsch, A., Arnold, J.G. 2002. Predicting sediment and phosphorus
loads in the Rock River Basin using SWAT. Transactions of the ASAE, 45: 1757–1769.
Krometis, L.-A.H., Dillaha, T.A., Love, N.G., Mostaghimi, S. 2009. Evaluation of
a filtration/dispersion method for enumeration of particle-associated Escherichia coli. Journal of Environment Quality, 38: 980.
Kushwaha, A., Jain, M.K. 2013. Hydrological Simulation in a Forest Dominated
Watershed in Himalayan Region using SWAT Model. Water Resources Management, 27: 3005–3023.
Li, X.Y., Yang, S.F. 2007. Influence of loosely bound extracellular polymeric
substances (EPS) on the flocculation, sedimentation and dewaterability of activated sludge. Water Research, 41: 1022–1030.
Liss, S.N., Droppo, I.G., Flannigan, D.T., Leppard, G.G. 1996. Floc Architecture
in Wastewater and Natural Riverine Systems. Environmental Science & Technology, 30: 680–686.
Mahler, B.J., Personné, J.C., Lods, G.F., Drogue, C. 2000. Transport of free and
particulate-associated bacteria in karst. Journal of Hydrology, 238: 179–193. Mikkelsen, O., Pejrup, M. 2001. The use of a LISST-100 laser particle sizer for
in-situ estimates of floc size, density and settling velocity. Geo-Marine Letters, 20: 187–195–195.
Ongley, E.D., Krishnappan, B.G., Droppo, I.G., Rao, S.S., Maguire, R.J. 1992.
Cohesive sediment transport: emerging issues for toxic chemical management. Hydrobiologia, 235-236: 177–187.
PFRA. 2001. PFRA's generalized land cover. Generalized land cover (l07g_pf0).
James Ashton. Version 1.
20
Pikuta, E.V., Hoover, R.B., Tang, J. 2007. Microbial Extremophiles at the Limits of Life. Critical Reviews in Microbiology, 33: 183–209.
Raddatz, R.L. 1998. Anthropogenic vegetation transformation and the potential
for deep convection on the Canadian prairies. Canadian Journal of Soil Science, 78: 657–666.
Santhi, C., Arnold, J.G., Williams, J.R., Hauck, L.M., Dugas, W.A. 2001. Application of a watershed model to evaluate management effects on point and nonpoint source pollution. Transactions of the ASABE, 44: 1559–1570.
Schiefer, E., Gilbert, R. 2008. Proglacial sediment trapping in recently formed
Silt Lake, upper Lillooet Valley, Coast Mountains, British Columbia. Earth Surface Processes and Landforms, 33: 1542–1556.
Schillinger, J.E., Gannon, J.J. 1985. Bacterial adsorption and suspended
particles in urban stormwater. Journal of the Water Pollution Control Federation, 57: 384–389.
Schindler, D.W. 2006. An impending water crisis in Canada's western prairie
provinces. Proceedings of the National Academy of Sciences, 103: 7210–7216.
Shook, K.R., Pomeroy, J.W. 2011. Memory effects of depressional storage in
Northern Prairie hydrology. Hydrological Processes, 25: 3890–3898. Simpson, K.A., Stasiuk, M., Shimamura, K., Clague, J.J., Friele, P. 2006.
Evidence for catastrophic volcanic debris flows in Pemberton Valley, British Columbia. Canadian Journal of Earth Sciences, 43: 679–689.
Soupir, M.L., Mostaghimi, S., Dillaha, T. 2010. Attachment of Escherichia coli
and Enterococci to particles in runoff. Journal of Environment Quality, 39: 1019–1027.
Soupir, M.L., Mostaghimi, S., Love, N.G. 2008. Method to partition between
attached and unattached E. coli in runoff From agricultural lands. Journal of the American Water Resources Association, 44: 1591–1599.
Viles, H.A. 2012. Microbial geomorphology: A neglected link between life and
landscape. Geomorphology, 157–158: 6–16. Viles, H.A., Naylor, L.A., Carter, N.E.A., Chaput, D. 2008. Biogeomorphological
disturbance regimes: progress in linking ecological and geomorphological systems. Earth Surface Processes and Landforms, 33: 1419–1435.
21
Woodward, J.C., Porter, P.R., Lowe, A.T., Walling, D.E., Evans, A.J. 2002. Composite suspended sediment particles and flocculation in glacial meltwaters: preliminary evidence from Alpine and Himalayan basins. Hydrological Processes, 16: 1735–1744.
Young, R.A., Onstad, C.A., Bosch, D.D., Anderson, W.P. 1989. AGNPS: A
nonpoint-source pollution model for evaluating agricultural watersheds. Journal of Soil and Water Conservation, 44: 168–173.
Zabawa, C.F. 1978. Microstructure of agglomerated suspended sediments in
Northern Chesapeake Bay estuary. Science, 202: 49–51.
22
2 LITERATURE REVIEW
2.1 The Presence of Bacteria in Agricultural and Alpine
Environments
The presence of bacteria in agricultural environments is accepted due to
the multitude of sources, including, but not restricted to, the manure of domestic
ungulates (Soupir et al. 2010). The effects of these bacteria are considered a
research priority due to their potential impact on human and livestock health
(Crabill et al. 1999). Bacteria are known to associate with sediment, and can
also be stored in fluvial bed sediments and resuspended during high flow events
(Jamieson et al. 2005b). This is of particular importance to the agricultural
environment that is found in southeast Saskatchewan, as flow regime is
dominated by nival or storm-induced event flows.
Although plentiful on the Earth, the impact of microbes on geomorphology
is often overlooked, perhaps due to size or perceived impact (Viles 2012).
Though small as individuals, it is estimated that prokaryotic organisms may
constitute up to 50% of the worlds biomass (Whitman et al. 1998). Even though
individual microbes are too small to see, the results of their colony formation can
often be viewed on the landscape. For example, biofilms are often noted over
large areas and thus their impact can be felt at a more comprehensible scale
(Viles 2012). Due to the presence of microbial organisms in all environments on
Earth (Pikuta et al. 2007), it should be expected that they may play a significant
23
role in shaping the Earth surface. However, quantification of the impact of
bacteria on geomorphological processes is notably limited in the literature.
While bacterial concentrations have been documented in agricultural
environments outside of Saskatchewan, comparatively little work has been done
in the alpine environment (Sharp et al. 1999). Preliminary studies have
confirmed the presence of bacteria in glacial environments (Sharp et al. 1999,
Skidmore et al. 2000, Lanoil et al. 2009). The interaction of sediment with
bacteria is often assumed to be insignificant in glacial meltwaters (Woodward et
al. 2002), however it has been shown that bacterial populations have a positive
relationship with sediment concentration (Sharp et al. 1999). Work has been
done examining the process of flocculation in glacial lakes (Hodder 2009),
lowland rivers (Droppo and Ongley 2004) and estuaries (Droppo et al. 1998). As
of yet, a study examining the presence of bacteria-sediment associations in
alpine environments has not been reported in the literature.
Flocculation has been defined slightly differently by a variety of different
studies, but in its simplest form is the process of individual clastic particles,
organic matter and organisms becoming associated with one another or
themselves, ultimately forming a composite particle, classified as a floc (Nič et
al. 2009). Since bacterial exopolymers are thought to be instrumental in the
formation of flocs, the presence of flocs may be related to the presence of
bacteria (Droppo 2004). The composition of flocs varies widely depending on
location and environment. For example, in samples from a glacier-fed lake,
(Hodder and Gilbert 2007) observed minimal evidence for organic matter in
24
desiccated flocs examined under a microscope. In contrast, (Droppo 2001)
observed large quantities of bacteria and diatoms in wet flocculated particles
from a lowland fluvial environment.
In both alpine and agricultural environments, anthropogenic activities alter
the concentration of bacteria present. For example, treated effluent from
wastewater treatment plants is a point source of bacteria (Iwane et al. 2001).
Microbial loading in rural areas is often from non-point sources and is, therefore,
much more difficult to pinpoint and control. In urban areas, the microbial loading
of surface waters is directly linked with the amount of impervious surfaces
located within the watershed (Mallin et al. 2000). The addition of bacteria in
planktonic (unattached, individual bacterium) form ultimately affects the water
quality of local waterways as well as the presence of BSA.
2.2 Bacterial Properties and Attachment Methods
Though the processes by which bacteria attach to sediment in the natural
environment are dynamic (Guber et al. 2007), experiments in controlled
conditions have provided some insight into attachment methods and the
formation of flocculated particles (Li and Yang 2007, Xuan et al. 2010). The
majority of research that has been undertaken regarding bacterial attachment to
sediments has focused on the water treatment process. Flocculation of
sediments and bacteria is a key step in water treatment to remove particulates
and organisms (Bache and Gregory 2010). However, the concentration of
bacteria and the fluid conditions in a water treatment scenario are often
25
considerably higher than that of the natural environment and therefore those
studies may not accurately represent the processes occurring in natural areas.
Current research suggests that extracellular polymeric substances (EPS)
allow for attachment of bacteria to sediment and other organisms (Wotton 2011).
EPS, also known as mucus, exudates or slimes (Wotton 2011), act as a glue
and can facilitate bacterial binding to other bacteria or sediments. Exudates are
a major component that helps hold complex flocs together (Wotton 2004).
Petticrew et al. (2011) note that the formation of flocs in freshwater occurs due to
the presence of EPS promoting adhesion and assisting in floc creation. While
EPS can be exuded by living organisms (Wotton 2011), it also can result from
microbe decomposition (Petticrew et al. 2011). An increased presence of EPS in
water will lead to an increase in the creation of flocs (Petticrew et al. 2011),
assuming that the environment is not sediment-limited. A second method of
bacterial attachment is due to electrochemical attraction between bacteria and
sediment (Zabawa 1978).
The attachment of bacteria to sediment has been categorised as either
weak or strong attachment strengths. Weak electrochemical adsorption occurs
due to van der Waals forces exceeding any repulsive forces at the bacteria-
sediment interface (Berry and Hagedorn 1991). Due to the nature of van der
Waals forces, bacteria are not physically attached to the sediment particles, thus
allowing for easier separation and the designation as “weak”. In contrast, BSA
formed via attachment of bacteria to sediment through EPS or cellular
appendages, such as pili, are much stronger (Jamieson et al. 2005a). The
26
difference in attachment ‘strength’ between the different types of bacterial
adsorption, is enough that weak electrochemical interactions can be considered
as reversible whereas adsorption due to polymers can be considered largely
irreversible (Jamieson et al. 2005a). Negatively charged cell surfaces repel
negatively charged minerals such as kaolinite (Gallagher et al. 2013). The
presence of positively charged ions, such as calcium, helps to negate the
repellent forces present between negatively charged objects (Gallagher et al.
2013). In freshwater environments that have low ionic strengths, it is assumed
that irreversible adsorption via polymers will dominate (Jamieson et al. 2005a)
due to decreased electrochemical interactions.
Of particular interest in agricultural environments is the bacterial species
Escherichia coli. It is very prominent in animal feces and thus is very common in
agricultural environments. It is known that E. coli excrete EPS (Kay et al. 2011)
and therefore could facilitate the creation of flocs in agricultural runoff.
Contemporary discussion of bacterial attachment has centered on the presence
of fimbriae or pili (Soupir et al. 2010). Pili are organelles that are used for
adhering to objects or other bacteria and can be used for reproductive purposes
(Castelain et al. 2010). Environmental conditions, such as the salt
concentrations, may cause morphological changes in bacteria cells thus
changing their adhesion potential (Gallagher et al. 2013).
The above discussion highlights the contemporary understanding of
bacterial attachment processes. While the effect of changing water properties in
a controlled environment has been found to change the amount and properties
27
of EPS excreted (Li and Yang 2007), very little is known about how EPS
excretion changes in relation to natural stream conditions. Changes in EPS
excretion due to different water properties will ultimately lead to a change in
bacterial adsorption to sediment.
2.2.1 Adsorbed vs planktonic
A portion of available bacteria adsorb to sediment, while the remainder
exist in free-phase (planktonic) state. Although prior studies have shown a high
bacteria-sediment association (BSA) ratio, there is a large range in the fraction
value determined. The fraction of available bacteria adsorbed to sediment
changes with flow conditions. For example, Characklis et al. (2005) determined
sediment-associated fractions of microorganisms to range from 20-35% in
background samples to 50-70% adsorbed during storm events. The storm
events provide a major change in flow volume, velocity, turbulence and
suspended sediment. The species of bacteria (Characklis et al. 2005), as well as
the type, size and concentration of sediment (Sharp et al. 1999, Jamieson et al.
2005a, Soupir et al. 2010) are important factors to consider when examining the
values of sediment association as they will all alter the degree to which bacteria
associate with sediment particles.
Studies of microbial attachment to sediment in stormwater highlight the
varying results obtained using a filtration method versus a centrifugation method
for determining BSA. For example, Jeng et al. (2005) found a sediment-
adsorption ratio of 0.218 –0.314 for E. coli using a filtration method whereas
28
Characklis et al (2005) found a 0.2 – 0.55 ratio using a centrifugation technique
in a comparable environment (Table 2.1).
Although prior research has compared adsorption rates among bacteria
species (Characklis et al. 2005), often a single strain of bacteria, such as E. coli
or fecal coliforms, are used as an index of all bacteria species (Characklis et al.
2005, Soupir et al. 2010). The use of indicator organisms allows for an
economically viable method of estimating microbial concentrations. The use of
indicator organisms has, however, been questioned in the past due to the
varying fraction of adsorption which can be dependant on bacterial species.
Specifically, the use of fecal coliforms will underestimate the adsorption of other
microbial species (Schillinger and Gannon 1985). In the proposed regions of
study, the total bacterial counts, and total adsorption are of importance,
regardless of bacterial strain. Furthermore, since comparatively little research
has been conducted in alpine, glacier-fed systems, the range of species and
concentrations are relatively unknown (Lanoil et al. 2009). A summary of results
from prior studies on the sediment-adsorbed fraction of two types of microbes is
found in Table 2.2.
Of particular interest in the currently presented study is the contrast
between agricultural areas and alpine environments. Notably, a positive
correlation between suspended sediment availability and bacterial adsorption
has been reported (Goulder 1977, Sharp et al. 1999). Soupir et al. (2010) found
that the BSA ratio related to the total suspended solids measured for
enterococci, but not for E. coli which may suggest that TSS is not the limiting
29
factor in BSA formation in agricultural environments. An increase of total
suspended sediment generally leads to an increase in sediment-associated
bacterial concentrations although the degree of association depends on species
type and environmental conditions. An increase in sediment concentration
provides an increase in the number of potential attachment sites for bacteria.
However, Soupir et al. (2010) reported that, in an agricultural environment, a
lack of sediment attachment sites may not be a limiting factor in the formation of
BSA. The diameter of sediment also appears to impact the fraction of adsorption
in previous studies, with bacteria preferentially adsorbing to sediment of smaller
diameter, particularly to clay-sized particles (Ling et al. 2002, Jamieson et al.
2005a, Soupir et al. 2010).
30
Table 2.1 Methods used in past studies for determining BSA ratios
Soupir (2010) Soupir (2011)
Soupir et al (2008)
Characklis et al (2005)
Jamieson (2005) Fries et al (2006)
Purpose: To determine preferential size range for bacterial partitioning
To measure E.Coli and Enterococci attachment - grassland sediment
To measure E. coli attachment to sediment from agricultural land
To measure microbial attachment to sediment in stormwater
To measure the resuspension of E.coli in natural stream
To quantify partitioning of E.Coli and enterococci to bed sediment and water column
Method of determining attachment:
Fractional filtration (FF) with filters of 500, 63, 8 and 3 μm. Then centrifuged at 3043 g (4700 rpm) for 15s. Supernatants shaken enumerated and counted
Same as Soupir (2010). Enumeration used standard methods for the examination of water and wastewater.
Same as Soupir (2010)
Sample of water centrifuged at 1164 g (2000 rpm) for 10 min. Brake of 4. Temperature controlled to 4°C.
Specific size range of sediment. Fractional filtration. Placed filter in jar of sterile saline sol. Dispersant added to jar. Bacteria then enumerated and counted
250 ml bottles centrifuged at 500 g for 10min in 20°C swinging bucket rotor
Assumptions made:
Centrifuge did not remove any planktonic bacteria. Rinsing of filters removed all filtered particles. 8 μm filter separated planktonic from attached
Same as Soupir (2010)
Same as Soupir (2010)
Particle-cell collisions doesn't lead to increased particle-associated fraction
Saline solution rinsed all sediment-associated bacteria from filter.
Gentle spinning used to limit particle-size distribution and collisions etc.
Notes
Used latex and glass beads to determine centrifuge regime. Approx. 80% latex remained suspended
Took ~15 minutes for equilibrium to be reached - at that time attached vs free-phase equil. was reached.
Found that centrifugation technique did not alter population sizes
31
Table 2.1 Continued
Goulder (1977) Jeng et al (2005) Krometis (2009) Ling et al (2002) Pandey et al (2012)
Purpose: To measure microbe association with sediment in an estuary
To quantify Indicator organism association with sediment in stormwater
To evaluate the best method of determining particle association
To measure adsorption of indicator organisms to sediment
To test accuracy of model for resuspension of E. coli from streambeds
Method of determining attachment:
Fractional filtration (3 μm). Enumerated using fluorescence microscopy.
Fractional filtration using filters of 100, 50, 35, 5, and 0.45 μm. Filters removed and placed in phosphate solution then shaken for 2 minutes
One sample filtered through 8 μm filter then filtrate and filtered microbes counted. Other sample was filtered and then used dispersion technique (sonication or dispersant).
Weak adsorption sample was centrifuged at 48 g for 3 min. Strong adsorption mixed in saline then centrifuged at 200 g for 6 min
Used fractional filtration and enumerated using Standard Methods for the examination of water and wastewater.
Assumptions made:
No dispersion technique used (at least listed)
Shaking dispersed all attached microbes. No dispersant or sonication used
All particle associated bacteria did not pass through 8um filter.
More significant centrifuging adequately separated between strong and weak adsorption.
Smallest filter catches all sediment-associated. No small sediment-associated gets through.
Notes
Looked at E.coli, FC and enteroccoci using more complex bacterial analysis
Concluded that use of sonication or dispersant yielded statistically similar results.
Focused on soils application - not water quality
32
Table 2.2 A comparison of prior study results evaluating sediment adsorbed fractions of E. coli and Enterococci, and the methods used to determine sediment association.
Study Method E. coli particle-adsorbed fraction
Enterococci adsorbed fraction
Environment
Schillinger and Gannon (1985) Filtration 0.288 - 0.479 N/A Urban stormwater Jamieson et al (2005) Filtration 0.2 - 0.44 N/A Fluvial Characklis et al (2005) Centrifugation 0.2 - 0.55 0.2 - 0.55 Fluvial Jeng et al (2005) Filtration 0.218 - 0.304 0.083 - 0.115 Estuary Fries et al (2006) Centrifugation 0.28 0.09 Estuary
33
2.3 Water Quality Implications
Although BSA is known to occur in aquatic environments, models
examining water quality have yet to be adjusted to include bacteria that are not
free-phase (“planktonic”; (Rehmann and Soupir 2009)). Water quality models are
widely deployed to inform water management decisions (Huang and Xia 2001,
Benham et al. 2006, Droppo et al. 2011). Failure to include sediment associated
bacteria results in inaccurate predictions (Rehmann and Soupir 2009). The
inclusion of standard environmentally dependent BSA coefficients would be a
logical first step in the development of more accurate models. Water quality
models are important for health monitoring since the presence of pathogenic
bacteria can have serious implications for human and livestock health. Although
E. coli is not the only pathogenic bacteria of concern to human health, it is
commonly used as an indicator organism (Soupir et al. 2010). It has been
argued that water quality monitoring based exclusively on surface water, and
collected during quiescent conditions, ignore the major sink of bacteria in
sediments and ultimately do not represent the potential risk to human health
adequately (Jamieson et al. 2005b).
