The Pennsylvania State University
The Graduate School
Department of Ecosystem Science and Management
PHOSPHORUS UPTAKE BY STREAM BENTHIC BIOFILMS: EMPIRICAL AND
EXPERIMENTAL APPROACHES TO EXPLAINING VARIATION
A Dissertation in
Wildlife and Fisheries Science
by
Keith J. Price
2012 Keith J. Price
Submitted in Partial Fulfillment
of the Requirements
for the Degree of
Doctor of Philosophy
August 2012
ii
The dissertation of Keith J. Price was reviewed and approved* by the following:
Hunter J. Carrick
Professor of Aquatic Ecosystems Ecology
Dissertation Advisor
Chair of Committee
John E. Carlson
Professor of Molecular Genetics
Director, Schatz Center for Tree Molecular Genetics
Jonathan Lynch
Professor of Plant Nutrition
John M. Regan
Associate Professor of Environmental Engineering
Tyler Wagner
Adjunct Associate Professor of Fisheries Ecology
Assistant Unit Leader, PA Cooperative Fish and Wildlife Research Unit
Michael G. Messina
Professor of Forest Resources
Head, Department of Ecosystem Science and Management
*Signatures are on file in the Graduate School
iii
ABSTRACT
Elevated phosphorus (P) concentrations in streams are frequently linked with
eutrophication and diminished water quality. Stream biofilms appear to play important
roles in P assimilation thus representing a valuable transformation of nutrients in aquatic
ecosystems. However, little work has identified parameters explaining variation in
uptake rates, evaluated the effect of common disturbance techniques, addressed how
increasing P-loads affect assimilative abilities, or estimated the influence of initial
assimilatory processes on biofilm P dynamics. Therefore, there were four central
approaches (chapters) to this dissertation: 1) perform an assessment of peer-reviewed
literature reporting aquatic microbial P-uptake rates, 2) evaluate the effect of physical
disturbance techniques commonly used in benthic biofilm metabolic studies, 3) measure
P-uptake rates for benthic biofilms along an experimental and natural nutrient gradient,
and 4) evaluate spatio-temporal P fluxes in biofilms. Regarding the first research
approach, several ecological/experimental parameters were found significant in
describing and explaining observed variation in published aquatic P-uptake rates:
microbial group (benthic, planktonic), source (culture, wild), and sample time (long,
short). This underscored the varied nature of microbial assimilatory kinetics and
provided a quantitative synthesis of uptake rates thereby advancing nutrient dynamic
models. The second chapter showed that common biofilm sampling techniques (physical
disturbance) caused no differential effects on kinetic parameter estimates (t= 0.69, p=
0.492, df= 33), lending credence to numerous metabolic studies on benthic microbes
post-abrasion and highlighting the potential for microbial uptake following scouring
events. The third chapter identified the occurrence of P saturation in some stream
biofilms and quantified its effect on the uptake of new P additions, and further concluded
that nitrogen was a synergistic nutrient for resident benthic biofilms, particularly in
streams of higher productivity (P legacy effect). Lastly, the fourth chapter demonstrated
rapid P exchange processes occurring at early time periods (i.e., ≤ 5 minutes), the
magnitude of which seems to diminish over longer periods (i.e., 15 - 30 minutes), further
suggesting that experimental time periods scaled to hours or longer obscure such
fundamental short-term responses. Overall, the studies conducted here employ both
empirical and experimental techniques and help to explain ecological and biological
variation in biofilm P-uptake rates.
iv
TABLE OF CONTENTS
LIST OF FIGURES .................................................................................................... viii
LIST OF TABLES ....................................................................................................... xii
ACKNOWLEDGEMENTS ............................................................................................ xv
BACKGROUND ............................................................................................................ 1
LITERATURE CITED .............................................................................................. 5
CHAPTER 1. META-ANALYTICAL APPROACH TO EXPLAIN VARIATION IN
MICROBIAL PHOSPHORUS UPTAKE RATES IN AQUATIC ECOSYSTEMS ............... 14
1.1 ABSTRACT ................................................................................................... 15
1.2 INTRODUCTION ............................................................................................ 16
1.3 METHODS .................................................................................................... 18
1.4 RESULTS ...................................................................................................... 23
1.5 DISCUSSION ................................................................................................. 26
1.6 ACKNOWLEDGEMENTS ................................................................................. 34
1.7 LITERATURE CITED ...................................................................................... 34
CHAPTER 2. EFFECTS OF PHYSICAL DISTURBANCE ON PHOSPHORUS UPTAKE IN
TEMPERATE STREAM BIOFILMS ......................................................................... 58
2.1 ABSTRACT ................................................................................................... 59
2.2 INTRODUCTION ............................................................................................ 60
2.3 METHODS .................................................................................................... 62
• P-UPTAKE RATES OF INTACT VS. SCRAPED BIOFILMS (SINGLE TIME POINT) . 62
• M-M PARAMETERS ALONG DISTINCT TIME-COURSES AND SHORT-TERM
FLUX ESTIMATIONS ..................................................................................... 64
• ABIOTIC SORPTION, P STORAGE, AND CELL VIABILITY ................................ 64
• CLEAN TECHNIQUES .................................................................................... 66
v
• STATISTICAL ANALYSES ............................................................................. 68
2.4 RESULTS ...................................................................................................... 69
2.5 DISCUSSION ................................................................................................. 71
2.6 ACKNOWLEDGEMENTS ................................................................................. 75
2.7 LITERATURE CITED ...................................................................................... 75
CHAPTER 3. EFFECTS OF NUTRIENT LOADING ON PHOSPHORUS UPTAKE BY
BIOFILMS SITUATED ALONG A STREAM PRODUCTIVITY GRADIENT ................... 88
3.1 ABSTRACT ................................................................................................... 89
3.2 INTRODUCTION ............................................................................................ 90
3.3 METHODS .................................................................................................... 93
• STUDY SITES ............................................................................................... 93
• DESIGN OF FIELD EXPERIMENTS- IN SITU ENRICHMENT SYSTEM (ISES) ...... 94
• ANALYTICAL MEASUREMENTS- BIOMASS AND AREAL NUTRIENT
CONCENTRATIONS ...................................................................................... 95
• ANALYTICAL MEASUREMENTS- P-UPTAKE ................................................. 96
• STATISTICAL ANALYSES- PARAMETRIC STATISTICS .................................... 98
• STATISTICAL ANALYSES- MULTILEVEL (HIERARCHICAL) MODEL ............. 100
3.4 RESULTS .................................................................................................... 101
• PHYSICOCHEMICAL CONDITIONS ............................................................... 101
• EFFECTS OF INCREASING P-LOADINGS ON BIOFILM P-UPTAKE .................. 102
• EFFECTS OF N WITH P-LOADINGS ON BIOFILM P-UPTAKE ......................... 104
• EFFECTS OF STREAM PRODUCTIVITY AND CELLULAR STOICHIOMETRY ON
BIOFILM P-UPTAKE ................................................................................... 105
3.5 DISCUSSION ............................................................................................... 107
• EVIDENCE FOR SATURATION OF BIOFILM P-UPTAKE.................................. 107
vi
• INTERACTIVE EFFECTS OF N AND P-LOADINGS ON BIOFILM P-UPTAKE ..... 112
3.6 ACKNOWLEDGEMENTS ............................................................................... 117
3.7 LITERATURE CITED .................................................................................... 117
CHAPTER 4. QUALITATIVE EVALUATION OF SPATIO-TEMPORAL PHOSPHORUS
FLUXES IN STREAM BIOFILMS.......................................................................... 144
4.1 ABSTRACT ................................................................................................. 145
4.2 INTRODUCTION .......................................................................................... 146
4.3 METHODS .................................................................................................. 148
• STATISTICAL ANALYSES ........................................................................... 150
4.4 RESULTS .................................................................................................... 151
4.5 DISCUSSION ............................................................................................... 154
• INITIAL ASSIMILATORY PROCESSES ........................................................... 154
• NUTRIENT LEGACIES ................................................................................. 160
• SPATIAL-TEMPORAL EFFECTS ................................................................... 163
4.6 ACKNOWLEDGEMENTS ............................................................................... 165
4.7 LITERATURE CITED .................................................................................... 165
CONCLUSION ......................................................................................................... 191
LITERATURE CITED .......................................................................................... 194
APPENDICES ........................................................................................................... 197
A. SUMMARY OF EXPERIMENTS TESTING EFFECTS OF PHYSICAL DISTURBANCE
ON P-UPTAKE (CHAPTER 2) .............................................................................. 197
B. DESCRIPTIVE STATISTICS FOR INTACT VS. SCRAPED STREAM BIOFILM P-
UPTAKE (CHAPTER 2) ....................................................................................... 198
C. GENERA LIST (100X) FROM INCREASING LEVELS OF DISTURBANCE
EXPERIMENT (CHAPTER 2) ............................................................................... 199
vii
D. GENERA LIST (400X) FROM INCREASING LEVELS OF DISTURBANCE
EXPERIMENT (CHAPTER 2) ............................................................................... 200
E. FRACTION OF BIOFILM GROWTH FORMS RECOVERED FROM DISTURBANCE
TREATMENTS (CHAPTER 2) .............................................................................. 201
F. MICHAELIS-MENTEN STATISTICS FOR SCRAPED BIOFILM ASSEMBLAGES (60
MIN) (CHAPTER 2) ............................................................................................ 202
G. MICHAELIS-MENTEN STATISTICS FOR INTACT BIOFILM ASSEMBLAGES (60
MIN) (CHAPTER 2) ............................................................................................ 203
H. MICHAELIS-MENTEN STATISTICS FOR SCRAPED BIOFILM ASSEMBLAGES (5 -
12 MIN) (CHAPTER 2) ....................................................................................... 204
I. MICHAELIS-MENTEN STATISTICS FOR INTACT BIOFILM ASSEMBLAGES (5 - 12
MIN) (CHAPTER 2) ............................................................................................ 205
J. MICHAELIS-MENTEN STATISTICS FOR SCRAPED BIOFILM ASSEMBLAGES (30 -
60 MIN) (CHAPTER 2) ....................................................................................... 206
K. MICHAELIS-MENTEN STATISTICS FOR INTACT BIOFILM ASSEMBLAGES (30 -
60 MIN) (CHAPTER 2) ....................................................................................... 207
L. PENNSYLVANIA MAP DEPICTING LOCATIONS OF 47 STREAMS FROM WHICH
EIGHT WERE SELECTED TO DEPLOY ISES (CHAPTER 3) .................................... 208
M. SUMMARY TABLE OF ISES EXPERIMENTS (CHAPTER 3) ............................. 209
N. SCATTER PLOT OF AREAL CARBON VS. AREAL CHLOROPHYLL-A FROM ISES
(CHAPTER 3) .................................................................................................... 210
O. TOTAL ALKALINITY MEASURED FROM EIGHT PENNSYLVANIA STREAMS
(CHAPTER 4) .................................................................................................... 211
P. SUMMARY TABLE OF SPATIO-TEMPORAL EXPERIMENTS (CHAPTER 4).......... 213
Q. SCATTER PLOT OF P-EFFLUX VS. POLY-P (CHAPTER 4) ................................ 214
viii
LIST OF FIGURES
Figure 1.1. Boxplot of log10-transformed P-uptake rates for the best predictor
parameters determined using mixed modeling: sample time, source, and
microbial group ................................................................................................. 57
Figure 2.1. Stacked column chart of benthic poly-P (mgP/mgChl) vs. disruption
treatment according to growth form ................................................................. 86
Figure 2.2. Thellier plots used to estimate Cmin for scraped (left) and intact (right)
biofilm assemblages .......................................................................................... 87
Figure 3.1. Mid-Atlantic region (left) and Pennsylvania statewide map (right)
showing the location of the eight ISES experiments in both of the major
physiographic provinces in Pennsylvania (Appalachian Plateau and
Piedmont) ........................................................................................................ 140
Figure 3.2. Scatter plot of Chl-specific P-uptake (log10(nmolP/μgChl/day)) versus
P-loading (log10(µg[PO4]/day)) for each stream tested with ISES (n= 10 per
stream) fit with linear regression and 95% confidence intervals .................... 141
Figure 3.3. Regression slopes (left) and intercepts (right) modeled as linear
functions of percent agriculture across eight streams sampled using ISES .... 142
ix
Figure 3.4. Scatter plot of log P-uptake (log10(nmolP/μgChl/day)) vs. log10 N:P
(top) and log10 C:P (bottom) ratios fit with linear regression at subgroups (N-
load) for streams in each province .................................................................. 143
Figure 4.1. Scatter plot of P flux (log10(µgP/µgChl)) vs. time (minute) conducted
on intact biofilm assemblages from eight Pennsylvania streams of varying
productivity in Summer 2010 ......................................................................... 183
Figure 4.2. Scatter plot of P flux (log10(µgP/µgChl)) vs. time (minute) conducted
on intact biofilm assemblages from eight Pennsylvania streams of varying
productivity in Spring 2010 ............................................................................ 184
Figure 4.3. Scatter plot of P flux (log10(µgP/µgChl)) vs. time (minute) conducted
on intact biofilm assemblages from eight Pennsylvania streams of varying
productivity in Winter 2010 ............................................................................ 185
Figure 4.4. Scatter plot of P flux (log10(µgP/µgChl)) vs. time (minute) conducted
on intact biofilm assemblages from eight Pennsylvania streams of varying
productivity in Fall 2009 ................................................................................ 186
x
Figure 4.5. Scatter plot with linear regression (solid-line) and LOESS (dashed-
line) of (log10) P-uptake (µgP/µgChl/min) (top) and (log10) P-efflux
(µgP/µgChl/min) (bottom) vs. (log10) part-P (mgP/m2) (left) and (log10) Chl
accumulation (mgChl/m2/d) (right) for intact biofilm assemblages established
on artificial substrata (tiles) from eight Pennsylvania streams of varying
productivity over four seasons 2009 - 2010 ................................................... 187
Figure 4.6. Scatter plot with linear regression (solid-line) and LOESS (dashed-
line) of (log10) P-uptake (µgP/µgChl/min) (top) and (log10) P-efflux
(µgP/µgChl/min) (bottom) vs. (log10) part-P (mgP/m2) (left) and (log10) Chl
(mgChl/m2) (right) for biofilm assemblages established on natural substrata
(rocks) from eight Pennsylvania streams of varying productivity over four
seasons 2009 - 2010 ........................................................................................ 188
Figure 4.7. Box plot of (log10) P-uptake (top) and (log10) P-efflux (bottom)
(µgP/µgChl/min) across eight streams situated in two geological provinces in
Pennsylvania over four seasons (n= 64) ......................................................... 189
Figure 4.8. Scatter plot with linear regression of (log10) assimilatory (stable) P-
uptake (30 minutes) vs. (log10) initial P-uptake (≤5 minutes) (µgP/µgChl/min)
for intact biofilm assemblages established on artificial substrata (tiles) from
eight Pennsylvania streams of varying productivity over four seasons 2009 -
2010 (n= 64) ................................................................................................... 190
xi
Figure C.1. Schematic of my four chapter dissertation research illustrating the
refinement and concentration of hypotheses through the successive
applications of preceding research findings ................................................... 196
xii
LIST OF TABLES
Table 1.1. Phosphorus uptake rates (μg P μg Chl a–1
d–1
) measured for two
microbial functional groups and five other experimental variables analyzed in
the meta-analysis with associated reference ..................................................... 51
Table 1.2. Model selection results for predicting the model probability ................ 54
Table 1.3. ANOVA results examining differences among predictors of log10-
transformed P-uptake rate (μg P μg Chl a–1
d–1
): microbial group, source, and
sample time ....................................................................................................... 55
Table 1.4. Descriptive statistics for all experimental parameters tested in the
meta-analysis of P-uptake rates (μg P μg Chl a–1
d–1
) ...................................... 56
Table 2.1. Three-way mixed-model ANOVA testing the main fixed effects of
treatment (n= 2) and experiment (n= 3) and random effect of site (n= 2) on P-
uptake rates (μgP/μgChl/d) for intact vs. scraped stream biofilms ................... 85
Table 3.1. A summary of geographic and biogeochemical characteristics for
streams where ISES experiments were conducted in 2008 ............................ 136
Table 3.2. Regression statistics for biofilm Chl-specific P-uptake
(log10(nmolP/μgChl/day)) versus P-loading (log10(µg[PO4]/day)) determined
from ISES experiments carried out in streams of varying nutrient content in
both Plateau and Piedmont provinces ............................................................. 137
xiii
Table 3.3. Likelihood ratio test results from iterative model building in
determining important parameters in explaining P-uptake against P-loading 138
Table 3.4. Regression statistics for biofilm Chl-specific P-uptake
(log10(nmolP/μgChl/day)) versus P-loading (log10(µg[PO4]/day)) without (n=
5) or with (n= 5) simultaneous N-loading ...................................................... 139
Table 4.1. P-uptake and P-efflux estimates from LOESS plots of P flux
(log10(µgP/µgChl)) vs. time (minute) and breakpoint estimates from the
‘segmented’ package in R for experiments conducted on intact biofilm
assemblages from eight Pennsylvania streams of varying productivity over
four seasons 2009 - 2010 ................................................................................ 177
Table 4.2. Linear regression statistics for (log10) P-uptake (µgP/µgChl/min) and
(log10) P-efflux (µgP/µgChl/min) vs. (log10) part-P (mgP/m2) and (log10) Chl
accumulation (mgChl/m2/d) for biofilm assemblages established on tiles, and
(log10) part-P (mgP/m2) and (log10) Chl (mgChl/m
2) for biofilm assemblages
established on natural substrata (rocks) from eight Pennsylvania streams of
varying productivity over four seasons 2009 - 2010 ...................................... 178
Table 4.3. Two-way ANOVA results for P-uptake (log10(µgP/µgChl/min)) and P-
efflux (log10(µgP/µgChl/min)) by season and province for replicate
measurements on intact biofilms across eight Pennsylvania streams of
varying productivity over four seasons 2009 - 2010 ...................................... 179
xiv
Table 4.4. Homogenous subsets based on Tukey's HSD post-hoc test for (log10)
P-uptake by season ......................................................................................... 180
Table 4.5. Homogenous subsets based on Tukey's HSD post-hoc test for (log10)
P-efflux by season ........................................................................................... 180
Table 4.6. Descriptive statistics for P-uptake and P-efflux (µgP/µgChl/min) for
biofilms on duplicate tiles in eight streams situated across two geologic
provinces over four seasons (n= 64) ............................................................... 181
Table 4.7. Pearson correlation matrix of all LOESS derived uptake estimations
and biochemical parameters tested in eight Pennsylvania streams over four
seasons (n= 32) ............................................................................................... 182
xv
ACKNOWLEDGEMENTS
First, I wish to thank my advisor Dr. Hunter Carrick for affording me this
tremendous opportunity and providing me profound insights on the methods and
principles of academic research. I am greatly appreciative to my committee members
Drs. John Carlson, Jonathan Lynch, John Regan, and Tyler Wagner for their generosity
with their time and expertise. They have been supportive in many aspects of this work
and I have learned much from each of them through enjoyable and valuable discussions.
I offer my gratitude to the Department of Ecosystem Science and Management for
providing financial support through teaching assistantships. To my FRB compatriots:
Andrew, Aubrey, Becca, Erin, and Melissa, I give mad props for filling in my blanks. I
received satisfying technical, moral, and (especially) corporeal support from these and
other sexy colleagues for which I am quite obliged. Additionally, I am eternally grateful
to Curious George for providing me towels, blood plasma, and pool floats during an
inadvertent stay inside a Peruvian prison.
Finally, and most importantly, I want to express my gratitude to my parents,
Joseph and Margaret Price, the keystone of my education, who taught me in high school
that “the tie makes the man”. I attribute this achievement to their humor, wisdom, and
generosity which provided motivation for me to complete this work. Quite fittingly, I
dedicate this dissertation to them.
Peace.
1
BACKGROUND
Photosynthesis, an essential physiological process that involves the capture of
light energy from the sun and its conversion into stored chemical energy, is understood to
be the process by which Earth’s atmosphere has been oxygenated and biological diversity
has proliferated (Raymond et al. 2002, Bekker et al. 2004). Plants capable of
photosynthesis require adequate light, water, and nutrients from the soil/sediment for
growth. Plant growth and reproduction thus depends on the availability and accessibility
of essential nutrients (Knecht and Göransson 2004, Zhou and Hosomi 2008). According
to the single limiting nutrient theorem (i.e., Liebig's Law of the Minimum), only one
nutrient is typically in shortest supply and thus controls production. Plant growth,
however, can be maximized through allocation of biomass to resource (nutrient)-
acquiring areas (Tilman 1988). Nutrients have been found to not only serve an important
role in growth and respiration but also in modulating gene expression and generating
hormonal response (Takei et al. 2002).
Similarly, algal production is both dependent on and stimulated by nutrient
availability (Schindler et al. 2001), and nutrient ratios of terrestrial plants are similar to
ratios found in plankton (Knecht and Göransson 2004). However, nutrient enrichment of
streams and lakes due to human activities can to lead to excessive algal biomass
(Carpenter et al. 1998, Smith et al. 1999) which can alter community structure (Miltner
and Rankin 1998) and in extreme cases, deplete dissolved oxygen (Dodds and Welch
2000). This can reduce suitable habitat conditions for fish and invertebrate survival
(Welch 1992). Much focus, therefore, has been aimed at reducing nutrient loadings to
freshwaters in an effort to reduce the primary production that drives eutrophication
2
(Bowes et al. 2007), the most widespread water quality problem in the United States
(Carpenter et al. 1998). Increased nutrient loadings may also affect in-stream nutrient
uptake capacity (Young and Huryn 1999). This is of particular concern because of
streams’ terminal export to downstream waters (e.g., lakes and oceans) (Carpenter et al.
1998, Correll 1998, Dodds and Welch 2000). Mulholland et al. (2008) suggested that
management of nutrient loading to streams is imperative to maintaining their nutrient
removal functions.
Phosphorus (P) is a major plant macronutrient, composing approximately 0.2% of
plant dry weight (Schachtman et al. 1998). Phosphorus is likewise an essential element
for algal growth, involved in structural and metabolic processes (Scinto and Reedy 2003,
Steinman and Mulholland 2006). It is required in energy molecules (e.g., ATP),
synthesis of macromolecules (e.g., DNA), photosynthesis and respiration (Raghothama
1999), and phosphorylation of sugars (Steinman and Mulholland 2006). Increased
anthropogenic inputs of P are of concern because large quantities (soluble inorganic P) in
lake, reservoir, and stream systems are the most common cause of eutrophication (Likens
1972, Moss et al. 1986, Correll 1998). Numerous studies have closely tied excessive
concentrations of P to eutrophication of freshwater stream systems (Van Nieuwenhuyse
and Jones 1996, Mainstone and Parr 2002, Withers and Hodgkinson 2009). Nitrogen (N)
is also important in regulating algal productivity (Nydick et al. 2004); it is required for
synthesis of amino acids and protein production (Colla et al. 2007). There is abundant
data indicating that supplies of N and/or P limit primary production in freshwater (Elser
et al. 1990). Furthermore, studies have found that combined enrichment with both N and
P can produce greater algal biomass compared to additions of N or P alone, suggesting
3
co-limitation may be typical in lotic systems (Francoeur 2001, Liess and Hillebrand
2006). In fact, N and P have been found to be the most pervasive stressors of streams in
the US (USEPA 2006). However, despite the importance of N in limiting biomass, N-
fixation by some aquatic microbes and N2 exchange between the atmosphere and water
suggests that a stronger focus on P transport management is most important to limiting
freshwater nutrient loading (Sharpley and Rekolainen 1997) and preventing freshwater
eutrophication (McDowell 2003, Withers and Jarvie 2008). Nevertheless, simply
reducing P-loads may not directly translate to improved water quality. It is necessary to
first understand how P can limit microbial metabolism and how in-stream processes can
modify P before the effects of nutrient loading can be realized (Withers and Jarvie 2008,
Marcarelli et al. 2009).
The effects of nutrient loads on a stream depend on the amount of nutrients that
are extracted from the water by biota (taking up inorganic nutrients) (Nijboer and
Verdonschot 2004). Stream biofilms (benthic bacteria, fungi, and algae associated with
submerged substrata, e.g., rocks, sand, plants (Lock et al. 1984)) play an important role in
P-uptake, assimilation, and retention in aquatic ecosystems (Pringle et al. 1988, Lind et
al. 1992, Lampert and Sommer 1997, McCormick et al. 1997, Reddy et al. 1999, Dodds
2003) and function as the primary link between dissolved inorganic nutrients in the water
and higher trophic levels (Hynes 1970). P-uptake, where P is transported from the water
column into the benthos (Dodds 2003), is thought to be the principle pathway by which P
is accumulated by benthic biofilms (Lean 1973). Biofilms have a high affinity for P and
have been linked to increased uptake efficiency and rapid nutrient recycling in P-limited
systems (Schindler 1977, Sand-Jensen 1983; Wetzel 1996, Havens et al. 1999). House
4
and Casey (1989) suggest benthic (biofilms) and/or phytoplankton communities may
account for 10 - 15% of riverine P flux and McColl (1974) found that as much as 97% of
added PO43-
was taken up by stream-bed associated biota (i.e., biofilms) under low flow
conditions in a semi-natural small stream system. Biofilms can intercept nutrients
leached from underlying substrata and/or take-up nutrients in overlying water (Riber et
al. 1983, Carlton and Wetzel 1988) and thereby strip the water column of nutrients and
reduce transported loads (Kim et al. 1990, McCormick and Scinto 1999, Noe et al. 2002,
Dodds 2003). P-loads can be controlled by biofilms as they influence rates of P retention
through uptake (Pringle et al. 1988, Dodds 2003) and storage (Costerton et al. 1987,
Freeman et al. 1995). Retention can buffer the impacts of high P-loadings on
downstream communities (Svendsen et al. 1995) and thus, the integrity and efficiency of
this process is crucial (Aldridge et al. 2009). Biofilms therefore have a significant role in
stream P dynamics and fundamental biogeochemical processes (Ryder and Miller 2005)
and can be a significant stream buffer (biotic sink) against the effects of increasing
nutrient loads and eutrophication (Reddy et al. 1999, Stevenson 2001, Dodds 2003,
Steinman and Mulholland 2006). Since nutrient uptake describes the rate of an important
stream process, it provides a measure of the performance of a stream system (Bunn et al.
1999). Determining nutrient uptake rates for stream biofilms can thus provide useful
information in assessing the effects of nutrient loads (Meyer et al. 2005, Steinman and
Mulholland 2006, Marcarelli et al. 2009).
Given the seriousness of nutrient loading to streams, particularly the importance
of P and N in generating potentially harmful effects to aquatic systems and the role of
benthic biofilms in mitigating such effects, my research here will focus on empirical and
5
experimental approaches that help to understand ecological and biological variation of
biofilm P-uptake rates. This dissertation is divided into four chapters. Chapter 1 applies
a meta-analytical approach to explain variation in microbial P-uptake rates. This study
was conducted to highlight the various ecological and experimental parameters that can
influence biotic assimilation of P. Chapter 2 focuses on the effect of procedural
techniques (physical disturbance) on P-uptake rates and cell viability. Chapter 3 explores
the effects of nutrient loading on P-uptake by biofilms situated along a stream
productivity gradient. Specifically this research shows how point phosphorus and
nitrogen loadings can moderate biofilm response to new P and the role of legacy effects
in generating different physiological uptake responses. Lastly, Chapter 4 examines
biofilm response to new P across spatial and temporal gradients; emphasis is given to
initial assimilatory kinetics and integration of the singular and interactive effects of space
and time on P fluxes.
LITERATURE CITED
Aldridge, K.T., Brookes, J.D., and Ganf, G.G. 2010. Changes in abiotic and biotic
phosphorus uptake across a gradient of stream condition. River Research and
Applications 26:636-649.
Bekker, A., Holland, H.D., Wang, P.L., Rumble, D. 3rd, Stein, H.J., Hannah, J.L.,
Coetzee, L.L., and Beukes, N.J. 2004. Dating the rise of atmospheric oxygen.
Nature 427:117-120.
6
Bowes, M.J., Smith, J.T., Hilton, J., Sturt, M.M., and Armitage, P.D. 2007. Periphyton
biomass response to changing phosphorus concentrations in a nutrient-impacted
river: A new methodology for phosphorus target setting. Canadian Journal of
Fisheries and Aquatic Sciences 64:227-238.
Bunn, S.E., Davies, P.M., and Mosisch, T.D. 1999. Ecosystem measures of river health
and their response to riparian and catchment degradation. Freshwater Biology
41:333-345.
Carlton, R.G. and Wetzel, R.G. 1988. Phosphorus flux from lake sediments: Effect of
epipelic algal oxygen production. Limnology and Oceanography 33:562-570.
Carpenter, S.R., Caraco, N.F., Correll, D.L., Howarth, R.W., Sharpley, A.N., and Smith,
V.H. 1998. Nonpoint pollution of surface waters with phosphorus and nitrogen.
Ecological Applications 8:559-568.
Colla, L.M., Reinehr, C.O., Reichert, C., and Costa, J.A.V. 2007. Production of biomass
and nutraceutical compounds by Spirulina platensis under different temperature
and nitrogen regimes. Bioresource Technology 98:1489-1493.
Correll, D.L. 1998. The role of phosphorus in the eutrophication of receiving waters: A
review. Journal of Environmental Quality 27:261-266.
Costerton, J.W., Cheng, K.J., Geesey, G.G., Ladd, T.I., Nickel, J.C., Dasgupta, M., and
Marrie, T. 1987. Bacterial biofilms in nature and disease. Annual Review of
Microbiology 41:435-464.
Dodds, W.K. 2003. The role of periphyton in phosphorus retention in shallow freshwater
aquatic systems. Journal of Phycology 39:840-849.
7
Dodds, W.K. and Welch, E. 2000. Establishing nutrient criteria in streams. Journal of the
North American Benthological Society 19:186-196.
Elser, J.J., Marzolf, E.R., and Goldman, C.R. 1990. Phosphorus and nitrogen limitation of
phytoplankton growth in the freshwaters of North America: A review and critique
of experimental enrichments. Canadian Journal of Fisheries and Aquatic Sciences
47:1468-1477.
Francoeur, S.N. 2001. Meta-analysis of lotic nutrient amendment experiments: Detecting
and quantifying subtle responses. Journal of the North American Benthological
Society 20:358-368.
Freeman, C., Chapman, P.J., Gilman, K., Lock, M.A., Reynolds, B., and Wheater, H.S.
1995. Ion exchange mechanisms and the entrapment of nutrients by river
biofilms. Hydrobiologia 297:61-65.
Havens, K.E., East, T.L., Hwang, S.J., Rodusky, A.J., Sharfstein, B., and Stienman, A.D.
1999. Algal responses to experimental nutrient addition in the littoral community
of a subtropical lake. Freshwater Biology 42:329-344.
House, W.A. and Casey, H. 1989. Transport of phosphorus in rivers. In: H. Tiessen, J.K.
Syers, L. Ryczkowski, and H.L Golterman (Ed.), Phosphorus cycles in terrestrial
and aquatic ecosystems. Regional SCOPE/UNEP workshop 1: Europe. University
of Saskatchewan, Saskatoon, Canada Institute of Pedology, p. 253-282.
Hynes, H.B.N. 1970. The ecology of running waters. Liverpool.
Kim, B.K., Jackman, A.P., and Triska, F.J. 1990. Modeling transient storage and nitrate
uptake kinetics in a flume containing a natural periphyton community. Water
Resources Research 26:505-514.
8
Knecht, M.F. and Göransson, A. 2004. Terrestrial plants require nutrients in similar
proportions. Tree Physiology 24:447-460.
Lampert, W. and Sommer, U. 1997. Limnoecology: The ecology of lakes and streams.
Oxford University Press, New York.
Lean, D.R.S. 1973. Movements of phosphorus between its biologically important forms
in lake water. Journal of the Fisheries Research Board of Canada 30:1525-1536.
Liess, A. and Hillebrand, H. 2006. Role of nutrient supply in grazer-periphyton
interactions: Reciprocal influences of periphyton and grazer nutrient
stoichiometry. Journal of the North American Benthological Society 25:632-642.
Likens, G.E. 1972. Eutrophication and aquatic ecosystems. In: G.E. Likens (Ed.),
Nutrients and eutrophication: The limiting-nutrient controversy. Proceedings of
the American Society of Limnology and Oceanography Special Symposia, Vol. 1.
Allen Press, Lawrence, Kansas, p. 3-13.
Lind, O.T., Doyle, R., Vodopich, D.S., and Trotter, B.G. 1992. Clay turbidity: Regulation
of phytoplankton production in a large, nutrient-rich tropical lake. Limnology and
Oceanography 37:549-565.
Lock, M.A., Wallace, R.R., Costerton, J.W., Ventullo, R.M., and Charlton, S.E. 1984.
River epilithon: Toward a structural-functional model. Oikos 42:10-22.
Mainstone, C.P. and Parr, W. 2002. Phosphorus in rivers—ecology and management. The
Science of the Total Environment 282-283:25-47.
Marcarelli, A.M., Bechtold, H.A., Rugenski, A.T., and Inouye, R.S. 2009. Nutrient
limitation of biofilm biomass and metabolism in the Upper Snake River basin,
southeast Idaho, USA. Hydrobiologia 620:63-76.
9
McColl, R.H.S. 1974. Self-purification of small freshwater streams: Phosphate, nitrate
and ammonia removal. New Zealand Journal of Marine and Freshwater Research
8:375-388.
McCormick, P.V. and Scinto, L.J. 1999. Influence of phosphorus loading on wetlands
periphyton assemblages: A case study from the Everglades. In: K.R. Reddy, G.A.
O’Connor, and C.L. Schelske (Ed.), Phosphorus biogeochemistry in subtropical
ecosystems. CRC Press, Boca Raton, FL, p. 301-319.
McCormick, P.V., Chimney, M.J., and Swift, D.R. 1997. Diel oxygen profiles and water
column community metabolism in the Florida Everglades, USA. Archiv für
Hydrobiologie 140:117-129.
McDowell, R.W. 2003. Sediment phosphorus chemistry and microbial biomass along a
lowland New Zealand stream. Aquatic Geochemistry 19:19-40.
Meyer, J.L., Paul, M.J., and Taulbee, W.K. 2005. Stream ecosystem function in
urbanizing landscapes. Journal of the North American Benthological Society
24:602-612.
Miltner, R.J. and Rankin, E.T. 1998. Primary nutrients and the biotic integrity of rivers
and streams. Freshwater Biology 40:145-158.
Moss, B.B.H., Irvine, K., and Stansfield, J. 1986. Restoration of two lowland lakes by
isolation from nutrient-rich water sources with and without removal of sediment.
Journal of Applied Ecology 23:391-414.
10
Mulholland, P.J., Helton, A.M., Poole, G.C., Hall, R.O., Hamilton, S.K., Peterson, B.J.,
Tank, J.L., Ashkenas, L.R., Cooper, L.W., Dahm, C.N., Dodds, W.K., Findlay,
S.E.G., Gregory, S.V., Grimm, N.B., Johnson, S.L., McDowell, W.H., Meyer,
J.L., Valett, H.M., Webster, J.R., Arango, C.P., Beaulieu, J.J., Bernot, M.J.,
Burgin, A.J., Crenshaw, C.L., Johnson, L.T., Niederlehner, B.R., O’Brien, J.M.,
Potter, J.D., Sheibley, R.W., Sobota, D.J., and Thomas, S.M. 2008. Stream
denitrification across biomes and its response to anthropogenic nitrate loading.
Nature 452:202-205.
Nijboer, R.C. and Verdonschot, P.F.M. 2004. Variable selection for modelling effects of
eutrophication on stream and river ecosystems. Ecological Modelling 177:17-39.
Noe, G.B., Childers, D.L., Edwards, A.L., Gaiser, E., Jayachandran, K., Lee, D., Meeder,
J., Richards, J., Scinto, L.J., Trexler, J.C., and Jones, R.D. 2002. Short-term
changes in phosphorus storage in an oligotrophic Everglades wetland ecosystem
receiving experimental nutrient enrichment. Biogeochemistry 59:239-267.
Nydick, K.R., Lafrancois, B.M., Baron, J.S., and Johnson, B.M. 2004. Nitrogen
regulation of algal biomass, productivity, and composition in shallow mountain
lakes, Snowy Range, Wyoming, USA. Canadian Journal of Fisheries and Aquatic
Sciences 61:1256-1268.
Pringle, C.M., Naiman, R.J., Bretschko, G.K., Karr, J.R., Oswood, M.W., Welcomme,
R.L., and Winterbourn, M.J. 1988. Patch dynamics in lotic systems. Journal of the
North American Benthological Society 7:503-524.
Raghothama, K.G. 1999. Phosphate acquisition. Annual Review of Plant Physiology and
Plant Molecular Biology 50:665-693.
11
Raymond, J., Zhaxybayeva, O., Gogarten, J.P., Gerdes, S.Y., and Blankenship, R.E.
2002. Whole-genome analysis of photosynthetic prokaryotes. Science 298:1616-
1620.
Reddy, K.R., Kadlec, R.H., Flaig, E., and Gale, P.M. 1999. Phosphorus retention in
streams and wetlands: A review. Critical Reviews in Environmental Science and
Technology 29:83-146.
Riber, H.H., Sorensen, J.P., and Kowalczewski, A. 1983. Exchange of phosphorus
between water, macrophytes and epiphytic periphyton in the littoral of
Mikołajskie Lake, Poland. In: R.G. Wetzel (Ed.), Periphyton of freshwater
ecosystems. W. Jung Publishers, The Hague, p. 235-243.
Ryder, D.S. and Miller, W. 2005. Setting goals and measuring success: Linking patterns
and processes in stream restoration. Hydrobiologia 552:147-158.
Sand-Jensen, K. 1983. Physical and chemical parameters regulating growth of periphytic
communities. In: R.G. Wetzel (Ed.), Periphyton of freshwater ecosystems. W.
Jung Publishers, The Hague, p. 89-96.
Schachtman, D.P., Reid, R.J., and Ayling, S.M. 1998. Phosphorus uptake by plants: From
soil to cell. Plant Physiology 116:447-453.
Schindler, D.E., Knapp, R.A., and Leavitt, P.R. 2001. Alteration of nutrient cycles and
algal production resulting from fish introductions into mountain lakes.
Ecosystems 4:308-321.
Schindler, D.W. 1977. Evolution of phosphorus limitation in lakes. Science 195:260-262.
Scinto, L.J. and Reddy, K.R. 2003. Biotic and abiotic uptake and release of phosphorus
by periphyton in a subtropical freshwater wetland. Aquatic Biology 77:203-222.
12
Sharpley, A.N. and Rekolainen, S. 1997. Phosphorus in agriculture and its environmental
implications. In: H. Tunney, O.T. Carton, P.C. Brookes, and A.E. Johnston (Ed.),
Phosphorus loss from soil to water. CAB International, Wallingford, p. 1-54.
Smith, V.H., Tilman, G.D., and Nekola, J.C. 1999. Eutrophication: Impacts of excess
nutrient inputs on freshwater, marine, and terrestrial ecosystems. Environmental
Pollution 100:179-196.
Steinman, A.D. and Mulholland, P.J. 2006. Phosphorus limitation, uptake, and turnover
in benthic stream algae. In: F.R. Hauer and G.A. Lamberti (Ed.), Methods in
stream ecology, 2nd
Ed. Academic Press, p. 187-212.
Stevenson, R.J. 2001. Using algae to assess wetlands with multivariate statistics,
multimetric indices, and an ecological risk assessment framework. In: R.R. Rader,
D.P. Batzer, and S.A. Wissinger (Ed.), Biomonitoring and management of North
American freshwater wetlands. Wiley, p. 113-140.
Svendsen, L.M., Kronvang, B., Kristensen, P., and Græsbøl, P. 1995. Dynamics of
phosphorus compounds in a lowland river system: Importance of retention and
non-point sources. Hydrological Processes 9:119-142.
Takei, K., Takahashi, T., Sugiyama, T., Yamaya, T., and Sakakibara, H. 2002. Multiple
routes communicating nitrogen availability from roots to shoots: A signal
transduction pathway mediated by cytokinin. Journal of Experimental Botany
53:971-977.
Tilman, D. 1988. Plant strategies and the structure and dynamics of plant communities.
Princeton University Press, Princeton, NJ.
13
USEPA (United States Environmental Protection Agency). 2006. Wadeable streams
assessment: A collaborative survey of the nation’s streams. US Environmental
Protection Agency Report EPA 841-B-06-002, Office of Water, Washington, DC.
Van Nieuwenhuyse, E.E. and Jones, J.R. 1996. Phosphorus-chlorophyll relationship in
temperate streams and its variation with stream catchment area. Canadian Journal
of Fisheries and Aquatic Sciences 53:99-105.
Welch, E.B. 1992. Ecological effects of wastewater. 2nd
Ed. Chapman and Hall, London,
UK.
Wetzel, R.G. 1996. Benthic algae and nutrient cycling in lentic freshwater ecosystems.
In: R.J. Stevenson, M.L. Bothwell, and R.L. Lowe (Ed.), Algal ecology:
Freshwater benthic ecosystems. Academic Press, New York, p. 641-667.
Withers, P.J.A. and Hodgkinson, R.A. 2009. The effect of farming practices on
phosphorus transfer to a headwater stream in England. Agriculture, Ecosystems
and Environment 131:347-355.
Withers, P.J.A. and Jarvie, H.P. 2008. Delivery and cycling of phosphorus in rivers: A
review. The Science of the Total Environment 400:379-395.
Young, R.G. and Huryn, A.D. 1999. Effects of land use on stream metabolism and
organic matter turnover. Ecological Applications 9:1359-1376.
Zhou, S. and Hosomi, M. 2008. Nitrogen transformations and balance in a constructed
wetland for nutrient-polluted river water treatment using forage rice in Japan.
Ecological Engineering 30:147-155.
14
CHAPTER 1
META-ANALYTICAL APPROACH TO EXPLAIN VARIATION IN MICROBIAL PHOSPHORUS
UPTAKE RATES IN AQUATIC ECOSYSTEMS1
1 Price KJ, Carrick HJ (2011) Meta-analytical approach to explain variation in microbial phosphorus
uptake rates in aquatic ecosystems. Aquatic Microbial Ecology 65:89-102.
15
1.1 ABSTRACT
Despite the fact that microbial uptake represents an important transformation of
nutrients in aquatic ecosystems, few comprehensive studies have identified key
parameters and evaluated their relative importance in explaining variation in uptake rates.
Therefore, I performed an assessment of peer-reviewed literature that reported aquatic
microbial phosphorus (P) uptake rates. The search yielded 36 different papers which
presented results of 102 uptake estimates. I then constructed a meta-analysis to examine
the effects of key parameters on uptake. Microbial group (benthic, planktonic), source
(culture, wild), and sample time (long, short) were significant parameters in explaining
observed variation in published P-uptake rates. Planktonic microbes had higher P-uptake
rates (65 μg P μg Chl a–1
d−1), compared with benthic (9 μg P μg Chl a
–1 d−1
). Lower
affinity for P by benthic microbes could be attributed to adnate growth forms, which can
create boundary layers separating cells from ambient P and promoting internal P cycling.
Cultured microbes exhibited higher P-uptake rates compared with wild samples, although
this trend was not significant (F= 2.63, p= 0.108), suggesting that cultured microbes in
these studies represented reasonable analogues. Shorter sampling times yielded over 3-
fold higher P-uptake rates (58 μg P μg Chl a–1
d−1
) and appear to represent more accurate
estimates of gross uptake. Microbes subject to longer time regimes may be
physiologically altered by experimental conditions; estimates might therefore not reflect
instantaneous uptake. My results highlight the influence of ecological variation on P
assimilation and provide criteria for developing a general model to predict observed
variation in microbial P-uptake rates.
16
1.2 INTRODUCTION
Despite the fact that microbial uptake represents an important transformation of
nutrients in aquatic ecosystems, few comprehensive studies exist that identify key
parameters and evaluate their relative importance in explaining variation in uptake rates.
Microbial uptake is a key component contributing to phosphorus (P) cycling models in
aquatic environments (Webster et al. 2009). Microbial uptake and other in-stream
processes resulted in annual removal of 33% of the soluble reactive phosphorus (SRP)
entering a first-order forested stream (Mulholland 2004). Additionally, periphyton on
submerged artificial substrata eliminated 0.83 mg P m-2
d-1
from eutrophic lakes (Jöbgen
et al. 2004). While benthic microbes have a substantial ability to alter P fluxes (Dodds
2003), their uptake velocities have been found to be lower than those of planktonic algae
and/or bacteria owing perhaps to boundary-layer constraints (Riber and Wetzel 1987,
Hwang et al. 1998). Conversely, benthic biofilms are typically surrounded by a
mucilaginous matrix of extracellular polymeric substances (e.g., polysaccharides;
Hoagland et al. 1993) which can aid in sequestration of nutrients from the environment.
For instance, polymeric secretions can act as a ‘sorptive sponge’ binding and
concentrating ions in proximity to cells (Decho and Herndl 1995, Decho 2000),
potentially leading to elevated uptake. These varying ideas in the literature suggest that
synthesizing kinetic ecological data on these separate but interrelated microbial groups
could shed light on P dynamics in aquatic systems.
While a number of studies have measured P-uptake rates for different microbial
groups, little has been done to integrate and evaluate uptake rates within and among
groups. Bacteria are particularly efficient in removing P at ambient concentrations owing
17
to their small size and high surface area to volume ratios and have proven superior
competitors in terms of uptake kinetics especially at low orthophosphate concentrations
(Currie and Kalff 1984a,b, Rosenberg and Ramus 1984). However, phytoplankton
display significant P-uptake kinetics as well (Tarapchak and Moll 1990) and can have
higher maximum uptake velocities (Vmax) compared with bacteria (e.g., Thingstad et al.
1993). That said, a synthesis of uptake rate values of these microbes could quantify a
breadth of rate estimations to determine general differences and lead to a better
understanding of variation in reported P-uptake rates. In addition to aforementioned
biological differences, previous uptake studies have found that variations in rates may
arise due to the effects of experimental and environmental variables (Reynolds and
Kristensen 2008). For example, microbial source (i.e., cultures vs. wild populations) or
duration of experimental manipulation can lead to variation which may influence the
reported results (Osenberg et al. 1999). This is not ideal because accurate uptake rates
are needed for the development of nutrient dynamics (retention) models for aquatic
systems (e.g., Marcé and Armengol 2009). Identifying potential sources of variation in
microbial P-uptake studies will promote a more comprehensive understanding and could
advance standardization of experimental techniques. Moreover, analysis on the
interactive effects of key ecological and experimental variables on P-uptake rates could
provide further insight into the interrelatedness of factors affecting uptake measurements,
which, to my knowledge, has not been previously attempted.
Meta-analysis is a method for quantitative synthesis and analysis of results and
has been used widely in ecology (Hays et al. 2005). Meta-analyses have been effective in
providing a quantitative and statistically valid method of comparing findings and
18
exploring variation among multiple ecological studies (Hedges and Olkin 1985, Arnqvist
and Wooster 1995, Gurevitch and Hedges 1999, Myers et al. 1999, Francoeur 2001,
Stewart 2010). I used a meta-analysis here because it allowed us to summarize and
statistically analyze a collection of independent data across a range of research studies
(e.g., Hedges and Olkin 1985, Gurevitch and Hedges 1999, Osenberg et al. 1999); this
approach further allowed us to examine key factors in explaining variation in P-uptake
rates. Specifically, I evaluated variation in P-uptake across a series of ecological and
experimental parameters (i.e., microbial group, source, experimental sample time,
system, region, and non-biological sorption) to explore natural and methodological
influences on reported uptake rates.
1.3 METHODS
I performed a comprehensive assessment of the published, peer-reviewed
literature to identify experimentally-derived estimates of P-uptake by aquatic microbes. I
used mainstream search engines and electronic databases to retrieve data published in the
primary scientific literature (i.e., Google Scholar, Biological Abstracts, Web of Science,
BioOne, and JSTOR). Additionally, a few journals (e.g., Journal of Phycology) were
manually searched from 1970 to 2009. I included in my analysis studies that met the
following criteria: (1) uptake was measured for one or both of the specific microbial
groups; (2) uptake was derived using P-loss from water or direct microbial P-uptake
experiments; and (3) uptake values were able to be normalized to common biomass units.
While a larger number of uptake measurements have been published, many were
excluded due to the absence of chlorophyll a (Chl a) data for biomass normalizations
19
(e.g., Rigler 1956). Adherence to these a priori conditions insured that statistical
comparisons were made on a common set of standard uptake quantities.
All uptake data were normalized for biomass and expressed in common units, μg
P μg Chl a–1
d−1
. Some studies that published biomass-normalized uptake estimates were
excluded due to expression in incomparable units (e.g., g P cell−1
h−1
) owing to scale
differences (Nan and Dong 2004). P-uptake rate (effect size) extracted from each study
was the response variable subjected to further analysis. Chl a values were converted
from dry weight (DW) using a literature derived average carbon (C):Chl a ratio of 49.38
for planktonic algae (Bothwell 1985, Riemann et al. 1989, Weisse et al. 1990, Cloern et
al. 1995, Coveney and Wetzel 1995) and 68.22 for benthic algae (de Jonge 1980,
Bothwell 1985, Gould and Gallagher 1990, Romaní and Sabater 2000) and assuming
carbon content of 41.94% for both microbial groups (Strickland 1965, Mayzaud and
Martin 1975, Härdstedt-Roméo 1982, Andersen and Hassen 1991, Jørgensen et al. 1991,
Klumpp et al. 1992, Anderson 1995, Qin et al. 2007). Chl a values were converted from
ash-free dry mass (AFDM) using an AFDM:Chl a ratio of 150, which has been used
before for periphyton (Warwick 2000); 4 values were converted from these units, all of
which were for benthic algae.
I selected key ecological/experimental parameters to test in my analyses, based on
primary research. ‘Microbial group’ was selected to test my a priori hypotheses
concerning distinct physiologies between these assemblages (see 1.2 Introduction).
‘Sample time’ was added to the model analysis due to its reported influence on uptake
kinetics (e.g., Harrison et al. 1989), and ‘source’ was added to test whether or not
cultured microbes adequately represent natural kinetics (Portielje and Lijklema 1994,
20
Hwang et al. 1998). Factors of ‘region’ and ‘system’ were also added to the analysis to
test for the influence of geographical and aquatic origin on predicting P-uptake,
respectively. Finally, testing for ‘abiotic sorption’ was factored into the analysis because
of the varied literature on abiotic processes. For instance, Klotz (1985) found that the
contribution of biotic processes to P cycling was minimal in comparison to abiotic
processes. However, Khoshmanesh et al. (1999) and Scinto and Reddy (2003) found
biotic processes accounted for a much larger proportion of P-uptake (45% and 83%,
respectively).
Data were then classified into microbial group (benthic and planktonic), source
(cultured and wild), sample time (long and short), region (North America and other),
system (lentic or lotic), and control for abiotic sorption (yes and no) according to
parameters reported. The source parameter was divided into wild samples and culture,
which included single species isolates. Sample time was divided into short (0 - 10
minutes) and long (> 10 minutes) because of a clear break point in the data set and
evidence for the importance of early sampling on the order of minutes or less in kinetic
studies (e.g., Goldman et al. 1981, Goldman and Glibert 1982). Finer resolution among
parameters was avoided to prevent rank deficiency in statistical tests. All data were
obtained from results reported in the publication text or extracted from published tables
and figures. For multiple uptake values reported in a single publication, I used the range
for each experiment as stand-alone data entries for the meta-analysis.
Meta-analysis was used to compare the variety of reported uptake rates for all
selected ecological/ experimental factors. Some have cautioned against the use of
parametric tests for meta-analyses due to variations between values from different
21
experiments and within experiments (Gurevitch and Hedges 1993, 1999, Stram 1996).
The distribution of these data was evaluated using a normal probability plot and an
Anderson-Darling normality test statistic. The plot of raw uptake data revealed a strong
skew and a heavy right tail, clearly showing that the data are inconsistent with a normal
distribution. The data were therefore log transformed (1+x) to base 10 to minimize
heteroscedasticity, attain a more normal distribution, and reduce the influence of outliers
(Zar 1974). While data transformation was implemented and successful in meeting the
homogeneity of variances assumption, the within-study variance (reference parameter)
could not be eliminated. Therefore, a mixed-model analysis was used because the
heteroscedasticity of the reference parameter violated assumptions of parametric
statistical tests (e.g., ANOVA), a common problem in meta-analyses owing to multiple
studies with dissimilar variances and sample sizes (Hedges and Olkin 1985). Gurevitch
and Hedges (1993) and Stram (1996) have proposed the use of mixed models in meta-
analysis as a way to combine the advantages of random and fixed effects models as they
incorporate a component of between-study variation into the estimates (Gurevitch and
Hedges 1999).
I constructed several mixed models in an effort to determine the best combination
of fixed and random effects in determining P-uptake rates. A null hypothesis testing
approach was used to determine the significance of fixed and random effects in models.
An additive approach was followed to the mixed-model fitting, where a series of
increasingly complex models were fit and significance of the added effects determined by
Bayesian information criterion (BIC) (Kass and Raftery 1995, Wasserman 2000). BIC is
used as criterion in model selection where a smaller BIC indicates better fit between the
22
model and the data; it can measure and compare the degree of support (evidence) in the
data for multiple competing models. The BIC approach to hypothesis testing may be an
improvement over statistical significance testing using p-values because of the ability to
test a number of alternative models (Raftery 1995). Further, BIC imposes a higher
penalty on the number of parameters (compared to AIC), and therefore leads to the
selection of less complicated models (Lin et al. 2009). I used model difference
calculations to determine statistical significance of the evidence in favor of one or the
other model hypotheses (Neuenschwander et al. 2003). When comparing alternative
models, BIC differences of 0 to 2 were interpreted as weak evidence of a given model
selection (i.e., impossible to discriminate between models), 6 to 10 showed strong
evidence, and > 10 showed very strong evidence (Raftery 1995).