Aside from the impact of E. coli and other bacteria on human health, there
are also potential impacts on the health of other fauna such as livestock. Many
organisms live in surface waters and are impacted by changes in water quality.
Organisms that live in small creeks and streams are particularly vulnerable to
changes in water quality as the volume of water is such that little dilution of
34
contaminants occurs. In agricultural areas there are many small, potentially
ephemeral, streams with inputs of fecal coliforms. The storage and transport of
bacteria in agricultural environments is an area of growing research interest
(Soupir and Mostaghimi 2011).
Understanding and predicting water quality through the application of
numerical models is an increasingly common occurrence in watershed
management. To accurately model the impact of BSA within surface waters,
background research must be undertaken to quantify their presence within a
variety of different environments. Field measurements of BSA provide both a
better understanding of bacterial and sediment transport and, in doing so, allow
for the development of improved water quality models (Soupir et al. 2010).
2.4 The Impact of Bacteria on Sediment Characteristics
The interaction between bacteria and sediment can be examined via two
distinct views. First, bacteria impact sedimentary processes such as settling
velocity, primarily due to the presence of flocculation (Droppo et al. 1997).
Secondly, the presence of sediment and the attachment of bacteria to sediment
can significantly alter the decay rate of bacteria (Russo et al. 2011).
The settling velocity of discrete clastic particles was first estimated by
(Stokes 1845). However, the equation developed by Stokes (1845) is entirely
theoretical and does not account for a number of conditions present outside the
laboratory environment. One limitation of Stokes law is that the equation
assumes all particles are perfectly spherical and do not interact with one
35
another. The shape and roundness of a particle are known to alter its settling
velocity (Hodder 2007) and as such Stokes’ law has limited application under
anything other than idealized laboratory conditions (Hodder 2009). A further
complication in the estimation of particle settling velocities is the presence of
flocculated particles.
Flocs, due to their increased surface area and assumed presence of
bacterial polymers, are competent at transporting nutrients and pollutants
(Wotton 2011). The creation of flocs alters sediment properties such as surface
area, settling rates and particulate densities while the sediment is in suspension.
The presence of these flocs alters the rate of settling in many glacial lakes, such
that theoretical predictions (e.g. Stokes’ Law) are inappropriate (Hodder 2009).
The increase in settling rate will also lead to enhanced deposition of flocs on
floodplains when rivers overflow their banks (Petticrew et al. 2011). Figure 2.1,
from Droppo (2001), shows an environmental scanning electron micrograph
(ESEM) of a floc from a highly modified fluvial environment, highlighting the
presence of bacteria through use of an arrow. Individual constituent particles in
flocs will settle slower alone than when part of a composite particle. The actual
settling rate depends on conditions such as flow velocity, fluid viscosity and floc
structure (Soupir et al. 2010, Petticrew et al. 2011). The increased settling rate of
flocs compared to individual particles is especially important in low velocity
mediums where the competence for transport is lower.
The presence of bacteria and bacterial polymers can continue to affect
sediment characteristics even after deposition has occurred. The presence of
36
EPS can make sedimentary deposits more cohesive and stable (Crabill et al.
1999, Lau and Droppo 2000), with a corresponding effect on bed sediment
erodibility. Bacterial interaction with sediments alters the properties and
characteristics of sediment particles, primarily through the creation of particles
with greater mass and volume, typically with a much lower density than that of
the constituent grains, thus resulting in a change of geomorphic process rates.
Altering the rate of sediment deposition will directly affect the creation of
laminated bed sediments, a paleoenvironmental proxy. The creation of
laminated sediments is acknowledged to occur in the context of a multivariate
process-network (Hodder and Gilbert 2013). However, the influence of bacteria
on the creation and sedimentation of lacustrine sediments has not been
addressed. A better understanding of the processes affecting the creation of
laminated sediments will allow for more accurate paleoenvironmental
reconstruction.
Composite particles have a greater settling velocity than their constituent
planktonic bacteria and sediment grains, and – all other factors being equal – the
composite particles is therefore more likely to be deposited. However, following
deposition, the composite particle does not necessarily remain stationary.
Increased bacterial concentration in flows is common when there is a period of
high-flow, potentially due to storm events. The effect of high-flow on bacterial
resuspension has been illustrated in natural rivers (Jamieson et al. 2005b),
overland runoff (Soupir et al. 2010) and urban storm-drain water (Characklis et
al. 2005). The increase in bacterial concentrations in water is primarily attributed
37
to the resuspension of bacterial colonies that have been protected in the bed.
Once resuspended, the bacteria impacts water quality. Due to their negligible
mass and constant supply of moving water, it is assumed that free-phase
bacteria are more mobile than sediment-associated bacteria (Characklis et al.
2005).
38
Figure 2.1. An ESEM image of a floc, with arrow labelling a bacterium. The largest objects visible in the image are diatoms. Source: Figure 3B in Droppo (2001; pg 1556).
39
2.4.1 The Effect of Sediment on Bacterial Properties
The interaction between bacteria and sediment also impacts bacterial
properties. For example, when bacteria are sediment associated, the sediment
can shield bacteria from predators (Yeung-Cheung et al. 2010). The shielding
that sediment-associated bacteria receive can slow decay rates, thus allowing
bacteria to persist longer in the environment (Russo et al. 2011). Although flocs
offer protection from predation, some organisms do feed on flocs and their
constituent particles (Logan and Hunt 1987). The increased concentrations of
bacteria that are found in bed sediments, relative to the water column, can be
resuspended and cause a human health risk if the water is used for drinking or
recreation (Yeung-Cheung et al. 2010).
The environment in which bacteria grow has the ability to alter their
characteristics, especially with regards to the production of EPS (Henry 2004).
Prior studies have illustrated that specific environmental conditions can lead to
both promotion of pili growth (Schillinger and Gannon 1985) and an increase in
the production of EPS (LeChevallier et al. 1984). The ability for bacteria to
produce fimbrae and EPS varies by species. The environment, therefore, may
alter the ability of bacteria to associate with sediment particles.
Another environmental stressor that can alter E coli decay rates is
temperature. When bacteria are sediment-associated, especially in bed
sediments, they are sheltered from sudden changes in temperature. Pachepsky
and Shelton (2011) found that bacteria present in finer sediments are less
sensitive to temperature changes than those in coarser sediments, thus those
40
found in finer sediments persist longer. The size of sediment that E. coli is
associated with has been shown to have direct impacts on the decay rate with
smaller, clay-sized particles leading to the slowest decay rate (i.e. longest
persistence); this has been attributed by some authors to increased protection
from predation in finer sediments (Wotton 2004, Droppo et al. 2009). Bacteria
associated with sediment are also protected from UV radiation, prolonging their
lifespan (Hartz et al. 2008, Xuan et al. 2010).
The presence of sediment has been found to promote bacterial growth.
Desmarais et al. (2002) found that the addition of sterile sediment to tropical
river water caused a significant increase in the amount of E coli and enterococci
over a 24-hour period. In an in situ study of Lake Onalaska in Northeastern USA,
(LaLiberte and Grimes 1982) found E. coli growth within lake bottom sediment
was higher than in the overlying water column.
2.5 Separating Adsorbed and Free-Phase Bacteria
Although there is awareness that bacteria and sediment interactions have
implications for water quality, the hydrodynamics of sediment transport and
bacterial survival, the process remains poorly understood. There is no standard
method to assess BSA. Recent studies have used unique methods (Table 2.1)
for determining the rate of partitioning and therefore it is difficult to compare the
results (Table 2.2).
Two distinct approaches have emerged as the predominant methods for
separating sediment-associated bacteria and planktonic bacteria. In some
41
studies fractional filtration and a chemical dispersant are used to separate
attached bacteria (Schillinger and Gannon 1985, Krometis et al. 2009) whereas
in some other studies a centrifuge is used (Characklis et al. 2005, Fries et al.
2006). The centrifugation technique relies on the difference in mass density
between bacteria (≈1.16 g/cm3 for E. Coli; Godin et al. 2007) and sediment (≈
2.65 g/cm3
Table 2.1
) to separate. While this allows for a quick sample processing, it does
not take into account the possibility of bacterial attachment to small particles that
will not be removed from suspension via centrifugation. There has yet to be a
standard time or force in studies using the centrifugation technique ( ).
In contrast, the filtration method relies on particle diameter instead of
differential density to separate sediment-associated bacteria and planktonic
bacteria. The use of the filtration method permits the capture of all cells that are
attached to particles of a given size. The method for enumerating bacteria is a
much more standardised procedure (Franson et al. 1998). Determination of the
sediment adsorbed fraction through the use of filtration methods is more uniform
and the methods have been more rigorously compared (Krometis et al. 2009,
Soupir et al. 2010). The sample containing sediment and bacteria is drawn
through a filter of known pore size that allows the passage of free-phase
bacteria, but traps particle associated bacteria. The size of 8 µm was used by
Krometis et al (2009) as E. coli cells have a dimension of approximately 1.1-1.5
µm x 2.0-6.0 µm (Holt et al. 2000). After filtration, a dispersant/surfactant, such
as Tween-85, is used to separate the bacteria from associated particles or
physical dispersion via sonication can also be used (Soupir et al. 2008). The
42
previously adsorbed bacteria are then available to be analyzed for the amount of
bacteria present using methods such as epifluorescence microscopy (Franson et
al. 1998).
Contemporary research on this topic has been limited mostly to
agricultural areas (Soupir et al. 2010, Soupir and Mostaghimi 2011) and primarily
to E. coli. Of interest to water quality modellers is the effect that different
environments will have on the magnitude of adsorption. Furthermore, although
E. coli is used frequently as an indicator species and has specific human health
impacts, the total concentration of bacteria is also of interest, as presumably
many species of bacteria are involved in the creation of composite particles – not
only E. coli.
There have been some recent attempts to incorporate sediment
associated bacteria into water quality models (Hipsey et al. 2008, Pandey et al.
2012). While the studies incorporated sediment attachment into their models, the
ranges of data provided made for a weak model. For example, the data used by
Hipsey et al (2008) was obtained from a single reservoir (Hipsey et al. 2006). As
the input data into the model is limited, the model is therefore unable to be
applied in different environments. Similarly, research done by Pandey et al
(2012) uses data from a single environment to estimate BSA within bed
sediments. Without more research being undertaken in different environments,
these models will be applicable only to very specific environments and
conditions. By quantifying the presence of BSA in contrasting environments, it is
43
hoped that this study will partially address this shortcoming of quantitative data
and provide a method that can be applied in future studies.
2.6 Agricultural beneficial management practices
In regions with high spatial concentrations of agriculture, the impact of
farming practices on the environment can be very large. Furthermore, certain
agricultural practices can have a negative economic impact on farmers due to
inefficiencies. For both of these reasons, beneficial management practices
(BMPs) are developed to minimize environmental impact and/or economic
losses. Environment-focused BMPs often are developed to address the
contribution of both point source and non-point source (NPS) pollution. The
realisation that non-point source pollution was having a major effect on water
quality, specifically with regards to phosphorus in lakes (Schindler 1974),
increased the profile and importance of developing BMPs. It is important to note
that not all BMPs provide environmental benefits. Some BMPs, such as winter
bale grazing, provide economic benefits to the farmer with unknown or
potentially negative environmental impacts (Jungnitsch et al. 2011).
2.6.1 Winter bale grazing
Cow manure is commonly used as a fertilizer, delivering nutrients to crops
during the summer growing season. The collection and distribution of this
manure is a labour and cost-intensive process. For this reason, many farmers
have started to winter their cows in crop fields over winter, largely reducing the
need for manure collection and spraying during the summer (Jungnitsch et al.
44
2011). The environmental impacts of this practice have been documented in
temperate regions and include soil compaction (Trimble and Mendel 1995,
Nguyen et al. 1998), increased presence of bacteria (Belsky et al. 1999) and
increased sediment mobilization from hoof erosion (Trimble and Mendel 1995,
Nguyen et al. 1998, Belsky et al. 1999).
Wintering cattle in-field is not a new practice, but it has recently returned
to favour due to increasing fertilizer and transportation costs (Jungnitsch et al.
2011). When both nutrient and transport costs were cheaper during the mid-20th
century, winter bale grazing was less common (Vanderholm 1979). Although this
practice does provide economic benefits, the environmental impacts are largely
unknown, especially in the Canadian Prairies where the manure is deposited on
snow or frozen soil until the spring nival melt occurs.
45
2.7 Literature cited
Bache, D.H., Gregory, R. 2010. Flocs and separation processes in drinking water treatment: a review. Journal of Water Supply: Research and Technology—AQUA, 59: 16.
Belsky, A.J., Matzke, A., Uselman, S. 1999. Survey of livestock influences on
stream and riparian ecosystems in the western United States. Journal of Soil and Water Conservation, 54: 419–431.
Benham, B.L., Baffaut, C., Zeckoski, R.W., Mankin, K.R., Pachepsky, Y.A.,
Sadeghi, A.M., Brannan, K.M., Soupir, M.L., Habersack, M.J. 2006. Modeling bacteria fate and transport in watersheds to support TMDLs. Transactions of the ASABE, 49: 987–1002.
Berry, D.F., Hagedorn, C. 1991. Soil and groundwater transport of
microorganisms. In, Assessing ecological risks of biotechnology. Butterworth-Heinemann, Boston. Edited by L.R. Ginzburg.
Castelain, M., Sjöström, A., Fällman, E., Uhlin, B., Andersson, M. 2010.
Unfolding and refolding properties of S pili on extraintestinal pathogenic Escherichia coli. European Biophysics Journal, 39: 1105–1115.
Characklis, G.W., Dilts, M.J., Simmins, O.D., III, Likirdopulos, C.A., Krometis, L.-
A.H., Sobsey, M.D. 2005. Microbial partitioning to settleable particles in stormwater. Water Research, 39: 1773–1782.
Crabill, C., Donald, R., Snelling, J., Foust, R., Southam, G. 1999. The impact of
sediment fecal coliform reservoirs on seasonal water quality in Oak Creek, Arizona. Water Research, 33: 2163–2171.
Desmarais, T.R., Solo-Gabriele, H.M., Palmer, C.J. 2002. influence of soil on
fecal indicator organisms in a tidally influenced subtropical environment. Applied and Environmental Microbiology, 68: 1165–1172.
Droppo, I.G. 2001. Rethinking what constitutes suspended sediment.
Hydrological Processes, 15: 1551–1564. Droppo, I.G. 2004. Structural controls on floc strength and transport. Canadian
Journal of Civil Engineering, 31: 569–578. Droppo, I.G., Jeffries, D., Jaskot, C., Backus, S. 1998. The prevalence of
freshwater flocculation in cold regions: a case study from the Mackenzie River Delta, Northwest Territories, Canada. Arctic, 51: 155–164.
46
Droppo, I.G., Krishnappan, B.G., Liss, S.N., Marvin, C., Biberhofer, J. 2011. Modelling sediment-microbial dynamics in the South Nation River, Ontario, Canada: Towards the prediction of aquatic and human health risk. Water Research, 45: 3797–3809.
Droppo, I.G., Leppard, G.G., Flannigan, D.T., Liss, S.N. 1997. The freshwater
floc: A functional relationship of water and organic and inorganic floc constituents affecting suspended sediment properties. Water, Air, and Soil Pollution, 99: 43–54.
Droppo, I.G., Liss, S.N., Williams, D., Nelson, T., Jaskot, C., Trapp, B. 2009.
Dynamic existence of waterborne pathogens within river sediment compartments. Implications for water quality regulatory affairs. Environmental Science & Technology, 43: 1737–1743.
Droppo, I.G., Ongley, E.D. 2004. Flocculation of suspended sediment in rivers of
southearstern Canada. Water Resources, 28: 1799–1809. Franson, M.A.H., Clesceri, L.S., Greenberg, A.E., Eaton, A.D. (Editors). 1998.
Standard methods for the examination of water and wastewater. 20 edition. American Public Health Association, Washington, DC.
Fries, J.S., Characklis, G.W., Noble, R.T. 2006. Attachment of fecal indicator
bacteria to particles in the Neuse River Estuary, N.C. Journal of Environmental Engineering, 132: 1338–1345.
Gallagher, D.L., Lago, K., Hagedorn, C., Dietrich, A.M. 2013. Effects of Strain
Type and Water Quality on Soil-Associated Escherichia coli. International Journal of Environmental Science and Development, 4: 25–31.
Godin, M., Bryan, A.K., Burg, T.P., Babcock, K., Manalis, S.R. 2007. Measuring
the mass, density, and size of particles and cells using a suspended microchannel resonator. Applied Physics Letters, 91: 123121.
Goulder, R. 1977. Attached and free bacteria in an estuary with abundant
suspended solids. Journal of Applied Bacteriology, 43: 399–405. Guber, A.K., Pachepsky, Y.A., Shelton, D.R., Yu, O. 2007. Effect of bovine
manure on fecal coliform attachment to soil and soil particles of different sizes. Applied and Environmental Microbiology, 73: 3363–3370.
Hartz, A., Cuvelier, M., Nowosielski, K., Bonilla, T.D., Green, M., Esiobu, N.,
McCorquodale, D.S., Rogerson, A. 2008. Survival potential of Escherichia coli and Enterococci in subtropical beach sand: implications for water quality managers. Journal of Environment Quality, 37: 898.
47
Henry, L.-A. 2004, April 27. Partitioning between the soil-adsorbed and planktonic phases of Escherichia coli. MSc Thesis. Virginia Tech. Blacksburg, Virginia.
Hipsey, M.R., Antenucci, J.P., Brookes, J.D. 2008. A generic, process-based
model of microbial pollution in aquatic systems. Water Resources Research, 44: 1–26.
Hipsey, M.R., Brookes, J.D., Regel, R.H., Antenucci, J.P., Burch, M.D. 2006. In
situ Evidence for the Association of Total Coliforms and Escherichia coli with Suspended Inorganic Particles in an Australian Reservoir. Water, Air, and Soil Pollution, 170: 191–209.
Hodder, K.R. 2007, June. The process-network of alpine, glacier-fed lakes with
particular reference to flocculation. PhD thesis. Queen's University. Kingston, Ontario.
Hodder, K.R. 2009. Flocculation: a key process in the sediment flux of a large,
glacier-fed lake. Earth Surface Processes and Landforms, 24: 1151–1163. Hodder, K.R., Gilbert, R. 2007. Evidence for flocculation in glacier-fed Lillooet
Lake, British Columbia. Water Research, 41: 2748–2762. Hodder, K.R., Gilbert, R. 2013. Physical properties of lake sediments. In,
Encyclopedia of Quaternary Science. Elsevier, Amsterdam. Edited by S.A. Elias.
Holt, J.G., Krieg, N.R., Sneath, P.H.A., Staley, J.T., Williams, S.T. (Editors).
2000. Bergey’s manual of determinative bacteriology. 9 edition. Lippincott Williams & Williams, Philadelphia, PA.
Huang, G.H., Xia, J. 2001. Barriers to sustainable water-quality management.
Journal of Environmental Management, 61: 1–23. Iwane, T., Urase, T., Yamamoto, K. 2001. Possible impact of treated wastewater
discharge on incidence of antibiotic resistant bacteria in river water. Water Science and Technology, 43: 91–99.
Jamieson, R., Joy, D.M., Lee, H., Kostachuk, R., Gordon, R. 2005a. Transport
and deposition of sediment-associated Escherichia Coli in natural streams. Water Research, 39: 2665–2675.
Jamieson, R.C., Joy, D.M., Lee, H., Kostachuk, R., Gordon, R.J. 2005b.
Resuspension of sediment-associated Escherichia Coli in a natural stream. Journal of Environmental Quality, 34: 581–589.
48
Jungnitsch, P.F., Schoenau, J.J., Lardner, H.A., Jefferson, P.G. 2011. Winter feeding beef cattle on the western Canadian prairies: Impacts on soil nitrogen and phosphorus cycling and forage growth. Agriculture, Ecosystems & Environment, 141: 143–152.
Kay, M.K., Erwin, T.C., McLean, R.J.C., Aron, G.M. 2011. Bacteriophage
ecology in Escherichia coli and Pseudomonas aeruginosa mixed-biofilm communities. Applied and Environmental Microbiology, 77: 821–829.
Krometis, L.-A.H., Dillaha, T.A., Love, N.G., Mostaghimi, S. 2009. Evaluation of
a filtration/dispersion method for enumeration of particle-associated Escherichia coli. Journal of Environment Quality, 38: 980.
LaLiberte, P., Grimes, D.J. 1982. Survival of Escherichia coli in lake bottom
sediment. Applied and Environmental Microbiology, 43: 623–628. Lanoil, B., Skidmore, M., Priscu, J.C., Han, S., Foo, W., Vogel, S.W., Tulaczyk,
S., Engelhardt, H. 2009. Bacteria beneath the West Antarctic Ice Sheet. Environmental Microbiology, 11: 609–615.
Lau, Y.L., Droppo, I.G. 2000. Influence of antecedent conditions on critical shear
stress of bed sediments. Water Research, 34: 663–667. LeChevallier, M.W., Hassenauer, T.S., Camper, A.K., McFeters, G.A. 1984.
Disinfection of bacteria attached to granular activated carbon. Applied and Environmental Microbiology, 48: 918–923.
Li, X.Y., Yang, S.F. 2007. Influence of loosely bound extracellular polymeric
substances (EPS) on the flocculation, sedimentation and dewaterability of activated sludge. Water Research, 41: 1022–1030.
Ling, T.Y., Achberger, E.C., Drapcho, C.M., Bengston, R.L. 2002. Quantifying
adsorption of an indicator bacteria in a soil-water system. Transactions of the ASAE, 45: 669–674.
Logan, B.E., Hunt, J.R. 1987. Advantages to microbes of growth in permeable
aggregates in marine systems. Limnology and Oceanography, 32: 1034–1048.
Mallin, M.A., Williams, K.E., Esham, E.C., Lowe, R.P. 2000. Effect of human
development on bacteriological water quality in coastal watersheds. Ecological Applications, 10: 1047–1056.
49
Nguyen, M.L., Sheath, G.W., Smith, C.M., Cooper, A.B. 1998. Impact of cattle treading on hill land: 2. Soil physical properties and contaminant runoff. New Zealand Journal of Agricultural Research, 41: 279–290.
Nič, M., Jirát, J., Košata, B., Jenkins, A., McNaught, A. (Editors). 2009. IUPAC
Compendium of Chemical Terminology. 2nd edition. IUPAC, Research Triangle Park, NC.
Pandey, P.K., Soupir, M.L., Rehmann, C.R. 2012. A model for predicting
resuspension of Escherichia coli from streambed sediments. Water Research, 46: 115–126.
Petticrew, E.L., Rex, J.F., Albers, S.J. 2011. Bidirectional delivery of organic
matter between freshwater and marine systems: the role of flocculation in Pacific salmon streams. Journal of the North American Benthological Society, 30: 779–786.
Pikuta, E.V., Hoover, R.B., Tang, J. 2007. Microbial extremophiles at the limits of
life. Critical Reviews in Microbiology, 33: 183–209. Rehmann, C.R., Soupir, M.L. 2009. Importance of interactions between the
water column and the sediment for microbial concentrations in streams. Water Research, 43: 4579–4589.
Russo, S.A., Hunn, J., Characklis, G.W. 2011. Considering bacteria-sediment
associations in microbial fate and transport modeling. Journal of Environmental Engineering, 137: 697–706.
Schillinger, J.E., Gannon, J.J. 1985. Bacterial adsorption and suspended
particles in urban stormwater. Journal of the Water Pollution Control Federation, 57: 384–389.
Schindler, D.W. 1974. Eutrophication and recovery in Experimental Lakes:
implications for lake management. Science, 184: 897–899. Sharp, M., Parkes, J., Cragg, B., Fairchild, I.J., Lamb, H., Tranter, M. 1999.
Widespread bacterial populations at glacier beds and their relationship to rock weathering and carbon cycling. Geology, 27: 107–110.
Skidmore, M.L., Foght, J.M., Sharp, M.J. 2000. Microbial Life beneath a high
arctic glacier. Applied and Environmental Microbiology, 66: 3214–3220. Soupir, M.L., Mostaghimi, S. 2011. Escherichia coli and Enterococci attachment
to particles in runoff from highly and sparsely vegetated grassland. Water, Air, and Soil Pollution, 216: 167–178.
50
Soupir, M.L., Mostaghimi, S., Dillaha, T. 2010. Attachment of Escherichia coli and Enterococci to particles in runoff. Journal of Environment Quality, 39: 1019–1027.
Soupir, M.L., Mostaghimi, S., Love, N.G. 2008. Method to partition between
attached and unattached E. coli in runoff From agricultural lands. Journal of the American Water Resources Association, 44: 1591–1599.
Stokes, G.G. 1845. On the effect of the internal friction of fluids on the motion of
pendulums. Transactions of the Cambridge Philosophical Society, 9: 8–106. Trimble, S.W., Mendel, A.C. 1995. The cow as a geomorphic agent — A critical
review. Geomorphology, 13: 233–253. Vanderholm, D.H. 1979. Handling of manure from different livestock and
management systems. Journal of Animal Science, 48: 113–120. Viles, H.A. 2012. Microbial geomorphology: A neglected link between life and
landscape. Geomorphology, 157–158: 6–16. Whitman, W.B., Coleman, D.C., Wiebe, W.J. 1998. Prokaryotes: The unseen
majority. Proceedings of the National Academy of Sciences, 95: 6578–6583. Woodward, J.C., Porter, P.R., Lowe, A.T., Walling, D.E., Evans, A.J. 2002.
Composite suspended sediment particles and flocculation in glacial meltwaters: preliminary evidence from Alpine and Himalayan basins. Hydrological Processes, 16: 1735–1744.
Wotton, R.S. 2004. The ubiquity and many roles of exopolymers (EPS) in
aquatic systems. Scientia Marina, 68: 13–21. Wotton, R.S. 2011. EPS (Extracellular Polymeric Substances), silk, and chitin:
vitally important exudates in aquatic ecosystems. Journal of the North American Benthological Society, 30: 762–769.
Xuan, W., Bin, Z., Zhiqiang, S., Zhigang, Q., Zhaoli, C., Min, J., Junwen, L.,
Jingfeng, W. 2010. The EPS characteristics of sludge in an aerobic granule membrane bioreactor. Bioresource Technology, 101: 8046–8050.
Yeung-Cheung, A.K., Chu, P., Dega, J. 2010. Comparison of bacterial levels
from water and sediments among upper and lower areas of Guion Creek. Proceedings of the Annual International Conference on Soils, Sediments, Water and Energy: Vol. 14, Article 25.
Zabawa, C.F. 1978. Microstructure of agglomerated suspended sediments in
Northern Chesapeake Bay estuary. Science, 202: 49–51.
51
3 THE IMPACT OF MANAGEMENT PRACTICES ON THE
PREVALENCE OF BACTERIA-SEDIMENT ASSOCIATIONS IN
SOUTHEASTERN SASKATCHEWAN
David C Barrett* and Kyle R. Hodder
Prairie Environmental Process Laboratory, Department of Geography, University
of Regina, Regina, SK
S4S0A2
*Corresponding author [[email protected]]
3.1 Abstract
Agricultural watersheds often have many sources of nonpoint source
pollution, especially in the form of bacteria. These bacteria when in the presence
of sediment often interact to form complex, cohesive particles, or flocs. The
movement of these complex particles through fluvial systems is poorly
understood and lacks a standardized technique for measurement. This study
evaluated the feasibility of applying a novel method of quantifying bacteria-
sediment associations (BSA) in an agricultural watershed of Southeastern
Saskatchewan, Canada. This method is aimed at addressing the gap in previous
literature by providing a standardized, simple method for quantifying BSA in all
types of environments. Samples taken from 12 microwatersheds and from within
Pipestone Creek of southeast Saskatchewan, Canada show that BSA are both
present and abundant with the percentage of bacteria associated with sediment
ranging from 2% within farm microwatersheds up to 70% in Pipestone Creek.
52
Results from the study also reveal that winter bale grazing (WBG) of cattle, when
compared with fall manure spreading (FMS) practices and no manure spreading,
leads to a larger export of total bacteria during the spring melt period.
Furthermore, the practice of winter bale grazing was found to affect sediment
characteristics; namely an increase in total sediment concentration but an overall
decrease in mean-particle diameter. The quantification of BSA in this study helps
to evaluate the environmental impacts of WBG in a Canadian prairie
environment and leads to a better understanding of bacteria-sediment transport
dynamics.
3.2 Introduction
Due to increasing stresses on freshwater from both population growth and
environmental change (Arnell 1999, Vörösmarty et al. 2000, Petticrew et al.
2011) the attention to freshwater contamination is becoming ever more important
(Arnell 2004, Li and Yang 2007). On the Canadian Prairies, intensive agricultural
practices have negative impacts on the water quality of surface water and lakes
(Hall et al. 1999, Characklis et al. 2005). Of major concern to water users
downstream of agricultural areas is the presence of pathogenic bacteria, such as
Escherichia coli, found in the manure distributed over fields as a fertilizer
(Jamieson et al. 2005a, Soupir et al. 2010, Kelln et al. 2011, Jungnitsch et al.
2011). Knowing the path and transport processes of bacteria from source to sink
is increasingly important and aids in the creation and honing of water quality
modelling programs.
53
While it is known that bacteria are often associated with sediment particles
(Schillinger and Gannon 1985, Ling et al. 2002, Characklis et al. 2005, Jamieson
et al. 2005b, Fries et al. 2006, Soupir et al. 2010, Soupir and Mostaghimi 2011),
there is yet to be a standard method developed to measure such associations.
However, some commonalities are observed between these studies. Most prior
studies employed either a) centrifugation or b) fractional filtration, to separate
planktonic organisms from those that were sediment associated. Studies using
centrifugation (eg. Ling et al. 2002, Characklis et al. 2005, Fries et al. 2006,
Krometis et al. 2009) separate free-phase bacteria from sediment-associated
bacteria by density. In contrast, studies that use fractional filtration (eg
(Schillinger and Gannon 1985, Auer and Niehaus 1993, Mahler et al. 2000, Jeng
et al. 2005, Soupir and Mostaghimi 2011) separate purely based on organism
size, operating under the assumption that most bacteria are approximately 1-2
μm in size (Henry 2004). Bacteria-sediment associations have been found to be
present in a range of different environments. In many cases, it has been noted
that bacteria preferentially associate with sediment particles, if available, as
opposed to remaining planktonic (Soupir et al. 2008).
Prior to this study, it appears that all previous studies were undertaken at
laboratory-proximal sites and therefore could take advantage of a fairly short
duration between sample collection and laboratory analysis. The attempt at
creating a standard method by Soupir (2008) could not be applied in this study,
as it did not allow for account for samples requiring extended transport before
undergoing lab analysis. E. coli experience a quick cell division rate,with division
54
occurring at a rate of 0.05 – 1 divisions per hour (Kubitschek and Freedman
1971) and as such lab analysis is time sensitive unless the organisms are fixed
to prevent growth and decay.
The lack of a standard method for evaluating bacteria-sediment
associations presents a challenge for comparing prior study results. Remote
environments that are not within close proximity to a laboratory require a
different technique. Furthermore there is a need for a method of measuring the
partitioning between sediment associated and planktonic bacteria that can be
used in remote areas without immediate access to laboratory microscopy
equipment and electricity.
The aims of this study were: 1) to develop a method for separating free-
phase bacteria from sediment-associated that is suitable for remote
environments, in the absence of electricity and microscopy equipment, 2) to use
the new technique to quantify the presence of BSA in an agricultural watershed
and 3) to examine the impact of WBG on the presence of BSA.
Furthermore, by measuring the interaction of bacteria with sediment
particles, the authors further document the presence of complex flocculated
particles in freshwater environments. The BSA ratio is calculated by comparing
the number of planktonic organisms to those that are sediment-associated
(Equation 3.1). While the presence of flocculated particles in freshwater lakes
(Hodder and Gilbert 2007) and rivers (Droppo and Ongley 1992, Jamieson et al.
2005b) is acknowledged, the majority of research on flocs has been undertaken
in salt-water environments such as estuaries (Goulder 1977, Jiang et al. 2011)
55
and estuaries (Droppo et al. 1998). The presence of bacteria within flocculated
particles, and as such BSA, has been documented in lowland fluvial
environments (Droppo and Ongley 2004, Droppo et al. 2009, Soupir et al. 2010,
Wotton 2011). Nevertheless, examination of the watershed-scale spatial and
temporal variability of BSA has yet to be undertaken.
Equation 3.1 The equation used for calculating the bacteria-sediment association ratio (BSA).
𝐵𝑆𝐴 =𝑆𝑒𝑑𝑖𝑚𝑒𝑛𝑡 − 𝑎𝑠𝑠𝑜𝑐𝑖𝑎𝑡𝑒𝑑 𝑜𝑟𝑔𝑎𝑛𝑖𝑠𝑚𝑠
𝑇𝑜𝑡𝑎𝑙 𝑜𝑟𝑔𝑎𝑛𝑖𝑠𝑚𝑠
Within agricultural watersheds, non-point source pollution is a major
concern (Logan 1993, Nagels et al. 2002). Non-point source pollution in
agricultural regions is due to the presence of pastured or wild animals and the
distribution of manure on fields as a fertiliser (Gagliardi and Karns 2000). The
development of agricultural beneficial management practices (BMPs) by
regulating bodies, such as governments, aims to promote practices that are a
positive benefit to producers both environmentally and economically. The
practice of winter bale-grazing cattle on pasturelands has been undertaken for
almost a century, however, the environmental impacts have only recently been
examined in detail (Jungnitsch et al. 2011) and remain fairly poorly understood
(Owens and Shipitalo 2009). Cattle wintered in-field naturally distribute their
feces on fields. Since this manure is rich in nutrients, it acts as a natural fertiliser
and is a desirable alternative to costly chemical fertilisers for farmers. Also, the
wintering of cattle directly on pastureland, as opposed to a confined system,
56
reduces the cost of transporting and manually distributing the manure as
fertiliser.
While studies such as Powell et al. (1998) noted the increase in crop
growth associated with direct depositing of feces on cropland, little is discussed
regarding the export of sediment or bacteria from this practice. White et al.
(2001) found that dairy cow excretions were spread evenly over pasture areas,
especially during winter periods. It should be noted, however, that both the
aforementioned studies took place in regions not representative of the Canadian
Prairies. To the knowledge of the author, the only study examining the viability of
in-field wintering of cattle in the semi-arid Canadian Prairies focused on nutrient
cycling (Jungnitsch et al. 2011). Jungnitsch et al. (2011) acknowledge that there
may be an increased potential for nutrient runoff in locations that have
seasonally frozen ground but this issue awaits further investigation.
Intensive cattle operations are known to have negative impacts on the
environment. The grazing of cattle on unfrozen soil is known to compact the
ground thereby reducing infiltration (Trimble and Mendel 1995, Nguyen et al.
1998). Furthermore, an increase in the delivery of bacteria has been noted due
to the presence of grazing cattle (Belsky et al. 1999). Hoof erosion, or ‘pugging’,
also leads to an increase in sediment mobilisation to nearby fluvial environments
(Trimble and Mendel 1995, Nguyen et al. 1998, Belsky et al. 1999). Other
negative impacts on the environment can include an increase in water
temperature due to loss of riparian vegetation, a potential decrease in dissolved
57
oxygen levels and reduced streambank stability in nearby fluvial systems (Belsky
et al. 1999).
In temperate regions, the environmental impacts of over-wintering cattle in
fields are better studied. The major issues in these environments are soil
compaction, heavy spring precipitation and a lack of vegetation during winter
(Warren et al. 1986, Owens et al. 1997, Owens and Shipitalo 2009, Flores and
Tracy 2012). The Canadian Prairies have very different climatic conditions
during winter months and as such these previous studies may not be entirely
applicable. The presence of frozen soil and persistent snowcover may cause
different environmental issues to be more prominent. For example, factors such
as freeze-thaw cycles (Fouli et al. 2013) and magnitude of snowmelt may be of
greater importance in the Canadian Prairies. Although impacts in temperate
regions are well documented in the literature, previous studies examining the
viability of winter-bale grazing cattle in cold weather environments have not
discussed them in great detail. The lack of prior research is of concern as WBG
becomes a more prominent management practice in the Canadian Prairies
(Jungnitsch et al. 2011).
3.3 Study site
This study took place within the Pipestone Creek watershed, located in
southeastern Saskatchewan (Figure 1.1). The watershed drains 3684 km2 of low
relief prairie pothole topography. Land use within the watershed is primarily
agricultural, with crop, forage and grasslands constituting 92% of the total area
58
(PFRA, 2001). The main waterway within the watershed is Pipestone Creek,
which has discharge driven primarily by precipitation with peak discharge
occurring during snowmelt (Cade-Menun et al. 2013). In conjunction with the
Watershed Evaluation of Beneficial Management Practices (WEBs) project lead
by Agriculture and AgriFood Canada, there were 6 in-field sample collection
locations used. Of the six sites in the study, 2 had winter-bale grazing occurring
on them, 2 had a fall manure-spreading regime applied and 2 were control sites
with no manure application (Table 3.1).
To examine spatial changes of BSA within the watershed, 3 locations
within Pipestone Creek were sampled. The first sample site in the Pipestone
Creek was located near the study farms. The furthest downstream location was
chosen downstream of all in-field sites, near Moosomin Reservoir in an attempt
to capture watershed-scale spatial variability. The second sample location was
chosen as a middle point between the other two sample locations. To facilitate
sampling each Pipestone Creek sample was within 10 metres of a local road
crossing.
59
Table 3.1 The agricultural practices undertaken at each of the field locations
Farm Site Area (ha) Treatment
M 2 1.94 Winter bale-grazing
3 1.21 Winter bale-grazing
4 1.52 Perennial pasture control
5 0.48 Spreading of manure
6 0.71 Spreading of manure
B 14 2.24 Perennial pasture control
60
3.4 Methods
Past studies employing fractional filtration methods were used to develop
the technique for separating bacteria-associated sediment from free-phase
organisms (Auer and Niehaus 1993, Soupir et al. 2008, Krometis et al. 2009,
Soupir et al. 2010). The initial bacterial treatment which allowed for remote
sampling and less time-sensitive samples was adapted from Franson et al.
(1998). The suggested method for adding fixative from Franson et al. (1998), did
not account for the storage of membranes that were used to quantify sediment-
associated bacteria. As such, phosphate buffer was used in lieu of liquid sample
to keep cells on the membrane hydrated during transport. Methods for
enumerating bacteria using acridine orange and fluorescence microscopy were
as outlined in Franson et al. (1998). The evaluation of these different methods
combined with in-lab testing allowed for the novel method, described below, to
be developed and applied.
3.4.1 Sample Collection
The majority of samples were collected using ISCO 6712 automatic water
samplers. While some disturbance of the largest agglomerations of sediment
and bacteria is expected during sampling of any kind (Eisma 1986, Hodder and
Gilbert 2007) it is assumed that the majority of bacteria-sediment associations
will be unaffected by sampling due to their strong cohesive bonds (Droppo et al.
2009, Wotton 2011).
61
Flow-triggered automatic water samplers collected field runoff hourly
during snowmelt- or rainfall-induced runoff. To ensure adequate capture of
runoff, flumes were constructed at a central sampling point. Grab samples were
also taken at three locations (Figure 1.1; Locations 1, 2, 3) along Pipestone
Creek. These grab samples were collected at three week intervals during the
summer of 2012 to capture seasonal shifts in water quality within the main stem.
Both automated samples and grab samples were transported from the collection
sites to the laboratory within 24-hours, at which time they were processed using
the procedures outlined below.