The ‘base’ model was simply a null model containing only a fixed intercept used
for comparison purposes after addition of terms. Afterwards, a second model with a
random intercept (reference) was fit to test the hypothesis that there is significant
variation among references in average log(uptake rates). The ‘base’ model was nested
within the second model in that it only differed by a single random effect. In this case, a
simple (null) model was compared to a more complex (reference effect) model to see if
the added parameter should be used in later analyses. The random intercept (reference)
model was a linear mixed model fit using restricted maximum likelihood (REML). A
fixed grouping (categorical) microbial group term was then added as I proceeded fitting
progressively more complex models by adding fixed effect predictor variables. The
intent of this additive fitting was to attain the most parsimonious model (most complex,
23
yet simplest). Model assumptions were assessed, specifically normal distribution of
residuals, using quantile-quantile (Q-Q) plots of residuals and leverage plots.
Once the best model was derived, those parameters in the model were tested for
significance using parametric means (ANOVA). ANOVA (analysis of variance) has
been found to be a useful statistical test in meta-analyses (e.g., Brett and Goldman 1997).
I used ANOVA here given that my data met all assumptions for equality of variance
(among my groups) and normal, Gaussian data distribution (Gurevitch and Hedges 1999).
For instance, Bartlett’s and Levene’s tests (equal variances) confirmed homogeneity of
variance across all parameters (group, source, and sample time) examined in the meta-
analysis (p> 0.05). Mixed-model analyses were performed using R (R Development
Core Team 2006). Graphing was carried out using Minitab 16 software. Descriptive
statistics and parametric tests were performed using SPSS version 18.0. Mean values and
standard error of the mean are reported.
1.4 RESULTS
My review of the uptake literature identified 102 individual microbial P-uptake
experiments recorded in 36 different papers published over a 35 yr period (Table 1.1).
Results from the mixed-model BIC analysis showed that the microbial group model
(fixed intercept, random reference, and fixed microbial group) was favored (Δi= 0.00),
and thus had the greatest degree of support (evidence) in predicting P-uptake (Table 1.2).
The models were ranked in terms of performance in predicting P-uptake rates and further
showed that the model containing source was below the suggested maximum for
determining strong Bayesian evidence; the model containing the sample time parameter
24
was also smaller than the determination value, and therefore was analyzed further using
parametric means (see below). The system parameter was dropped from analysis due to
inherent correlation with microbial group. Region and abiotic mixed models were not
well supported by the Bayesian evidence.
Of the 102 individual P-uptake values reported and analyzed in this meta-analysis,
33 controlled for abiotic sorption, 29 of which were from wild microbial sources. While
some experiments controlled for abiotic sorption, values were typically not reported (e.g.,
Bothwell 1985, Hwang et al. 1998), limiting quantitative analyses. Background
phosphorus concentration ([P], in μg l−1
; range= <0.001 to 24.2 mg l−1
) was extracted
from 44 of the 102 experiments and subsequently regressed against corresponding P-
uptake values. P-uptake (log transformed) showed a significant linear regression against
[P] (log transformed) (r2= 0.14, F= 7.01, p= 0.011). Uptake vs. [P], split by microbial
group, showed a slight, negative relationship in benthic experiments and a slight, positive
relationship in planktonic experiments; however, neither regression was significant (p>
0.05). The biogeochemical conditions in which the studies were conducted varied greatly
and prohibited categorization for statistical analysis; experiment locations included the
Laurentian Great Lakes, limestone streams, wetlands, and marine areas (e.g., North Sea).
Of the 102 individual P-uptake estimates, 60% derived uptake from disappearance of P
from water while 40% derived from uptake of P directly into aquatic microbes. Mean P-
uptake rates were not different between these experiment methods (F= 0.14, p= 0.708).
Relatedly, 63% of the experiments used radiotracers in estimating P-uptake rates. Uptake
rates between tracer experiments (x = 56.3 μg P μg Chl a−1
d−1
) vs. non-tracer experiments
(x = 15.1 μg P μg Chl a−1
d−1
) were not significantly different (F= 1.92, p= 0.169). As
25
stated in methods, microbial group data were split into two groups to prevent rank
deficiency (interaction term with <1 observation for every combination of the factor
levels) which prevents full matrix calculations and analysis of all interactions in a three-
way ANOVA. Nonetheless, I parsed microbial group data into three groups (benthic
microbes, planktonic microbes, and bacterioplankton) to estimate the effect size of
bacteria on P-uptake. Rank deficiencies were present; I therefore analyzed these data
using a one-way ANOVA. No data were available for benthic bacteria, thus limiting my
analysis to planktonic bacterial groups. Bacterioplankton had 30 observations in the
dataset (x = 94.4 ± 25.8 μg P μg Chl a−1
d−1
). A post-hoc Tukey’s HSD test showed that
bacterial uptake was significantly different from benthic groups (p< 0.001) but not
different from planktonic groups (p= 0.112), thus warranting consolidation.
Three-way ANOVA results yielded a significant difference in mean P-uptake rate
between microbial groups (F= 21.13, p< 0.001) (Table 1-3). Interactions of microbial
group and sample time (F= 4.80, p= 0.031) and source and sample time (F= 4.22, p=
0.043) were also significant in the analysis. A three-way interaction of microbial group,
source, and sample time (F= 3.68, p= 0.058) was marginally significant. The mean P-
uptake rate for planktonic groups (65.2 ±14.4 μg P μg Chl a−1
d−1
) was 7.3 times greater
than benthic groups (8.9 ±4.1 μg P μg Chl a−1
d−1
) (Table 1-4, Figure 1.1). Experiments
using cultured microbes showed higher mean P-uptake (52.2 ±12.6 μg P μg Chl a−1
d−1
)
compared to wild microbes (31.0 ±12.2 μg P μg Chl a−1
d−1
). Experiments with shorter
sample times (0 - 10 minutes) had threefold higher mean P-uptake rates (58.0 ±14.3 μg P
μg Chl a−1
d−1) than longer experiments (17.5 ±5.4 μg P μg Chl a
−1 d−1
).
26
1.5 DISCUSSION
Bayesian information criterion supported my research hypothesis that microbial
group would be the strongest single predictor of P-uptake in the meta-analysis. This
finding was further supported in the literature as microbial uptake has been seen to
represent an important and significant biological mechanism for P removal in aquatic
systems (Confer 1972, McColl 1974, Tarapchak and Moll 1990, McCormick et al. 2006).
Planktonic microbes exhibited higher P-uptake rates than benthic microbes, which was in
accordance with my hypothesis. Since these two distinct communities occupy different
niches in aquatic systems (benthic vs. pelagic), there is no reason to believe that their
physiological ecology should be similar (Reuter et al. 1986). Additionally, the P
concentration required to saturate growth for benthic microbes is much greater compared
to planktonic (Reynolds 2006), likely due to the growth form of benthic assemblages
(Hill et al. 2009). Early P cycling research found that most orthophosphate uptake was
associated with the smallest aquatic particles (Rigler 1956). Moreover, some planktonic
groups have a relatively low P storage capacity (Vadstein et al. 1988), and competition
between different aquatic microbial groups for P is a function of both uptake and storage
capacity (Kilham 1978).
The lower uptake rates demonstrated by benthic microbes may be attributable to
boundary layer formation and internal recycling (Riber and Wetzel 1987). A diffusive
boundary layer has been shown to form around intact microbial biofilms (Jørgensen and
Revsbech 1985). The formation of these physical boundary layers atop intact benthic
films may insulate microbes and inhibit nutrient uptake from the water column into biota
within the film (Riber and Wetzel 1987, Reuter and Axler 1992, Hwang et al. 1998)
27
thereby reducing the ability of benthic microbes to compete for water column nutrients.
Kinetic calculations have shown that internal P recycling is a more significant source of
nutrients than external sources for benthic microbes (Riber and Wetzel 1987). These
results suggest that there is an intrinsic, physical, and resultant chemical conditioning in
benthic microbes that favors internal nutrient recycling over ambient water column
uptake. Nonetheless, benthic microbes can be competitive with planktonic groups for
pelagic P supply under some circumstances (e.g., Axler and Reuter 1996). Further,
benthic microbes on organic substrates can sequester nutrients associated with the
benthos (sediment-water interface) and limit the amount that is released into the water
column (Pringle 1990, Hagerthey and Kerfoot 1998, Woodruff et al. 1999). In this
regard, benthic microbes may outcompete planktonic groups for available nutrients
(Hansson 1990). That said, the substantial range of P-uptake reported in the literature for
both benthic and planktonic microbial groups may give some insight into the highly
dynamic nature of these assemblages, slowing development of robust dissolved nutrient-
biomass models for aquatic systems (Biggs 2000). When splitting the dataset into three
microbial groups, bacterioplankton showed the greatest uptake rates. This is in accord
with past research demonstrating that prokaryotes are capable of rapid and considerable
phosphate uptake in aquatic systems, especially during periods of low P levels (Jackim et
al. 1977). While bacteria may show higher specific orthophosphate uptake rates under P-
limiting conditions, phytoplankton in oligotrophic aquatic systems can substitute
phospholipids with non-P membrane lipid molecules (Van Mooy et al. 2009), effectively
reducing required P and thus remaining competitive with bacteria. Normalization to Chl
a would bias uptake rates if bacteria were the principal constituents, but their abundance
28
would be included in carbon, AFDM, and DW estimations (Podgórska et al. 2008,
Pusceddu and Danovaro 2009), all of which were included in this meta-analysis.
The sample time parameter provided moderately strong evidence for predicting P-
uptake, suggesting it is an important source of variation in experiments (Osenberg et al.
1999). Short sample times averaged over 3 times higher uptake rates than long
incubations, a finding accurately supported by other research. For instance, time-course
experiments of phosphate uptake in plankton showed that, over a shorter interval (0 - 60
minutes) rates were nearly threefold greater than those recorded over the longer interval
(60 - 120 minutes) (Suttle et al. 1988). Furthermore, the lower P-uptake rates found in
longer sampling times might suggest a biotic acclimation to nutrients over time and/or
oversight of instantaneous/‘surge’ uptake (Goldman et al. 1981, Harrison et al. 1989);
Vmax may decrease with increasing incubation time, especially for phosphate (Harrison et
al. 1989). Uptake parameters in cells can change rapidly based on the concentration of
pulsed nutrients; therefore, longer incubation periods and sampling times might neglect
changes in uptake occurring within minutes or seconds (Conway et al. 1976, Goldman
and Glibert 1982, Parslow et al. 1985). Longer incubation times represent uptake that
approximates net rates rather than gross rates. For example, incubation times of 30 to 40
minutes may ignore biotic uptake of remineralized/ excreted P and would result in
underestimates of gross uptake (Barlow-Busch et al. 2006). In shorter term experiments,
such nutrient recycling is probably minimized and estimates are assumed to represent
total (gross) uptake (Steinman and Mulholland 1996). Phosphate can be released from
intercellular P pools, as cytoplasmic P is readily exchangeable with external P (Cembella
et al. 1984b); P exchange may even exceed net uptake (Lean and Nalewajko 1976).
29
Further, consumer (e.g., herbivores) excretion could also regenerate bioavailable P
(Fisher and Lean 1992), thereby complicating net uptake estimates. While this would
certainly be evident in longer term experiments, short term experiments (<10 minutes)
would likely not have been influenced by these artifacts to as great a degree and thus may
represent membrane transport and initial (gross) uptake rates (see Wheeler et al. 1982).
Cultured microbes showed higher, albeit non-significant, P-uptake rates compared
to wild microbes. These findings suggest that extrapolation of P-uptake values obtained
using single species isolates or cultured organisms to natural environments may be
reasonable. Though extrapolation of single species P-uptake kinetic estimates to wild
communities may be dubious (e.g., Portielje and Lijklema 1994), present study and others
have found that cultured microbes represent a reasonable analogue for wild types. For
instance, competitive properties of single isolates of marine microbes in culture have
been found to agree with natural conditions (Russell and Fielding 1974). Further, P
limitation in freshwater ecosystems has been demonstrated at several levels of
complexity, from algal cultures to whole lakes (Hecky and Kilham 1988), suggesting
some consistency in physiological and environmental responses to phosphorus.
However, extrapolation is cautioned against, particularly for benthic groups, as spatial
microbial segregation can affect physical transport processes (Portielje and Lijklema
1994). The significant interaction of group and source/sample time may have been
driven by the large values observed for short sampling periods using cultured microbes
(x = 83.0 μg P μg Chl a–1
d−1
) and/or competition among wild microbes. Competition for
phosphorus among freshwater phytoplankton has been adequately modeled with internal
storage capacity and cell size parameters (Smith and Kalff 1983). In cultured samples
30
and single species isolates, diversity and competition is irrelevant; however, in wild,
natural samples there is likely to be strong competition for resources. Competitive
responses to nutrient alterations may require some phenotypic modification by microbes
(see Demars and Edwards 2007) and if time-dependent, may explain my significant
source/sample interaction. Further, given the controlled conditions in which culture
studies are typically performed (e.g., Plato and Denovan 1974), shorter time sampling
periods may be more practicable in these situations than in wild studies. For instance,
research with marine phytoplankton culture collections grown to steady state found that
the appropriate temporal scale of NH4+ uptake may be on the order of seconds (Goldman
and Glibert 1982). However, in a mixed species community (wild) and patchy environs,
different microbes must adjust their P-uptake system to external P fluctuations
(Istvánovics and Herodek 1995); rapid/short sampling uptake estimates therefore may not
be representative of an entire mixed assemblage, thereby yielding moderate rates.
The type of uptake experiments being executed was considered in the analysis and
showed that rates were similar between PO4-3
disappearance studies and gross/net uptake
studies of radiolabeled substrate. While many more experiments derived uptake
measurements from disappearance of P from the water, the actual rates reported were not
statistically different. Similarly, Burmaster and Chisholm (1979), in their comparison of
the two methods (isotopic tracer and disappearance experiments) for measuring PO4-3
uptake, found they yield comparable results and complement each other in practicality.
These findings suggest that between these studies, P efflux and/or remineralization
processes were negligible. Further, my findings may be related to the marginally
significant three-way interaction found between microbial group, source, and sample
31
time, which indicated unsystematic changes in the data for each microbial group across
source and time and suggested that there are additional complexities in interpreting
differences between experiment types. Experiments that used tracers (regardless of
experiment type) did not show significantly different P-uptake rates from those that did
not use tracers, suggesting that both methods are equally sensitive in estimating the
magnitude of P movements.
Despite my care in conducting this meta-analysis, it likely contained biases
(Osenberg et al. 1999). There were certain limitations and inherent difficulties in this
meta-analysis, including incomplete data reporting, lack of independence among effect-
size estimates (although neutralized through mixed modeling), publication bias, and
research bias (see Gurevitch and Hedges 1999). Investigators performing the research
may have introduced bias or preference in site/species selection depending on the
hypotheses being tested (Dodds and Welch 2000). In addition, methodologies,
investigator rigor, and spatial/temporal scales varied widely. As a result, there remains
some uncertainty with reference to microbial differences. Other factors, such as nutrient
transport limitation, abiotic uptake, and biphasic uptake can further affect uptake kinetics
in both of these microbial organisms (Dodds and Biggs 2002). With regards to abiotic
uptake, as incubation duration decreases, the importance of biological processes relative
to physical processes will also decrease (Collos 1983), and passive phenomena (e.g., ion
absorption) may be more responsible for measured uptake. I examined the abiotic
component of uptake and found it was not a significant predictor of P-uptake, suggesting
that sorptive processes may make up a small fraction of total uptake (e.g., Hwang et al.
1998). This finding is in accord with previous studies showing that biological demand
32
for P (uptake) exceeds abiotic sorption, which accounts for <15% of water column P
removal (Scinto and Reddy 2003). Nonetheless, the relative importance of adsorptive
processes depends on biogeochemistry and varies across stream condition gradients as
changes in land-use affect the affinity of abiotic processes for phosphorus (Aldridge et al.
2010). Attention to time-course experiments and the interpretation of results in light of
applied experimental conditions (i.e., substrate concentration, incubation, and experiment
duration) is important when deducing kinetic rates. Meta-analyses such as these can
provide an information platform on which to build a greater understanding of complex
environmental systems.
Despite these limitations, which are intrinsic to most meta-analytical approaches,
my study revealed important differences in uptake kinetics between aquatic microbes,
sample times, and microbial sources. These results indicated that both microbial groups
act as important drivers in aquatic P dynamics, although benthic microbes may be less
expeditious at removing nutrients from water column sources. The present synthesis and
quantitative analysis of P-uptake rates across a range of ecological and experimental
parameters has never been previously attempted to my knowledge. The present work is
an integrative and important step in realizing the potential of aquatic microbes to function
in nutrient transformations and thereby to mitigate nutrient transport to downstream
ecosystems (e.g., lakes, oceans). Further, this analysis shows that microbial group,
sample time, and microbial source are highly variable; likely dependent on study
objectives. Not surprisingly, the range of uptake values spanned over 3 orders of
magnitude in accord with previous discussions of variability in P-uptake rates of
microbes (e.g., Cembella et al. 1984a). While natural physiology may explain some of
33
this tremendous variation (Auer and Canale 1982), this meta-analysis sheds light on the
potential of different ecological and experimental approaches and parameters to drive
such varied rates of uptake. This approach has left the kinetic uptake field with a range
of ‘quantitative’ data that is difficult to interpret and may lead to misleading estimates of
true uptake (cf. Strayer 1985).
Microbes represent important P storage pools in aquatic systems; accurate
estimates on the dynamics of nutrient pools can aid in better understanding nutrient
compartments in flux models. Uptake can control dissolved nutrient concentrations as
well as influence regeneration/ remineralization, fundamental processes that maintain
most primary productivity in aquatic environments (Dodds 1993, Hudson et al. 1999).
Additionally, P-uptake rates are useful in modeling the distribution of nuisance algal
species associated with nutrient perturbations in aquatic systems (e.g., Auer and Canale
1980). Therefore, there is a need for accurate uptake estimates for models predicting the
effects of nutrient management strategies. Differences in experimental conditions may be
driving some of the variation in uptake rates observed in this meta-analysis, and,
therefore, it may be practical in future experiments to use standardized methodologies so
as to create more useful and relatable results (Hein et al. 1995). My results presented
here should help in development of aquatic nutrient models by offering insight into some
of the ecological factors important in predicting observed variation in microbial P-uptake
rates.
34
1.6 ACKNOWLEDGMENTS
I thank T. Wagner for valuable assistance with statistical analyses, M. J.
McCarthy for constructive comments on a previous version of this manuscript, and J. T.
Scannell for assistance with data entry. Funding for this research was provided to H.J.C.
by the Pennsylvania Department of Environmental Protection (Grant No. 4100034506).
1.7 LITERATURE CITED
Aldridge, K.T., Brookes, J.D., and Ganf, G.G. 2010. Changes in abiotic and biotic
phosphorus uptake across a gradient of stream condition. River Research and
Applications 26:636-649.
Andersen, T. and Hassen, D.O. 1991. Carbon, nitrogen, and phosphorus content of
freshwater zooplankton. Limnology and Oceanography 36:807-814.
Anderson, L.A. 1995. On the hydrogen and oxygen content of marine phytoplankton.
Deep Sea Research 43:1675-1680.
Arnqvist, G. and Wooster, D. 1995. Meta-analysis: Synthesizing research findings in
ecology and evolution. Trends in Ecology and Evolution 10:236-240.
Auer, M.T. and Canale, R.P. 1980. Phosphorus uptake dynamics as related to
mathematical modeling of Cladophora at a site on Lake Huron. Journal of Great
Lakes Research 6:1-7.
Auer, M.T. and Canale, R.P. 1982. Ecological studies and mathematical modeling of
Cladophora in Lake Huron. 2. Phosphorus uptake kinetics. Journal of Great Lakes
Research 8:84-92.
35
Auer, M.T. and Forrer, B.E. 1998. Development and parameterization of a kinetic
framework for modeling light- and phosphorus-limited phytoplankton growth in
Cannonsville Reservoir. Lake and Reservoir Management 14:290-300.
Axler, R.P. and Reuter, J.E. 1996. Nitrate uptake by phytoplankton and periphyton:
Whole-lake enrichments and mesocosm-15
N experiments in an oligotrophic lake.
Limnology and Oceanography 41:659-671.
Barlow-Busch, L., Baulch, H.M., and Taylor, W.D. 2006. Phosphate uptake by seston
and epilithon in the Grand River, Southern Ontario. Aquatic Sciences 68:181-192.
Berman, T. 1985. Uptake of [32
P]orthophosphate by algae and bacteria in Lake Kinneret.
Journal of Plankton Research 7:71-84.
Biggs, B.J.F. 2000. Eutrophication of streams and rivers: Dissolved nutrient-chlorophyll
relationships for benthic algae. Journal of the North American Benthological
Society 19:17-31.
Borchardt, M.A., Hoffmann, J.P., and Cook, P.W. 1994. Phosphorus uptake kinetics of
Spirogyra fluviatilis (Charophyceae) in flowing water. Journal of Phycology
30:403-417.
Bothwell, M.L. 1985. Phosphorus limitation of lotic periphyton growth rates: An intersite
comparison using continuous-flow troughs (Thompson River system, British
Columbia). Limnology and Oceanography 30:527-542.
Brett, M.T. and Goldman, C.R. 1996. Consumer versus resource control in freshwater
pelagic food webs. Science 275:384-386.
36
Burmaster, D.E. and Chisholm, S.W. 1979. A comparison of two methods for measuring
phosphate uptake by Monochrysis lutheri Droop grown in continuous culture.
Journal of Experimental Marine Biology and Ecology 39:187-202.
Cembella, A.D., Antia, N.J., and Harrison, P.J. 1984a. The utilization of inorganic and
organic phosphorous compounds as nutrients by eukaryotic microalgae: A
multidisciplinary perspective: Part I. Critical Reviews in Microbiology 10:317-
391.
Cembella, A.D., Antia, N.J., and Harrison, P.J. 1984b. The utilization of inorganic and
organic phosphorus compounds as nutrients by eukaryotic microalgae: A
multidisciplinary perspective. Part II. Critical Reviews in Microbiology 11:13-81.
Cloern, J.E., Grenz, G., and Vidergar-Lucas, L. 1995. An empirical model of the
phytoplankton chlorophyll:carbon ratio—the conversion factor between
productivity and growth rate. Limnology and Oceanography 40:1313-1321.
Collos, Y. 1983. Transient situations in nitrate assimilation by marine diatoms. 4. Non-
linear phenomena and the estimation of the maximum uptake rate. Journal of
Plankton Research 5:677-691.
Confer, J.L. 1972. Interrelations among plankton, attached algae, and the phosphorus
cycle in artificial open systems. Ecological Monographs 42:1-23.
Conway, H.L., Harrison, P.J., and Davis, C.O. 1976. Marine diatoms grown in
chemostats under silicate or ammonium limitation. II. Transient responses of
Skeletonema costatum to a single addition of the limiting nutrient. Marine Biology
35:187-199.
37
Corning, K., Duthie, H., and Paul, B. 1989. Phosphorus and glucose uptake by seston and
epilithon in boreal forest streams. Journal of the North American Benthological
Society 8:123-133.
Cotner, J.B. and Wetzel, R.G. 1992. Uptake of dissolved inorganic and organic
phosphorus compounds by phytoplankton and bacterioplankton. Limnology and
Oceanography 37:232-243.
Coveney, M.F. and Wetzel, R.G. 1995. Biomass, production, and specific growth rate of
bacterioplankton and coupling to primary production in an oligotrophic lake.
Limnology and Oceanography 40:1187-1200.
Currie, D.J. and Kalff, J. 1984a. A comparison of the abilities of freshwater algae and
bacteria to acquire and retain phosphorus. Limnology and Oceanography 29:298-
310.
Currie, D.J. and Kalff, J. 1984b. The relative importance of bacterioplankton and
phytoplankton in phosphorus uptake in freshwater. Limnology and Oceanography
29:311-321.
Davis, J.C. and Minshall, G.W. 1999. Nitrogen and phosphorus uptake in two Idaho
(USA) headwater wilderness streams. Oecologia 119:247-255.
Davis, L.S., Hoffmann, J.P., and Cook, P.W. 1990. Production and nutrient accumulation
by periphyton in a wastewater treatment facility. Journal of Phycology 26:617-
623.
de Jonge, V.N. 1980. Fluctuations in the organic carbon to chlorophyll a ratios for
estuarine benthic diatom population. Marine Ecology Progress Series 2:345-353.
38
Decho, A.W. 2000. Microbial biofilms in intertidal systems: An overview. Continental
Shelf Research 20:1257-1273.
Decho, A.W. and Herndl, G.J. 1995. Microbial activities and the transformation of
organic matter within mucilaginous material. The Science of the Total
Environment 165:33-42.
Demars, B. and Edwards, A. 2007. Tissue nutrient concentrations in freshwater aquatic
macrophytes: High inter-taxon differences and low phenotypic response to
nutrient supply. Freshwater Biology 52:2073-2086.
Dodds, W.K. 1993. What controls levels of dissolved phosphate and ammonium in
surface waters? Aquatic Sciences 55:132-142.
Dodds, W.K. 2003. The role of periphyton in phosphorus retention in shallow freshwater
aquatic systems. Journal of Phycology 39:840-849.
Dodds, W.K. and Biggs, B.J.F. 2002. Water velocity attenuation by stream periphyton
and macrophytes in relation to growth form and architecture. Journal of the North
American Benthological Society 21:2-15.
Dodds, W.K. and Welch, E. 2000. Establishing nutrient criteria in streams. Journal of the
North American Benthological Society 19:186-196.
Falkner, G., Strasser, P., and Graffius, D. 1984. Phosphate uptake by blue green algae in
vitro and in a lake during an algal bloom: Useful application of a force-flow
relationship. Hydrobiologia 108:265-271.
Fisher, T.R. and Lean, D.R.S. 1992. Interpretation of radiophosphate dynamics in lake
waters. Canadian Journal of Fisheries and Aquatic Sciences 49:252-258.
39
Francoeur, S.N. 2001. Meta-analysis of lotic nutrient amendment experiments: Detecting
and quantifying subtle responses. Journal of the North American Benthological
Society 20:358-368.
Goldman, J.C. and Glibert, P.M. 1982. Comparative rapid ammonium uptake by four
species of marine phytoplankton. Limnology and Oceanography 27:814-827.
Goldman, J.C., Taylor, C.D., and Glibert, P.M. 1981. Nonlinear time-course uptake of
carbon and ammonium by marine phytoplankton. Marine Ecology Progress Series
6:137-148.
Gould, D.M. and Gallagher, E.D. 1990. Field measurement of specific growth rate,
biomass, and primary production of benthic diatoms of Savin Hill Cove, Boston.
Limnology and Oceanography 35:1757-1770.
Gurevitch, J. and Hedges, L.V. 1993. Meta-analysis: Combining the results of
independent studies in experimental ecology. In: S.M. Scheiner and J. Gurevitch
(Ed.), The design and analysis of ecological experiments. Chapman and Hall,
New York, New York, USA, p. 378-398.
Gurevitch, J. and Hedges, L.V. 1999. Statistical issues in ecological meta-analyses.
Ecology 80:1142-1149.
Hagerthey, S.E. and Kerfoot, W.C. 1998. Groundwater flow influences the biomass and
nutrient ratios of epibenthic algae in a north temperate seepage lake. Limnology
and Oceanography 43:1227-1242.
Hansson, L-A. 1990. Quantifying the impact of periphytic algae on nutrient availability
for phytoplankton. Freshwater Biology 24:265-273.
40
Härdstedt-Roméo, M. 1982. Some aspects of the chemical composition of plankton from
the North-Western Mediterranean Sea. Marine Biology 70:229-236.
Harrison, P.J., Parslow, J.S., and Conway, H.L. 1989. Determination of nutrient uptake
kinetic parameters: A comparison of methods. Marine Ecology Progress Series
52:301-312.
Harrison, W.G., Azam, F., Renger, E.H., and Eppley, R.W. 1977. Some experiments on
phosphate assimilation by coastal marine plankton. Marine Biology 40:9-18.
Hays, G.C., Richardson, A.J., and Robinson, C. 2005. Climate change and marine
plankton. Trends in Ecology and Evolution 20:337-344.
Healey, F. 1973. Characteristics of phosphorus deficiency in Anabaena. Journal of
Phycology 9:383-394.
Hecky, R.E. and Kilham, P. 1988. Nutrient limitation of phytoplankton in freshwater and
marine environments: A review of recent evidence on the effects of enrichment.
Limnology and Oceanography 33:796-822.
Hedges, L.V. and Olkin, I. 1985. Statistical methods for meta-analysis. Academic Press,
Orlando, Florida.
Hein, M., Folager Pedersen, M., and Sand-Jensen, K. 1995. Size-dependent nitrogen
uptake in micro- and macroalgae. Marine Ecology Progress Series 118:247-253.
Hill, W.R., Fanta, S.E., and Roberts, B.J. 2009. Quantifying phosphorus and light effects
in stream algae. Limnology and Oceanography 54:368-380.
Hoagland, K.D., Rosowski, J.R., Gretz, M.R., and Roemer, S.C. 1993. Diatom
extracellular polymeric substances: Function, fine structure, chemistry and
physiology. Journal of Phycology 29:537-566.
41
Hudson, J.J., Taylor, W.D., and Schindler, D.W. 1999. Planktonic nutrient regeneration
and cycling efficiency in temperate lakes. Nature 400:659-661.
Hwang, S.J., Havens, K.E., and Steinman, A.D. 1998. Phosphorus kinetics of planktonic
and benthic assemblages in a shallow subtropical lake. Freshwater Biology
40:729-745.
Istvánovics, V. and Herodek, S. 1995. Estimation of net uptake and leakage rates of
orthophosphate from 32
P-uptake kinetics by a linear force-flow model. Limnology
and Oceanography 40:17-32.
Istvánovics, V., Shafik, H., Présing, M., and Juhos, S. 2000. Growth and phosphate
uptake kinetics of the cyanobacterium, Cylindrospermopsis raciborskii
(Cyanophyceae) in throughflow cultures. Freshwater Biology 43:257-275.
Jackim, E., Morisson, G., and Steel, R. 1977. Effects of environmental factors on
radiocadmium uptake by four species of marine bivalves. Marine Biology 40:303-
308.
Jöbgen, A.M., Palm, A., and Melkonian, M. 2004. Phosphorus removal from eutrophic
lakes using periphyton on submerged artificial substrata. Hydrobiologia 528:123-
142.
Jørgensen, B.B. and Revsbech, N.P. 1985. Diffusive boundary layers and the oxygen
uptake of sediments and detritus. Limnology and Oceanography 30:111-122.
Jørgensen, S.E., Nielsen, S.N., and Jørgensen, L.A. 1991. Handbook of Ecological
Parameters and Ecotoxicology. Elsevier, Amsterdam.
Karlson, B. 1989. Seasonal phosphate uptake by size-fractionated plankton in the
Skagerrak. Journal of Experimental Marine Biology and Ecology 127:141-154.
42
Kass, R.E. and Raftery, A.E. 1995. Bayes factors. Journal of the American Statistical
Association 90:773-795.
Khoshmanesh, A., Hart, B.T., Duncan, A., and Beckett, R. 1999. Biotic uptake and
release of phosphorus by a wetland sediment. Environmental Technology 29:85-
91.
Kilham, S.S. 1978. Nutrient kinetics of freshwater planktonic algae using batch and
semicontinuous methods. Internationale Vereinigung für Theoretische und
Angewandte Limnologie, Mitteilungen 21:147-157.
Klotz, R.L. 1985. Factors controlling phosphorus limitation in stream sediments.
Limnology and Oceanography 30:543-553.
Klumpp, D.W., Salita-Espinosa, J.S., and Fortes, M.D. 1992. The role of epiphytic
periphyton and macroinvertebrate grazers in the trophic flux of a tropical seagrass
community. Aquatic Botany 43:327-349.
Lean, D. and White, E. 1983. Chemical and radiotracer measurements of phosphorus
uptake by lake plankton. Canadian Journal of Fisheries and Aquatic Sciences
40:147-155.
Lean, D.R. and Nalewajko, C. 1976. Phosphate exchange and organic phosphorus
excretion by freshwater algae. Journal of the Fisheries Research Board of Canada
33:1312-1323.
Lin, D., Shkedy, Z., Burzykowski, T., Aerts, M., Gohlmann, H.W.H., De Bondt, A.,
Perera, T., Geerts, T., Van den Wyngaert, I., and Bijnens, L. 2009. Classification
of trends in dose-response microarray experiments using information theory
selection methods. The Open Applied Informatics Journal 3:34-43.
43
Lohman, K. and Priscu, J.C. 1992. Physiological indicators of nutrient deficiency in
Cladophora (Chlorophyta) in the Clark Fork of the Columbia River, Montana.
Journal of Phycology 28:443-448.
Marcé, R. and Armengol, J. 2009. Modeling nutrient in-stream processes at the watershed
scale using nutrient spiralling metrics. Hydrology and Earth System Sciences
13:953-967.
Mayzaud, P. and Martin, J.L.M. 1975. Some aspects of the biochemical mineral
composition of marine plankton. Journal of Experimental Marine Biology and
Ecology 17:297-310.
McColl, R.H.S. 1974. Self-purification of small freshwater streams: Phosphate, nitrate,
and ammonia removal. New Zealand Journal of Marine and Freshwater Research
8:375-388.
McCormick, P.V., Shuford III, R.B.E., and Chimney, M.J. 2006. Periphyton as a
potential phosphorus sink in the Everglades Nutrient Removal Project. Ecological
Engineering 27:279-289.
Mulholland, P.J. 2004. The importance of in-stream uptake for regulating stream
concentrations and outputs of N and P from a forested watershed: Evidence from
long-term chemistry records from Walker Branch Watershed. Biogeochemistry
70:403-426.
Myers, R.A., Bowen, K.G., and Barrowman, N.J. 1999. The maximum reproductive rate
of fish at low population sizes. Canadian Journal of Fisheries and Aquatic
Sciences 56:2404-2419.
44
Nan, C. and Dong, S. 2004. Comparative studies on phosphorus uptake and growth
kinetics of the microalga Tetraselmis subcordiformis and the macroalga Ulva
pertusa. Journal of Ocean University of China 3:56-59.
Neuenschwander, S., Brinkhof, M.W.G., Kölliker, M., and Richner, H. 2003. Brood size,
sibling competition, and the cost of begging in great tits (Parus major).
Behavioral Ecology 14:457-462.
Newman, S., Aldridge, F., Phlips, E., and Reddy, R. 1994. Assessment of phosphorus
availability for natural phytoplankton populations from a hypereutrophic lake.
Archiv für Hydrobiologie 130:409-427.
Nyholm, N. 1977. Kinetics of phosphate limited algal growth. Biotechnology and
Bioengineering 19:467-492.
Okada, M., Sudo, R., and Aiba, S. 1982. Phosphorus uptake and growth of blue-green
alga, Microcystis aeruginosa. Biotechnology and Bioengineering 24:143-152.
Osenberg, C.W., Sarnelle, O., Cooper, S.D., and Holt, R.D. 1999. Resolving ecological
questions through meta-analysis: Goals, metrics and models. Ecology 80:1105-
1117.
Parslow, J.S., Harrison, P.J., and Thompson, P.A. 1985. Ammonium uptake by
phytoplankton cells on a filter: A new high-resolution technique. Marine Ecology
Progress Series 25:121-129.
Pauli, A. and Kaitala, S. 1997. Phosphate uptake kinetics by Acinetobacter isolates.
Biotechnology and Bioengineering 53:304-309.
45
Perrin, C.J. 1993. Pilot fertilization of the Nechako River II: A test of nutrient deficiency
and periphyton response to nutrient addition. Nechako Fisheries Conservation
Program, Technical Report No. RM88-3.
Perry, M.J. 1976. Phosphate utilization by an oceanic diatom in phosphorus-limited
chemostat culture and in the oligotrophic waters of the Central North Pacific.
Limnology and Oceanography 21:88-107.
Planas, D., Maberly, S., and Parker, J. 1996. Phosphorus and nitrogen relationships of
Cladophora glomerata in two lake basins of different trophic status. Freshwater
Biology 35:609-622.
Plato, P. and Denovan, J.T. 1974. The influence of potassium on the removal of 137
Cs by
live Chlorella from low level radioactive wastes. Radiation Botany 14:37-41.
Podgórska, B., Mudryk, Z.J., and Skórczewski, P. 2008. Abundance and production of
bacteria in a marine beach (Southern Baltic Sea). Polish Journal of Ecology
56:405-411.
Portielje, R. and Lijklema, L. 1994. Kinetics of luxury uptake of phosphate by algae-
dominated benthic communities. Hydrobiologia 275/276:349-358.
Prieto, B., Pardo, M., Garbisu, C., Llama, M., and Serra, J. 1997. Phosphate uptake by
phosphorus-starved cells of the cyanobacterium Phormidium laminosum. World
Journal of Microbiology and Biotechnology 13:699-705.
Pringle, C.M. 1990. Nutrient spatial heterogeneity: Effects on community structure,
physiognomy, and diversity of stream algae. Ecology 71:905-920.
46
Pusceddu, A. and Danovaro, R. 2009. Exergy, ecosystem functioning and efficiency in a
coastal lagoon: The role of auxiliary energy. Estuarine, Coastal and Shelf Science
84:227-236.
Qin, P., Mayer, C.M., Schulz, K.L., Ji, X., and Ritchie, M.E. 2007. Ecological
stoichiometry in benthic food webs: Effect of light and nutrients on periphyton
food quantity and quality in lakes. Limnology and Oceanography 42:1436-1443.
R Development Core Team 2006. R: A language and environment for statistical
computing. R Foundation for Statistical Computing, Vienna, Austria.
Raftery, A.E. 1995. Bayesian model selection in social research. In: P.V. Marsden (Ed.),
Sociological methodology. Blackwell, Cambridge, MA, p. 111-163.
Reuter, J.E. and Axler, R.P. 1992. Physiological characteristics of inorganic nitrogen
uptake by spatially separate algal communities in a nitrogen-deficient lake.
Freshwater Biology 27:227-236.
Reuter, J.E., Loeb, S.L., and Goldman, C.R. 1986. Inorganic nitrogen uptake by epilithic
periphyton in a N-deficient lake. Limnology and Oceanography 31:149-160.
Reynolds, C.K. and Kristensen, N.B. 2008. Nitrogen recycling through the gut and
nitrogen economy of ruminants: An asynchronous symbiosis. Journal of Animal
Science 86:E293-E305.
Reynolds, C.S. 2006. The ecology of phytoplankton (ecology, biodiversity and
conservation). Cambridge University Press, Cambridge.
Riber, H.H. and Wetzel, R.G. 1987. Boundary layer and internal diffusion effects on
phosphorus fluxes in lake periphyton. Limnology and Oceanography 32:1181-
1194.
47
Riegman, R. and Mur, L. 1986. Phytoplankton growth and phosphate uptake (for P
limitation) by natural phytoplankton populations from the Loosdrecht lakes (The
Netherlands). Limnology and Oceanography 31:983-988.
Riemann, B., Simonsen, P., and Stensgaard, J. 1989. The carbon and chlorophyll content
of phytoplankton from various nutrient regimes. Journal of Plankton Research
11:1037-1045.
Rigler, F.H. 1956. A tracer study of the phosphorus cycle in lake water. Ecology 37:550-
562.
Romaní, A.M. and Sabater, S. 2000. Influence of algal biomass on extracellular enzyme
activity in river biofilms. Microbial Ecology 40:16-24.
Rosemarin, A.S. 1982. Phosphorus nutrition of two potentially competing filamentous
algae, Cladophora glomerata (L.) Kütz and Stigeoclonium tenue (Agardh) Kütz.
from Lake Ontario. Journal of Great Lakes Research 8:66-72.
Rosenberg, G. and Ramus, J. 1984. Uptake of inorganic nitrogen and seaweed surface
area:volume ratios. Aquatic Botany 19:65-72.
Runcie, J., Ritchie, R., and Larkum, A. 2004. Uptake kinetics and assimilation of
phosphorus by Catenella nipae and Ulva lactuca can be used to indicate ambient
phosphate availability. Journal of Applied Phycology 16:181-194.
Russell, G. and Fielding, A.H. 1974. The competitive properties of marine algae in
culture. Journal of Ecology 62:689-698.
Scinto, L.J. and Reddy, K.R. 2003. Biotic and abiotic uptake of phosphorus by
periphyton in a subtropical freshwater wetland. Aquatic Botany 77:203-222.
48
Smith, R.E.H. and Kalff, J. 1983. Competition for phosphorus among co-occurring
freshwater phytoplankton. Limnology and Oceanography 28:448-464.
Sorokin, Yu I. and Dallocchio, F. 2008. Dynamics of phosphorus in the Venice lagoon
during a picocyanobacteria bloom. Journal of Plankton Research 30:1019-1026.
Steinman, A.D. and Boston, H.L. 1993. The ecological role of aquatic bryophytes in a
woodland stream. Journal of the North American Benthological Society 12:17-26.
Steinman, A.D., Meeker, R.H., Rodusky, A.J., Davis, W.P., and Hwang, S.J. 1997.
Ecological properties of charophytes in a large, subtropical lake. Journal of the
North American Benthological Society 16:781-793.
Steinman, A.D. and Mulholland, P.J. 1996. Measurement of phosphorus limitation,
uptake, and turnover in stream algae. In: F.R. Hauer and G.A. Lamberti (Ed.),
Methods in stream ecology. Academic Press, San Diego, CA, p. 161-190.
Steinman, A.D., Mulholland, P.J., and Kirschtel, D.B. 1991. Interactive effects of nutrient
reduction and herbivory on biomass, taxonomic structure, and P uptake in lotic
periphyton communities. Canadian Journal of Fisheries and Aquatic Sciences
48:1951-1959.
Stewart, G. 2010. Meta-analysis in applied ecology. Biology Letters 6:78-81.
Stram, D.O. 1996. Meta-analysis of published data using a linear mixed-effects model.
Biometrics 52:536-544.
Strayer, D. 1985. The benthic micrometazoans of Mirror Lake, New Hampshire. Archiv
für Hydrobiologie, Supplement 72:287-426.
49
Strickland, J.D.H. 1965. Production of organic matter in the primary stages of the marine
food chain. In: J.P. Riley and G. Skirrow (Ed.), Chemical oceanography, Vol. 1.
Academic Press, London, p. 477-610.
Suttle, C.A., Stockner, J.G., Shortreed, K.S., and Harrison, P.J. 1988. Time-courses of
size-fractionated phosphate uptake: Are larger cells better competitors for pulses
of phosphate than smaller cells? Oecologia 74:571-576.
Sweerts, J-P.R.A., Gons, H.J., and Rijkeboer, M. 1986. Phosphate uptake capacity of
summer phytoplankton of the Loosdrecht Lakes in relation to phosphorus loading
and irradiance. Hydrobiological Bulletin 20:101-107.
Tarapchak, S.J. and Moll, R.A. 1990. Phosphorus sources for phytoplankton and bacteria
in Lake Michigan. Journal of Plankton Research 12:743-758.
Thingstad, T.F., Skjoldal, E.F., and Bohne, R.A. 1993. Phosphorus cycling and algal-
bacterial competition in Sandsfjord, western Norway. Marine Ecology Progress
Series 99:239-259.
Vadstein, O., Jensen, A., Olsen, Y., and Reinertsen, H. 1988. Growth and phosphorus
status of limnetic phytoplankton and bacteria. Limnology and Oceanography
33:489-503.
Van Mooy, B.A.S., Fredricks, H.F., Pedler, B.E., Dyhrman, S.T., Karl, D.M., Koblížek,
M., Lomas, M.W., Mincer, T.J., Moore, L.R., Moutin, T., Rappé, M.S., and
Webb, E.A. 2009. Phytoplankton in the ocean use non-phosphorus lipids in
response to phosphorus scarcity. Nature 458:69-72.
Warwick, J.J. 2000. Calleguas Creek Data Analysis in Support of TMDL Determination.
Report. Prepared for Larry Walker and Associates, Davis, CA.
50
Wasserman, L. 2000. Bayesian model selection and model averaging. Journal of
Mathematical Psychology 44:92-107.
Webster, J.R., Newbold, J.D., Thomas, S.A., Valett, H.M., and Mulholland, P.J. 2009.
Nutrient uptake and mineralization during leaf decay in streams—a model
simulation. International Review of Hydrobiology 94:372-390.
Weisse, T., Müller, H., Pinto-Coelho, R.M., Schweizer, A., Springman, D., and
Baldringer, G. 1990. Response of the microbial loop to the phytoplankton spring
bloom in a large prealpine lake. Limnology and Oceanography 35:781-794.
Wheeler, P.A., Glibert, P.M., and McCarthy, J.J. 1982. Ammonium uptake and
incorporation by Chesapeake Bay phytoplankton: Short term uptake kinetics.
Limnology and Oceanography 27:1113-1128.
Woodruff, S.L., House, W.A., Callow, M.E., and Leadbeater, B.S.C. 1999. The effects of
biofilms on chemical processes in surficial sediments. Freshwater Biology 41:73-
89.
Zar, J.H. 1974. Biostatistical analysis. Prentice-Hall, Inc, Englewood Cliffs, New Jersey.
51
Table 1.1. Phosphorus uptake rates (μg P μg Chl a–1
d–1
) measured for two microbial functional
groups and five other experimental variables analyzed in the meta-analysis with associated reference.
Original unit abbreviations: dry weight (DW), chlorophyll a (Chl a), carbon (C), ash-free dry mass
(AFDM). Chl a values were converted from DW using an average C:Chl a ratio of 49.38 for
planktonic algae and 68.22 for benthic algae and assuming a carbon content of 41.94% for both
microbial groups (see ‘Methods’ for details and references). Chl a values were converted using an
AFDM:Chl a ratio of 150.
Uptake
Values
Original
Units
Microbial
Group
Source Sample
Time
Region System Abiotic
Sorption
Reference
0.652 DW Benthic Culture Long N. America Lentic No Auer and Canale 1982
7.320 DW Benthic Culture Long N. America Lentic No Auer and Canale 1982
1.300 Chl a Planktonic Wild Short N. America Lentic No Auer and Forrer 1998
11.400 Chl a Planktonic Wild Short N. America Lentic No Auer and Forrer 1998
0.370 Chl a Planktonic Wild Short Other Lentic Yes Berman 1985
15.030 Chl a Planktonic Wild Short Other Lentic Yes Berman 1985
1.257 DW Benthic Culture Short N. America Lotic No Borchardt et al. 1994
1.847 DW Benthic Culture Short N. America Lotic No Borchardt et al. 1994
2.058 DW Benthic Culture Short N. America Lotic No Borchardt et al. 1994
3.135 DW Benthic Culture Short N. America Lotic No Borchardt et al. 1994
2.640 Chl a Benthic Wild Short N. America Lotic Yes Bothwell 1985
26.400 Chl a Benthic Wild Short N. America Lotic Yes Bothwell 1985
0.576 Chl a Benthic Wild Short N. America Lotic Yes Corning et al. 1989
1.920 Chl a Benthic Wild Short N. America Lotic Yes Corning et al. 1989
0.690 DW Planktonic Wild Short N. America Lentic No Cotner and Wetzel 1992
10.239 DW Planktonic Wild Short N. America Lentic No Cotner and Wetzel 1992
0.692 DW Planktonic Wild Short N. America Lentic No Cotner and Wetzel 1992
5.858 DW Planktonic Wild Short N. America Lentic No Cotner and Wetzel 1992
5.645 Chl a Benthic Wild Short N. America Lotic No Davis and Minshall 1999
23.285 Chl a Benthic Wild Short N. America Lotic No Davis and Minshall 1999
0.061 DW Benthic Wild Long N. America Lotic No Davis et al. 1990
0.126 DW Benthic Wild Long N. America Lotic No Davis et al. 1990
59.462 Chl a Planktonic Culture Short Other Lentic No Falkner et al. 1984
62.436 Chl a Planktonic Culture Short Other Lentic No Falkner et al. 1984
26.758 Chl a Planktonic Culture Short Other Lentic No Falkner et al. 1984
29.731 Chl a Planktonic Culture Short Other Lentic No Falkner et al. 1984
0.149 Chl a Planktonic Wild Long N. America Lentic Yes Harrison et al. 1977
0.795 Chl a Planktonic Wild Long N. America Lentic Yes Harrison et al. 1977
64.766 DW Planktonic Culture Long N. America Lentic No Healey 1973
0.754 DW Benthic Wild Long N. America Lentic Yes Hwang et al. 1998
52
30.905 DW Benthic Wild Long N. America Lentic Yes Hwang et al. 1998
0.480 DW Benthic Wild Long N. America Lentic Yes Hwang et al. 1998
177.511 DW Benthic Wild Long N. America Lentic Yes Hwang et al. 1998
0.158 DW Planktonic Wild Short N. America Lentic Yes Hwang et al. 1998
44.950 DW Planktonic Wild Short N. America Lentic Yes Hwang et al. 1998
0.424 DW Planktonic Wild Short N. America Lentic Yes Hwang et al. 1998
612.931 DW Planktonic Wild Short N. America Lentic Yes Hwang et al. 1998
247.572 C Planktonic Culture Short Other Lentic No Istvánovics et al. 2000
252.786 C Planktonic Culture Short Other Lentic No Istvánovics et al. 2000
305.998 C Planktonic Culture Short Other Lentic No Istvánovics et al. 2000
67.196 C Planktonic Culture Short Other Lentic No Istvánovics et al. 2000
361.343 C Planktonic Culture Short Other Lentic No Istvánovics et al. 2000
105.950 C Planktonic Culture Short Other Lentic No Istvánovics et al. 2000
228.847 C Planktonic Culture Short Other Lentic No Istvánovics et al. 2000
0.060 Chl a Planktonic Wild Long Other Lentic Yes Karlson 1989
0.260 Chl a Planktonic Wild Long Other Lentic Yes Karlson 1989
4.602 Chl a Planktonic Wild Short N. America Lentic No Lean and White 1983
64.800 Chl a Planktonic Wild Short N. America Lentic No Lean and White 1983
0.242 DW Benthic Culture Long N. America Lotic No Lohman and Priscu 1992
0.648 DW Benthic Culture Long N. America Lotic No Lohman and Priscu 1992
0.399 DW Benthic Culture Long Other Lentic No Nan and Dong 2004
14.400 Chl a Planktonic Wild Long N. America Lentic No Newman et al. 1994
57.600 Chl a Planktonic Wild Long N. America Lentic No Newman et al. 1994
62.172 DW Planktonic Culture Short Other Lentic No Nyholm 1977
53.694 DW Planktonic Culture Short Other Lentic No Nyholm 1977
14.978 DW Planktonic Culture Long Other Lentic No Okada et al. 1982
20.065 DW Planktonic Culture Long Other Lentic No Okada et al. 1982
0.283 DW Planktonic Culture Long Other Lentic No Okada et al. 1982
0.848 DW Planktonic Culture Long Other Lentic No Okada et al. 1982
33.912 DW Planktonic Culture Long Other Lentic No Pauli and Kaitala 1997
15.543 DW Planktonic Culture Long Other Lentic No Pauli and Kaitala 1997
16.391 DW Planktonic Culture Long Other Lentic No Pauli and Kaitala 1997
10.456 DW Planktonic Culture Long Other Lentic No Pauli and Kaitala 1997
2.400a Chl a Benthic Wild Short N. America Lotic Yes Perrin 1993
9.600a Chl a Benthic Wild Short N. America Lotic Yes Perrin 1993
106.510 Chl a Planktonic Culture Short N. America Lentic Yes Perry 1976
0.533 Chl a Planktonic Culture Short N. America Lentic Yes Perry 1976
142.040 Chl a Planktonic Culture Short N. America Lentic Yes Perry 1976
122.540 Chl a Planktonic Culture Short N. America Lentic Yes Perry 1976
0.860 Chl a Planktonic Wild Short N. America Lentic Yes Perry 1976
1.100 Chl a Planktonic Wild Short N. America Lentic Yes Perry 1976
6.329 DW Benthic Culture Long Other Lentic No Planas et al. 1996
9.596 DW Benthic Culture Long Other Lentic No Planas et al. 1996
53
11.513 DW Benthic Culture Long Other Lentic No Planas et al. 1996
11.174 DW Benthic Culture Long Other Lentic No Planas et al. 1996
1.163 DW Benthic Culture Long Other Lentic No Planas et al. 1996
4.536b Chl a Benthic Wild Long N. America Lotic Yes Price and Carrick 2008
1.440b Chl a Benthic Wild Long N. America Lotic Yes Price and Carrick 2008
0.072 Chl a Planktonic Culture Long Other Lentic No Prieto et al. 1997
0.041 Chl a Planktonic Culture Long Other Lentic No Prieto et al. 1997
7.200 Chl a Planktonic Wild Long Other Lentic No Riegman and Mur 1986
50.400 Chl a Planktonic Wild Long Other Lentic No Riegman and Mur 1986
136.800 Chl a Planktonic Wild Long Other Lentic No Riegman and Mur 1986
19.647 C Benthic Culture Short N. America Lentic No Rosemarin 1982
0.953 C Benthic Culture Short N. America Lentic No Rosemarin 1982
3.023 DW Benthic Culture Short Other Lentic No Runcie et al. 2004
8.001 DW Benthic Culture Short Other Lentic No Runcie et al. 2004
1.507 DW Benthic Culture Short Other Lentic No Runcie et al. 2004
0.439 DW Benthic Culture Short Other Lentic No Runcie et al. 2004
5.372 DW Benthic Wild Long N. America Lentic Yes Scinto and Reddy 2003
3.631 DW Benthic Wild Long N. America Lentic Yes Scinto and Reddy 2003
1.741 DW Benthic Wild Long N. America Lentic Yes Scinto and Reddy 2003
115.200 Chl a Planktonic Wild Short Other Lentic Yes Sorokin and Dallocchio 2008
151.200 Chl a Planktonic Wild Short Other Lentic Yes Sorokin and Dallocchio 2008
0.108 AFDM Benthic Wild Short N. America Lotic No Steinman and Boston 1993
0.648 AFDM Benthic Wild Short N. America Lotic No Steinman and Boston 1993
0.054 AFDM Benthic Wild Short N. America Lotic No Steinman et al. 1991
0.216 AFDM Benthic Wild Short N. America Lotic No Steinman et al. 1991
1.406 DW Benthic Wild Long N. America Lentic No Steinman et al. 1997
3.179 DW Planktonic Wild Short N. America Lentic No Steinman et al. 1997
27.554 DW Planktonic Wild Short N. America Lentic No Steinman et al. 1997
31.200 Chl a Planktonic Wild Long Other Lentic No Sweerts et al. 1986 a P-uptake rates extrapolated from Bothwell (1985) b Unpublished data
54
Table 1.2. Model selection results for predicting the model probability. K is the number of
parameters in the model including an intercept, log(L) is the log likelihood value for each model, BIC
is the Bayesian information criterion, and ∆i is the difference between the BIC value for each model
and the model with the lowest BIC (BICi - BICmin). Models were calculated using log10-transformed
P-uptake. The intercept/reference model was used as the reduced ‘base’ model, and thus not listed.