3.4.2 Partitioning sediment-associated bacteria from free-phase
bacteria
Samples were filtered through a polycarbonate membrane with 8 μm
pores using a Millipore Sterifil Aseptic system according to the method of
Krometis et al. (2009). The methods for this study, and that done by Krometis et
al (2009), were developed based on Escherichia coli as a fecal coliform indicator
species and widely documented presence of this species in agricultural
environments (Baxter-Potter and Gilliland 1988, Collins and Rutherford 2004,
Guber et al. 2007). Furthermore, a majority of bacterial species have a long-axis
length of a few microns, similar to E. coli (Madigan et al. 2009). An 8 μm pore
ensures planktonic E. coli pass through the filter, whereas particle-associated
bacteria are retained on the membrane surface (Mahler et al. 2000). For this
reason, the filtrate that passed through the membrane was assumed to
62
exclusively contain planktonic bacteria whereas those that were retained on the
membrane were assumed to be particle-associated. Mahler et al. (2000)
reported that 99% of planktonic E. Coli passed through an 8 μm membrane. Post
filtration, the membrane and 9 ml of filtrate were placed in separate glass vials to
which 9 ml of phosphate buffer (KH2PO4
To these vials, a solution of glutaraldehyde was added at a final
concentration of 0.5% (w/v) as a fixative. Cells that are immersed in fixative do
not exhibit significant changes in frequency for up to three weeks (Hopwood
1969, 1972, Geesey et al. 1977, Kepner and Pratt 1994, Franson et al. 1998,
Bacallao et al. 2006). All samples were transported from the field and analysed
within this three-week period. The majority of samples were analysed within 48
hours of sample collection. The maximum interval between fixation and analysis
was 5 days.
) was added to keep cells hydrated
(Franson et al. 1998).
3.4.3 Enumeration and quantification of total bacteria
The procedure for doing total microbial counts followed the technique
outlined by Franson et al. (1998). All samples were enumerated using
fluorescence microscopy within the three-week window for sample analysis. To
the fixed samples, acridine orange fluorochrome was added to a final
concentration of 0.05% (w/v). A two minute fluorochrome exposure period was
used to ensure adequate attachment of dye to bacterial RNA and DNA (Jones
1974, Hobbie et al. 1977, Kepner and Pratt 1994). The solutions were then
vacuum-filtered through a 0.2 μm polycarbonate membrane and the filtration
63
apparatus was rinsed with phosphate buffer. Between samples, filtration
equipment was flame sterilized to reduce cross-contamination.
Air-dried 0.2 μm membranes (n = 96) were mounted on microscope slides
using Cargille type-A immersion oil (refractive index = 1.604) and examined at
1000x magnification. To induce acridine orange fluorescence, a fluorescent lamp
(mercury bulb) was paired with a 465-495 nm excitation filter, and images of the
fluorescing bacteria were captured using an Olympus DP70 camera. Once a
visual comparison of at least 10 fields of view established that bacteria were
relatively evenly distributed and visible on the membrane, images of three fields
of view were captured. Three different fields on the membrane were captured
chosen in an attempt to capture small variations present due to unequal
dispersion.
Images were processed using a semi-automated process involving the
ImageJ software package (http://rsbweb.nih.gov/ij/) to obtain total bacteria
counts. To ensure optimal detection, automated counts were manually verified to
ensure the software detected all visible bacteria. This verification was
undertaken for all images captured. Bacterial counts, per unit area of membrane,
were then used to calculate the total bacteria concentration per unit volume.
3.4.4 Measurement of particle size, suspended sediment
concentration and water quality
A Laser In-Situ Scattering and Transmissometry (LISST-C) instrument
(Sequoia Scientific) was used to measure particle-size distribution. The LISST
64
device uses laser backscatter measurements to measure the concentration of
suspended particles across 32 logarithmically-spaced size classes (Agrawal and
Pottsmith 2000). Particle-sizes reported in this study are the median values (n in
μm) for each of the size classes. The laser scattering is then used to calculate
the volume concentration (VC in μl l-1
Equation 3.2
) in each of the output size classes. From
the LISST device output matrices, the volume moment mean diameter for each
sample was calculated (VMM) using (adapted from (Hodder 2009);
variable titles adapted to make consistent within this paper).
Equation 3.2 Calculation of volume moment mean diameter (VMM). VC is volume concentration (μl l-1
), n is median size for each of the 32 bins (μm)
𝑉𝑀𝑀 = � 𝑉𝐶𝑛
500
𝑛=2.5
𝑛 ∕ � 𝑉𝐶𝑛
500
𝑛=2.5
Each sample was processed using the LISST device, initially without
dilution to obtain a particle-size distribution (PSD). If the optical transmission was
<30% the samples were diluted to allow for adequate laser transmission. The
dilution factor was used to determine undiluted concentration. Samples were
vacuum filtered through pre-weighed 0.4 μm polycarbonate membranes and air-
dried for 24 hours to determine suspended sediment concentration (SSC) using
a precision balance (± 0.2 mg).
3.5 Results
The sampling undertaken in this study focused on examining temporal
trends of BSA both at the in-field and in-reach scale. Samples from the
65
Pipestone creek exhibited BSA ratios that were much greater than in-field
samples (Figure 3.1). No spatial pattern was detectable from in-field sample
locations in 2012. Within Pipestone Creek, however, some spatiotemporal
patterns are visible from the tri-weekly sampling regime. Table 3.2 shows that
creek samples exhibited a decreasing BSA ratio with distance downstream on
May 3rd, however the opposite trend (increasing BSA) was evident in the May
24th samples.
66
Table 3.2 A comparison of BSA ratios at three sample locations within the Pipestone Creek main stem on three sample days. Missing samples due to analysis errors resulting in unusable data.
April 12,
2012 May 3,
2012 May 24,
2012
Pipestone Creek sample location 1 - 0.54 0.19
Pipestone Creek sample location 2 0.77 0.51 0.64
Pipestone Creek sample location 3 - 0.44 0.69
67
Figure 3.1 BSA from all in-field sampling locations and the Pipestone Creek during 2012 and 2013. Red line indicates mean, black line indicates median, box is 25-75th percentile, whiskers represent 90th and 10th percentile, black dots are outliers.
68
Within the watershed, there was a notable spatial distinction between the
in-field samples and samples from within the main stem. At the smallest class
size, in-field samples showed a weak negative correlation between
concentration and BSA (Figure 3.2) whereas main stem samples showed a
positive correlation (Figure 3.3). Conversely, for particles within the size class of
88.2 μm, an increase in concentration related to an increase in BSA for the main
stem samples and the in-field samples. Both sample scales showed a decrease
in BSA corresponding to a decrease in volume concentration at the 462 μm size
class.
One of the in-field sample sites captured a mid-summer storm event
(Figure 3.4) and a definitive peak in BSA occurs at hour 8 of the storm with a
rapid decrease immediately following. Peak BSA values that occur during the
rainstorm are below the mean value found during snowmelt.
Samples processed from 2013 snowmelt showed no significant difference
(P>0.05; T-test) between sites that had winter bale grazing (WBG) occurring
compared to fall manure spreading (FMS) and control (no manure) sites (Figure
3.5). There was, however, a statistically significant (p=0.035; T-test) difference
between WBG and fall manure spreading sites with respect to total organism
concentration with WBG sites having higher concentrations of organisms per liter
sampled (Figure 3.6). Total organism transport from in-field locations varied
temporally at the WBG sites, FMS sites and control sites (Figure 3.6) with all
displaying an increase in organism mobilization as snowmelt proceeded.
69
Figure 3.2 The relationship between BSA and volume concentration within the Pipestone Creek watershed, at the edge-of-field scale for three different particle sizes. Linear fit curve applied; n=20.
r = 0.15
r = 0.12
r = 0.36
70
Figure 3.3 The relationship between BSA and volume concentration within Pipestone Creek for three different particle sizes. Linear fit curve applied; n=7.
r = 0.09
r = 0.27
r = 0.46
71
Figure 3.4 BSA and local precipitation from an in-field location during a storm runoff event in summer, 2012
72
Figure 3.5 Bacteria-sediment associations measured in winter bale grazing, manure spreading and control sites during spring snowmelt. Box plot parameters same as Figure 3.1.
73
Figure 3.6 Concentrations of organisms per unit volume at winter bale grazing, manure spreading and control sites on during spring snowmelt. Box plot parameters same as Figure 3.1.
74
The different in-field management practices also resulted in statistically
significant change in sediment concentration and particle size. Total sediment
concentrations at the WBG locations were significantly (p=.031) greater than at
FMS or control sites (Figure 3.7). Notably, at the WBG sites, the VMM was
significantly smaller than at the FMS or control sites (p=<0.001; Figure 3.8)
indicating that while there was an increased volume of sediment being
transported from the WBG sites, the average particle size was 1.2 times smaller
than the FMS sites.
75
Figure 3.7 The impact of WBG, FMS and control sites on total concentrations of particles during the 2013 snowmelt. Box plot parameters same as Figure 3.1.
76
Figure 3.8 The impact of WBG, FMS and control sites on volume moment mean calculations during the 2013 snowmelt. Box plot parameters same as Figure 3.1.
77
3.6 Discussion
The results of this study indicate that WBG as an agricultural
management practice may have effects on the environmental conditions of
watersheds in the Canadian Prairies. When compared with FMS, snowmelt
runoff from fields that had WBG employed contained higher concentrations of
bacteria. Also, WBG sites were found to have differences in both particle-size
and particle-concentration with an increase in quantity and a decrease in mean
size being observed. Although other factors such as micrometeorology and local
differences in drainage may have impacted the sediment properties, the sample
locations were close enough spatially to negate the majority of these factors. At
the in-field scale, WBG was found to have no impact on the BSA ratio compared
to FMS. Further research into the effects of WBG and FMS on soil properties
and bacterial loading is vital to understanding the environmental impacts of both
management practices. Spatially, there was a considerably higher BSA ratio
observed within Pipestone Creek than at in-field locations.
The application of the novel method for quantifying BSA proved to be
successful in this area. Samples could be processed in the laboratory in an
adequate timeframe. The method proved to be relatively simple and allowed for
bacteria to be viewed and counted using epifluorescence microscopy. A
limitation of this method is the inability to discern between a) microbial species
and b) viable vs non-viable organisms. However, obtaining total bacterial counts
via this new method proved to be successful. This method can be applied to
other remote environments where immediate analysis of samples cannot occur.
78
Agricultural practices are sources of bacteria (Guber et al. 2007) within a
watershed. These sources can be either point (feed lots; Puckett 1995) , or non-
point (grazing; Tian et al. 2002) sources depending on a variety of factors,
including management practices. As landuse in the Pipestone Creek watershed
is primarily agricultural, the importance of bacteria-sediment associations and
their influence upon sediment hydrodynamics (Wotton 2004, Petticrew et al.
2011, Wotton 2011) and bacterial survival (Burton et al. 1987, Droppo et al.
2009) are important aspects of water quality within the watershed. The
contamination of surface water from agricultural practices within Pipestone
Creek watershed impacts the usability of the water for personal, irrigation and
recreation uses.
This study also illustrates that the practice of WBG may influence the
geomorphic processes occurring within a watershed. Increased particle
concentrations at the in-field scale may lead to an increase in suspended
sediment concentration in larger drainage pathways. Increased sediment loads
in fluvial environments has both biological and geomorphic implications (Belsky
et al. 1999). Increased sediment levels combined with increased bacterial loads
results in the high BSA ratio observed in Pipestone Creek. Since BSA create a
form of floc, it can be assumed that increased sedimentation may occur due to
the increased settling velocity of flocs as compared to their constituent particles
(Gibbs 1985, Droppo 2001, Hodder 2009). The presence of bacteria and flocs in
bed sediments will also lead to an increase in bed stability (Jamieson et al.
2005b).
79
Of particular interest is the large difference in BSA ratio that occurs
between in-field sites and the main creek stem sample locations. This highlights
organisms are likely exported from these agricultural sites predominantly in free
phase (Figure 3.1). The properties of sediment at the different spatial scales may
also account for the variation in BSA ratios measured. These results are in
contrast to Soupir et al (2010), which found that in some cases, 40% of E. coli
and Enterococci were particle-associated in runoff. The order of magnitude
difference in BSA between this study and Soupir et al. (2010) is likely due to the
concentration of manure used in their study. While Soupir et al. (2010) measured
runoff from individual cow pats, similar measurements were not undertaken in
this study.
The delivery of free-phase organisms in runoff directly modifies the BSA
ratio (Equation 3.1). Since the delivery of free-phase organisms is apparent from
the in-field sample locations and there is a noticeably higher BSA ratio measured
in Pipestone Creek, it is possible that there is a period required for bacteria
association to occur. This difference would indicate that the method of bacterial
transport within a watershed might be scale-dependent. One likely explanation of
this could be that flocculation is occurring between the in-field sample locations
and Pipestone Creek. Flocculation in-creek may also explain the large difference
in measured BSA values.
The relationship between particle size and BSA varied between in-field
samples (Figure 3.2) and Pipestone Creek samples (Figure 3.3). At the in-field
locations, the BSA ratio was positively correlated with volume concentration in
80
the 32.7 μm size class likely due to the presence of flocculated particles. A
similar correlation is evident within the Pipestone creek when comparing BSA
ratios and the volume concentration of sediment in the 14.4 μm size class.
Samples from the in-field sample locations frequently had visible organic
particles present which were primarily within the largest size class of the LISST
device. These particles were primarily removed from agricultural fields during
snowmelt and were observed to be composed largely of organic particulates.
The organic particles may have acted as attachment sites for bacteria, thus
accounting for the stronger positive correlation for larger particle sizes in
comparison with in-field samples. At both scales, the correlation coefficients
were relatively low and it is suggested that further research be undertaken to
better understand the relationship between BSA and particle sizes.
The spatial variation of BSA ratios occurring in the watershed illustrates
that there is a delay between the addition of free-phase bacteria and the
formation of flocculated particles. To the knowledge of the author, there are no
reports of the rate of flocculation in fluvial environments. Results from in-field
sampling suggest that flocs account for lower portion of the suspended sediment
load, but at a far lower concentration; however, flocs are prevalent in Pipestone
Creek (Figure 3.1). While planktonic organisms are transported through the
watershed, those that are transported as part of a flocculated particle have a
lower die-off rate due to protection from predators and changing environmental
conditions (Wotton 2011). The presence of BSA in Pipestone Creek is potentially
of concern to the regions downstream relying on Moosomin reservoir as a
81
source of drinking water. The prolonged storage and survival of bacteria when
associated with sediment may lead to water quality degradation during storm
events (Jamieson et al. 2005b, Pandey et al. 2012). Furthermore, the increased
survival rates of particle-associated bacteria in the water column and due to
resuspension from sediments pose human health threats if in high enough
concentration (Droppo et al. 2011).
Aeolian processes are known to deposit sediment particles on snow
(Carling et al. 2012). However, the lack of winter dust storms in the region
(Wheaton 1992) would suggest that the snowpack contained negligible amounts
of sediment. Relatively low levels of SSC measured in snowmelt runoff also
supported low levels of sediment within the snowpack. Frozen soils and
snowcover in the area also limit the delivery of sediment to the snowpack. The
relatively low BSA ratio found at the WBG, FMS and pasture sample locations
(Figure 3.5) during snowmelt is likely due to the limited presence of sediment
atop or within winter snowpacks and the frozen soil over which most of the
snowmelt occurs. Although particle concentrations were found to be significantly
higher at WBG sites than manure spreading sites (Figure 3.7; t-test; P=0.031), it
is expected that this is due to the presence of residual crop material from cow
excretions. This hypothesis is further supported by the VMM of the measured
particles being smaller at the WBG sites than others (Figure 3.8).
The significantly higher total organism export from WBG sites during
snowmelt events (Figure 3.6) is likely due to the location of the cow excretions
(feces and urine) within the snowpack. As snowmelt proceeds, fecal bacteria are
82
mobilised at the WBG sites whereas bacteria from manure at the FMS sites are
located beneath the snowpack and are likely not mobilised until melt nears
completion. Of significance to water quality and physical water quality models is
the decreased die-off rate when bacteria are associated with particles as
opposed to planktonic form (Gerba and McLeod 1976, Sherer et al. 1992). The
results of this study could be used to further tune these surface water quality
models (Hipsey et al. 2008, Pandey et al. 2012) by providing BSA ratios for
different environments. The values reported here could be assumed to be
representative of other similar environments. Furthermore, the difference of BSA
values between in-field locations and Pipestone Creek will allow for a more
accurate representation of BSA at different scales. Also, a better understanding
of land use impacts on BSA will allow for more accurate model projections.
Within Pipestone Creek, there was no discernible temporal trend visible.
This may be due to the dynamic nature of relatively small, highly modified prairie
streams (Dodds et al. 2004). Further data collection would be beneficial in
helping to confirm the absence of a temporal trend. Table 3.2 shows the patterns
of BSA that were present in the Pipestone Creek after the annual snowmelt had
occurred. On 3 May 2012 there was an inverse relationship between distance
downstream and BSA whereas on 24 May, the opposite was observed. These
results confirm that BSA ratio in the creek is not static over space or time.
Although the reason(s) for such variability are unknown, one possible
explanation is resuspension of bacteria and sediment from the streambed. The
samples taken on 3 May were obtained during a rain event as discharge in
83
Pipestone Creek was increasing. Precipitation across the watershed and the
resulting overland flow may have delivered more bacteria and sediment to
Pipestone Creek, thus resulting in the relatively equal BSA values found along
the stream (Table 3.2). In contrast, samples taken on 24 May were obtained
during a period of low flow and as such the pulse of bacteria and sediment may
have just been finishing moving through the creek resulting in low BSA values at
the upstream sample site and higher at the downstream sample site (Table 3.2).
Previous studies have found E. coli survive in bed sediments for more
than 3 weeks (Jamieson et al. 2004). Resuspension of cohesive particles from
bed sediments could cause a ‘pulse’ of sediment with a high BSA ratio to move
downstream. Precipitation events occurred on both days, with the larger (12.8
mm) occurring on 3 May and the smaller (3.6 mm) occurring on 24 May. Heavy
precipitation also causes runoff from agricultural fields and therefore can
transport free-phase bacteria at the edge-of-field scale. Due to the sample
locations not being hydrologically linked, it is unlikely that bacteria from the in-
field locations would reach Pipestone creek. However, in an environment where
surface runoff from agricultural areas does contribute to larger waterways, either
naturally or through engineered solutions such as pipe drains, we expect that the
delivery of bacteria would alter the BSA ratio. The mobilisation of sediment-
associated bacteria during a storm event has a distinct peak that occurs once
the runoff has reached an intensity at which it is able to entrain and remove
sediment and bacteria (Figure 3.4). The large difference in BSA between in-field
locations and major water pathways in the watershed suggests that bacteria and
84
sediment transport transitions from being primarily individual particles/organisms
to composite particles.
This study, to the authors’ knowledge is the first to undertake and
document: a) a novel technique to measure BSA that is not as time-sensitive as
previously document methods, b) the presence of BSA in an agricultural region
of the Canadian Prairies and c) the impact of management practices on the
presence of BSA.
3.7 Conclusions
Bacteria-sediment associations are an important factor in the transport of
bacteria and the deposition of sediment but are poorly documented. This study
found that there are significant spatiotemporal trends in BSA present between
and within watersheds.
Within sub-watersheds, the BSA ratio was low but increased significantly
in the main stem, indicating that BSAs require time to form and are potentially
impacted by flow conditions as well as the presence of bacterial sources.
Although not hydrologically linked, the microwatersheds and the main stem
provide example environments in which the BSA ratio is significantly different.
The contrast between in-field and main stem values is consistent with flocculated
particles requiring a certain time and conditions to develop. We expect that this
time requirement would be present regardless of whether the different scales are
hydrologically linked.
85
The environmental impacts of winter-bale grazing cattle as opposed to fall
manure spreading in the Canadian Prairies are relatively understudied
(Jungnitsch et al. 2011). This study shows that the practice of WBG does lead to
an increased transport of bacteria from field locations during spring snowmelt.