Parameter K log(L) BIC Δi
Microbial Group 4 -185.8 390.17 0.00*
Source 5 -183.9 391.02 0.84*
Sample Time 6 -183.2 394.22 4.04*
Region 7 -183.0 398.34 8.16
Abiotic Sorption 8 -182.9 402.73 12.56
* Parameters preserved for analysis with parametric statistics based on strength of evidence
55
Table 1.3. ANOVA results examining differences among predictors of log10-transformed P-uptake
rate (μg P μg Chl a–1
d–1
): microbial group, source, and sample time. Note: mean difference is
significant at α= 0.05 level.
Variable Type III Sum
of Squares
df Mean
Square
F p
Corrected Model 21.581 7 3.083 7.708 0.000
Intercept 75.866 1 75.866 189.671 0.000
Group 8.450 1 8.450 21.127 0.000
Source 1.052 1 1.052 2.631 0.108
Sample Time 1.247 1 1.247 3.117 0.081
Group * Source 1.273 1 1.273 3.182 0.078
Group * Sample Time 1.919 1 1.919 4.799 0.031
Source * Sample Time 1.690 1 1.690 4.224 0.043
Group * Source * Sample Time 1.471 1 1.471 3.679 0.058
Error 37.599 94 0.400
Total 152.324 102
Corrected Total 59.180 101
56
Table 1.4. Descriptive statistics for all experimental parameters tested in the meta-analysis of P-
uptake rates (μg P μg Chl a–1
d–1
). Missing fields are due to rank deficiency.
Microbial
Group
Source Sample
Time
Region System Abiotic
Sorption
N Mean Std.
Deviation
Benthic Culture Long N. America Lentic No 2 3.99 4.72
Lotic No 2 0.45 0.29
Other Lentic No 6 6.70 4.94
Short N. America Lentic No 2 10.30 13.22
Lotic No 4 2.07 0.78
Other Lentic No 4 3.24 3.34
Wild Long N. America Lentic No 1 1.41 -
Yes 7 31.48 65.28
Lotic No 2 0.09 0.05
Yes 2 2.99 2.19
Short N. America Lotic No 6 4.99 9.22
Yes 6 7.26 9.90
Planktonic Culture Long N. America Lentic No 1 64.77 -
Other Lentic No 10 11.26 11.20
Short N. America Lentic Yes 4 92.91 63.27
Other Lentic No 13 143.38 117.63
Wild Long N. America Lentic No 2 36.00 30.55
Yes 2 0.47 0.46
Other Lentic No 4 56.40 56.44
Yes 2 0.16 0.14
Short N. America Lentic No 10 13.03 19.90
Yes 6 110.07 246.99
Other Lentic Yes 4 70.45 74.17
57
Group
Source
Time
PlanktonicBenthic
WildCultureWildCulture
ShortLongShortLongShortLongShortLong
3.0
2.5
2.0
1.5
1.0
0.5
0.0
(lo
g)
P U
pta
ke
Ra
te
Figure 1.1. Boxplot of log10-transformed P-uptake rates for the best predictor parameters
determined using mixed modeling: sample time, source, and microbial group. Boxplots show the
median value (line) 25 and 75% quartiles (box), upper and lower limits (whiskers), and outliers (>1.5
times interquartile range; asterisks).
58
CHAPTER 2
EFFECTS OF PHYSICAL DISTURBANCE ON PHOSPHORUS UPTAKE IN TEMPERATE
STREAM BIOFILMS
59
2.1 ABSTRACT
The evaluation of microbial biofilm nutrient uptake kinetics can provide insight
into uptake mechanisms that regulate stream productivity. While kinetic uptake
experiments have been performed on stream biofilms, there has not been an evaluation of
disturbance (removal) techniques and the effects that such abrasive methods may have on
benthic microbes. Therefore, the goal of this study was to evaluate common removal
techniques used in metabolic studies on benthic biofilms to determine effects on
phosphorus (P) uptake rate, physiological capability, and abiotic sorption. Artificial
substrata were collected from two reaches along a temperate stream; resident biofilms
were either removed via scraping or left intact. A series of short-term radiotracer
(H333
PO4) experiments were then conducted to measure P-uptake. In vivo
autofluorescence was measured as a proxy of algal physiological condition. The
experiments showed no difference in P-uptake rates (μgP/μgChl/d) between the scraped
(x = 0.77 ±0.11 (SE) μgP/μgChl/d) and intact (x = 0.91 ±0.17 (SE) μgP/μgChl/d) biofilms
(t= 0.69, p= 0.492, df= 33). Further, microbial physiology was not depressed by physical
disturbance. While killed samples yielded significantly lower uptake compared to live
biota (F= 17.51, p= 0.001), abiotic sorption still accounted for a moderate fraction of total
uptake and thus warrants estimation in metabolic studies. Overall, these findings lend
credence to the numerous experiments that investigate benthic microbial physiologic
responses post-disturbance and highlight the importance of uptake following common
physical disturbances that occur in turbulent environments.
60
2.2 INTRODUCTION
Streams are dynamic features of the landscape in part because they serve as
sediment and nutrient transport avenues (Hall et al. 2002); specifically, small streams
(width ≤10 m) represent up to 85% of total stream length in most watersheds, provide a
crucial link between terrestrial and aquatic environs, and are key elements of nutrient
transformation and downstream transport (Peterson et al. 2001, Sweeney et al. 2004).
Benthic stream biofilms are capable of assimilating and effectively retaining nutrients
that would otherwise transport downstream (Dodds 2003). Biotic uptake/assimilation
transfers inorganic nutrients into particulate form and in so doing can buffer downstream
ecosystems from more soluble reactive forms (Svendsen et al. 1995). As such, biofilms
are critical in the removal of dissolved phosphorus (P) from stream water and are major
components in stream self-purification (Sabater et al. 2002, Covich et al. 2004).
Nonetheless, owing to their complex structure, some biofilm components may be more
susceptible to physical disturbance and removal (i.e., top layer of biofilm matrix);
periodic sloughing and storm spates can act to remove portions of the biofilm and
transport them downstream (Peterson and Stevenson 1990, Biggs 1996). While physical
disturbance and removal of the biofilm is a natural occurrence and previous physiological
assays have employed physical scraping/disturbance as a biofilm extraction method (e.g.,
Tank and Webster 1998, Thompson and Sinsabaugh 2000, Miranda et al. 2007), the
effect that such disruption has on cell physiology and viability is largely unknown. That
said, researchers have inferred uptake rates from direct incorporation of radiolabel into
microbial components (e.g., Hwang et al. 1998, Scinto and Reddy 2003), while others
have used the loss of activity from the overlying water over time (e.g., Odum et al. 1958,
61
Steinman and Mulholland 2006); such differences in experimental methodologies may
produce variation in uptake rates across varying temporal and spatial scales (cf. Price and
Carrick 2011).
Nutrient retention in aquatic systems is hence a consequence of this active
biological uptake but also non-biological sorption (Haggard et al. 1999); the latter factor
can be considerable, if not the dominant process, under some conditions (Rejmánková
and Komárková 2000, Aldridge et al. 2009). However, few studies routinely test for
abiotic sorption (e.g., Steinman and Mulholland 2006) or provide measured abiotic
sorption fractions in uptake experiments, and quantitative reports on non-biological
sorption rates are varied. For example, Klotz (1985) found that the contribution to P
cycling by abiotic processes was much greater in comparison to biotic processes.
However, Mulholland et al. (1983) and Scinto and Reddy (2003) found that abiotic
process accounted for a much smaller proportion of P-uptake (10.3 and <15%,
respectively).
Michaelis-Menten (M-M) kinetics have been used to describe nutrient uptake into
biofilms (Reuter et al. 1986), facilitating estimations of the maximum uptake rate (Vmax)
and half-saturation constant (Km), although there are many instances where this model is
not supported (e.g., Tarapchak and Herche 1986). Similarly, nutrient uptake by higher
plants typically follows M-M saturation kinetics and is described by the parameters Vmax,
Km, and Cmin, the minimum nutrient concentration required for uptake to occur (Nielsen
1979, Akhtar et al. 2007). The Cmin or ‘threshold’ value has not been widely estimated
before in the aquatic phycology literature (Aubriot et al. 2000, Wagner and Falkner
2001), but may be an important factor in uptake models.
62
Therefore, the objective of this study was to examine the effect of procedural
techniques (physical disturbance) on P-uptake rate (μgP/μgChl/d); specifically I sought to
estimate: 1) differences in uptake between physically scraped and intact biofilms through
single endpoint techniques (Collos 1983) and 2) M-M kinetic parameters (Vmax, Km, and
Cmin) along varying time-courses. In addition, cell viability post-disturbance and abiotic
sorption was also investigated.
2.3 METHODS
P-uptake rates of intact vs. scraped biofilms (single time point)
A set of six unglazed ceramic tiles (surface area= 8.42 cm2) were secured to
cement blocks and established in an upstream (40.7786, -77.7696) and downstream
(40.8222, -77.8369) reach in a 3rd
order stream in a mixed land-use watershed in
Pennsylvania (USA) that eventually drains into the Chesapeake Bay (Spring Creek)
(Appendix A). Unglazed tiles have been used extensively as standardized substrate for
biofilm colonization and development to reduce sample variability (Lamberti and Resh
1985). The tiles were incubated for 30 days to allow the development of mature biofilm
assemblages (Biggs 1988), after which the tiles with intact biofilms were placed into 60
mL translucent polypropylene (Nalgene) incubation jars filled with site-specific water.
Once returned to the laboratory, three tiles from each site were scraped with a stiff
bristled brush and three tiles were left intact. For the scraped samples, 1 mL of biofilm
material with water (slurry) was removed from each replicate chamber using a sterile
repeat pipette 10 minutes after tracer (H333
PO4) injection. Samples were then injected
into a 12-place Millipore filter manifold fitted with 0.20 μm Versapor filters and 15 mL
63
glass water collection tubes. Vacuum filters and water were separately placed into
labeled 20 mL scintillation vials filled with 5 mL Ecolume scintillation cocktail (ICN
Pharmaceuticals, Costa Mesa, CA, USA) and read for activity using liquid scintillation
counting (LSC) (model LS 6000 IC; Beckman-Coulter, Fullerton, CA). For the intact
samples, 1 mL of water (overlying biofilms) was removed from each replicate chamber
using a sterile pipette, again 10 minutes after tracer injection. No obvious signs of seston
were present in the incubation jars; however, filtration was still performed prior to
removing water samples using a 5cc BD Luer-lock syringe with interchangeable filters
(Acrodisc 13 mm syringe filter with 0.2 µm Supor membrane; Pall Corp., Ann Arbor,
MI, USA). The radioactivity in the samples was determined again by LSC. Uptake rates
were calculated as ln [P0/(P0-X)] over time, where P0 is the total activity in 1mL of
radiolabelled sample and X is the radioactivity on the filter (biota), following Hwang et
al. (1998). For all samples, counts per minute (CPM) were converted into disintegrations
per minute (DPM) using an internal quench curve. These experiments differ from kinetic
experiments in that only a single time point was sampled (limiting the ability to estimate
kinetic parameters); rather the goal of these experiments was to validate single time point
estimates (e.g., Collos 1983) of P-uptake between scraped and intact benthic biofilms.
These experiments were performed in triplicate and conducted on 10/17/08, 6/3/09, and
6/11/09. Chlorophyll-a (Chl) concentrations were determined following standard
fluorometric methods using a Turner 10-AU fluorometer (Carrick et al. 1993) and
converted to areal densities (mg/m2).
64
M-M parameters along distinct time-courses and short-term flux estimations
Similar to the single-time point estimation experiments, an additional set of 20
unglazed ceramic tiles were secured to cement blocks and established in an upstream
(40.7786, -77.7696) reach as before (Spring Creek, USA). Again, tiles supporting intact
biofilms were placed into incubation jars after 30 days of residence in the stream. Once
tiles were returned to the laboratory, intact biofilms were physically removed from ten of
the collected tiles while the other ten tiles were left with intact biofilms. Next, all 20
experimental containers were amended with increasing concentrations of KH2PO4 (0 -
500 µgP/L). Duplicate experimental jars were amended with five [P]: 0 (control), 10, 20,
200, and 500 µgP/L (n= 10). For the scraped samples, 1 mL of biofilm material with
water (slurry) was removed from each replicate chamber using a sterile repeat pipette 5,
12, 18, 30, and 60 minutes after tracer (H333
PO4) injection. As before, samples were
injected into a filter manifold and subsequently placed into labeled 20 mL scintillation
vials filled with scintillation cocktail and read for activity using LSC. For the intact
samples, 1 mL of water (overlying biofilms) was removed from each replicate chamber
using a sterile pipette, again 5, 12, 18, 30, and 60 minutes after tracer injection. Seston
uptake was measured in a previous experiment and determined to be a negligible
component; however, filtration was still performed prior to removing water samples. The
radioactivity in the samples was determined again by LSC.
Abiotic sorption, P storage, and cell viability
The non-biological (abiotic) sorption of phosphorus by stream benthic biofilms
was also investigated using natural rock substrates collected from the same upstream and
downstream reach (Spring Creek, USA). On 17 June 2009, biofilms were physically
65
removed from rock substrates in the field, washed into labeled containers, and returned to
the laboratory. Slurries from each site (n= 16) were incubated with the following agents:
1) 10% formaldehyde, 2) 3% glutaraldehyde (Wolfstein et al. 2002), and 3) control (no
inhibitory agents). Sixty minutes after agents were injected into incubation jars, 0.50 mL
carrier-free H333
PO4 radiotracer was injected into each sample. After 10 minutes, slurry
was removed from each incubation jar and filtered. Filters were thoroughly rinsed of any
non-specific 33
P binding, placed into scintillation vials filled with 5 mL Ecolume
scintillation cocktail, and activities estimated via LSC. It should be noted that this was a
first-order estimation, as the effect of formaldehyde/glutaraldehyde fixation on the
possible sorption of P is unknown.
In addition, the effect of increasing levels of disturbance on biofilms was
examined in the field using three different physical scraping techniques with a stiff-
bristled brush (three rocks/ technique): one pass (low disturbance), two passes (medium
disturbance), and multiple passes (high disturbance). For identification of soft algae, 20
mL from each disturbance treatment (n= 9) was preserved with 1% formalin and cells
were identified to genus under 100x and 400x magnifications using a Leica light
microscope (Carrick and Steinman 2001). Large (≈ 300µm) cells were identified to
genus first under 100x magnification using a Palmer-Maloney counting chamber (394 sq.
mm in area, 0.1 mL volume). Remaining cells were counted at 400x magnification
(random fields) to a minimum of 200 cells until 400 cells total were reached under both
100x and 400x. Cells were then grouped by growth form (filamentous, stalked/erect, and
prostrate/adnate). Physiognomic (growth form) classifications of microbial genera
followed Graham and Vinebrooke (1998), Wellnitz and Ward (2000), and Passy (2007).
66
Samples from experiments were filtered (Whatman EPM 2000 glass fiber) for
polyphosphate (poly-P) via hot-water extraction and analyzed using spectrophotometry
following standard methods (Fitzgerald and Nelson 1966, Eixler et al. 2005). These
samples were also analyzed for Chl content using a Turner 10-AU fluorometer (Carrick
et al. 1993).
Autofluorescence was measured (via fluorometry) as a proxy of algal
physiological condition (n= 9, three rocks/disturbance treatment). Sample was poured
into a cuvette and read in vivo on a fluorometer. Cuvettes were then placed in dark
conditions for one minute and 0.1 mL of DCMU (3-(3,4-dichlorophenyl)-1,1-
dimethylurea) –a photosynthetic inhibitor– stock (10-3
M) was added to each sample
(final concentration= 10-5
M) which blocks electron transport from photosystem II to
photosystem I resulting in maximum fluorescence (Prezelin 1981). Cellular fluorescence
capacity (CFC) (proportion of absorbed light being used in photosynthesis) was
calculated as:
(Fa – Fb)/ Fa (Eq. 1)
where Fa is in vivo fluorescence post DCMU addition and Fb is in vivo fluorescence
(Vincent 1980, Vyhnalek et al. 1993). CFC should vary directly with photosynthetic
activity (physiological condition) and inversely with negative effects on photosynthetic
activity (e.g., physical disturbance) (Vincent 1981, Thompson 1997).
Clean techniques
Clean techniques were strictly followed due to the sensitive nature of
radioisotopic experimentation and to avoid any potential contamination which could
influence biofilm P-uptake. These techniques will be briefly discussed here. Small
67
amounts of trace elements can affect phytoplankton (Fitzwater et al. 1982) and
periphyton (Hill et al. 2000) metabolic processes. Contamination during uptake
experiments using radiotracers may include impurities such as metals, inadvertently
added during sampling which may reduce the precision of uptake measurements.
Therefore, clean techniques and sterile instrumentation/apparatus were followed and used
in order to reduce any potential contamination and generate the most accurate uptake
measurements. Following suggestion from Fitzwater et al. (1982), biofilm communities
were incubated in polycarbonate jars instead of glass. Polycarbonate has been shown to
adsorb very little manganese (Mn) and zinc (Zn) compared to glass containers that can
rapidly bind trace metals (Fitzwater et al. 1982). Incubation jars used for each
experiment were cleaned in phosphate free detergent, acid-washed (10% hydrochloric
acid at 20° C for at least 24 hours), and rinsed with deionized water from a Barnstad
Nano-Pure system (Singh et al. 2006) prior to use. Sterile pipette tips constructed of inert
virgin CFR 21 compliant polypropylene (VWR Int.) were used to inject tracer into
samples (Rehder and Borges 2010). Tracer stock solution was stored in a high-density
polyethylene wide mouth bottle at 4° C until use. Additionally, disposable powder-free
gloves (nitrile) and laboratory coats were worn during experiments to protect against
biohazards and radioactivity as well as to prevent any contamination from clothing and/or
skin. By employing clean sampling procedures, using polycarbonate incubation vessels,
and sterile equipment, P-uptake experiments can avoid some of the most serious sources
of trace metal contamination (Fitzwater et al. 1982).
68
Statistical analyses
A mixed-model three-way analysis of variance (ANOVA) was performed to
evaluate variation in P-uptake rate (μgP/μgChl/d) between disturbance treatment (scraped
and intact biofilms), site location (up and downstream), and experiment time (sampling
dates: 17 October 2008, 3 June 2009, and 11 June 2009). Site was treated as a random
effect here, as there were no a priori predictions regarding differences among particular
stream locations, therefore, a mixed-model ANOVA approach was most applicable
(McKone and Lively 1993). For statistical comparisons, I determined the uptake into the
biota from the intact samples by subtracting the activity measured in the water and
activity on the filter (seston) from the total activity. ANOVA was also used to analyze
the differences between controls and poisoning agents in estimating biotic vs. abiotic
uptake. All descriptive statistics and ANOVA analyses were performed using SPSS
software version 19.0 (SPSS Inc., Chicago, IL, USA). M-M parameters (Vmax and Km)
were estimated using an iterative ‘nls’ function (nonlinear least-squares regression) in the
statistical package R (R Development Core Team 2006, Marino et al. 2010).
Observations with DFFITS values greater than 2√(k/n), where k is the number of
predictors (including constant), were removed from regression analyses, following
Belsley et al. (1980). To estimate Cmin, or the threshold [P] at which net incorporation of
P by the biofilm ceases due to insufficient available energy (Aubriot et al. 2000), a plot of
the uptake rate versus the logarithm of the external P concentration (Thellier plot;
Thellier 1970) was made; the intercept on the log [P] axis corresponds to the logarithmic
P threshold concentration (i.e., Cmin) (Wagner et al. 1995).
69
2.4 RESULTS
In general, no overall difference in P-uptake rate (μgP/μgChl/d) estimates was
detected between the two methods, scraped versus intact biofilms (t= 0.69, p= 0.492, df=
33) (Appendix B). The scraping technique yielded a relatively lower coefficient of
variation (x = 18.1%) among P-uptake estimates by the biofilms, while the P-uptake by
intact biofilms yielded a comparably higher coefficient of variation (x = 25.4%).
Because P-uptake was measured for both intact and scraped assemblages across
spatial and temporal scales, I evaluated these influences on the resulting rates (Table 2.1).
As above, scraping was found to have no effect on P-uptakes rates, and thus no treatment
effect (intact, disturbed) was observed. There was a significant temporal component to
the variation (F2, 28= 22.38, p= 0.043), such that uptake by biofilms during the fall
experiment on 10/17/08 (x = 1.03 (±0.57 SD) μgP/μgChl/d) was greater compared with
uptake by biofilms sampled from both the first (x = 0.75 (±0.77 SD) μgP/μgChl/d) and
second (x = 0.68 (±0.58 SD) μgP/μgChl/d) spring experiments on 6/3/09 and 6/11/09,
respectively. Physical disturbance (scraping) did not appear to negatively affect cellular
function as expressed through function of the phototsystems; ANOVA results showed no
significant differences in CFC values among disturbance intensities (F2, 6= 2.73, p=
0.144). The various physical disturbance techniques segregated organisms within the
biofilm (possessing specific internal P storage capacities). Biologically bound P (poly-P)
varied significantly among treatments (F2, 6= 11.34, p= 0.009); specifically the low
treatment averaged 0.133 (±0.028 SD), the medium averaged 0.051 (±0.003 SD), and the
high averaged 0.080 (±0.024 SD) mgP/mgChl. The higher poly-P content in the low
disturbance treatment was likely linked to the higher presence of filamentous taxa in the
70
upper strata of the biofilm (Figure 2.1) (Appendix C, D, E). Killed controls showed that
abiotic sorption yielded significantly lower uptake compared to biotic uptake (F2, 9=
17.51, p= 0.001); specifically P-uptake rates averaged 0.068 (±0.017 SD), 0.029 (±0.01
SD), and 0.025 (±0.00 SD) μgP/μgChl/d by control, formaldehyde, and glutaraldehyde
treatments, respectively. Interestingly, average abiotic sorption (glutaraldehyde and
formaldehyde) across both sites sampled accounted for approximately 40% of total P-
uptake relative to control treatments.
The nonlinear (weighted) least-squares analysis of the M-M model for the scraped
assemblages yielded a Vmax of 2.25 (±0.77 SE) µgP/µgChl/d and Km of 231.36 (±140.92
SE) μgP/L (Appendix F); intact assemblages yielded a Vmax of 2.33 (±0.63 SE)
µgP/µgChl/d and Km of 293.87 (±173.95 SE) µgP/L (Appendix G). A one-way ANOVA
of P-uptake rate (μgP/μgChl/day) versus treatment showed that there was no difference
(F= 0.72, p= 0.408) in uptake rates between scraped and intact assemblages over the 60
minute period. I then tested re-estimated uptake constants (k) by splitting samples
between short time periods (i.e., 5 - 12 minutes) and long time periods (i.e., 30 - 60
minutes) to estimate the effect of experiment duration on the uptake constant and M-M
parameters. For short time periods the scraped assemblages yielded a Vmax of 16.64
(±1.40 SE) µgP/µgChl/d and Km of 1457.10 (±139.35 SE) µgP/L (Appendix H); intact
assemblages yielded a Vmax of 2.31 (±4.70 SE) µgP/µgChl/d and Km of 23.44 (±85.59
SE) µgP/L (Appendix I). For long time periods the scraped assemblages yielded a Vmax
of 8.42 (±16.61 SE) µgP/µgChl/d and Km of 1236.55 (±2831.18 SE) µgP/L (Appendix J);
intact assemblages yielded a Vmax of 1.11 (±1.25 SE) µgP/µgChl/d and Km of 653.45
(±960.90 SE) µgP/L (Appendix K). Additionally, I tested the soil nutrient uptake
71
formulation of Barber (1995) for benthic biofilms by estimating Cmin, the minimum [P]
required for uptake to transpire. The Thellier plots used to estimate Cmin yielded a [P] of
6.80 and 9.95 µg/L for the scraped and intact biofilm treatments, respectively (Figure
2.2).
2.5 DISCUSSION
Measurements made between benthic biofilms subject to physical disturbance
(abrasion/scraping) and undisturbed intact biofilms, showed that single time point
estimates of P-uptake did not vary significantly between the treatments. These findings
indicate that scraping does not negatively affect the ability of biofilms to assimilate P
during brief (≤ 30 minutes) single time-course experiments. In disturbed treatments,
uptake was estimated through direct incorporation of tracer into microbes (on filter),
while in undisturbed treatments, uptake was inferred through tracer loss from overlying
water; either way, both methods appear to be practical and comparable during this
experimental period. A meta-analysis on P-uptake in aquatic microbes supports this
finding, as experiments that derive uptake from water were comparable with those that
derive uptake from direct analyses (Price and Carrick 2011). Further, my data here lend
credence to the numerous experiments that investigate benthic microbial physiologic
responses post-disturbance (e.g., Bothwell 1985, Reuter et al. 1986, Scinto and Reddy
2003, Chapter 3). The M-M parameter estimates did show relative differences when split
by short and long time samplings, with the shorter yielding higher Vmax; this finding
corresponds to earlier research and likely indicates initial transport vs. assimilation
kinetics (Flynn 1998, Price and Carrick 2011). In fact, the rapid uptake revealed during
72
my short time period (i.e., 5 - 12 minutes) is directly in line with Taft et al. (1975), who
found that initial rapid phosphate uptake (attributed to membrane-transport) declined
after 15 - 60 minutes in phytoplankton in the Chesapeake Bay estuary. My Km estimates
were similar to those reported for freshwater periphyton (508 µgP/L) (Scinto and Reddy
2003). In my study, the estimated Km was slightly lower for scraped assemblages and
suggests that microbes liberated from the diffusive boundary layer (see Larned et al.
2004) via physical disturbance, expressed higher affinity (Furihata et al. 1992) for
inorganic P compared to those microbes within an intact biofilm; however, low
replication of these individual experiments precluded more definitive statistical analysis.
Similarly, no level of imposed disturbance caused physiological changes in the
microbes removed from the biofilms (as estimated by CFC). This indicates that, during
brief incubation periods, there are no obvious deleterious effects arresting or impairing
the rate of electron flow between photosystem II and photosystem I and, by extension, the
ability of the scraped biofilms to assimilate P (e.g., Rueter and Ades 1987). Viable cells
sloughing from periphyton (benthic biofilms) have been observed before through
microscopic examination (Naiman and Sibert 1978), and as stream benthic microbial
biofilms are continually subject to the abrasive effects of suspended solids and bed
sediments and consequently periodic sloughing losses (Biggs and Close 1989), it seems
reasonable that they have adapted mechanisms to remain viable during such processes.