Furthermore, the practice of WBG alters the properties of sediment exported
from agricultural fields, including particle-size distribution, total-sediment
concentration and volume moment mean diameter. The results of this study are
consistent with those from more temperate regions where the grazing of cattle
has increased sediment erosion. However, this study is the first one in the
Canadian prairies to observe changes in particle size and volume concentration.
An increase in particle concentration with a decrease in the volume moment
mean size is noted from WBG sites. Although the WBG beneficial management
practice undoubtedly has positive economic benefits, it is also associated with
an increase of bacteria in runoff. We expect these results are relevant in other
regions where WBG is occurring on frozen soil. We also expect that the
implementation of riparian protection practices will aid in mitigating the impacts
of WBG.
Further research into the impact of cattle on frozen, snow-covered ground
and soil stability would provide a better understanding of sediment mobilisation
during spring snowmelt periods. Finally, laboratory study into the impact of
bacteria on the settling velocities of sediment will help to further our
understanding of bacterial transport dynamics in agricultural watersheds.
86
3.8 Literature cited
Agrawal, Y.C., Pottsmith, H.C. 2000. Instruments for particle size and settling velocity observations in sediment transport. Marine Geology, 168: 89–114.
Arnell, N.W. 1999. Climate change and global water resources. Global
Environmental Change, 9: 31–49. Arnell, N.W. 2004. Climate change and global water resources: SRES emissions
and socio-economic scenarios. Global Environmental Change, 14: 31–52. Auer, M.T., Niehaus, S.L. 1993. Modeling fecal coliform bacteria—I. Field and
laboratory determination of loss kinetics. Water Research, 27: 693–701. Bacallao, R., Sohrab, S., Phillips, C. 2006. Guiding Principles of Specimen
Preservation for Confocal Fluorescence Microscopy. In, Handbook Of Biological Confocal Microscopy. Springer US, Boston, MA. Edited by J.B. Pawley.
Baxter-Potter, W.R., Gilliland, M.W. 1988. Bacterial Pollution in Runoff from
Agricultural Lands. Journal of Environment Quality, 17: 27–34. Belsky, A.J., Matzke, A., Uselman, S. 1999. Survey of livestock influences on
stream and riparian ecosystems in the western United States. Journal of Soil and Water Conservation, 54: 419–431.
Burton, G.A., Gunnison, D., Lanza, G.R. 1987. Survival of pathogenic bacteria in
various freshwater sediments. Applied and Environmental Microbiology, 53: 633–638.
Cade-Menun, B.J., Bell, G., Baker-Ismail, S., Fouli, Y., Hodder, K., McMartin,
D.W., Perez-Valdivia, C., Wu, K. 2013. Nutrient loss from Saskatchewan cropland and pasture in spring snowmelt runoff. Canadian Journal of Soil Science, 93: 445–458.
Carling, G.T., Fernandez, D.P., Johnson, W.P. 2012. Dust-mediated loading of
trace and major elements to Wasatch Mountain snowpack. Science of The Total Environment, 432: 65–77.
Characklis, G.W., Dilts, M.J., Simmins, O.D., III, Likirdopulos, C.A., Krometis, L.-
A.H., Sobsey, M.D. 2005. Microbial partitioning to settleable particles in stormwater. Water Research, 39: 1773–1782.
Collins, R., Rutherford, K. 2004. Modelling bacterial water quality in streams
draining pastoral land. Water Research, 38: 700–712.
87
Dodds, W.K., Gido, K., Whiles, M.R., Fritz, K.M., Matthews, W.J. 2004. Life on the edge: the ccology of Great Plains prairie streams. BioScience, 54: 205–216.
Droppo, I.G. 2001. Rethinking what constitutes suspended sediment.
Hydrological Processes, 15: 1551–1564. Droppo, I.G., Jeffries, D., Jaskot, C., Backus, S. 1998. The prevalence of
freshwater flocculation in cold regions: a case study from the Mackenzie River Delta, Northwest Territories, Canada. Arctic, 51: 155–164.
Droppo, I.G., Krishnappan, B.G., Liss, S.N., Marvin, C., Biberhofer, J. 2011.
Modelling sediment-microbial dynamics in the South Nation River, Ontario, Canada: Towards the prediction of aquatic and human health risk. Water Research, 45: 3797–3809.
Droppo, I.G., Liss, S.N., Williams, D., Nelson, T., Jaskot, C., Trapp, B. 2009.
Dynamic existence of waterborne pathogens within river sediment compartments. Implications for water quality regulatory affairs. Environmental Science & Technology, 43: 1737–1743.
Droppo, I.G., Ongley, E.D. 1992. The state of suspended sediment in the
freshwater fluvial environment: a method of analysis. Water Research, 26: 65–72.
Droppo, I.G., Ongley, E.D. 2004. Flocculation of suspended sediment in rivers of
southearstern Canada. Water Resources, 28: 1799–1809. Eisma, D. 1986. Flocculation and de-flocculation of suspended matter in
estuaries. Netherlands Journal of Sea Research, 20: 183–199. Flores, J.P., Tracy, B. 2012. Impacts of winter hay feeding on pasture soils and
plants. Agriculture, Ecosystems & Environment, 149: 30–36. Fouli, Y., Cade-Menun, B.J., Cutforth, H.W. 2013. Freeze–thaw cycles and soil
water content effects on infiltration rate of three Saskatchewan soils. Canadian Journal of Soil Science, 93: 1–12.
Franson, M.A.H., Clesceri, L.S., Greenberg, A.E., Eaton, A.D. (Editors). 1998.
Standard methods for the examination of water and wastewater. 20th
edition. American Public Health Association, Washington, DC.
Fries, J.S., Characklis, G.W., Noble, R.T. 2006. Attachment of fecal indicator bacteria to particles in the Neuse River Estuary, N.C. Journal of Environmental Engineering, 132: 1338–1345.
88
Gagliardi, J.V., Karns, J.S. 2000. Leaching of Escherichia coli O157:H7 in diverse soils under various agricultural management practices. Applied and Environmental Microbiology, 66: 877–883.
Geesey, G.G., Richardson, W.T., Yeomans, H.G., Irvin, R.T., Costerton, J.W.
1977. Microscopic examination of natural sessile bacterial populations from an alpine stream. Canadian Journal of Microbiology, 23: 1733–1736.
Gerba, C.P., McLeod, J.S. 1976. Effect of sediments on the survival of
Escherichia coli in marine waters. Applied and Environmental Microbiology, 32: 114–120.
Gibbs, R.J. 1985. Estuarine flocs: Their size, settling velocity and density.
Journal of Geophysical Research: Oceans, 90: 3249–3251. Goulder, R. 1977. Attached and free bacteria in an estuary with abundant
suspended solids. Journal of Applied Bacteriology, 43: 399–405. Guber, A.K., Pachepsky, Y.A., Shelton, D.R., Yu, O. 2007. Effect of bovine
manure on fecal coliform attachment to soil and soil particles of different sizes. Applied and Environmental Microbiology, 73: 3363–3370.
Hall, R.I., Leavitt, P.R., Quinlan, R., Dixit, A.S., Smol, J.P. 1999. Effects of
agriculture, urbanization, and climate on water quality in the northern Great Plains. Limnology and Oceanography, 44: 739–756.
Henry, L.-A. 2004, April 27. Partitioning between the soil-adsorbed and
planktonic phases of Escherichia coli. MSc Thesis. Virginia Tech. Blacksburg, Virginia.
Hipsey, M.R., Antenucci, J.P., Brookes, J.D. 2008. A generic, process-based
model of microbial pollution in aquatic systems. Water Resources Research, 44: 1–26.
Hobbie, J.E., Daley, R.J., Jasper, S. 1977. Use of nuclepore filters for counting
bacteria by fluorescence microscopy. Applied and Environmental Microbiology, 33: 1225–1228.
Hodder, K.R. 2009. Flocculation: a key process in the sediment flux of a large,
glacier-fed lake. Earth Surface Processes and Landforms, 24: 1151–1163. Hodder, K.R., Gilbert, R. 2007. Evidence for flocculation in glacier-fed Lillooet
Lake, British Columbia. Water Research, 41: 2748–2762.
89
Hopwood, D. 1969. Fixatives and fixation: a review. The Histochemical Journal, 1: 323–360.
Hopwood, D. 1972. Theoretical and practical aspects of glutaraldehyde fixation.
The Histochemical Journal, 4: 267–303. Jamieson, R., Gordon, R., Joy, D., Lee, H. 2004. Assessing microbial pollution of
rural surface waters: A review of current watershed scale modeling approaches. Agricultural Water Management, 70: 1–17.
Jamieson, R., Joy, D.M., Lee, H., Kostachuk, R., Gordon, R. 2005a. Transport
and deposition of sediment-associated Escherichia Coli in natural streams. Water Research, 39: 2665–2675.
Jamieson, R.C., Joy, D.M., Lee, H., Kostachuk, R., Gordon, R.J. 2005b.
Resuspension of sediment-associated Escherichia Coli in a natural stream. Journal of Environmental Quality, 34: 581–589.
Jeng, H., England, A., Bradford, H. 2005. Indicator organisms associated with
stormwater suspended particles and estuarine sediment. Journal of Environmental Science and Health, 40: 779–791.
Jiang, X., Cheng, J., Guo, X. 2011. The flocculation of fine-grained suspended
sediment and its impact on spectral characteristics based on in situ measurement in the Changjiang Estuary, China. Canadian Journal of Remote Sensing, 37: 404–412.
Jones, J.G. 1974. Some observations on direct counts of freshwater bacteria
obtained with a fluorescence microscope. Limnology and Oceanography, 19: 540–543.
Jungnitsch, P.F., Schoenau, J.J., Lardner, H.A., Jefferson, P.G. 2011. Winter
feeding beef cattle on the western Canadian prairies: Impacts on soil nitrogen and phosphorus cycling and forage growth. Agriculture, Ecosystems & Environment, 141: 143–152.
Kelln, B.M., Lardner, H.A., McKinnon, J.J., Campbell, J.R., Larson, K., Damiran,
D. 2011. Effect of winter feeding system on beef cow performance, reproductive efficiency, and system cost. The Professional Animal Scientist, 27: 410–421.
Kepner, R.L., Pratt, J.R. 1994. Use of Fluorochromes for Direct Enumeration of
Total Bacteria in Environmental Samples: Past and Present. Microbiology and Molecular Biology Reviews, 58: 603–615.
90
Krometis, L.-A.H., Dillaha, T.A., Love, N.G., Mostaghimi, S. 2009. Evaluation of a filtration/dispersion method for enumeration of particle-associated Escherichia coli. Journal of Environment Quality, 38: 980-986.
Kubitschek, H.E., Freedman, M.L. 1971. Chromosome Replication and the
Division Cycle of Escherichia coli B/r. Journal of Bacteriology, 107: 95–99. Li, X.Y., Yang, S.F. 2007. Influence of loosely bound extracellular polymeric
substances (EPS) on the flocculation, sedimentation and dewaterability of activated sludge. Water Research, 41: 1022–1030.
Ling, T.Y., Achberger, E.C., Drapcho, C.M., Bengston, R.L. 2002. Quantifying
adsorption of an indicator bacteria in a soil-water system. Transactions of the ASAE, 45: 669–674.
Logan, T.J. 1993. Agricultural best management practices for water pollution
control: current issues. Agriculture, Ecosystems & Environment, 46: 223–231.
Madigan, M.T., Martinko, J.M., Dunlap, P.V. 2009. Brock biology of
microorganisms. 12 edition. Pearson/Benjamin Cummings, San Francisco, California.
Mahler, B.J., Personné, J.C., Lods, G.F., Drogue, C. 2000. Transport of free and
particulate-associated bacteria in karst. Journal of Hydrology, 238: 179–193. Nagels, J.W., Davies-Colley, R.J., Donnison, A.M., Muirhead, R.W. 2002. Faecal
contamination over flood events in a pastoral agricultural stream in New Zealand. Water Science and Technology, 45: 45–52.
Nguyen, M.L., Sheath, G.W., Smith, C.M., Cooper, A.B. 1998. Impact of cattle
treading on hill land: 2. Soil physical properties and contaminant runoff. New Zealand Journal of Agricultural Research, 41: 279–290.
Owens, L.B., Edwards, W.M., Van Keuren, R.W. 1997. Runoff and sediment
losses resulting from winter feeding on pastures. Journal of Soil and Water Conservation, 52: 194–197.
Owens, L.B., Shipitalo, M.J. 2009. Runoff quality evaluations of continuous and
rotational over-wintering systems for beef cows. Agriculture, Ecosystems & Environment, 129: 482–490.
Pandey, P.K., Soupir, M.L., Rehmann, C.R. 2012. A model for predicting
resuspension of Escherichia coli from streambed sediments. Water Research, 46: 115–126.
91
Petticrew, E.L., Rex, J.F., Albers, S.J., (null), (null). 2011. Bidirectional delivery of organic matter between freshwater and marine systems: the role of flocculation in Pacific salmon streams. Journal of the North American Benthological Society, 30: 779–786.
PFRA. 2001. PFRA's generalized land cover. Generalized land cover (l07g_pf0).
James Ashton. Version 1. Powell, J.M., Ikpe, F.N., Somda, Z.C., Fernández-Rivera, S. 1998. Urine effects
on soil chemical properties and the impact of urine and dung on pearl millet yield. Experimental Agriculture, 34: 259–276.
Puckett, L.J. 1995. Identifying the major sources of nutrient water pollution.
Environmental Science & Technology, 29: 408–414. Schillinger, J.E., Gannon, J.J. 1985. Bacterial adsorption and suspended
particles in urban stormwater. Journal of the Water Pollution Control Federation, 57: 384–389.
Sherer, B.M., Miner, J.R., Moore, J.A., Buckhouse, J.C. 1992. Indicator bacterial
survival in stream sediments. Journal of Environment Quality, 21: 591. Soupir, M.L., Mostaghimi, S. 2011. Escherichia coli and Enterococci attachment
to particles in runoff from highly and sparsely vegetated grassland. Water, Air, and Soil Pollution, 216: 167–178.
Soupir, M.L., Mostaghimi, S., Dillaha, T. 2010. Attachment of Escherichia coli
and Enterococci to particles in runoff. Journal of Environment Quality, 39: 1019–1027.
Soupir, M.L., Mostaghimi, S., Love, N.G. 2008. Method to partition between
attached and unattached E. coli in runoff From agricultural lands. Journal of the American Water Resources Association, 44: 1591–1599.
Tian, Y.Q., Gong, P., Radke, J.D., Scarborough, J. 2002. Spatial and temporal
modeling of microbial contaminants on grazing farmlands. Journal of Environmental Quality, 31: 860–869.
Trimble, S.W., Mendel, A.C. 1995. The cow as a geomorphic agent — A critical
review. Geomorphology, 13: 233–253. Vörösmarty, C.J., Green, P., Salisbury, J., Lammers, R.B. 2000. Global water
resources: vulnerability from climate change and population growth. Science, 289: 284–288.
92
Warren, S.D., Thurow, T.L., Blackburn, W.H., Garza, N.E. 1986. The influence of livestock trampling under intensive rotation grazing on soil hydrologic characteristics. Journal of Range Management, 39: 491–495.
Wheaton, E.E. 1992. Prairie dust storms — A neglected hazard. Natural
Hazards, 5: 53–63. White, S.L., Sheffield, R.E., Washburn, S.P., King, L.D., Green, J.T. 2001.
Spatial and time distribution of dairy cattle excreta in an intensive pasture system. Journal of Environment Quality, 30: 2180.
Wotton, R.S. 2004. The ubiquity and many roles of exopolymers (EPS) in
aquatic systems. Scientia Marina, 68: 13–21. Wotton, R.S. 2011. EPS (Extracellular Polymeric Substances), silk, and chitin:
vitally important exudates in aquatic ecosystems. Journal of the North American Benthological Society, 30: 762–769.
93
4 THE PRESENCE OF BACTERIA-SEDIMENT ASSOCIATIONS IN AN ALPINE ENVIRONMENT AND THEIR EFFECT ON THE
SETTLING VELOCITY OF SEDIMENT
David C Barrett* and Kyle R. Hodder
Prairie Environmental Process Laboratory, University of Regina, Regina, SK
S4S0A2
*Corresponding author [[email protected]]
4.1 Abstract
A key aspect for undertaking paleoenvironmental reconstructions from
laminated lake sediments is understanding the geomorphological processes by
which they formed. While bacteria are noted to enhance formation of flocculated
particles in engineered systems, the impacts in natural systems have yet to be
quantified. The impact of bacteria on the settling rate of sediment in the alpine
environment and associated impact on the formation of laminated sediments
have also yet to be measured. This study addresses key gaps in previous
literature by employing a novel technique to measure the presence of bacteria-
sediment associations (BSA) in a predominantly remote, alpine watershed in the
Southern Coast Mountains of British Columbia, Canada. The impact of bacteria-
sediment associations on the creation of flocculated particles and thus the
settling velocity of sediment is quantified using a laser transmissometer. Results
from the study indicate that BSA are present in an alpine watershed and exhibit
spatial and temporal trends. The percentage of bacteria associated with
94
sediment particles was found to range from 1% up to 40% depending on time
and location within the watershed. Major sources of planktonic bacteria such as
agricultural land and wastewater treatment outflow lead to large decreases in the
BSA ratio. Laboratory results showed that an increase in the concentration of
bacteria was associated with an increase in some size classes and a decrease
in others consistent with the formation of flocs. Bacterial concentrations were
also associated with an increase in settling velocity, also consistent with
flocculation.
4.2 Introduction
Biogeomorphology is an area of study that overlaps ecology and
geomorphology (Naylor et al. 2002). Under this loose definition, the role of
bacteria in fluvial and lacustrine geomorphologic processes would be considered
as a biogeomorphological process. However, past studies have viewed bacterial
transport dynamics solely as a water quality problem, ignoring the potential
impact of bacteria on geomorphic processes. One such process that is poorly
understood is the association of bacteria to sediment particles present in fluvial
environments. This paper intends to address this research gap by providing
quantitative results illustrating the importance of bacteria on geomorphic
processes.
Although bacteria-sediment interactions are known to occur, (Schillinger
and Gannon 1985, Jamieson et al. 2005a, Soupir and Mostaghimi 2011), the
topic is under-studied. To compound the problem, past studies that have
95
attempted to quantify the presence of BSA have used a variety of different
methods therefore making inter-study comparisons difficult (Schillinger and
Gannon 1985, Schindler 2001, Jamieson et al. 2005a, Krometis et al. 2009,
Soupir et al. 2010, Soupir and Mostaghimi 2011). Also, past studies have used
time-sensitive methods that cannot be applied in remote regions, thus limiting
their use to laboratory-proximal environments. The calculation of a BSA ratio is
done by comparing the concentration of planktonic organisms to those that are
sediment-associated (Equation 4.1)
Equation 4.1 The equation used for calculating the bacteria-sediment association ratio (BSA)
𝐵𝑆𝐴 =𝑆𝑒𝑑𝑖𝑚𝑒𝑛𝑡 − 𝑎𝑠𝑠𝑜𝑐𝑖𝑎𝑡𝑒𝑑 𝑜𝑟𝑔𝑎𝑛𝑖𝑠𝑚𝑠
𝑇𝑜𝑡𝑎𝑙 𝑜𝑟𝑔𝑎𝑛𝑖𝑠𝑚𝑠
The impact of bacteria as biogeomorphic agents is largely unknown (Viles
et al. 2008). While megafauna have been largely studied and discussed as
geomorphic agents (Arnell 1999, Hall et al. 1999, Butler and Malanson 2005,
Petticrew and Albers 2010, Albers and Petticrew 2013), the impact of microfauna
on geomorphological processes has been left notably under-studied (Schillinger
and Gannon 1985, Jamieson et al. 2005a, Soupir and Mostaghimi 2011, Viles
2012). This is likely due to their size and perceived geomorphic power (Viles et
al. 2008). However, with bacteria being found in all environments of Earth
(Rothschild and Mancinelli 2001), their potential impact as geomorphic agents
should be investigated.
Flocculated particles are present in a wide range of environments and
many studies have found bacteria to be present with flocs (Zabawa 1978, Eisma
96
1986, Liss et al. 1996, Droppo et al. 1998, Droppo 2001). Hodder and Gilbert
(2007) provided evidence of flocs in Lillooet Lake, however, the presence of
bacteria was not noted. It is well documented that sediment-settling velocities
are altered when flocs are present (Gibbs 1985, Droppo et al. 1997, Hodder
2009, Bache and Gregory 2010). The increased settling velocity compared to the
individual constituent particles of a floc can be attributed to an increase in overall
mass. This alteration of sediment transport properties within lakes can affect the
creation of laminated sediments at the lake bed (Hodder 2009). While laminated
lake sediments are often used as a proxy for hydroclimatic reconstruction, the
role of bacteria in the creation of this proxy remains unknown (Hodder et al.
2007; Figure 1).
This study applies a novel method for quantifying BSA to determine if
spatiotemporal trends are apparent. Furthermore, this study quantifies how the
presence of bacteria affects the settling velocity of sediment. Results from this
study aid in a better understanding of BSA in the natural environment. It also
allows for a more holistic understanding of the factors influencing the creation of
varved sediments.
4.3 Study site
The focus of this study was the Lillooet River, from the headwater at
Lillooet Glacier downstream to Lillooet Lake. Lillooet Glacier is located in the
Coast Mountain range of British Columbia and is the headwater for Lillooet
River. The river drains first into Silt Lake and eventually to Lillooet Lake,
97
approximately 100 km downstream (Figure 1.3). Aside from Silt Lake and the
floodplain of Lillooet River, there are no other significant sediment sinks within
the Lillooet River Valley (Hodder and Gilbert 2007). Lillooet Glacier is the largest
of the valley glaciers in the region, terminating at an elevation of roughly 950
masl (Gilbert 1975, Schiefer and Gilbert 2008). Like most glaciers in the area,
glacial retreat in recent years has been well documented (Schiefer et al. 2007,
Schiefer and Gilbert 2007)
The glacier itself is inaccessible by road and as such, is considered to be
free of direct human influence. Logging activities have progressed to the vicinity
of point F on Figure 1.3 and are present in the downstream direction until point
M, just outside Pemberton. There are also areas of intensive agriculture
upstream of Lillooet Lake, stretching up the valley to approximately point I on
Figure 1.3. The confluence of the Green and Lillooet rivers is found
approximately 12 km upstream of Lillooet Lake. Although it was historically
believed that the majority of the sediment travelling through the watershed is due
to glacial erosion (Jordan and Slaymaker 1991, Desloges and Gilbert 1994), a
recent study by Friele et al. (2005) suggests that much of the sediment delivery
to Lillooet Lake is supplied by the Mt Meager volcanic complex. The contribution
of volcanic sediment is of greatest significance immediately following major
slides or mass wasting events, such as that in 2010 (Guthrie et al. 2012).
Regardless of the source, the Lillooet River between the glacier and Lillooet
Lake can be considered sediment-rich with concentrations reaching up to 400
98
mgL-1 at Silt Lake (Schiefer and Gilbert 2008) and peaks of 800 mgL-1
The Lillooet Valley is experiencing increasing anthropogenic influence due
to intensifying land-use change (Gilbert 1975, Schiefer and Gilbert 2008).
Logging in the valley is expected to contribute to the high sediment
concentrations found within the valley (Jordan and Slaymaker 1991). Also, much
of the land located within the valley falls within the agricultural land reserve
(ALR). Fertile floodplain soils have facilitated agriculture in the Valley (Jordan
and Slaymaker 1991). The population of the village of Pemberton has increased
by 8.1% between 2006 and 2011 (Statistics Canada, 2012) and as such the
anthropogenic influence in the area has correspondingly increased, including the
development of a wastewater treatment plant in 2008.
at the
inflow to Lillooet Lake during snowmelt (Gilbert 1975).
Many glaciers and glacier-fed lakes in the Southern Coast mountains of
British Columbia have been extensively studied with a specific focus on
hydroclimatic reconstructions. The study site of the Lillooet Valley was chosen
partially due to the number of prior geomorphological studies that have occurred
at proglacial Silt Lake (Schiefer and Gilbert 2008), within the glacial valley
(Reyes and Clague 2004, Schiefer and Gilbert 2007), in Lillooet Lake (Gilbert
1975, Desloges and Gilbert 1994, Menounos et al. 2005, Gilbert et al. 2006,
Hodder and Gilbert 2007) and in the basin as a whole (Jordan and Slaymaker
1991, Friele et al. 2005). A notable omission from past studies done in the river
basin is the presence and geomorphic consequences of bacteria, a knowledge
99
gap first raised by Hodder and Gilbert (2007). This study aims to bridge that
knowledge gap.
4.4 Methods
A new method for quantifying BSA was applied, combining fractional
filtration and fluorescence microscopy. The method was developed in such a
way as to allow it to be applied in remote areas where immediate laboratory
analysis was unavailable. Previous studies using fractional filtration methods
were used to develop the technique (Auer and Niehaus 1993, Soupir et al. 2008,
Krometis et al. 2009, Soupir et al. 2010). Enumerating bacteria was done
through the application of fluorescence microscopy (Franson et al. 1998, Droppo
et al. 2009). A hydrologic monitoring station on the Lillooet River, near
Pemberton (WSC station 08MG005) provided discharge values.
4.4.1 Sample Collection
Water samples were obtained using both an ISCO 6712 automatic
sampler at location B (Figure 1.3) and by collecting grab samples at the other
locations. Although both methods might disrupt the largest macroflocs (Eisma
1986, Hodder and Gilbert 2007), we assumed that the majority of smaller flocs
will remain intact (Eisma 1986). While large macroflocs are relatively fragile,
smaller flocs are likely to remain undisturbed due to their strong electrochemical
and biologically cohesive bonds (Droppo et al. 2009, Wotton 2011). Grab
samples were collected using 1L polycarbonate Nalgene bottles and standard
hand grab sample methods (Canadian Council of the Ministers of the
100
Environment 2011). All samples were taken over an 8-day period in June and
July, 2012, in an attempt to capture peak glacial melt.
The automatic sampler was installed along the main river channel of the
Lillooet River, proximal to Lillooet Glacier (location B; Figure 1.3). Two hundred
millilitre samples were collected every 2 hours for a 48-hour period and then
every 6 hours for an additional 48 hours. Grab samples were retrieved at
multiple locations along the Lillooet River (Figure 1.3), including at the edge of
glacier and the first major sediment sink, Silt Lake. Grab samples were also
retrieved downstream of Silt Lake along the Lillooet River at sites upstream of
Lillooet Lake. These grab samples were taken to capture spatial variability as
well as the impact of agriculture, which becomes more concentrated in the
downstream direction.
The automatic sampler was also employed to collect samples for
suspended sediment concentration (SSC) and particle-size distribution. Samples
for SSC analysis were collected every hour for a period of 40 hours at location B
(Figure 1.3). The automatic sampler intake was placed directly in a main branch
of the Lillooet River. The flow velocity was such that it ensured that the intake
remained suspended ~5 cm above the river bottom. No attempt was made to
differentiate suspended or bed load.
4.4.2 Quantifying BSA
Separating free-phase (planktonic) bacteria from those associated with
sediment particles was done through fractional filtration. Initial processing of all
samples had to be done in the field and then further analysis was completed
101
upon returning to the laboratory. An 8 μm polycarbonate membrane was used
with the assumption (Schillinger and Gannon 1985, Krometis et al. 2009) that
any bacteria retained on the membrane were sediment-associated whereas
those which passed through the membrane could be considered planktonic.
Using E. coli, (Mahler et al. 2000) found that 99% of planktonic organisms
passed through the 8 μm filter and consequently were captured in the filtrate.
After completing the fractional filtration, a mixture of glutaraldehyde (0.5% w/v)
and phosphate buffer (KH2PO4
Enumeration of the bacteria was completed using fluorescence
microscopy and a semi-automated cell counting program. A solution of acridine
orange fluorochrome was added to the cells (Jones 1974, Hobbie et al. 1977) at
a final concentration of 0.05% (w/v). An exposure period of two minutes was
used to ensure adequate attachment of the solution to bacterial RNA and DNA
(Jones 1974, Hobbie et al. 1977, Kepner and Pratt 1994). The sample was
subsequently vacuum-filtered through a 0.2 μm polycarbonate membrane to
capture all planktonic organisms. The filtration apparatus was rinsed with
phosphate buffer before and after sample addition to maximize organism
capture. Between samples, the filtration equipment was flame sterilized to
prevent cross-contamination.
) was used to fix cells (Franson et al. 1998).
Air dried 0.2 μm membranes were mounted on microscope slides using
Cargille type-A immersion oil (refractive index = 1.694) and examined at 1000x
magnification. Fluorescence was induced using a fluorescent lamp and a 465-
495 nm excitation filter, and digital images were captured using a microscope-
102
mounted Olympus DP70 camera. In an attempt to minimize miscalculation due
to different bacterial densities on the membrane, a minimum of three fields of
view was captured for each slide. A visual comparison of at least 10 fields of
view was undertaken to establish that bacteria were equally distributed on the
membrane.
The digital images were then processed using ImageJ
(http://rsbweb.nih.gov/ij/) to obtain total bacteria counts. To ensure optimal
detection, manual verification of the automated counts was done to ensure all
visible organisms were counted. Manual verification was done for each individual
image captured. Once total bacterial counts were measured, concentration of
organisms per unit volume of sample could be calculated.
4.4.3 Measurement of particle size and SSC
Measuring of particle size was done using a Sequoia Scientific brand
Laser In-Situ Scattering and Transmissometry (LISST-C) instrument. The LISST
device measures particle-size distributions and volume concentrations over a
range 2.5 – 500 microns in diameter by measuring the laser diffraction and the
optical transmission (Fugate and Friedrichs 2002). Output values are binned
into 32 size bins ranging from 2.73 μm to 462 μm diameter. In this paper the bin
sizes are referred to by their median diameter. Within each of the size classes,
concentrations of sediment are measured in μl/l.
Each of the 40 SSC samples was processed using the LISST device,
after ensuring an optical transmission of >30%. All grab samples obtained were
also processed using the LISST-C device to obtain particle-size distributions.
103
Samples were vacuum filtered through pre-weighed 0.4 μm polycarbonate
membranes and air-dried for 24 hours to determine SSC using a precision
balance (±0.2 mg).
4.4.4 Measuring settling velocities
Settling velocity was measured using a settling tube attached to the
LISST and a 50% optical path reduction module. Specific concentrations of
bacterial solution inoculum were added to a constant concentration of sediment.
Mechanically homogenized bottom sediment from Lillooet Lake was suspended
in reverse osmosis (RO) water and immersed in an ultrasonic bath for 1 minute
as per Sperazza et al. (2004). The prepared sediment was then added to the
settling chamber for a final concentration of 120 mgL-1. The concentration of 120
mgL-1
Fixed quantities of bacteria inoculants were created using Polyseed®
bacterial capsules (http://polyseed.com) and a nutrient water solution created
following manufacturer protocols. Containers and equipment exposed to bacteria
inoculant were triple rinsed with RO water and were sterilised with anti-bacterial
soap and water and then with 10% HCl solution. Fixed volumes of inoculant
were added to the settling chamber along with the sediment slurry. Settling
experiments were initiated immediately following the addition of bacteria and
sediment slurry. Each settling experiment run consisted of 83 size distributions
was used as it was representative of values measured in Silt Lake
(Schiefer and Gilbert 2008) and other glacier-fed lakes in similar environments
(Gilbert and Desloges 1987). It was also near the maximum concentration that
allowed for adequate optical transmission in the LISST-ST device.
104
collected during a 22-hour period, and each settling experiment was repeated at
least once.
4.5 Results
4.5.1 Lillooet River
The majority of in situ samples examined were from the main stem of the
Lillooet River approximately 2 km downstream of the Lillooet Glacier snout
(Location B; Figure 1.3). The initial sampling interval of 2 hours showed changes
in measured bacteria-sediment associations over the 48-hour sampling interval,
as did the samples taken every 4 hours (Figure 4.1). Ratios of sediment-
associated to free-phase ranged from a minimum of 0.04 to a maximum of 0.40
during this 72- hour sampling period. Total microorganism counts also varied
during this time period (Figure 4.1). The highest peaks occurred during late
evening to early night hours. A trend of increasing BSA occurring at or around
1800 hours each day is visually observed but not statistically supported.
In Lillooet River, the ratio of sediment-associated to free phase bacteria
ranged from a minimum of 0.063 to a maximum of 0.38 (Figure 4.2). Samples
that were taken at locations immediately downstream of bacteria sources,
including the agricultural region and wastewater treatment plant (Bernhard et al.
2003), were associated with a decrease in BSA. Samples obtained downstream
of bacteria sources exhibited bacteria counts that were higher than at other
locations (Figure 4.2). The total organisms measured in samples also increased
105
in the downstream direction. Total organism concentration was linearly
correlated to distance downstream from Lillooet Glacier (R2 Figure 4.2=0.51; ).
At sample location B (Figure 1.3), the recorded SSC values ranged from
roughly 160 – 675 mgL-1 with a mean value of 281 mgL-1 Figure 4.3 ( ); values
consistent with those measured at Silt Lake by (Schiefer and Gilbert 2008). Peak
SSC values were noted at about 1800 hours during the sampling period.
All fifteen of the grab samples taken along the Lillooet River were
processed through the LISST device to obtain particle-size distributions of the
sediment in the river. Comparison of BSA values with the concentration of a
variety of particle sizes revealed trends in the alpine environment (Figure 4.4).
Within the watershed, as the concentration of very fine particles (2.73 μm)
increases, the BSA decreases. In contrast, there is a link between the
concentrations of 88.2 μm particles and the BSA (Figure 4.4). Low volume
concentrations of 462 μm particles, many at or near 0 μl/l, caused there to be
very little correlation between particle concentration and BSA (Figure 4.4).
106
Figure 4.1 Temporal changes in bacteria-sediment associations at the Lillooet River location over a 72-hour sampling period in early summer, 2012.
107
Figure 4.2 A transect along the Lillooet River showing calculated BSA and total organisms per litre. Dashed line is a linear regression curve showing an increase (R2
=0.51) in organisms in a downstream direction.
108
Figure 4.3 Suspended sediment concentrations of the Lillooet River, proximal to Lillooet Glacier, measured over a 40 hour period. Outliers were a result of sampler intake being submerged in riverbed sediment.
109
Figure 4.4 The relationship between BSA and particle concentration along Lillooet River, at three different particle size bins
110
4.5.2 Settling velocity and particle-size distribution
Within a controlled laboratory setting, it was recorded that an increased
concentration of bacteria resulted in an increase in settling velocities. Similarly,
an increase in bacterial concentration also resulted in a change in particle-size
distribution.
The addition of a bacterial solution to a sediment slurry resulted in a
change in particle-size distribution when compared to a control run to which no
bacteria were added (Figure 4.5, Figure 4.6). At the beginning of the experiment
the addition of bacteria resulted in an increase of particle concentration in the 7-
26 μm size range. During the same time period, a decrease of the smallest
recordable particles (2.5-7 μm) was observed. This trend occurred for the first 30
minutes of the 22-hour experiment duration. Observations collected after the 4-
hour point showed and increase in the smallest particle sizes. Above the 26 μm
particle size class, there was generally no change in particle concentration over
the 22-hour period. This trend was evident regardless of bacterial concentration,
however a higher the concentration of bacterial mixture resulted in a greater
magnitude of change.
The increase in particle size resulted in changes in the settling velocities
measured by the LISST-ST. A general trend of increasing settling velocities with
increasing bacterial concentrations was observed (Figure 4.7). Although an
increase in settling velocities was noted as bacterial concentrations were
increased, they never exceeded the calculated Stokes’ settling velocity
calculated for the density of quartz (2.65 g/cm3). With the exception of particles
111
in the 49.4 μm size bin, the measured settling velocities were at least an order of
magnitude below those estimated by Stokes’ law.
112
Figure 4.5 Difference in particle-size distribution between mixture with 40 ml of bacterial solution and one to which no bacteria were added. Positive values denote a (higher/lower) concentration in the run containing bacteria. Sample numbers represent logarithmically spaced sample intervals over 23-hour sample period.
113
Figure 4.6 Difference in particle-size distribution between mixture with 10 ml of bacterial solution and one to which no bacteria were added. Positive values denote a (higher/lower) concentration in the run containing bacteria. Sample numbers represent logarithmically spaced sample intervals over 23-hour sample period.
114
Figure 4.7 A comparison of settling velocities without bacteria, and at 6 Polyseed dilutions. Also shown is the calculated Stokes' settling velocity for quartz minerals.
115
4.6 Discussion
The results of this study provide evidence of previously undocumented
bacteria-sediment associations within an alpine watershed. Distinct spatial and
temporal trends were present in the BSA measurements. Furthermore, the
laboratory analysis of the effect of bacterial concentrations on sediment settling
velocities and particle-size distributions (PSD) provided previously
undocumented results.
A noticeable spatial trend was visible from glacier snout to Lillooet Lake.
Samples that were taken immediately downstream of areas with intense
agricultural activity showed a decrease in the bacteria-sediment association ratio
and an increase in total organisms (Figure 4.2). This is likely occurring due to the
fact that organisms being transported from these agricultural sites are
predominantly in free phase. This is supported by results found in a heavily
agricultural area in Southeastern Saskatchewan (Barrett and Hodder, chapter 3)
where BSA ratios were low at locations within agricultural fields. The increase in
free-phase organisms directly impacts the BSA ratio (Equation 4.1). Notably, a
delayed response appears to be present where the BSA ratio increases
downstream of locations where free-phase bacteria are introduced. One likely
explanation of this could be that flocculation is occurring between sample
locations. With alpine streams having high sediment concentrations (Church and
Gilbert 1975), an introduction of free-phase bacteria would provide ample
opportunity to interact with sediment to create or enhance floc formation.
116
As expected, total organism concentration increased in the downstream
direction. The increase in anthropogenic activity in the lower reaches of Lillooet
River corresponds with an increase in total bacteria. Figure 4.2 illustrates the
linear relationship between distance downstream and the total organism counts.
As previously discussed, peaks in organism counts increase significantly near
perceived sources of planktonic bacteria such as heavily agricultural lands and
downstream of the wastewater treatment plant outfall.
This is the first study to record the formation of BSA, and inherently
flocculated particles, over a known distance in the natural environment. In the
given conditions, flocculated particles were being formed over a distance of
14km (between sites L and M; Figure 1.3). With an average flow velocity of
approximately 2.19 m/s this suggests that flocs are forming over a time period of
1.78 hours in these conditions. The dynamic nature of natural systems likely
means that this rate of flocculation will be widely variable and dependant on local
conditions such as flow velocity, bed roughness and channel type. The drop in
total organisms (Figure 4.2) between sites L and M would also indicate that a
large number of these flocs are settling to the riverbed and becoming part of the
bed sediments, removing the bacteria associated with them from suspension.
This creates a sink at the bottom of rivers, which can consequently be
resuspended during periods of very high flow (Nagels et al. 2002, Jamieson et
al. 2003, 2005b). Due to the decreased die-off rate of bacteria when associated
to sediment (Davies et al. 1995, Pachepsky and Shelton 2011), resuspended
bacteria can pose major health concern during periods of high flow.
117
Temporally, the variations of the BSA range from <0.05 to 0.4 (Figure
4.1). The diurnal trend that occurs, with peaks in BSA at around 18:00, could be
caused by a number of different factors including diurnal animal activity (Green
and Bear 1990) and diurnal temperature fluctuations and resulting glacial melt
(Röthlisberger and Lang 1987, Hock 2003). An increase in animal activity at or
around the glacier or meltwater stream during daylight hours would introduce a
wide variety of microorganisms into the area due to defecation. The diurnal
melting cycle of the alpine glaciers in the region, during the summer months, can
lead to the release of stored organisms of subglacial (Sharp et al. 1999) and
supraglacial origin (Anesio et al. 2009) suggesting that glaciers may be a source
of bacteria. A delayed release of bacteria or other potential contaminants stored
within glaciers is becoming of greater concern due to climate change (Blais et al.