The quantity of biologically stored P (poly-P) varied among layers (growth form)
within the biofilm with the highest concentration found in top (filamentous forms; 64%)
and lower for both middle (stalked forms; 16%) and bottom (prostrate forms; 20%).
Cells can store a large amount of phosphorus in poly-P granules (Jacobson and Halman
73
1982) and immediate accumulation in microbes has been observed during P-surplus
(Casadevall et al. 1985, Zeng and Wang 2009). Higher poly-P concentrations in the low
disturbance treatment suggest that surplus P is available to filamentous forms in the
biofilm. This was expected, as filamentous microbes in a biofilm generally have greater
biomass exposed to overlying waters where nutrients may be more bio-available and
further supports work showing that access to phosphate supplies from the water depends
on the position of microbial cells in the biofilm (see Burkholder et al. 1990).
While I found that biofilms subject to both formaldehyde and glutaraldehyde
treatments showed significantly reduced uptake rates compared to controls, the data
suggest that abiotic uptake of P could be important in streams like the one studied here.
These estimates for the abiotic fraction of uptake did show considerable P sorption (40%
of total) and were comparable to previous literature. For example, Aldridge et al. (2010)
found that abiotic interception accounted for more than 70% of the total P-uptake by
epilithic communities across a gradient of unmodified and modified streams. My abiotic
sorption estimate may be artificially elevated due to the absence of flow in my
experiments. That is, a key mechanism of abiotic P retention includes adsorption by
sediments (Reddy et al. 1999) and therefore in a flowing system, that specific mechanism
would be diminished due to added transport and fewer opportunities for nutrient-
sediment encounters (Triska et al. 1989). Abiotic uptake is also dependent on
biogeochemistry and sediment composition of the stream (Stone and English 1993).
Here I measured a single system, but an analysis of abiotic uptake across spatial and
temporal gradients should be further considered to refine models (Dodds et al. 2002).
Despite this need, Price and Carrick (2011) found that only 32% of uptake experiments
74
factor an abiotic fraction of uptake into analysis; I would therefore suggest this be
performed on a more regular basis to avoid overestimation of biotic uptake rates.
Concentrations of substrate below which cells cannot acquire nutrients reflect
important and potentially ecologically meaningful threshold concentrations (Istvánovics
and Herodek 1995). Due to energetic constraints, biotic uptake is then only possible
when external [P] surpasses this threshold, or Cmin, which, in P-limited communities, has
been found to be in the nanomolar range (Falkner et al. 1989). My values averaged 219
nM for the scraped biofilm treatment and 321 nM for the intact treatment and thus appear
to be within ranges previously reported. Previous work has suggested that the threshold
concentration in P-deficient lakes is close to the ambient P concentration (lstvánovics and
Herodek 1995). Biofilms for this experiment were retrieved from an upstream site in
Spring Creek, Pennsylvania where orthophosphate concentrations are <10 µg/L and
biofilm nutrient ratios indicate P limitation (Godwin and Carrick 2008); these conditions
therefore provide further creditability to my results. Nevertheless, rapid nutrient cycling
within biofilms (Riber and Wetzel 1987) may essentially act to insulate microbes and
reduce dependency on bulk surface water nutrients; in such circumstances, Cmin may be
an extraneous parameter. Lastly, given that the magnitude of the Cmin estimates are so
low (nM), including it in the M-M model would not likely alter the shape of the simple
rectangular hyperbolic (Isvánovics et al. 1993) in this set of experiments. Thus my
results seem to indicate that while Cmin parameter estimates can be practical, their
determination and usage in M-M models may not fundamentally alter benthic biofilm
kinetic rate estimates.
75
2.6 ACKNOWLEDGMENTS
I thank B. Eckert and J. Regan for valuable feedback and helpful comments on
this manuscript. Funding for this research was provided to H.J.C. by the Pennsylvania
Department of Environmental Protection (Grant No. 4100034506).
2.7 LITERATURE CITED
Akhtar, M.S., Oki, Y., Adachi, T., and Khan, Md.H.R. 2007. From fate to faith and soil
to cell: Estimation of phosphate (Pi) uptake rate, Pi-uptake kinetics and relative
growth rate are important parameters to scavenge Pi by Brassica cultivars under
P-starved environment. Journal of the Faculty of Environmental Science and
Technology, Okayama University 12:99-106.
Aldridge, K.T., Brookes, J.D., and Ganf, G.G. 2009. Changes in abiotic and biotic
phosphorus uptake across a gradient of stream condition. River Research and
Applications 26:636-649.
Aubriot, L., Wagner, F., and Falkner, G. 2000. The phosphate uptake behaviour of
phytoplankton communities in eutrophic lakes reflects alterations in the phosphate
supply. European Journal of Phycology 35:255-262.
Belsley, D., Kuh, E., and Welsch, R. 1980. Regressions Diagnostics: Identifying
Influential Data and Sources of Collinearity. New York: John Wiley and Sons.
Biggs, B.J.F. 1988. Artificial substrate exposure times for periphyton biomass estimates
in rivers. New Zealand Journal of Marine and Freshwater Research 22:507-515.
76
Biggs, B.J.F. 1996. Patterns in benthic algae of streams. In: R.J. Stevenson, M.L.
Bothwell, and R.L. Lowe (Ed.), Algal ecology: Freshwater benthic ecosystems.
Academic Press, San Diego, California, p. 31-56.
Biggs, B.J.F. and Close, M.E. 1989. Periphyton biomass dynamics in gravel bed rivers:
The relative effects of flows and nutrients. Freshwater Biology 22:209-231.
Bothwell, M.L. 1985. Phosphorus limitation of lotic periphyton growth rates: An intersite
comparison using continuous-flow troughs (Thompson River system, British
Columbia). Limnology and Oceanography 30:527-542.
Burkholder, J.M., Wetzel, R.G., and Klomparens, K.L. 1990. Direct comparison of
phosphate uptake by adnate and loosely attached microalgae within an intact
biofilm matrix. Applied and Environmental Microbiology 56:2882-2890.
Carrick, H.J. and Steinman, A.D. 2001. Variation in periphyton biomass and species
composition in Lake Okeechobee, Florida (USA): Distribution of algal guilds
along environmental gradients. Archiv Für Hydrobiologie 152:411-438.
Carrick, H.J., Aldridge, F.J., and Schelske, C.L. 1993. Wind influences phytoplankton
biomass and composition in a shallow, productive lake. Limnology and
Oceanography 38:1179-1192.
Casadevall, E., Dif, D., Largeau, C., Gudin, C., Chamount, D., and Desanti, O. 1985.
Studies on batch and continuous culture of Botryococcus braunii: Hydrocarbon
production in relation to physiological state, cell ultrastructure and phosphate
nutrition. Biotechnology and Bioengineering 27:286-295.
77
Collos, Y. 1983. Transient situations in nitrate assimilation by marine diatoms. 4. Non-
linear phenomena and the estimation of the maximum uptake rate. Journal of
Plankton Research 5:677-691.
Covich, A.P., Austen, M.C., Barlocher, F., Chauvet, E., Cardinale, B.J., Biles, C.L.,
Inchausti, P., Dangles, O., Solan, M., Gessner, M.O., Statzner, B., and Moss, B.
2004. The role of biodiversity in the functioning of freshwater and marine benthic
ecosystems. BioScience 54:767-775.
Dodds, W.K., López, A.J., Bowden, W.B., Gregory, S., Grimm, N.B., Hamilton, S.K.,
Hershey, A.E., Martí, E., McDowell, W.H., Meyer, J.L., Morrall, D., Mulholland,
P.J., Peterson, B.J., Tank, J.T., Valett, H.M., Webster, J.R., and Wollheim, W.
2002. N uptake as a function of concentration in streams. Journal of the North
American Benthological Society 21:206-220.
Eixler, S., Selig, U., and Karsten, U. 2005. Extraction and detection methods for
polyphosphate storage in autotrophic planktonic organisms. Hydrobiologia
533:135-143.
Falkner, G., Falkner, R., and Schwab, A.J. 1989. Bioenergetic characterization of
transient state phosphate uptake by the cyanobacterium Anacystis nidulans.
Theoretical and experimental basis for a sensory mechanism adapting to varying
environmental phosphate levels. Archives of Microbiology 152:353-361.
Fitzgerald, G.P. and Nelson, T.C. 1966. Extraction and enzymatic analyses for limiting or
surplus phosphorus in algae. Journal of Phycology 2:32-37.
Fitzwater, S.E., Knauer, G.A., and Martin, J.H. 1982. Metal contamination and its effect
on primary production measurements. Limnology and Oceanography 27:544-551.
78
Flynn, K.J. 1998. Estimation of kinetic parameters for the transport of nitrate and
ammonium into marine phytoplankton. Marine Ecology Progress Series 169:13-
28.
Furihata, T., Suzuki, M., and Sakurai, H. 1992. Kinetic characterization of two phosphate
uptake systems with different affinities in suspension-cultured Catharanthus
roseus protoplasts. Plant and Cell Physiology 33:1151-1157.
Godwin, C.M. and Carrick, H.J. 2008. Spatio-temporal variation of periphyton biomass
and accumulation in a temperate spring-fed stream. Aquatic Ecology 42:583-595.
Graham, M.D. and Vinebrooke, R.D. 1998. Trade-offs between competiveness and
herbivore resistance in periphyton of acidified lakes. Canadian Journal of
Fisheries and Aquatic Sciences 55:806-814.
Haggard, B.E., Stanley, E.H., and Hyler, R. 1999. Sediment-phosphorus relationships in
three northcentral Oklahoma streams. Transactions of the American Society of
Agricultural Engineers 42:1709-1714.
Hall, R.O., Bernhardt, E.S., and Likens, G.E. 2002. Relating nutrient uptake with
transient storage in forested mountain streams. Limnology and Oceanography
47:255-265.
Hill, W.R., Bednarek, A.T., and Larsen, I.L. 2000. Cadmium sorption and toxicity in
autotrophic biofilms. Canadian Journal of Fisheries and Aquatic Sciences 57:530-
537.
Hwang, S-J., Havens, K.E., and Steinman, A.D. 1998. Phosphorus kinetics of planktonic
and benthic assemblages in a shallow subtropical lake. Freshwater Biology
40:729-745.
79
Istvánovics, V. and Herodek, S. 1995. Estimation of net uptake and leakage rates of
orthophosphate from 32
P-uptake kinetics by a linear force-flow model. Limnology
and Oceanography 40:17-32.
Istvánovics, V., Pettersson, K., Rodrigo, M.A., Pierson, D., and Padisák, J. 1993.
Gloeotrichia echinulata, a colonial cyanobacterium with unique phosphorus
uptake and life strategy. Journal of Plankton Research 15:531-552.
Jacobson, L. and Halman, M. 1982. Polyphosphate metabolism in the blue-green alga
Microcystis aeruginosa. Journal of Plankton Research 4:481-488.
Lamberti, G.A. and Resh, V.H. 1985. Comparability of introduced tiles and natural
substrates for sampling lotic bacterial, algae and macroinvertebrates. Freshwater
Biology 15:21-30.
Larned, S.T., Nikora, V.I., and Biggs, B.J.F. 2004. Mass-transfer-limited nitrogen and
phosphorus uptake by stream periphyton: A conceptual model and experimental
evidence. Limnology and Oceanography 49:1992-2000.
Marino, G., Aqil, M., and Shipley, B. 2010. The leaf economics spectrum and the
prediction of photosynthetic light–response curves. Functional Ecology 24:263-
272.
McKone, M.J. and Lively, C.M. 1993. Statistical analysis of experiments conducted at
multiple sites. Oikos 67:184-186.
Miranda, M.R., Guimarães, J.R.D., and Coelho-Souza, A.S. 2007. [3H]Leucine
incorporation method as a tool to measure secondary production by periphytic
bacteria associated to the roots of floating aquatic macrophyte. Journal of
Microbiological Methods 71:23-31.
80
Mulholland, P.J., Newbold, J.D., Elwood, J.W., and Hom, C.L. 1983. The effect of
grazing intensity on phosphorus spiralling in autotrophic streams. Oecologia
58:358-366.
Naiman, R.J. and Sibert, J.R. 1978. Transport of nutrients and carbon from the Nanaimo
River to its estuary. Limnology and Oceanography 23:1183-1193.
Nielsen, N.E. and Schjørring, J.K. 1983. Efficiency and kinetics of phosphorus uptake
from soil by various barley genotypes. Plant and Soil 72:225-230.
Odum, E.P., Kuenzler, E.J., and Blunt, Sr. M.X. 1958. Uptake of P32
and primary
productivity in marine benthic algae. Limnology and Oceanography 3:340-345.
Passy, S.I. 2007. Diatom ecological guilds display distinct and predictable behavior along
nutrient and disturbance gradients in running waters. Aquatic Botany 86:171-178.
Peterson, B.J., Wollheim, W.H., Mulholland, P.J., Webster, J.R., Meyer, J.L., Tank, J.L.,
Martí, E., Bowden, W.B., Valett, H.M., Hershey, A.E., McDowell, W.H., Dodds,
W.K., Hamilton, S.K., Gregory, S., and Morrall, D.D. 2001. Control of nitrogen
export from watersheds by headwater streams. Science 292:86-90.
Peterson, C.G. and Stevenson, R.J. 1990. Post-spate development of epilithic algal
communities in different current environments. Canadian Journal of Botany
68:2092-2102.
Prézelin, B.B. 1981. Light reactions in photosynthesis. In: T. Platt (Ed.), Physiological
bases of phytoplankton ecology. Canadian Bulletin of Fisheries and Aquatic
Sciences 210:1-43.
81
Price, K.J. and Carrick H.J. 2011. Meta-analytical approach to explain variation in
microbial phosphorus uptake rates in aquatic ecosystems. Aquatic Microbial
Ecology 65:89-102.
R Development Core Team. 2006. R: A language and environment for statistical
computing. R Foundation for Statistical Computing, Vienna, Austria.
Reddy, K.R., Kadlec, R.H., Flaig, E., and Gale, P.M. 1999. Phosphorus retention in
streams and wetlands: A review. Critical Reviews in Environmental Science and
Technology 29:83-146.
Rehder, D.S. and Borges, C.R. 2010. Cysteine sulfenic acid as an intermediate in
disulfide bond formation and nonenzymatic protein folding. Biochemistry
49:7748-7755.
Rejmánková, E. and Komárková, J. 2000. A function of cyanobacterial mats in
phosphorus-limited tropical wetlands. Hydrobiologia 431:135-153.
Reuter, J.E., Loeb, S.L., and Goldman, C.R. 1986. Inorganic nitrogen uptake by epilithic
periphyton in an N-deficient lake. Limnology and Oceanography 31:149-160.
Riber, H.H. and Wetzel, R.G. 1987. Boundary-layer and internal diffusion effects on
phosphorus fluxes in lake periphyton. Limnology and Oceanography 32:1181-
1194.
Rueter, J.G. and Ades, D.R. 1987. The role of iron nutrition in photosynthesis and
nitrogen assimilation in Scenedesmus quadricauda (Chlorophyceae). Journal of
Phycology 23:452-457.
82
Sabater, S., Guasch, H., Romaní, A., and Muñoz, I. 2002. The effect of biological factors
on the efficiency of river biofilms in improving water quality. Hydrobiologia
469:149-156.
Scinto, L.J. and Reddy, K.R. 2003. Biotic and abiotic uptake of phosphorus by
periphyton in a subtropical freshwater wetland. Aquatic Botany 77:203-222.
Singh, S.P., Ma, L.Q., and Hendry, M.J. 2006. Characterization of aqueous lead removal
by phosphatic clay: Equilibrium and kinetic studies. Journal of Hazardous
Materials 136:654-662.
Steinman, A.D. and Mulholland, P.J. 2006. Phosphorus limitation, uptake, and turnover
in benthic stream algae. In: F.R. Hauer and G.A. Lamberti (Ed.), Methods in
stream ecology, 2nd
Ed. Academic Press, p. 187-212.
Stone, M. and English, M.C. 1993. Geochemical composition, phosphorus speciation and
mass transport characteristics of fine-grained sediment in two Lake Erie
tributaries. Hydrobiologia 253:17-29.
Svendsen, L.M., Kronvang, B., Kristensen, P., and Græsbøl, P. 1995. Dynamics of
phosphorus compounds in a lowland river system: Importance of retention and
non-point sources. Hydrological Processes 9:119-142.
Sweeney, B.W., Bott, T.L., Jackson, J.K., Kaplan, L.A., Newbold, J.D., Standley, L.J.,
Hession, W.C., and Horwitz, R.J. 2004. Riparian deforestation, stream narrowing,
and loss of stream ecosystem services. Proceedings of the National Academy of
Sciences of the United States of America 101:14132-14137.
Taft, J.L., Taylor, W.R., and McCarthy, J.J. 1975. Uptake and release of phosphorus by
phytoplankton in the Chesapeake Bay estuary, USA. Marine Biology 33:21-32.
83
Tank, J.L. and Webster, J.R. 1998. Interaction of substrate and nutrient availability on
wood biofilm processes in streams. Ecology 79:2168-2179.
Tarapchak, S.J. and Herche, L.R. 1986. Phosphate uptake by microorganisms in lake
water: Deviations from simple Michaelis-Menten kinetics. Canadian Journal of
Fisheries and Aquatic Sciences 43:319-328.
Thellier, M. 1970. An electrokinetic interpretation of the functioning of biological
systems and its application to the study of mineral salt absorption. Annals of
Botany 34:983-1009.
Thompson, A.J. and Sinsabaugh, R.L. 2000. Matric and particulate phosphatase and
aminopeptidase activity in limnetic biofilms. Aquatic Microbial Ecology 21:151-
159.
Thompson, J.A. 1997. Cellular fluorescence capacity as an endpoint in algal toxicity
testing. Chemosphere 35:2027-2037.
Triska, F.J., Kennedy, V.C., Avanzino, R.J., Zellweger, G.W., and Bencala, K.E. 1989.
Retention and transport of nutrients in a third-order stream: Channel processes.
Ecology 70:1877-1892.
Vincent, W.F. 1980. Mechanisms of rapid photosynthetic adaptation in natural
phytoplankton communities. II. Changes in photochemical capacity as measured
by DCMU-induced chlorophyll fluorescence. Journal of Phycology 16:568-577.
Vincent, W.F. 1981. Photosynthetic capacity measured by DCMU-induced chlorophyll
fluorescence in an oligotrophic lake. Freshwater Biology 11:61-78.
84
Vyhnalek, V., Fisar, Z., Fisarova, A., and Komarkova J. 1993. In vivo fluorescence of
chlorophyll a: Estimation of phytoplankton biomass and activity in Rimov
Reservoir (Czech Republic). Water Science and Technology 28:29-33.
Wagner, F. and Falkner, G. 2001. Phosphate limitation. In: L.C. Rai and J.P. Gaur (Ed.),
Algal adaptation to environmental stresses-physiological, biochemical and
molecular mechanisms. Springer-Verlag, New York, p. 65-110.
Wagner, F., Falkner, R., and Falkner, G. 1995. Information about previous phosphate
fluctuations is stored via an adaptive response of the high-affinity uptake system
of the cyanobacterium Anacystis nidulans. Planta 197:147-155.
Wellnitz, T.A. and Ward, J.V. 2000. Herbivory and irradiance shape periphytic
architecture in a Swiss alpine stream. Limnology and Oceanography 45:64-75.
Wolfstein, K., de Brouwer, J.F.C., and Stal, L.J. 2002. Biochemical partitioning of
photosynthetically fixed carbon by benthic diatoms during short-term incubations
at different irradiances. Marine Ecology Progress Series 245:21-31.
Zeng, J. and Wang, W-X. 2009. The importance of cellular phosphorus in controlling the
uptake and toxicity of cadmium and zinc in Microcystis aeruginosa, a freshwater
cyanobacterium. Environmental Toxicology and Chemistry 28:1618-1626.
85
Table 2.1. Three-way mixed-model ANOVA testing the main fixed effects of treatment (n= 2) and
experiment (n= 3) and random effect of site (n= 2) on P-uptake rates (μgP/μgChl/d) for intact vs.
scraped stream biofilms.
Source DF Sequential
sums of squares
Adjusted
mean square
F p
Site 1 10.7368 10.6698 42.83 0.683
Treatment 1 0.2018 0.2493 0.36 0.657
Experiment 2 0.9915 0.4958 22.38 0.043
Site * Treatment 1 0.7827 0.6970 1.48 0.347
Site * Experiment 2 0.0443 0.0222 0.05 0.955
Treatment * Experiment 2 1.4666 0.7333 1.55 0.393
Site * Treatment * Experiment 2 0.9476 0.4738 14.49 0.000
Error 28 0.9158 0.0327
Total 39 16.0873
86
Figure 2.1. Stacked column chart of benthic poly-P (mgP/mgChl) vs. disruption treatment according
to growth form. Estimates were made by multiplying mean percent growth form from each
physiognomic classification by the mean poly-P concentration for each disturbance treatment.
87
Figure 2.2. Thellier plots used to estimate Cmin for scraped (left) and intact (right) biofilm
assemblages. The fitted linear regression intercepts the log [P] axis at the logarithmic threshold
concentration (Aubriot et al. 2000).
88
CHAPTER 3
EFFECTS OF NUTRIENT LOADING ON PHOSPHORUS UPTAKE BY BIOFILMS SITUATED
ALONG A STREAM PRODUCTIVITY GRADIENT
89
3.1 ABSTRACT
Biofilms play an important role in the assimilation of phosphorus (P) in streams;
however, they are often treated as a “black box” and relatively few direct measurements
have linked nutrient loadings with assimilation. Herein, I measured P-uptake rates for
resident, benthic biofilms along a stream productivity gradient (n= 8 streams). Realistic
point source nutrient loadings were applied using an in situ enrichment system (ISES):
vials spiked with increasing concentrations of P with and without nitrogen (N) (n= 70 for
each experiment). Vials were capped with porous disks that were removed with attached
biofilms after three weeks of in-stream incubation. A series of short-term radiotracer
(H333
PO4) experiments were then conducted to measure P-uptake by resident biofilms
that grew on the disks collected from each stream experiment. P-uptake versus P-loading
(log10 transformed) was explained by simple linear regression for all streams. Similarly,
uptake efficiency declined with increased P-loading, and the overall slope of the
regressions for these relationships also declined as a function of stream productivity.
Specifically, biofilms in low-productivity streams showed significantly higher uptake
rates for moderate P-loads compared to high-productivity streams (e.g., 1.29 vs. 0.84
(log10(nmolP/μgChl/day)), respectively) (F1,30= 6.21, p= 0.018), indicating that these
communities were more physiologically poised to respond to new P additions. Biofilms
in productive streams appeared to be P-saturated (presumably from P legacy effects), and
thereby had lower demand for experimental P additions. P-uptake rates were greater for
P+N versus sole P-loads; these results indicated that N had a synergistic effect on biofilm
P-uptake ability, and may have been secondarily limiting in streams of higher
productivity (P legacy effect). My results identified the occurrence of P saturation in
90
streams and quantified its effect on the uptake of new P additions, as well as the reality of
N as a synergistic nutrient for these resident biofilms.
3.2 INTRODUCTION
Phosphorus (P) and nitrogen (N) are key nutrients that can limit terrestrial
(Vitousek and Howarth 1991) and aquatic primary production (Francouer 2001). P is an
essential nutrient for all living organisms; it is involved in the formulation of the nucleic
acids deoxyribonucleic acid (DNA) and ribonucleic acid (RNA) and the “energy
currency” adenosine triphosphate (ATP) (Versaw and Harrison 2002, Dodson 2005,
Oelkers and Valsami-Jones 2008). Specifically, in primary producers, ADP (adenosine
diphosphate) binds to inorganic P to make ATP which then drives energy-dependent
processes. In aquatic ecosystems, increased anthropogenic inputs of P can be closely tied
to eutrophication (Correll 1998, Mainstone and Parr 2002); however, recent focus on
managing aquatic ecosystems for P alone, without specific concern for N (see Schindler
et al. 2008) has shown great limitations (see Paerl et al. 2010) which underscore the need
for combined P and N management to control freshwater eutrophication (Paerl 2009,
Scott and McCarthy 2010). Strong synergistic effects of simultaneous P and N
enrichment have been identified in other studies (e.g., Elser et al. 2007), such that both
nutrients need to be considered together as important resources that influence the
biological integrity and sustainability of aquatic ecosystems (Karr 1991, Franklin et al.
2005).
Nitrogen is an important nutrient that can regulate primary production (LeBauer
and Treseder 2008), as it is required for biosynthesis of amino acids and proteins as well
91
as photosynthetic pigments (e.g., chlorophyll-a, phycobilins) (Grula 2005). The proposed
mechanisms for the observed combined P and N synergistic effects on primary
production is that cellular stoichiometry of these nutrients are in close balance and that
single enrichments induce limitation by the alternative nutrient (Davidson and Howarth
2007, Elser et al. 2007). For example, Lapointe (1989) found that Gracilaria tikvahiae
(Division Rhodophyta, Order Gigartinales) was limited primarily by P and secondarily by
N when P supply increased. While much research has addressed the synergisms of P and
N additions, the work has focused largely on their effects on net primary productivity
(e.g., Elser et al. 2007), with little regard for physiological processes like uptake, the
initial stage in assimilation of nutrients into organic matter (Beardall et al. 2001).
Stream benthic biofilms have a high affinity for inorganic nutrients (Hoffmann
1998) and have been shown to act as dynamic nutrient sinks (Dodds 2003), much like soil
microbial communities in terrestrial ecosystems (Olander and Vitousek 2004). Biotic
uptake/assimilation by benthic algae transforms reactive inorganic nutrients into
particulate form, thereby rending them unavailable; this transformation thus regulates
nutrient bioavailability to downstream ecosystems (Svendsen et al. 1995). As such,
biofilms are critical in the removal of dissolved nutrients from stream water and key
elements in stream “self-purification” (Gantzer et al. 1988, Sabater et al. 2002, Covich et
al. 2004). Thus, while streams are significant landscape features owing to their role in
nutrient transformations (Peterson et al. 2001), the efficiency of nutrient uptake (relative
to flux) by resident biofilms is paramount in this regard; as such it is important to
understand environmental factors that might act to suppress such efficiency. For
instance, Earl et al. (2006) found that both high ambient [N] and experimental N
92
amendments to streams contributed to decreased uptake efficiency. Excessive N inputs
have generated N saturation responses in forest and aquatic ecosystems (Aber et al. 1989,
1998, Earl et al. 2006, De Schrijver et al. 2008); such that, N availability/supply in excess
of biotic demand limits the retainment of N within the biota and increases the potential
for losses to surrounding environs (Aber et al. 1989). Recent research has focused on
such saturation responses to elevated N loading in lotic ecosystems (e.g., Bernot and
Dodds 2005, Earl et al. 2006, O'Brien and Dodds 2010, Martin et al. 2011), however,
considering the importance of P as a bio-limiting nutrient and the present nutrient debates
(see Paerl 2009), it is critical that both N and P be examined with respect to how single
and dual nutrient loadings can affect biofilm uptake efficiencies.
The research I report on here was designed to evaluate variation in P-uptake as a
function of experimental P and N enrichment among streams of varying productivity; the
experiments were performed in situ to achieve a high degree of environmental realism.
My specific objectives of this study were: (1) derive regression models to describe the
relationship between P-uptake and P-loading, (2) measure the effect (synergistic, neutral,
antagonistic) of concurrent N-loadings with P on biofilm assimilative abilities, and (3)
examine biofilm assimilative capacities across a gradient of stream productivity to
determine the effect of nutrient legacies. I hypothesize that P-uptake by biofilms will
decline as a function of P-loading, and the rate of decline will become less pronounced in
streams that support higher productivity. Further, enrichment with N, a typical secondary
limiting nutrient in aquatic enrichment experiments (Stockner and Shortreed 1978, Marks
and Lowe 1993), will have a synergistic effect with P and augment biofilm response to P-
loading.
93
3.3 METHODS
Study sites
I carried out field experiments in eight streams that were a representative subset
from a previous study that evaluated relationships between benthic chlorophyll-a and
nutrient concentrations in 47 streams sampled in Pennsylvania (Carrick et al. 2009)
(Appendix L). Specifically, experiments were conducted in four streams located in each
of the two major geologic provinces that divide Pennsylvania along a southwest to
northeast axis (Appalachian Plateau, Ridge-Valley Piedmont, respectively (PASDA
2005)) (Appendix M). These eight streams were selected in order to capture the broad
range of environmental and biogeochemical conditions reflective of environments in
Pennsylvania and the larger, mid-Atlantic region (Figure 3.1). The streams examined
here exhibited a strong productivity gradient similar to that described elsewhere (Busse et
al. 2006). In fact, my range of chlorophyll-a concentrations reflect trophic conditions
ranging from oligotrophic (<50 mg/m2) to eutrophic (>100 mg/m
2) (see Dodds et al.
1998). Flow velocity can influence the biomass and taxonomic composition of stream
biofilms (Stevenson 1990, Hart and Finelli 1999), and mediate or even override the
effects of nutrient enrichment under certain conditions (e.g., Hullar and Vestal 1989,
Ghosh and Gauer 1994). In all cases, the experiments were conducted during base flow
conditions, where stream discharge was generally < 100 ft3/s and deviated little
throughout the experimental period; therefore, hydrologic mechanisms did not
differentially affect microbial biofilms in my study streams and further, flow rates were
likely too low to affect uptake through increasing basal nutrient flux (cf. Triska et al.
1990).
94
At each stream site, water physicochemical conditions (e.g., temperature,
conductivity, oxygen concentration) were monitored at the beginning and end of the
experiment using an YSI (Yellow Springs Instrument) data Sonde (model 6600); algal
growth (as chlorophyll-a) was also estimated at the start and finish of the experiment on
duplicate collections of natural rocks. Stream flow was measured by nearby United
States Geological Survey monitoring gage stations and recorded through the National
Water Information System. Water chemistry was measured at each site on samples
collected in acid-rinsed, 1-liter amber bottles that were kept cool during transport to the
laboratory (Millie et al. 2010). The concentration of water column phosphorus was
measured using standard digestions and colorimetric reactions (Wetzel and Likens 2000).
Design of field experiments- in situ enrichment system (ISES)
The P-uptake response of benthic biofilms to P-enrichment was directly measured
under natural stream conditions through a series of experiments carried out in 2008 using
an in situ enrichment system (ISES). The ISES design used small, polystyrene vials
(outer dimensions 2.7 x 8.0 cm) that were filled with ultra-pure agar (2% agar noble), to
which five levels of P (P1 - P5) was added (0.00 control, 0.005, 0.025, 0.05, and 0.50 M
as NaH2PO4). In addition, one set of vials received no additional N sources, while a
second set of vials received N additions (0.0 and 0.50 M as NaNO3, respectively). For
each experiment, a total of 70 vials were prepared (7 replicates of 10 treatments), and
each sealed by securing a porous, porcelain crucible disk (2.6-cm diameter, Leco
Corporation, St. Joseph, Michigan) to the top of each vial; the disks acted as both
colonizing substrate and point source for P and N released from the tubes over time
(Gibeau and Miller 1989). Experiments were deployed for a period of 15-18 days in each
95
stream during two intensive field expeditions in late summer 2008. Experiments were
conducted in the Piedmont streams from 11-12 August to 27-29 August, while
experiments were conducted in the Plateau streams from 21-22 August to 3-5 September.
This incubation period was selected to allow for adequate establishment of a growing
stream biofilm to estimate active uptake of inorganic nutrients, while minimizing the
chance of sloughing and space limitation on the experimental substrata (e.g., Stevenson
1996). Experimental vials were placed in epoxy-coated test tube racks, and these were in
turn placed in a protective acrylic sled that was secured to the streambed using rebar.
Following the incubation period, each experimental unit was removed from the
streambed and the experimental vials were processed streamside. Accumulated biofilm
material on each experimental crucible (disk) was scraped with a hard-bristled brush
(e.g., Carlisle and Clements 2003) and washed into labeled 50 mL polypropylene jars
containing ambient stream water; once the experimental samples were collected, the jars
were covered and stored in coolers for transport back to the laboratory. Nutrient loading
rates were estimated from the difference between initial and final concentrations of P and
N in the experimental vials (see Carrick and Price 2011).
Analytical measurements- Biomass and areal nutrient concentrations
Algal biomass was estimated for each crucible disk by concentrating subsamples
onto 47-mm glass microfiber filters (EPM-2000, Whatman International, Maidstone,
UK). The chlorophyll-a (Chl) concentration in each filter was determined using an
organic extraction procedure (50:50 mixture of 90% Acetone to DMSO); Chl
concentrations were subsequently measured using a standard fluorometric technique
(Carrick et al. 1993). Chl was used here as a proxy of the total biofilm biomass (e.g.,
96
Montuelle et al. 2010). However, because Chl has been found to vary in the proportion
of biomass it accounts for, areal carbon (g/m2) was also estimated, as it can provide a
more inclusive means of evaluating particulate organic carbon (Malone and Chervin
1979). Despite that, Chl was found to be strongly correlated with carbon (Pearson
correlation coefficient (r)= 0.768, p< 0.001, r2= 0.590), and therefore was considered an
accurate and appropriate measure of biomass (Appendix N). Total phosphorus (TP) was
measured using wet persulfate digestion (Menzel and Corwin 1965), where liberated
soluble reactive P (as PO43-
) was measured colorimetrically using a spectrophotometer
(American Public Health Association 1992). Carbon (C) and N were measured on
subsamples concentrated onto Whatman glass-fiber filters using a high-temperature
combustion carbon analyzer.
Analytical measurements- P-uptake
P-uptake was estimated from incorporation of carrier-free H333
PO4 into biofilms
exposed to increasing P-loads from the ISES experiments. Use of an isotopic radiotracer
is a common method for measuring phosphate uptake kinetics (e.g., Fuhs et al. 1972,
Cembella et al. 1984) and provides a sensitive marker of activity while adding
insubstantial amounts of P to natural assemblages (Burmaster and Chisholm 1979, Noe et
al. 2003). Additionally, application of radioactive phosphate does not significantly alter
the properties of the uptake system (Falkner et al. 1995). By exposing a biofilm to trace
amounts of 33
P, the isotope can mix with the pool of non-radioactive phosphorus and the
incorporation of nutrient in the cells can be determined. From a diluted stock of 33
P
(PerkinElmer), as carrier-free H333
PO4 (1mL 33
P and 100mL DI H2O), 1.0 mL was
injected into 50 mL, translucent polypropylene (Nalgene) incubation jars containing an
97
ISES disk collected from the experiments (activity of 33P stock= 10 μCi mL
-1). Although
past uptake experiments incubated periphyton communities with magnetic stirring bars
(e.g., Havens et al. 1999), for this experiment the contents of the incubation jars were
hand swirled approximately every minute. I assume that differences in the physical
mixing process were not significant over the short time period of the experiments (e.g.,
Caperon and Meyer 1972, Burmaster and Chisholm 1979). Experimental methods
followed procedures outlined by Steinmann and Mulholland (2006). The kinetics of P-
uptake was assessed by measuring P incorporation into the biofilm over time (30
minutes); this time point was chosen, because previous experiments showed that uptake
is linear up to 30 minutes (K.J. Price, unpublished data). Aliquots (1.0 mL) of microbial
slurry were removed from the incubation jars and injected into a Millipore model 1225
12-place sampling manifold (Millipore Corp., Bedford, MA). Filters (GF/F) were then
removed and placed into 20 mL glass scintillation vials filled with 7 mL Ecolume
scintillation cocktail (ICN Biomedicals) and analyzed for activity using liquid
scintillation counting (LSC) on a Beckman LS3801 liquid scintillation spectrometer
(Beckman Instruments Inc., counting efficiency 95.9%). Uptake is the net transfer of a
chemical constituent across a cell membrane and is typically a fast process that can be
analyzed over tens of minutes (Burmaster and Chisholm 1979). The first-order uptake
rate constant k (min-1
) was determined as the quotient of ln [P0/(P0-X)] and time, where
P0 is the total activity in 1 mL of radio-labeled water sample and X is the radioactivity on
the filter (Hwang et al. 1998, Havens et al. 1999). The rate constant k was multiplied by
the in situ stream phosphate concentration, which then estimates the phosphate uptake
rate, with appropriate corrections for the volume of water used (Lean and Pick 1981,
98
Corning et al. 1989, Havens et al. 1999). Measured uptake rates are expressed as log10
and normalized for chlorophyll-a (nmol P/μg Chl/day). Chl-specific P-uptake
(normalized per unit biomass) is defined here as the assimilation efficiency of stream
biofilms (e.g., Platt et al. 1982, Cloern et al. 1995).
Statistical analyses- Parametric statistics
Variation in the magnitude of P-loading among P treatments was assessed using
one-way analysis of variance (ANOVA) by comparing the overall average of P-loads
(results discussed in Carrick and Price 2011). Simple, descriptive statistics were
calculated for measurements of stream water quality and benthic biofilm biomass for
each of the eight streams where experiments were executed. Quantitative relationships
between the response variable (P-uptake) and the predictor variable (P-loading) were
assessed using simple linear regression; P-uptake was regressed against mass P-loading
rates from each experiment (separate analyses for each stream). The slope and y-
intercept values were estimated from these regression analyses to evaluate changes in P-
uptake vs. nutrient load (efficiency) along said productivity gradient. Slope gradients in
linear regression models relating uptake and nutrient amendments have been used to infer
saturation in streams (cf. Earl et al. 2006). Separate regressions were also performed for
P-uptake versus P-loading with (n= 5) or without (n= 5) simultaneous N- loading.
Biofilm Chl concentration estimates for all eight streams were the average of values
measured on rocks collected at the start and end of experiments and served as a proxy for
stream productivity (trophic state). Tukey multiple means comparisons test, widely used
for pairwise comparisons among group means (Games 1971), was used to evaluate the
least significant difference within the group means for each of the five loading
99
treatments, separated by province (Plateau vs. Piedmont). A paired t-test was performed
to test mean differences between P and P+N treatments for Chl-specific, and bulk P-
uptake. To estimate nutrient legacy effects on biofilm P-uptake I regressed biofilm P-
uptake rates from ISES P1 (control) disks against ambient stream conductivity (dissolved
ion content). I used Cook's distance (Di) estimates to indicate outliers in my least squares
regression analysis; outlier removal was based on Di >4/n (Bollen and Jackman 1990).
Red Clay (Di= 0.62) was the only observation greater than my exclusion Di (0.50) and
was thus removed from the analysis. To integrate potentially interacting effects of
nutrient stoichiometry, N-loading, and biogeochemistry (province) on biofilm P-uptake
capacity, I performed linear regressions on log10 P-uptake vs. log10 N:P and log10 C:P
ratios with discrete regressions for N vs. no N-load for each province. These data were
statistically evaluated using analysis of covariance (ANCOVA). Here, to measure the
influence of points on the fitted regression, mean log10 P-uptake vs. mean log10 N:P (n=
10) and mean log10 C:P (n= 10) observations with DFFITS values greater than 2√(k/n),
where k is the number of predictors (including constant), were removed from the
analysis, following Belsley et al. (1980). Again, Red Clay was the only stream to show a
DFFITS statistic for mean log10 N:P (1.05) and mean log10 C:P (1.27) greater than the
cutoff level (1.00) and was subsequently removed from the analysis. All data were log10-
transformed (for values < 1 I added an arbitrary constant [1] to the entire data set and
then log10 transformed to avoid negative numbers) to meet the assumptions of normality
and homogeneity of variances among treatments. A Kolmogorov-Smirnov test showed
that the data were normally distributed (p> 0.05). Levene's test indicated equal variances
at all predictor values (p> 0.05). I further calculated a White test for homoscedasticity
100
(constant residual variance), which indicated that the residuals were homoscedastic for
both the C:P (χ2= 3.29, df= 2, p> 0.05) and N:P ratios (χ
2= 3.64, df= 2, p> 0.05). Data
were analyzed using the statistical program SPSS 18.0 for Windows (SPSS, Inc.,
Chicago, IL).
Statistical analyses- Multilevel (hierarchical) model
Slope (change in P-uptake per unit change in loading intensity; efficiency) and
intercept (average P-uptake at 0 loading intensity) coefficients were extracted from linear
modeling; however, the values were varied across the tested streams. Since the model
coefficients have important ecological meanings, I fit a series of multilevel mixed models
to determine important factors explaining P-uptake - loading relationships for stream
biofilms (Qian et al. 2010). Multilevel modeling was utilized here to account for factors
operating at different spatial scales which limit the functionality of linear models. For
instance, measured P-uptake rates are affected both by predictors measured at equivalent
spatial scales (i.e., controlled loading within stream) and by predictors working at
different spatial scales (i.e., varying percent agriculture across stream). A null hypothesis
testing approach was used to determine the significance of fixed and random effects in
model building. Specifically, I was interested in assessing statistical significance of
random intercept and slope effects in describing variability in the P-uptake - loading
relationship. An additive approach to mixed model fitting was followed, where a series
of increasingly complex models were fit and a likelihood ratio test (LRT) was used to
determine the usefulness (importance) of the added variable against a reduced model.
The ‘base’ model was simply a null model containing only a fixed intercept; used for
comparison purposes after addition of terms. Afterwards, a second model with a random
101
intercept (stream effect) was fit to test the hypothesis that there is significant variation
among streams in average P-uptake. In this case, a simple (null) model was compared to
a more complex (stream effect) model to determine if the added parameter should be used
in later analyses. The third model fit contained a random stream and random stream ×
loading interaction. The second model was nested within the third model and thus
allowed use of LRT to analyze the interaction effect. The fourth model added a fixed
effect of percent agriculture and finally the fifth model added a fixed effect interaction of
percent agriculture × loading. The overall intent of this additive fitting was to attain the
most parsimonious model. Lastly, I modeled the coefficients of the best model as
functions of percent agriculture. Significance of fixed effects was determined by
examining the LRT statistics and the p-value for the Chi-square statistic. Statistical
significance for parameter inclusion in these multilevel mixed models was determined at
p< 0.10 because of reduced power to detect differences caused by small sample sizes
(Hebblewhite 2006). All linear mixed-effects models were fit using the R function lmer
(linear mixed effect regression). All mixed model analyses and graphing were performed
using R (R Development Core Team 2010).
3.4 RESULTS
Physicochemical conditions
The streams in which I conducted my ISES experiment exhibited a strong
productivity gradient (Table 3.1). Average biofilm Chl concentrations measured on
natural rocks in each stream spanned more than two orders of magnitude and were
significantly different among the eight streams (F= 5.38, p< 0.001). Specifically, average
102
Chl concentrations were 268.1 (± 393.0 SD) and 64.3 (± 108.0 SD) mg/m2 for streams in
the Piedmont and Plateau provinces, respectively. Conductivity (μS/cm) and salinity
(ppt) varied 7- and 10-fold, respectively, along the stream gradient (Table 3.1).
Conductivity (μS/cm) was plotted against Chl (mg/m2) and showed a significant linear
regression (r2= 0.422, F= 10.21, p= 0.006), suggesting it as a strong proxy for
productivity. A one-way ANOVA for each province showed that the differences between
stream conductivities were significant. Tukey's post hoc test was used to compare
differences between groups according to conductivity measurements and identified a
distribution of the groups into three homogeneous subsets: low (East Hickory and
Tionesta), medium (Cowanesque, Tunkhannock, and Cooks), and high conductivity
(productivity) (Penns, Red Clay, and Spring). Generally higher TN and TP values were
characteristic of Piedmont streams (Table 3.1). TP concentrations were exceedingly high
in Red Clay Creek during experiment retrieval for unknown reasons. Mean discharges
from streams during the ISES experimental period ranged from 11.8 to 72.2 ft3/s and
maximum discharge from the streams were all near base flow conditions (< 100 ft3/s,
range 30.0 to 99.0 ft3/s). Biofilm Chl concentrations on the ISES P1 (control) disks were
not significantly different from Chl concentrations measured from paired samples of
natural rocks collected at the time of the ISES retrieval (two-sample t-test, t= 1.87, p=
0.072), suggesting that ISES disks were a reasonable analogue for natural substrata (see
Carrick and Price 2011).
Effects of increasing P-loadings on biofilm P-uptake
The P treatments spanned an ecologically relevant range of loading rates: the
control (P1) had no appreciable P release from the ISES tubes, P2 and P3 were comparable
103
to nutrient export from more pristine, forested watersheds (Binkley et al. 2004), and P4
and P5 reflected loads typical in agricultural watersheds (Peterjohn and Correll 1984) (see
Carrick and Price 2011). The absolute P-uptake rates obtained from the ISES
experimental treatments (x = 0.45 ( 0.69 SD), range= 0.001 - 3.288 µg P/µg Chl/day)
were within the range for values reported in the literature for P-uptake by natural, lotic,
benthic assemblages (range= 0.054 - 5.645 µg P/µg Chl/day, for review see Price and
Carrick 2011). A series of simple, linear regression models were used to describe the
intracellular P-uptake (and incorporation) by stream biofilms as a function of P-loading.
The relative changes among treatments yielded predictable and statistically reliable
results. P-uptake was inversely related to P-loading, and the relationships in six of the
eight streams yielded significant regression models (Table 3.2). The effect of P-loading
on uptake was apparent, such that uptake rates measured in any individual stream were 2-
to 3-fold higher under low P-loading conditions (P1 treatment) compared with high P-
loading (P5 treatment). Background (ambient) P-uptake rates were estimated from the y-
intercepts (0 µg/day P-loading from experiment), and these values declined predictably
with increased stream productivity in both provinces (Figure 3.2). Moreover, background
P-uptake was nearly an order of magnitude higher in streams located in the Appalachian
Plateau (13.6 to 50.4 nmol P/µg Chl/day) compared with the Piedmont (0.5 to 76.6 nmol
P/µg Chl/day) province. For unknown reasons, the y-intercept in Red Clay was higher
relative to the other seven streams. Streams in the Appalachian Plateau yielded
significantly steeper slopes compared to those in the Piedmont (F= 46.94, p< 0.001);
specifically, P-uptake versus P-loading regressions in the Appalachian Plateau streams
yielded 2- to 4-fold steeper slopes (range -0.204 to -0.280) compared to those in the
104
Piedmont (range -0.046 to -0.109), and again, these values (slopes) declined with
increasing stream productivity. The steeper negative slopes in the Plateau province
provide evidence of a faster decrease in P-uptake efficiency, and thus more rapid
approach towards saturation, compared to the Piedmont province.
Results from the multilevel model analysis suggested that a model containing a
fixed interaction term for percent agriculture was the best model in describing P-uptake -
loading relationship for stream biofilms (Table 3.3). This model provided a significant
improvement in fit over the reduced (random stream effect) model and provided the best
overall explanation for these data (model 3 vs. model 5; x2= 5.95, df= 1, p= 0.015).
Regressions with percent agriculture as a group level predictor incorporated into the
model showed positive and negative associations with slopes and intercepts, respectively
(Figure 3.3). Further, the percent agriculture model accounted for a greater proportion of
variability in slopes (estimate= 0.003 (±0.001 SE)) compared to intercepts (estimate= -
0.006 (±0.01 SE)).
Effects of N with P-loadings on biofilm P-uptake
Streams in the Piedmont province showed stronger regressions when N was added
concurrently with P (Table 3.4). The addition of N with P-loading explained greater
variation in the physiological response (uptake) of biofilms to increasing P-loads, as
evidenced by the increased coefficient of determination (r2), in all four streams in the
Piedmont province. Conversely, in Plateau streams the addition of N with P-loading
resulted in a moderately reduced coefficient of determination for all four streams.
Ambient P-uptake, as estimated from the y-intercept of the linear regression equation,
increased when N was added along with P in the Piedmont province. Further, when N
105
was added concurrently with P, the regression slopes significantly decreased in all
Plateau streams (F= 7.83, p= 0.031) but significantly increased in all Piedmont streams
(F= 14.51, p= 0.009). Bulk P-uptake (not normalized to Chl) and Chl were significantly
higher in P+N treatments (t= -2.82, p= 0.007); suggesting N augments biomass and
biofilm response to P. However, a paired t-test with the data split by geological province,
indicated that biofilm bulk P-uptake showed a significant response to P+N only in the
Piedmont province (t= -2.49, p= 0.020).
Effects of stream productivity and cellular stoichiometry on biofilm P-uptake
Biofilm P-uptake was organized into two groups relative to P-loadings from ISES
experiments performed on the Plateau streams. The 0, 10, 30, and 100 μg/day loading
treatments were separated into one group according to the Tukey’s test, while 30, 100 and
1,180 μg/day loading treatments formed a second group. Alternatively, P-uptake in the
Piedmont province showed no grouping patterns according to P-load. I found a
significant negative linear regression between biofilm P-uptake rates on ISES P1 (control)
disks and ambient stream conductivity (r2= 0.594, F= 7.31, p= 0.040, slope= -1.35 (±
0.50 SE)). For the stoichiometric plot of P-uptake vs. N:P, concurrent N-loadings with P
appeared to have an antagonistic effect on biofilm assimilative abilities in the Plateau
region as the regression declined when N was loaded with P (r2= 0.081, F= 1.60, p=
0.223, slope= 0.254 (± 0.201 SE)) compared to sole P-loadings (r2= 0.324, F= 8.63, p=
0.009, slope= 0.570 (± 0.194 SE)) (Figure 3.4). However, N-loadings with P appeared to
have a synergistic effect in the Piedmont region, such that the regression improved when
N was loaded with P (r2= 0.485, F= 12.25, p= 0.004, slope= 0.426 (± 0.122 SE))
compared to sole P-loadings (r2= 0.245, F= 4.22, p= 0.061, slope= 0.314 (± 0.153 SE)).
106
Similarly, for the stoichiometric plot of P-uptake vs. C:P, concurrent N-loadings with P
appeared to have an antagonistic effect on biofilm assimilative abilities in the Plateau
region as the regression declined when N was loaded with P (r2= 0.014, F= 0.252, p=
0.622, slope= 0.102 (± 0.203 SE)) compared to sole P-loadings (r2= 0.263, F= 6.42, p=
0.021, slope= 0.493 (± 0.194 SE)) (Figure 3.4). Conversely, N-loadings with P again
appeared to have a synergistic effect in the Piedmont region, such that the regression
improved when N was loaded with P (r2= 0.560, F= 16.55, p= 0.001, slope= 0.500 (±
0.123 SE)) compared to sole P-loadings (r2= 0.260, F= 4.57, p= 0.052, slope= 0.338 (±
0.158 SE)). An ANCOVA testing the equality of the slopes for N-load and N:P ratio
showed a non-significant interaction term for the Piedmont (p= 0.538) and the Plateau
(p= 0.238), indicating the slopes of the N vs. no N-load regression lines were not
significantly different from each other. Similarly, an ANCOVA for N-load and C:P ratio
showed a non-significant interaction term for the Piedmont (p= 0.383) and for the Plateau
(p= 0.147). P-uptake vs. N:P ratio y-intercepts were 0.592 and 0.781 for N vs. no N-load,
respectively for the Plateau (ANCOVA p= 0.120) and 0.060 and 0.025 for N vs. no N-
load, respectively for the Piedmont (ANCOVA p= 0.657). P-uptake vs. C:P ratio y-
intercepts were 0.534 and 0.579 for N vs. no N-load, respectively for the Plateau
(ANCOVA p= 0.694) and -0.442 and -0.483 for N vs. no N-load, respectively for the
Piedmont (ANCOVA p= 0.591).
107
3.5 DISCUSSION
Evidence for saturation of biofilm P-uptake
The incorporation of new P into stream biofilms, as measured here, varied as a
function of P-loading, such that uptake declined in a linear manner. Similarly, in higher
plants, nutrient uptake is often negatively related to supply (Pennell et al. 1990,
Clemensson-Lindell and Asp 1995) owing to inhibitory feedbacks on uptake with
increased internal root nutrient concentration (Jensén and Ktönig 1982). The relationship
I identified here may also be controlled by feedback inhibition, whereby once an internal
cellular pool is filled (and cell quota is achieved) during loading, further uptake processes
are diminished (cf. Jansson 1993). The data further suggests that assemblages growing
on the lowest P treatments (P1 and P2) were likely P-deficient, and thus exhibited higher
uptake rates to overcome this deficiency, a common pattern observed for microorganisms
(see Lean and Pick 1981). This phenomenon has been described in the literature; high
and rapid P-uptake has been observed in algal cells cultivated under nutrient deficient
conditions (Kaya and Picard 1995). The gradient of P to which ISES biofilms were
exposed spanned well beyond a typical starvation (deficient) - satiation (sufficient)
gradient. For instance, under wastewater conditions, 5.5 mg P/L has been used as a
‘starvation’ growth medium for unicellular green microalgae with non-starved (nutrient
sufficient) algae grown under 25.0 mg P/L (4.5 times starvation [P]) (Kaya and Picard
1995). My lowest P treatments (P1, P2) provided 10 to 1000 times lower P-loadings
compared to my highest P treatments (P4, P5).
Organisms experiencing P deficiency generally exhibit rapid uptake of P once a
new source is supplied (Cembella et al. 1984, Raghothama 1999). Graziano et al. (1996)
108
found a 10-fold increase in maximum P-uptake rate when uptake was compared between
P-deficient versus P-sufficient cultures of Dunaliella tertiolecta (Division Chlorophyta,
Order Chlorococcales). Further, maximum uptake rates (Vmax) for P-deplete samples
exceeded Vmax of P-enriched samples of the green macroalga Ulva lactuca (Division
Chlorophyta, Order Ulvales) and an estuarine red algal epiphyte Catenella nipae
(Division Rhodophyta, Order Gigartinales) (Runcie et al. 2004). Specifically, Runcie et
al. (2004) found that P-deficient (starved) Catenella nipae exhibited 3-fold higher P-
uptakes compared to those that were P-enriched, a magnitude difference that is in-line
with my findings. In general, periods of P deficiency likely enhance the P-uptake
transport capacity (Jansson 1993). For instance, P transport in the bacterium Escherichia
coli (Division Proteobacteria, Order Enterobacteriales) is regulated by two separate
transport systems, one of which is induced by P deficiency (Rosenberg 1987). In fact,
uptake in P-starved algae can be 10 to 100 times higher than maximal growth rate
(Harrison et al. 1989). This disparity between uptake and growth can partly be attributed
to the ability of some microorganisms to store intracellular pools of P as polyphosphates
(e.g., Harold 1966). My results here suggest that biofilms in the ISES experiment are
non-homeostatic (i.e., plastic), allowing for luxury uptake (storage of P within biomass in
the form of polyphosphate bodies) (Powell et al. 2008, Webster et al. 2009). P-deficient
algae are capable of mobilizing internal stores of P (poly-P) in order to continue growth
and metabolic function (Rhee 1973); these cells can then capitalize on pulsed nutrient
events. This appears to be the case in this set of experiments, as poly-P storage
(discussed in Carrick and Price 2011) increased with P-load and was negatively
109
correlated with Chl-specific P-uptake rates (Pearson correlation coefficient (r)= -0.255,
p= 0.023, n= 80).