2001, Noyes et al. 2009), and is of further importance when flocculated particles
are present, as they provide an ideal transport mechanism (Droppo 2001).
The relationship between particle size and BSA within the watershed was
very dependant on what particle size was being examined. The negative link
between increasing sediment concentration in the 2.73 μm particle size range
and BSA can be explained by the size range being smaller than most flocculated
particles (Hodder and Gilbert 2007). With the assumption in this study being that
bacteria have a long axis of roughly 2 μm (fecal coliforms; Qualls et al. 1983) the
negative relationship between 2.73 μm particle concentration and BSA is to be
expected (Figure 4.4). Experiments ran in the laboratory confirmed that the
LISST-C device was able to measure single bacterium. For this reason,
118
increased volume concentration within the 2.73 μm size range may indicate an
increase in planktonic bacteria consequently resulting in a low BSA ratio. Hodder
and Gilbert (2007) reported that average flocculated particle sizes in Lillooet
Lake range from 15.1 μm to 24.5 μm diameter. Similar results are illustrated in
Figure 4.4, where an increasing concentration of particles in the 88.2 μm size
range correlates to an increase in BSA. This suggests that flocculated particles
are present in the Lillooet River and are being measured by the LISST device.
Although flocculated particles in the 462 μm size range may be present in situ,
they are considered to be fragile and subject to disturbance (Droppo et al. 1998,
Hodder and Gilbert 2007). Concentrations of particles in the 462 μm size range
were very low in Lillooet River (Figure 4.4) and as such no relation to BSA was
present, possibly due to disturbances during sample acquisition.
The addition of bacteria to the sediment slurry rich water in this study did
provide previously undocumented changes in particle settling velocities and
associated particle-size distributions. The addition of Polyseed® bacterial
solution to a standardised sediment concentration of 120 mgL-1
Figure 4.5
lead to changes
in both particle-size distributions ( , Figure 4.6) and settling velocities
(Figure 4.7). Although particle-size distributions changed relative to the amount
of bacterial solution added, with a greater magnitude of change occurring when
more bacteria was added, the general trend was the same regardless of
bacterial concentration. The increase of particles in the 13 μm size bin and the
concurrent decrease in the smallest size particles (3.48 μm) is consistent with
the formation of flocculated particles. The decrease in the concentration of small,
119
primary particles is indicative of them combining to form larger, flocculated
particles. Hodder and Gilbert (2007) found that flocculated particles in Lillooet
Lake that were formed by primary particles had a diameter of 10-35 μm. The
results from this study indicate that bacteria may influence the creation of
flocculated particles in alpine environments. This is in contrast to the
suggestions of Eisma (1997, cf Hodder and Gilbert 2007) who hypothesised that
a lack of organic material was partially the cause of limited flocculated particles
in glacial meltwater. The results from this study and Hodder and Gilbert (2007)
would suggest that flocculated particles are both present and prevalent in this
environment.
Although there is documented evidence for the presence of flocculated
particles in glacier-fed lakes (Hodder and Gilbert 2007), the impact of these
particles on the settling velocity of sediment had yet to be quantified. The settling
velocity of flocculated particles is known to be higher than those of the individual
constituent particles comprising the floc. This property of flocs is harnessed in
water treatment (Bache and Gregory 2010) and allows sediment, that may
otherwise not settle in lacustrine environments, to be deposited on lake beds
(Hodder 2009). The results displayed in Figure 4.7, Figure 4.5 and Figure 4.6
further support this notion by showing that higher concentrations of bacteria lead
to an increase in particle size and greater settling velocities.
Past studies have noted that flocculated particles generally settle at a rate
much lower than that suggested by Stokes’ Law (summarised by Hodder (2009),
figure 1). The results displayed in Figure 4.7 support the results of prior studies
120
by demonstrating that measured settling velocities of flocculated sediments were
lower than those estimated by Stokes’ settling velocity equation for quartz
particles (mass density of 2.65 g/cm3
The LISST measures a single axis of a particle that may, or may not be
the longest axis of the particle. Compounding that, the LISST device assumes a
spherical particle shape, which is known to not be representative of flocs and
natural grains (Agrawal and Pottsmith 2000, Pedocchi and García 2006, Hodder
2009). If the LISST device does not measure the long-axis of a particle, the
overall size and shape may be misrepresented. While these technological
limitations of the device are important, and discussed in more detail by Pedocchi
and García (2006), the results displayed in
). These results show that both floc density
and therefore settling velocity increase when bacterial concentrations are
increased. Although bacteria have previously been acknowledged to be present
in flocculated particles (eg Droppo 2001), the impact of increasing bacterial
concentrations on changes in density of the flocs has yet to be discussed
explicitly. Droppo (2001) did acknowledge that floc density is related directly to
porosity. As bacterial concentrations are increased, the individual bacterium
may, in fact, be joining previously flocculated material and therefore decreasing
porosity and increasing density. While more bacteria may affect other aspects of
the floc, such as shape, it would be expected that these parameters would have
less of an impact on settling velocity than changes in density.
Figure 4.7 further support prior
research suggesting that Stokes’ Law grossly overestimates the settling velocity
of sediment in lacustrine environments.
121
Although there are many aspects that influence the physical properties of
lake sediments (Hodder and Gilbert 2013, figure 1), the presence and
prevalence of microorganisms and their impact is often left undiscussed. As
noted by Hodder (2009), without flocs, many of the smallest primary particles
may not settle to the lake bottom. However, with flocs, such as those formed
through the presence of bacteria, these particles do settle out. As such, the
presence of bacteria and inherently flocs, change the physical properties of lake
sediments. O'Brien and Pietraszek-Mattner (1998) inferred the presence of
flocculated materials from the presence of random particle orientation in two
Pleistocene Glacial lakes bed sediments. However, beyond the inference of
flocculated particles, little discussion has occurred around the presence of
bacteria or flocs within the sedimentary record and more importantly what the
impacts of these processes is on paleoenvironmental reconstructions. For
example, the presence of flocs may lead to thicker laminated sediments being
created. While thicker sediments are often attributed solely to hydrolclimatic
conditions, the presence of bacteria and the creation of flocs may actually
explain some of the variation. During times of high bacterial concentrations, due
to contamination or warmer temperatures, there may be visible evidence of
increased flocculation present in the sedimentary structure of laminated
sediments.
4.6.1 Conclusions
Although noted to be present in agricultural regions, the presence of
bacteria-sediment associations in an alpine setting has gone undocumented.
122
This may be partially due to the fact that prior studies used sample collection
and analysis methods that were both a) time sensitive and b) required laboratory
equipment. The new method developed and used in this study was successfully
employed in a remote alpine environment. Although not species-specific, it is
suggested that this method could be adopted for analysis in other regions
therefore making inter-study results easier to compare.
Spatial results from this study illustrate that bacteria are associated with
sediment particles throughout the watershed, from headwater to major sink. The
BSA ratio, however, varies throughout the watershed, reaching its minimum
value immediately downstream of sources (eg agricultural land and wastewater
treatment release) of bacteria. This indicates that bacteria are being delivered
from these sources in planktonic form and are associating with sediment in the
fluvial environment. This provides evidence that bacteria and sediment are
flocculating over a reach of river approximately 14km in length.
The influence of bacteria on the formation of flocs in natural environments
had previously been acknowledged, but not quantified. Results from this study
provide evidence that bacteria are aiding in the formation of flocculated particles
and therefore changing the particle-size distribution of sediment in a laboratory
environment. It is expected that due to the presence of BSA, the changes of
particle-size distribution documented in this study would be applicable to the
natural conditions.
Alterations to the particle-size distributions of sediment in within the water
column is inherently linked to changes in the settling velocity of particles. This
123
study demonstrated that increases in the concentration of bacteria causes not
only an increase in the presence of flocculated particles, but also an increase in
settling velocity. The measured increase in settling velocity for particles of the
same size category is likely linked with an increase in density as more bacteria
are introduced. Understanding the processes that influence the formation of
laminated sediments helps encourage more holistic paleoenvironmental
reconstructions.
124
4.7 Literature cited
Agrawal, Y.C., Pottsmith, H.C. 2000. Instruments for particle size and settling velocity observations in sediment transport. Marine Geology, 168: 89–114.
Albers, S.J., Petticrew, E.L. 2013. Biogeomorphic impacts of migration and
disturbance: Implications of salmon spawning and decay. Geomorphology. In press.
Anesio, A.M., Hodson, A.J., Fritz, A., Psenner, R., Sattler, B. 2009. High
microbial activity on glaciers: importance to the global carbon cycle. Global Change Biology, 15: 955–960.
Arnell, N.W. 1999. Climate change and global water resources. Global
Environmental Change, 9: 31–49. Auer, M.T., Niehaus, S.L. 1993. Modeling fecal coliform bacteria—I. Field and
laboratory determination of loss kinetics. Water Research, 27: 693–701. Bache, D.H., Gregory, R. 2010. Flocs and separation processes in drinking
water treatment: a review. Journal of Water Supply: Research and Technology—AQUA, 59: 16.
Bernhard, A.E., Goyard, T., Simonich, M.T., Field, K.G. 2003. Application of a
rapid method for identifying fecal pollution sources in a multi-use estuary. Water Research, 37: 909–913.
Blais, J.M., Schindler, D.W., Muir, D.C.G., Sharp, M., Donald, D., Lafrenière, M.,
Braekevelt, E., Strachan, W.M.J. 2001. Melting glaciers: a major source of persistent organochlorines to subalpine Bow Lake in Banff National Park, Canada. Ambio, 30: 410–415.
Butler, D.R., Malanson, G.P. 2005. The geomorphic influences of beaver dams
and failures of beaver dams. Geomorphology, 71: 48–60. Canadian Council of the Ministers of the Environment. 2011. Protocols manual
for water quality sampling in Canada. http://www.ccme.ca/assets/pdf/protocols_document_e_final_101.pdf
Church, M., Gilbert, R. 1975. Proglacial fluvial and lacustrine environments. In,
Glaciofluvial and glaciolacustrine sedimentation Special Publication. The Society of economic paleontologists and mineralogists. Edited by A.V. Jopling and B.C. McDonald.
125
Davies, C.M., Long, J.A., Donald, M., Ashbolt, N.J. 1995. Survival of fecal microorganisms in marine and freshwater sediments. Applied and Environmental Microbiology, 61: 1888–1896.
Desloges, J.R., Gilbert, R. 1994. Sediment source and hydroclimatic inferences
from glacial lake sediments: the postglacial sedimentary record of Lillooet Lake, British Columbia. Journal of Hydrology, 159: 375–393.
Droppo, I.G. 2001. Rethinking what constitutes suspended sediment.
Hydrological Processes, 15: 1551–1564. Droppo, I.G., Jeffries, D., Jaskot, C., Backus, S. 1998. The prevalence of
freshwater flocculation in cold regions: a case study from the Mackenzie River Delta, Northwest Territories, Canada. Arctic, 51: 155–164.
Droppo, I.G., Leppard, G.G., Flannigan, D.T., Liss, S.N. 1997. The freshwater
floc: A functional relationship of water and organic and inorganic floc constituents affecting suspended sediment properties. Water, Air, and Soil Pollution, 99: 43–54.
Droppo, I.G., Liss, S.N., Williams, D., Nelson, T., Jaskot, C., Trapp, B. 2009.
Dynamic existence of waterborne pathogens within river sediment compartments. Implications for water quality regulatory affairs. Environmental Science & Technology, 43: 1737–1743.
Eisma, D. 1986. Flocculation and de-flocculation of suspended matter in
estuaries. Netherlands Journal of Sea Research, 20: 183–199. Franson, M.A.H., Clesceri, L.S., Greenberg, A.E., Eaton, A.D. (Editors). 1998.
Standard methods for the examination of water and wastewater. 20 edition. American Public Health Association, Washington, DC.
Friele, P.A., Clague, J.J., Simpson, K., Stasiuk, M. 2005. Impact of a Quaternary
volcano on Holocene sedimentation in Lillooet River valley, British Columbia. Sedimentary Geology, 176: 305–322.
Fugate, D.C., Friedrichs, C.T. 2002. Determining concentration and fall velocity
of estuarine particle populations using ADV, OBS and LISST. Continental Shelf Research, 22: 1867–1886.
Gibbs, R.J. 1985. Estuarine flocs: Their size, settling velocity and density.
Journal of Geophysical Research: Oceans, 90: 3249–3251. Gilbert, R. 1975. Sedimentation in Lillooet Lake, British Columbia. Canadian
Journal of Earth Sciences, 12: 1697–1711.
126
Gilbert, R., Crookshanks, S., Hodder, K., Spagnol, J., Stull, R. 2006. The record of an extreme flood in the sediments of montane Lillooet Lake, British Columbia: implications for paleoenvironmental assessment. Journal of Paleolimnology, 35: 737–745.
Gilbert, R., Desloges, J.R. 1987. Sediments of ice-dammed, self-draining Ape
Lake, British Columbia. Canadian Journal of Earth Sciences, 24: 1735–1747. Green, R.A., Bear, G.D. 1990. Seasonal cycles and daily activity patterns of
Rocky Mountain elk. The Journal of Wildlife Management, 54: 272–279. Guthrie, R.H., Friele, P., Allstadt, K., Roberts, N., Evans, S.G., Delaney, K.B.,
Roche, D., Clague, J.J., Jakob, M. 2012. The 6 August 2010 Mount Meager rock slide-debris flow, Coast Mountains, British Columbia: characteristics, dynamics, and implications for hazard and risk assessment. Natural Hazards and Earth System Science, 12: 1277–1294.
Hall, K., Boelhouwers, J., Driscoll, K. 1999. Animals as erosion agents in the
alpine zone: Some data and observations from Canada, Lesotho and Tibet. Arctic, Antarctic and Alpine Research, 31: 436–446.
Hobbie, J.E., Daley, R.J., Jasper, S. 1977. Use of nuclepore filters for counting
bacteria by fluorescence microscopy. Applied and Environmental Microbiology, 33: 1225–1228.
Hock, R. 2003. Temperature index melt modelling in mountain areas. Journal of
Hydrology, 282: 104–115. Hodder, K.R. 2009. Flocculation: a key process in the sediment flux of a large,
glacier-fed lake. Earth Surface Processes and Landforms, 24: 1151–1163. Hodder, K.R., Gilbert, R. 2007. Evidence for flocculation in glacier-fed Lillooet
Lake, British Columbia. Water Research, 41: 2748–2762. Hodder, K.R., Gilbert, R. 2013. Physical properties of lake sediments. In,
Encyclopedia of Quaternary Science. Elsevier, Amsterdam. Edited by S.A. Elias.
Hodder, K.R., Gilbert, R., Desloges, J.R. 2007. Glaciolacustrine varved sediment
as an alpine hydroclimatic proxy. Journal of Paleolimnology, 38: 365–394. Jamieson, R., Joy, D.M., Lee, H., Kostachuk, R., Gordon, R. 2005a. Transport
and deposition of sediment-associated Escherichia Coli in natural streams. Water Research, 39: 2665–2675.
127
Jamieson, R.C., Gordon, R.J., Tattrie, S.C., Stratton, G.W. 2003. Sources and persistence of fecal coliform bacteria in a rural watershed. Water Quality Resarch Journal of Canada, 38: 33–47.
Jamieson, R.C., Joy, D.M., Lee, H., Kostachuk, R., Gordon, R.J. 2005b.
Resuspension of sediment-associated Escherichia Coli in a natural stream. Journal of Environmental Quality, 34: 581–589.
Jones, J.G. 1974. Some observations on direct counts of freshwater bacteria
obtained with a fluorescence microscope. Limnology and Oceanography, 19: 540–543.
Jordan, P., Slaymaker, O. 1991. Holocene sediment production in Lillooet River
Basin, British Columbia: a sediment budget approach. Géographie physique et Quaternaire, 45: 45–57.
Kepner, R.L., Pratt, J.R. 1994. Use of fluorochromes for direct enumeration of
total bacteria in environmental samples: past and present. Microbiological Reviews, 58: 603–615.
Krometis, L.-A.H., Dillaha, T.A., Love, N.G., Mostaghimi, S. 2009. Evaluation of
a filtration/dispersion method for enumeration of particle-associated Escherichia coli. Journal of Environment Quality, 38: 980.
Liss, S.N., Droppo, I.G., Flannigan, D.T., Leppard, G.G. 1996. Floc Architecture
in Wastewater and Natural Riverine Systems. Environmental Science & Technology, 30: 680–686.
Mahler, B.J., Personné, J.C., Lods, G.F., Drogue, C. 2000. Transport of free and
particulate-associated bacteria in karst. Journal of Hydrology, 238: 179–193. Menounos, B., Clague, J.J., Gilbert, R., Slaymaker, O. 2005. Environmental
reconstruction from a varve network in the southern Coast Mountains, British Columbia, Canada. The Holocene, 15: 1163–1171.
Nagels, J.W., Davies-Colley, R.J., Donnison, A.M., Muirhead, R.W. 2002. Faecal
contamination over flood events in a pastoral agricultural stream in New Zealand. Water Science and Technology, 45: 45–52.
Naylor, L.A., Viles, H.A., Carter, N.E.A. 2002. Biogeomorphology revisited:
looking towards the future. Geomorphology, 47: 3–14.
128
Noyes, P.D., McElwee, M.K., Miller, H.D., Clark, B.W., Van Tiem, L.A., Walcott, K.C., Erwin, K.N., Levin, E.D. 2009. The toxicology of climate change: Environmental contaminants in a warming world. Environment International, 35: 971–986.
O'Brien, N.R., Pietraszek-Mattner, S. 1998. Origin of the fabric of laminated fine-
grained glaciolacustrine deposits. Journal of Sedimentary Research, 68: 832–840.
Pachepsky, Y.A., Shelton, D.R. 2011. Escherichia coli and fecal coliforms in
freshwater and estuarine sediments. Critical Reviews in Environmental Science and Technology, 41: 1067–1110.
Pedocchi, F., García, M.H. 2006. Evaluation of the LISST-ST instrument for
suspended particle size distribution and settling velocity measurements. Continental Shelf Research, 26: 943–958.
Petticrew, E.L., Albers, S. 2010. Salmon as biogeomorphic agents: temporal and
spatial effects on sediment quantity and quality in a northern British Columbia spawning channel. Proceedings of the ICCE symposium. IAHS Press, Warsaw, Poland.
Pikuta, E.V., Hoover, R.B., Tang, J. 2007. Microbial Extremophiles at the Limits
of Life. Critical Reviews in Microbiology, 33: 183–209. Qualls, R.G., Flynn, M.P., JR, J.M.C. 1983. The role of suspended particles in
ultraviolet disinfection. Journal of the Water Pollution Control Federation, 55: 1280–1285.
Reyes, A.V., Clague, J.J. 2004. Stratigraphic evidence for multiple Holocene
advances of Lillooet Glacier, southern Coast Mountains, British Columbia. Canadian Journal of Earth Sciences, 41: 903–918.
Rothschild, L.J., Mancinelli, R.L. 2001. Life in extreme environments. Nature,
409: 1092–1101. Röthlisberger, H., Lang, H. 1987. Glacial Hydrology. In, Glacio-fluvial sediment
transfer, an alpine perspective. Wiley, New York. Edited by A.M. Gurnell and M.J. Clark.
Schiefer, E., Gilbert, R. 2007. Reconstructing morphometric change in a
proglacial landscape using historical aerial photography and automated DEM generation. Geomorphology, 88: 167–178.
129
Schiefer, E., Gilbert, R. 2008. Proglacial sediment trapping in recently formed Silt Lake, upper Lillooet Valley, Coast Mountains, British Columbia. Earth Surface Processes and Landforms, 33: 1542–1556.
Schiefer, E., Menounos, B., Wheate, R. 2007. Recent volume loss of British
Columbian glaciers, Canada. Geophysical Research Letters, 34: L16503. Schillinger, J.E., Gannon, J.J. 1985. Bacterial adsorption and suspended
particles in urban stormwater. Journal of the Water Pollution Control Federation, 57: 384–389.
Schindler, D.W. 2001. The cumulative effects of climate warming and other
human stresses on Canadian freshwaters in the new millennium. Canadian Journal of Fisheries and Aquatic Sciences, 58: 18–29.
Sharp, M., Parkes, J., Cragg, B., Fairchild, I.J., Lamb, H., Tranter, M. 1999.
Widespread bacterial populations at glacier beds and their relationship to rock weathering and carbon cycling. Geology, 27: 107–110.
Soupir, M.L., Mostaghimi, S. 2011. Escherichia coli and Enterococci attachment
to particles in runoff from highly and sparsely vegetated grassland. Water, Air, and Soil Pollution, 216: 167–178.
Soupir, M.L., Mostaghimi, S., Dillaha, T. 2010. Attachment of Escherichia coli
and Enterococci to particles in runoff. Journal of Environment Quality, 39: 1019–1027.
Soupir, M.L., Mostaghimi, S., Love, N.G. 2008. Method to partition between
attached and unattached E. coli in runoff From agricultural lands. Journal of the American Water Resources Association, 44: 1591–1599.
Sperazza, M., Moore, J.N., Hendrix, M.S. 2004. High-resolution particle size
analysis of naturally occurring very fine-grained sediment through laser diffractometry. Journal of Sedimentary Research, 74: 736–743.
Statistics Canada. 2012. Pemberton, British Columbia (Code 5931012). Census
profile. 2011 Census. Statistics Canada catalogue no 98-316-XWE. Ottawa. Viles, H.A. 2012. Microbial geomorphology: A neglected link between life and
landscape. Geomorphology, 157–158: 6–16. Viles, H.A., Naylor, L.A., Carter, N.E.A., Chaput, D. 2008. Biogeomorphological
disturbance regimes: progress in linking ecological and geomorphological systems. Earth Surface Processes and Landforms, 33: 1419–1435.
130
Whitman, W.B., Coleman, D.C., Wiebe, W.J. 1998. Prokaryotes: The unseen majority. Proceedings of the National Academy of Sciences, 95: 6578–6583.
Wotton, R.S. 2011. EPS (Extracellular Polymeric Substances), silk, and chitin:
vitally important exudates in aquatic ecosystems. Journal of the North American Benthological Society, 30: 762–769.
Zabawa, C.F. 1978. Microstructure of agglomerated suspended sediments in
Northern Chesapeake Bay estuary. Science, 202: 49–51.
131
5 SYNTHESIS
The discipline of biogeomorphology is usually defined as an overlap
between ecology and geomorphology (Naylor et al. 2002), however, as the
results of this study illustrate, it is an area of study that intersects with many
others. The importance of microorganisms in the natural environment has been
a topic of particular interest for researchers, especially those found within water
due to the required usage of water as a resource. Until very recently, little
thought or discussion has occurred around the impact of these microscopic
organisms on the geomorphology of the world around us (Viles 2012).
As stated in chapter one of this thesis, there were three main objectives to
this study that are intricately linked. The three objectives examine questions that
examine process geomorphology and water quality analysis. While the
agricultural portion of this thesis (chapter 3) focused on the impacts of bacteria-
sediment associations and their potential impacts on water quality parameters,
the geomorphological impact is implicit. The impact of BSA on geomorphology is
more explicitly discussed in the results of the portion of the study undertaken in
an alpine environment. However, the presence of BSA in both, starkly
contrasting environments, illustrates that “microbial geomorphology” (Viles 2012)
is an area of study that overlaps many more disciplines. For example, results
from the Lillooet Valley illustrate the anthropogenic impact on the fluvial system
through the delivery of bacteria from agricultural practices and a wastewater
treatment plant. The increase in total organism concentration in Lillooet River
132
correlates with the increasing anthropogenic impact in the downstream reaches.
These bacteria go on to affect the settling rates of sediment and therefore the
creation of laminated sediments, a valuable paleoenvironmental proxy, in Lillooet
Lake. This study highlights the cross-discipline nature of biogeomorphology and
more specifically, microbial geomorphology.
With BSA values being quite high in main-stem fluvial environments (up to
0.7 in Pipestone Creek and 0.4 in Lillooet River), the potential impacts on water
quality are large. The decreased die-off rate of bacteria when associated with
sediment (Gerba and McLeod 1976) indicates that bacteria will be present within
the environment for a longer duration. An increased persistence of bacteria is of
major concern to humans when the water is used for consumption, recreation or
irrigation. Although bacteria-sediment associations have been observed for
decades (eg. Gerba and McLeod (1976), Schillinger and Gannon (1985)),
contemporary water quality models have yet to account for them.
The increased settling velocity observed in the laboratory experiments
conducted in this study indicate that the association of bacteria with sediment
will ultimately lead to flocculated particles settling out of suspension quicker.
Within heavily contaminated rivers this may be of concern as high-flow events
can lead to a resuspension of bacteria-rich bed sediment (McDonald et al. 1982,
Nagels et al. 2002, Jamieson et al. 2005). Once the sediment-associated
bacteria reaches a sink such as Moosomin Reservoir or Lillooet Lake, they have
an increased opportunity to settle to the bed. Again, the increased prevalence of
bacteria due to their presence within flocs is of concern to water quality,
133
especially in Moosomin Reservoir where the water has a variety of uses ranging
from irrigation to consumption and to recreation. Although the bacteria are
removed from the water column quicker than if they were in planktonic form, the
presence of potentially heavily contaminated bed sediments becomes of
significant concern.
The changes in settling velocity due to the presence of bacteria, and
therefore flocculation, are of particular interest when considering the dynamics of
sediment delivery within alpine-fed lacustrine environments. The laminated
sediments present in many of these lakes are often used as paleoenvironmental
proxies and as such understanding the processes that create them is
paramount. The results from this study indicate that the presence of bacteria
increased the occurrence of flocculation in laboratory simulations. With regards
to the creation of laminated sediments, the presence of flocs may significantly
alter the timing of sediment delivery and the sediment characteristics of the
sediment. While biological activity has been discussed as a factor influencing the
physical properties of lacustrine sediments (Hodder and Gilbert 2013), little
quantitative research has been done to better understand the magnitude of this
impact. Results delivered in this study would indicate that microorganisms may
have a greater impact on settling velocity, and therefore the creation of
laminated sediments, than previously documented. These results further the
understanding of laminated sediment creation and therefore allow for more
holistic paleoenvironmental reconstructions using this proxy.
134
Land use changes were found to impact the presence and magnitude of
BSA to some extent. Within the Pipestone Creek watershed, a large difference
was found between the in-field sample locations and the reach-scale sample
locations. The majority of bacteria collected at the in-field sites were in
planktonic form whereas samples from Pipestone Creek showed considerably
higher BSA ratios. These results were similarly confirmed in Lillooet River, where
samples taken immediately downstream of heavily agricultural areas had
considerably lower BSA ratios than those taken at other locations within the
watershed. This suggests that bacteria are removed from fields primarily in
planktonic form and become sediment-associated once in a larger fluvial
environment. Specific in-field land management practices appeared to have little
impact on BSA values, likely due to the low BSA values present. The practice of
winter bale grazing did, however, alter sediment and total bacteria values
observed during spring snowmelt, with both increasing in concentration
significantly when compared to fall-manure spreading and control sites.
The scale difference of BSA observed between the in-field sites and the
reach-scale sites would suggest there might be a process difference between
the two scales. Results from Lillooet River showed a marked increase in BSA
ratio approximately 11km downstream from an agricultural area hypothesised to
be contributing high amounts of planktonic bacteria. This would suggest that
bacteria require certain favourable conditions to become associated with
sediment and are forming flocs within an 11km stretch of river. The rate of
flocculated particle formation in natural environments is relatively unstudied and
135
this study only takes into account the distance over which flocs are forming.
Further research into this topic would be beneficial to link the BSA ratios at the
in-field and in-reach scales.
Although this study documents the presence of BSA in contrasting
environments, using a novel method of detection, it begs for further discussion
and research on a number of different topics. Detailed examination of laminated
lake sediments, with a specific focus on the presence of paleo-bacteria and the
preservation of flocs would further aid in a more holistic paleoenvironmental
reconstructions. Furthermore, a better understanding of the rate of floc formation
in different environments would be beneficial from both a geomorphological
standpoint as well as a water quality stand point. It is hoped that the new method
of quantifying BSA can be adopted in further studies as it can be applied in
remote environments. Also, the adoption of a standardized technique would
allow for better comparison of study results between studies. This would then aid
in the updating of contemporary water quality models to better account for the
presence of sediment-associated bacteria including their effect on bacterial
persistence and the settling velocities of sediment.
Further research into the processes surrounding the attachment of
bacteria to sediment would provide a better understanding of how these
interactions act in the natural environment. It is hoped that the successful
application of a novel method for measuring BSA will be used in future studies to
allow for inter-study comparisons. Further examination of the presence of flocs
and bacteria in varved sediments would link the process geomorphology
136
research done in this study to the paleoenvironmental reconstructions done in
other studies. Furthermore, to get a better understanding of the impact of
bacteria on the settling velocities of sediments, more laboratory analysis should
be undertaken. Using different bacteria phenotypes in the lab experiments would
help to understand the bacterial characteristics that most promote floc formation.
With regards to the specific study areas of this research, more research
would aid in a better understanding of the impacts of BSA at both locations. In
the Pipestone Creek, further hydrological research examining the connectivity of
the microwatersheds to Pipestone Creek itself may help to uncover the cause for
the significant in BSA noted at the different scales. Also, species-specific BSA
measurements, as opposed to the total counts documented in this study would
be beneficial for understanding the water quality and human health implications
of different field management practices. Within the Lillooet River watershed, this
study provides a brief discussion on the anthropogenic impact within the
watershed. Examination of the phenotype of bacteria being exported from
agricultural fields and the wastewater treatment plant in the area would help to
quantify any potential risk to human health.
The results presented in this dissertation are exploratory and lay the
groundwork for further studies in the field of microbial geomorphology and other
intersecting disciplines. It provides previously undocumented results on the
magnitude of bacteria-sediment associations in two different environments and
explores some of the water quality and geomorphological impacts associated
with BSA. By quantifying the presence of BSA and providing a new technique for
137
detection, it is hoped that the results of this study can be applied to develop
more accurate water quality models and aid in the interpretation of laminated
sediments as a paleoenvironmental proxy.
5.1 Literature Cited
Gerba, C.P., McLeod, J.S. 1976. Effect of sediments on the survival of Escherichia coli in marine waters. Applied and Environmental Microbiology, 32: 114–120.
Hodder, K.R., Gilbert, R. 2013. Physical properties of lake sediments. In,
Encyclopedia of Quaternary Science. Elsevier, Amsterdam. Edited by S.A. Elias.
Jamieson, R.C., Joy, D.M., Lee, H., Kostachuk, R., Gordon, R.J. 2005.
Resuspension of sediment-associated Escherichia Coli in a natural stream. Journal of Environmental Quality, 34: 581–589.
McDonald, A., Kay, D., Jenkins, A. 1982. Generation of fecal and total coliform
surges by stream flow manipulation in the absence of normal hydrometeorological stimuli. Applied and Environmental Microbiology, 44: 292–300.
Nagels, J.W., Davies-Colley, R.J., Donnison, A.M., Muirhead, R.W. 2002. Faecal
contamination over flood events in a pastoral agricultural stream in New Zealand. Water Science and Technology, 45: 45–52.
Naylor, L.A., Viles, H.A., Carter, N.E.A. 2002. Biogeomorphology revisited:
looking towards the future. Geomorphology, 47: 3–14. Schillinger, J.E., Gannon, J.J. 1985. Bacterial adsorption and suspended
particles in urban stormwater. Journal of the Water Pollution Control Federation, 57: 384–389.
Viles, H.A. 2012. Microbial geomorphology: A neglected link between life and
landscape. Geomorphology, 157–158: 6–16.
138
APPENDIX A – RELATIONSHIP BETWEEN PARTICLE SIZES AND BSA
139
Figure A. 1 Relationship between volume concentration of 32 particle size bins and associated R2 values from samples in the Lillooet River
140
Figure A. 2 Relationship between volume concentration of 32 particle size bins and associated R2 values from samples at in-field locations in the Pipestone Creek watershed
141
Figure A. 3 Figure A. 4 Relationship between volume concentration of 32 particle size bins and associated R2 values from samples taken in Pipestone Creek
142
APPENDIX B – SOP’S
Measuring BSA – field processing
- Modified from procedure 9216B in Franson (1998)
Required materials:
- vials; 25 ml with plastic caps, two per sample - RO-H2- Glutaraldehyde, 5% (w/v). Use phosphate buffer to dilute as necessary.
Target concentration for fixing samples: 0.5 % (w/v)
O
- 45mm diameter polycarbonate membranes with 8 μm pore diameter - Phosphate buffer (13.6 g KH2PO4- Eppendorf pipette (1-10 ml) and 10 ml tips
dissolved in 1 L RO water)
- Vacuum filtration unit for 47 mm membranes (milipor sterifil system or similar) and hand pump
Methods:
- Obtain sample using automatic sampler or grab sample technique - Thoroughly rinse vacuum filtration unit with RO-H2- Place polycarbonate membrane on filtration unit
O
- Filter as much sample as possible before filter clogs with sediment; add measured sample volumes incrementally
- Place 9 ml of filtrate in glass vial - Place membrane in separate vial and add 9 ml phosphate buffer - Add 1 ml of the 5% glutaraldehyde fixative mixture to both vials - Label vials appropriately
Three week storage clock starts now. Store samples cold (~ 4°C) until next step.
143
Figure A. 5 Materials required for separation of sediment-associated bacteria and planktonic bacteria in the field
144
Preparing slides for microscopy
Required materials:
- Alcohol lamp - Vacuum filtration unit (with 25 mm diameter membrane capacity) - 25 mm diameter polycarbonate membranes with 0.2 μm pore diameter - 25 mm diameter cellulosic membranes with 5 μm pore diameter - 3 ml syringes with syringe membranes (0.2 μm pore diameter) - Acridine orange fluorochrome, 0.1% (w/v) in phosphate buffer; target
concentration for staining sample is 0.05% (w/v) - Vortex mixer - Eppendorf pipette (1-10 ml) and 10 ml tips - Cargille Type-A Immersion oil with low fluorescence - Microscope slides (1 mm thickness) - Microscope coverslips (#1 thickness) - Membrane samples
Methods:
- Rinse using RO-H2
- Place cellulosic membrane on filtration unit
O and flame sterilize glass filtration unit before and after each sample
- Place polycarbonate membrane on top of cellulosic; cellulosic used as backer for support
- If processing sediment-associated sample, use vortex mixer for approximately 5 seconds to move particles from membrane to suspension
- Take 1 ml of fixed sample from the vial and add to filtration unit it - Filter 1 ml of acridine orange using syringe and 0.2 μm syringe filter - Add filtered fluorochrome to filtration unit - Let fluorochrome and sample stand for 2 minutes - Add 3 ml of phosphate buffer to promote more even cell distribution - Filter through apparatus using low vacuum pressure - Rinse unit with 5 ml of phosphate buffer to maximize capture of organisms
on the membrane - Remove polycarbonate membrane with tweezers and air dry - Discard cellulosic membrane - Repeat procedure with other membrane - Place small amount of immersion oil on clean microscope slide - Place air-dried membrane on slide - Place small amount of immersion oil on top of membrane - Apply coverslip -
145
Figure A. 6 Materials required for laboratory analysis of samples
146
Enumerating bacteria
Required materials:
- Prepared slides - Fluorescence microscope – digital image capture equipped - Excitation filters suitable for use with acridine orange - ImageJ software (http://rsbweb.nih.gov/ij/) - Immersion oil
Methods:
- Follow appropriate steps for starting and tuning specific microscope brand/model
- Place slide on stage and add small amount of immersion oil - Use 100x oil immersion lens and fluorescence microscope methods to
view organisms - View at least 10 different regions of the slide to determine how well
distributed organisms are on the membrane - Capture 3 digital images at random on the membrane - Open ImageJ - Open file to be analyzed in ImageJ - Set scale - Duplicate image and save said image - Auto adjust brightness/contrast (Image->Adjust->Brightness/contrast) - Set colour threshold (Image->Adjust->Colour threshold)
o Select so that only green is coloured black o Adjust brightness and saturation sliders until it appears to be
differentiating between red and green - Make into a binary - Erode - Analyze particles
o Size 2-Infinity o Circularity 0-1 o Show: Outlines o Ensure that the following options are active: Display results, Clear
Results, Summarize, record starts - Save all files as ImageX_____ (results, summary etc) - Calculate concentration of cells
147
Measuring the impact of bacteria on sediment settling velocity
Required materials:
- LISST-ST - Polyseed bacteria capsule - Nutrient pillows (suggest Hach brand for use) 1 for every 3 L of nutrient
water created); Follow manufacturers designated dilution requirements - Sediment (used lake bottom sediment from Lillooet Lake) - Low temperature (20°C) incubator - Mixing plate and magnetic stir bar - Aeration stone and bubbler/pump - Ultrasonic bath - Eppendorf pipette (1-10 ml) and 10 ml tips - RO-H2
O supply
Polyseed Method (Polyseed steps adapted from http://www.polyseed.com/videos/index.php)
- Sterilise storage bottles o Hot water in sink and soap – soak for 1hr. o Scrub storage bottle and ‘triple rinse’ with RO-H2o Fill half way with RO-H
O 2
o Add 10 ml 10% HCl acid – swirl O
o Fill storage bottle with RO-H2o Let sit for 24 hours then rinse
O
- Prepare nutrient water mix o Rinse container with RO-H2o Fill with 6L RO-H
O 2
o Add nutrient package and mix well O
o Incubate for 24 hours @ 20°C for 24 hours - Rehydrate Polyseed
o Shake nutrient water mix o Use 1L beaker
Record lot number of Polyseed o Fill 500 ml of nutrient water mix o Add one capsule of polyseed o Stir with stir plate and aerate with pump
Allow to stir and aerate for 1 hour before use o Allow stirred sample to settle for 15 minutes to clear filler/storage-
bran to separate 5)
148
Sediment Method - Manually homogenize
o Mechanically mix sediment sample - Create concentrated sediment slurry by mixing sediment with RO water
o target concentration of slurry was ~ 4 g/L - Sonicate samples
o Put samples in ultrasonic bath and sonicate for 1 minute (Sperazza 2004)
- Add desired bacteria concentration along with sediment sample to LISST settling tube
o Add aliquot of bacteria/nutrient mix o Add sediment aliquot
target concentration of sediment in chamber is XX to YY mg/L
o Fill ST with RO-H2o Run settling experiment for at least 2 runs (48 hours)
O
- Repeat with varying concentrations of bacteria
Background literature
Franson, M.A.H., Clesceri, L.S., Greenberg, A.E., Eaton, A.D. (Editors). 1998. Standard methods for the examination of water and wastewater. 20 edition. American Public Health Association, Washington, DC.
Hodder, K.R. 2007, June. The process-network of alpine, glacier-fed lakes with
particular reference to flocculation. PhD thesis. Queen's University. Kingston, Ontario.
Krometis, L.-A.H., Dillaha, T.A., Love, N.G., Mostaghimi, S. 2009. Evaluation of a
filtration/dispersion method for enumeration of particle-associated Escherichia coli. Journal of Environment Quality, 38: 980-986.
Sperazza, M., Moore, J.N., Hendrix, M.S. 2004. High-resolution particle size
analysis of naturally occurring very fine-grained sediment through laser diffractometry. Journal of Sedimentary Research, 74: 736–743.