Phosphorus has been implicated as the nutrient most often controlling (limiting)
primary production in freshwater systems (Schindler 1977) and therefore, the availability
of P in streams is a concern for downstream transport. The high P-uptake efficiency
(uptake/flux) I measured in my low P-loading treatments indicated that biofilm uptake
contributes significantly to the retainment (and thus net transformation into unavailable
soluble species) of dissolved P in these ecosystems. In a general sense, benthic biofilms
may play a similar central role in a variety of other aquatic systems (e.g., Reddy et al.
1999). However, the retentive capacity of streams has specific limits as biofilm P-uptake
was diminished with increasing P-loading; in a sense, P-uptake appeared to become
saturated (Martin et al. 2011). If the biofilms are unable to assimilate new surges of P,
for instance, during storm events (Powers et al. 2009) from point (e.g., domestic sewage)
and non-point (e.g., agricultural fertilizer) sources (Jaworski 1981), this may enhance
transport of P and negatively affect water quality in downstream reaches. These results
are consistent with the N saturation hypothesis proposed by Aber et al. (1998). The
hypothesis suggests that nutrient (N) inputs exceeding biotic (forest) uptake will
accumulate and eventually be in oversupply. In this situation, where nutrient supply is in
excess of demand, losses will increase (Friedland and Miller 1999). In the same manner,
I found that biofilms in low-productivity streams showed significantly higher uptake rates
for moderate P-loads compared with those in high-productivity streams. Chronic and
elevated nutrient loads, likely present in the high-productivity streams, can impose
physiological constraints on organisms (e.g., surpass storage capacity), leading to
110
saturation of biological uptake mechanisms (O'Brien and Dodds 2010); predictably, new
P pulses would be passed to downstream regions to a greater degree in the high-
productivity systems.
Biofilm P-uptake rates on ISES P1 (control) disks were negatively associated with
ambient stream conductivity, suggesting the occurrence of a nutrient legacy effect. The
negative relationship between these parameters suggests that as streams increase in
productivity (subjected to higher dissolved ion concentrations), biofilm ability to take-up
P declines. Thus nutrient legacy effects, stemming from land use/form differences across
these sampled provinces, influence microbial physiological responses and ultimately
stream ecosystem dynamics (cf. McColl 1974, Allan 2004). Stream biofilm assemblages
in the Plateau province showed two distinct groupings in uptake capacity according to P-
load, suggesting that a physiological maximum exists between 10 and 1,180 µg/day; a
Tukey post-hoc test suggested that this biological ‘saturation threshold’ can occur at low
P-loadings (i.e., between 10 and 30 µg/day). This was expected as streams in the Plateau
province are better buffered from anthropogenic nutrient inputs and thus biofilm
assemblages would experience lower ambient [P]. Pennsylvania's Allegheny
(Appalachian) Plateau region is heavily forested with white pine, beech, hemlock, and
mixed hardwoods of red maple, yellow and sweet birch, white ash, and black cherry
(Hough and Forbes 1943). Such forested ecosystems are vital in the retention and
entrapment of bio-limiting nutrients carried in surface runoff events that can cause high
productivity in aquatic systems (Bormann et al. 1968, Anbumozhi et al. 2005).
Conversely, in Pennsylvania's Ridge-Valley/Piedmont region, fragmented, mixed-
hardwood deciduous forest with large agricultural and urban clearings is the dominant
111
landscape (see Allmendinger et al. 2005). Thus there is a distinct difference in the
prevalence of agriculture between the two major provinces in Pennsylvania which
contributes to the strong gradient of phosphorus delivery to surface waters (Ribaudo and
Johansson 2006). P export rate often increases with land disturbance (agricultural/urban
developments) and proportion of impervious surfaces (Puckett 1995), which in turn can
enhance aquatic productivity (Harvey et al. 1998). The pattern of biofilm response to P-
loads within each province suggests that regional biogeochemical characteristics (e.g.,
geology, land use) are important factors in biofilm nutrient limitation and thus response
to loading events (e.g., Rüegg et al. 2011). My results here also emphasize the
importance of considering the nutrient legacy (cf. Verspoor et al. 2010) in a stream when
trying to predict uptake relative to loading and show that uptake can saturate and
efficiency can decline with even moderate (e.g., 10 and 30 µg/day) nutrient loadings
(Earl et al. 2006).
The final model from the multilevel analysis supports these ideas and suggests
that there is a strong effect of percent agriculture in determining the slope of the P-uptake
- loading relationship; that is, the efficiency of stream biofilms in assimilating a new P
source is explained by the amount of agriculture in the surrounding watershed. Steeper
negative slopes in biofilms situated in the Plateau province provide evidence of a faster
decrease in P-uptake efficiency, and thus a more rapid approach towards saturation,
compared to those in the Piedmont province. Microbial physiological adaptation to P
concentration fluxes depends on previous exposure episodes (Aubriot et al. 2011);
accordingly, those biofilms originating from stream systems adjacent to less agricultural
lands, and less subject to high nutrient loadings (David and Gentry 2000), would likely
112
saturate more quickly when exposed to a new nutrient source. Further, greater
physiological costs may be imposed on biofilms attuned to lower ambient [P] (i.e.,
located in minimal percent agriculture environs) when exposed to elevated P, owing to
added adjustment of more machinery (e.g., activation of P-uptake receptors) (e.g., Menge
et al. 2011), which may accelerate the decline in uptake efficiency. Such has been found
in higher plant research as well, where nutrient uptake efficiency decreases considerably
with increasing rates of nutrient loading (Cabrera 2003, Ristvey et al. 2007). Further,
while more variable compared to the slope estimates, the decline in ambient uptake
(intercepts) by biofilms across the agricultural gradient was expected as well. For
instance, studies have shown that [P] strongly affects P-uptake by microalgae such that P
starvation can stimulate uptake and phosphate transport rates (Singh et al. 2007). Thus
microbes located in streams subject to less agricultural surroundings would more likely
be starved for P and demonstrate higher ambient uptake as observed. These findings are
particularly significant given that the forests of the eastern United States, largely under
private ownership, are at high risk of continued fragmentation from parcelization and
Marcellus shale development (Li et al. 2010, PADCNR 2010), which would likely
enhance in-stream nutrient concentrations (Likens et al. 1970, Hobbs 1993).
Interactive effects of N and P-loadings on biofilm P-uptake
My results suggest that bulk P-uptake (uncorrected for Chl) was augmented by P-
enrichment with N; however, Red Clay was the only stream in the study where this effect
proved significant. Red Clay also had the highest ambient [P] which suggests that
biofilm assemblages growing in this stream were likely P-sufficient and only capable of
additional P-uptake when N was also loaded. There is abundant data indicating that
113
supplies of both P and N limit algal growth in freshwater ecosystems (Elser et al. 1990).
For instance, Francoeur (2001) performed a meta-analysis of 237 nutrient amendment
experiments, and found 39 (16.5%) displayed N-limitation, 43 (18.1%) displayed P-
limited, and 55 (23.2%) showed combined N- and P-limitation. Furthermore, Tyrell
(1999) dispelled the notion that only N is important in the world’s oceans when he
reviewed the relative importance both P and N have on oceanic primary production. My
study showed that bulk P-uptake was amplified by P+N additions, and suggests that one
nutrient may exert biotic influences on the other. Similarly, P-limitation in the Pearl
River (southern Hong Kong) resulted in low planktonic utilization of nitrate (NO3)
leading to excess NO3 in downstream transport; P regeneration may allow phytoplankton
to take-up additional N in excess of the cellular needs (Yin et al. 2004). I found here that
both P and N are important in mediating physiological response processes, and thus
should be concomitantly managed as “inter-nutrient feedbacks” are possible with single
nutrient reductions (Paerl 2009).
Several authors have discussed the importance of N for benthic algal production
in lotic ecosystems (Francouer 2001, Dodds et al. 2002) and N enrichment has been
shown to enhance certain cellular chemical element concentrations in the diatom
Thalassioseira pseudonana (Division Bacillariophyta, Order Centrales) (Rijstenbil et al.
1998). Cells growing under N enriched conditions may synthesize high concentrations of
proteins (Wang and Dei 2001, Rausch and Bucher 2002), and proteins, especially
membrane and transporter proteins, are critical biochemical compounds needed for
phosphate uptake (Smith 2003). Therefore, it is possible that biofilms growing under N
enrichment may have had higher protein synthesis available for expression of cell surface
114
transporters, thus facilitating efficient P-uptake. This P-uptake enhancement through N
enrichment phenomenon, while novel in benthic microbial ecology, is resonated in the
higher plant literature. For instance, P-uptake was positively influenced by N fertilization
in Spartina alterniflora (Division Angiospermae, Order Poales) whereby single P
additions resulted in a significantly lower P content compared to adding P fertilizer in
conjunction with N treatments (Huberty 2005). My data likewise showed that elevated
[N] provided biofilms a greater ability to respond to a given P-load, suggesting that N is a
controlling (limiting) resource, particularly when P is in high concentrations. For
instance, the average TN concentration in Piedmont streams was estimated to be 3.13
mg/L (Carrick et al. 2009), approximately 2-fold higher than the suggested boundary
between mesotrophic and eutrophic conditions for streams (Dodds et al. 1998). That
said, streams in the Piedmont province (Red Clay, Cooks, Penns, Spring) showed
superior response to highest P-loadings (e.g., P5) compared to Plateau streams, indicating
that such N sufficiency may induce a competitive advantage in P-uptake (Rhee 1974).
Relatedly, with biofilms in the Plateau streams, N-loading augmented P-uptake at highest
P-loading (P5). In higher plants, nitrogen (specifically as ammonium) can enhance
phosphorus uptake by increasing P solubility and transport in root systems or increasing
P-uptake efficiency of roots (Cole et al. 1963, Tisdale et al. 1985, Smith and Jackson
1987, Hoffman et al. 1994).
The effects of N-loadings on biofilm ability to take-up P thus differed across the
stream productivity gradient. For example, the addition of N with P-loading appeared to
hamper the ability of biofilms in Plateau streams to take-up P, as indicated by reduced
intercepts (Table 3.4). Lean and Pick (1981) have discussed energy trade-offs, as the
115
uptake of one resource (e.g., N) may reduce the uptake of a second resource (e.g., P). It
is possible that N-loading events may induce biofilms to dedicate greater cell surface area
to N-uptake, thereby reducing total uptake sites available for P (cf. Ward et al. 2011). As
the number of uptake sites is related to the cell surface area (see Aksnes and Egge 1991,
Kriest and Oschlies 2007), smaller celled microbes in Plateau streams (plankton mean
cell size is strongly correlated with chlorophyll concentrations (Harris et al. 1987)) would
comparatively exhibit a stronger reduction of P-uptake sites when faced with a N-load
and thus demonstrate a lessened uptake potential. In contrast, synergistic effects of N-
loading with P on P-uptake in the Piedmont province suggests that N likely plays a role in
regulating the production of benthic biofilms in these streams, and appears to augment P-
uptake. This phenomenon is reasonable as secondary nitrogen limitation may be more
prominent in eutrophic aquatic systems (e.g., Matthews et al. 2002) and Piedmont
streams were shown to be more productive. This is also in agreement with my
hypotheses, such that nutrient thresholds for the Piedmont region are 3-fold higher for TP
compared to the Plateau region (Carrick et al. 2009) suggesting biofilms here may have
more expansive stores of poly-P. Similarly, phytoplankton in hypereutrophic Lake
Apopka was found to contain large supplies of P as polyphosphates, usable for cellular
growth when supplied with N (Carrick et al. 1993). Polyphosphate storage can also be
important in altering biomass stoichiometry (Sterner and Elser 2002, Makino and Cotner
2004).
That said, the spatial heterogeneity (Plateau vs. Piedmont) in P-uptake I observed
in my experiment may have been modified in some measure by microbial stoichiometry,
such that the biofilm internal nutrient balance facilitated different physiological response
116
(uptake) according to nutrient load. It has been found that microbial nutrient
stoichiometry can influence nutrient processing (e.g., sequestration and cycling) (Sterner
and Elser 2002). For example, biofilm communities with low cellular N:P ratios would
demonstrate a higher demand for N relative to P, and potentially increase the transport
length of P downstream (cf. Frost et al. 2002). However, chronic nutrient loading may
complicate this response; the nutrient content in the biofilm can be enhanced by
experimental nutrient enrichment (Hillebrand and Kahlert 2001). Biogeochemical
differences between the provinces may also have differentially contributed to the N:P and
C:P ratios. For instance, parent material in the Piedmont province is dominated by
limestone-dolomite (Appendix O), which could stimulate sorption of orthophosphate -
calcium and precipitate as dicalcium phosphate (Hargreaves and Tucker 1996). This
particulate inorganic phosphorus complex would then have been present in the chemical
analysis of TP, potentially influencing areal N:P and C:P ratios. Nonetheless, while both
regressions showed that benthic biofilm P-uptake increases as the N:P and C:P
stoichiometric ratios increase (i.e., towards P-limitation), these patterns followed a
different trajectory when N was loaded concurrently with P, suggesting that N
enrichments produce a more complex P-uptake phenomenon mediated by
biogeochemistry (province) and cellular stoichiometric ratios. My results show that the
effects of P- and N-loading on biofilm uptake ability can be interactive, that these effects
can vary considerably across geologic province, and that some of this variation may be
related to microbial nutrient stoichiometry.
Uptake efficiency, however, in the Piedmont was greatly curtailed under higher
nutrient loads, as signified by the increased negative slopes when N was loaded with P.
117
These stronger slopes here indicate that concurrent loadings of N with P induce a more
rapid decrease in biofilm P-uptake and uptake efficiency. This then suggests that, while
N can facilitate P-uptake (particularly when [P] is high), biofilm uptake capacity still
saturates when exposed to chronic, high nutrient loadings and in so doing decreases
uptake efficiency and retainment of P (Earl et al. 2006). This is particularly important
considering that one (ultimate) destination for this nutrient transport is the Chesapeake
Bay, and recent preliminary estimates have shown that phosphorus loads to the
Chesapeake Bay from 2009 to 2010 increased from 9 to 16 million pounds (CBPO 2011).
My work here highlights the capacity of biofilms to moderate downstream fluxes of P
through assimilation and the importance of reducing both P- and N-loads in efforts to
increase stream uptake efficiency (retainment) and ultimately achieve sustainable aquatic
management.
3.6 ACKNOWLEDGMENTS
I thank T. Wagner for support with the multilevel model and A. Lashaway, A.
Scanlan, and R. Wagner for assistance in the field. This research was supported by grants
provided to H.J.C. from Pennsylvania Department of Environmental Protection (Grant
No. 4100034506) and Pennsylvania Sea Grant (Grant No. UP69VY0).
3.7 LITERATURE CITED
Aber, J., McDowell, W., Nadelhoffer, K., Magill, A., Bernston, G., Kamakea, M.,
McNulty, S., Currie, W., Rustad, L., and Fernandez, I. 1998. Nitrogen saturation
in temperate forest ecosystems. BioScience 48:921-934.
118
Aber, J.D., Nadelhoffer, K.J., Steudler, P., and Melillo, J.M. 1989. Nitrogen saturation in
northern forest ecosystems. BioScience 39:378-386.
Aksnes, D.L. and Egge, J.K. 1991. A theoretical model for nutrient uptake in
phytoplankton. Marine Ecology Progress Series 70:65-72.
Allan, J.D. 2004. Landscapes and rivers: The influence of land use on stream ecosystems.
Annual Review of Ecology, Evolution, and Systematics 35:257-284.
Allmendinger, N.E., Pizzuto, J.E., Potter, J.N., Johnson, T.E., and Hession, W.C. 2005.
The influence of riparian vegetation on stream width, eastern Pennsylvania, USA.
Geological Society of America Bulletin 117:229-243.
American Public Health Association. 1992. Standard methods for the examination of
water and waste water, 18th
Ed. The Association, New York.
Anbumozhi, V., Radhakrishnan, J., and Yamaji, E. 2005. Impact of riparian buffer zones
on water quality and associated management considerations. Ecological
Engineering 24:517-523.
Aubriot, L., Bonilla, S., and Falkner, G. 2011. Adaptive phosphate uptake behaviour of
phytoplankton to environmental phosphate fluctuations. FEMS Microbiology
Ecology 77:1-16.
Beardall, J., Young, E., and Roberts, S. 2001. Approaches for determining phytoplankton
nutrient limitation. Aquatic Sciences 63:44-69.
Belsley, D., Kuh, E., and Welsch, R. 1980. Regressions Diagnostics: Identifying
Influential Data and Sources of Collinearity. New York: John Wiley and Sons.
Bernot, M.J. and Dodds, W.K. 2005. Nitrogen retention, removal, and saturation in lotic
ecosystems. Ecosystems 8:442-453.
119
Binkley, D., Ice, G.G., Kaye, J., and Williams, C.A. 2004. Nitrogen and phosphorus
concentrations in forest streams of the United States. Journal of the American
Water Resource Association 40:1277-1292.
Bollen, K.A. and Jackman R.W. 1990. Regression diagnostics: An expository treatment
of outliers and influential cases. In: J. Fox and J.S. Long (Ed.), Modern methods
of data analysis. Sage Publications, Newbury Park, CA.
Bormann, F.H., Likens, G.E., Fisher, D.W., and Pierce, R.S. 1968. Nutrient loss
accelerated by clear-cutting of a forest ecosystem. Science 159:882-884.
Burmaster, D.E. and Chisholm, S.W. 1979. A comparison of two methods for measuring
phosphate uptake by Monochrysis lutheri Droop grown in continuous culture.
Journal of Experimental Marine Biology and Ecology 39:187-202.
Busse, L.B., Simpson, J.C., and Cooper, S.D. 2006. Relationships among nutrients, algae,
and land use in urbanized southern California streams. Canadian Journal of
Fisheries and Aquatic Sciences 63:2621-2638.
Cabrera, R.I. 2003. Nitrogen balance for two container-grown woody ornamental plants.
Scientia Horticulturae 97:297-308.
Caperon, J. and Meyer, J. 1972. Nitrogen-limited growth of marine phytoplankton-II.
Uptake kinetics and their role in nutrient-limited growth of phytoplankton. Deep
Sea Research and Oceanographic Abstracts 19:619-632.
Carlisle, D.M. and Clements, W.H. 2003. Growth and secondary production of aquatic
insects along a gradient of Zn contamination in Rocky Mountain streams. Journal
of the North American Benthological Society 22:582-597.
120
Carrick, H.J. and Price, K.J. 2011. Determining variation in TMDL reduction criteria.
Final Report, Pennsylvania Department of Environmental Protection, 84 p.
Carrick, H.J., Scanlan, A., and Wagner, R. 2009. Use of periphyton to estimate TMDL
end-points. Final Report, Pennsylvania Department of Environmental Protection,
57 p.
Carrick, H.J., Schelske, C.L., Aldridge, F.J., and Coveney, M.F. 1993. Assessment of
phytoplankton nutrient limitation in productive waters: Application of dilution
bioassays. Canadian Journal of Fish and Aquatic Sciences 50:2208-2221.
CBPO (Chesapeake Bay Program Office). 2011. Phosphorus loads and river flow to the
bay. www.chesapeakebay.net/status_phosphorus.aspx.
Cembella, A.D., Antia, N.J., and Harrison, P.J. 1984. The utilization of inorganic and
organic phosphorus compounds as nutrients by eukaryotic microalgae: A
multidisciplinary perspective. Part 2. Critical Reviews in Microbiology 11:13-81.
Clemensson-Lindell, A. and Asp, H. 1995. Fine-root morphology and uptake of 32
P and
35S in a Norway spruce (Picea abies (L.) Karst.) stand subjected to various
nutrient and water supplies. Plant and Soil 173:147-155.
Cole, C.V., Grunes, D.L., Porter, L.K., and Olsen, S.R. 1963. The effects of nitrogen on
short-term phosphorus absorption and translocation in corn (Zea Mays).
Proceedings - Soil Science Society of America 27:671-674.
Corning, K.E., Duthie, H.C., and Paul, B.J. 1989. Phosphorus and glucose uptake by
seston and epilithon in boreal forest streams. Journal of the North American
Benthological Society 8:123-133.
121
Correll, D.L. 1998. The role of phosphorus in the eutrophication of receiving waters: A
review. Journal of Environmental Quality 27:261-266.
Covich, A.P., Austen, M.C., Barlocher, F., Chauvet, E., Cardinale, B.J., Biles, C.L.,
Inchausti, P., Dangles, O., Solan, M., Gessner, M.O., Statzner, B., and Moss, B.
2004. The role of biodiversity in the functioning of freshwater and marine benthic
ecosystems. BioScience 54:767-775.
David, M.B. and Gentry, L.E. 2000. Anthropogenic inputs of nitrogen and phosphorus
and riverine export for Illinois, USA. Journal of Environmental Quality 29:494-
508.
Davidson, E.A. and Howarth, R.W. 2007. Environmental science: Nutrients in synergy.
Nature 449:1000-1001.
De Schrijver, A., Verheyen, K., Mertens, J., Staelens, J., Wuyts, K., and Muys, B. 2008.
Nitrogen saturation and net ecosystem production. Nature 451:E1.
Dodds, W.K. 2003. The role of periphyton in phosphorus retention in shallow freshwater
aquatic systems. Journal of Phycology 39:840-849.
Dodds, W.K., Jones, J.R., and Welch, E.B. 1998. Suggested classification of stream
trophic state: Distributions of temperate stream types by chlorophyll, total
nitrogen, and phosphorus. Water Research 32:1455-1462.
Dodds, W.K., Smith, V.H., and Lohman, K. 2002. Nitrogen and phosphorus relationships
to benthic algal biomass in temperate streams. Canadian Journal of Fisheries and
Aquatic Sciences 59:865-874.
Dodson, S.I. 2005. Introduction to Limnology. McGraw Hill.
122
Earl, S.R., Valett, H.M., and Webster, J.R. 2006. Nitrogen saturation in stream
ecosystems. Ecology 87:3140-3151.
Elser, J.J., Bracken, M.E.S., Cleland, E.E., Gruner, D.S., Harpole, W.S., Hillebrand, H.,
Ngai, J.T., Seabloom, E.W., Shurin, J.B., and Smith, J.E. 2007. Global analysis of
nitrogen and phosphorus limitation of primary producers in freshwater, marine
and terrestrial ecosystems. Ecology Letters 10:1135-1142.
Elser, J.J., Marzolf, E.R., and Goldman, C.R. 1990. Phosphorus and nitrogen limitation of
phytoplankton growth in the freshwaters of North America: A review and critique
of experimental enrichments. Canadian Journal of Fisheries and Aquatic Sciences
47:1468-1477.
Francoeur, S.N. 2001. Meta-analysis of lotic nutrient amendment experiments: Detecting
and quantifying subtle responses. Journal of the North American Benthological
Society 20:358-368.
Franklin, D.H., Cabrera, M.L., and Calvert, V.H. 2005. Fertilizer source and soil aeration
effects on runoff volume and quality. Soil Science Society of America Journal
70:84-89.
Friedland, A.J. and Miller, E.K. 1999. Major-element cycling in a high-elevation
Adirondack forest: Patterns and changes, 1986–1996. Ecological Applications
93:958-967.
Frost, P.C., Stelzer, R.S., Lamberti, G.A., and Elser, J.J. 2002. Ecological stoichiometry
of trophic interactions in the benthos: understanding the role of C:N:P ratios in
lentic and lotic habitats. Journal of the North American Benthological Society
21:515-528.
123
Fuhs, G.W., Demmerle, S.D., Canelli, E., and Chen, M. 1972. Characterization of
phosphorus-limited plankton algae. In: G.E. Likens (Ed.), Nutrients and
eutrophication: The limiting-nutrient controversy. Proceedings of the American
Society of Limnology and Oceanography Special Symposia, Vol. 1. Allen Press,
Lawrence, Kansas, p. 113-133.
Games, P.A. 1971. Multiple comparisons of means. American Educational Research
Journal 8:531-565.
Gantzer, C.J., Rittmann, B.E., and Herricks, E.E. 1988. Mass transport to streambed
biofilms. Water Research 22:709-722.
Ghosh, M. and Gaur, J.P. 1994. Algal periphyton of an unshaded stream in relation to in
situ nutrient enrichment and current velocity. Aquatic Botany 47:185-189.
Gibeau, G.G. and Miller, M.C. 1989. A micro-bioassay for epilithon using nutrient
diffusing artificial substrata. Journal of Freshwater Ecology 5:171-176.
Graziano, L.M., La Roche, J., and Geider, R.J. 1996. Physiological responses to
phosphorus limitation in batch and steady-state cultures of Dunaliella tertiolecta
(Chlorophyta): A unique stress protein as an indicator of phosphate deficiency.
Journal of Phycology 32:825-838.
Grula, J.W. 2005. Evolution of photosynthesis and biospheric oxygenation contingent
upon nitrogen fixation? International Journal of Astrobiology 4:251-257.
Hargreaves, J.A. and Tucker, C.S. 1996. Evidence for control of water quality in channel
catfish Ictalurus punctatus ponds by phytoplankton biomass and sediment
oxygenation. Journal of the World Aquaculture Society 27:21-29.
124
Harold, F.M. 1966. Inorganic polyphosphates in biology: Structure, metabolism, and
function. Microbiology and Molecular Biology Reviews 30:772-794.
Harris, G.P., Ganf, G.G., and Thomas, D.P. 1987. Productivity, growth rates and cell size
distributions of phytoplankton in the SW Tasman Sea: Implications for carbon
metabolism in the photic zone. Plankton Research 9:1003-1030.
Harrison, P.J., Parslow, J.S., and Conway, H.L. 1989. Determination of nutrient uptake
kinetic parameters: A comparison of methods. Marine Ecology Progress Series
52:301-312.
Hart, D.D. and Finelli, C.M. 1999. Physical-biological coupling in streams: The
pervasive effects of flow on benthic organisms. Annual Review of Ecology and
Systematics 30:363-395.
Harvey, C.J., Peterson, B.J., Bowden, W.B., Hershey, A.E., Miller, M.C., Deegan, L.A.,
and Finlay, J.C. 1998. Biological responses to fertilization of Oksrukuyik Creek, a
tundra stream. Journal of the North American Benthological Society 17:190-209.
Havens, K.E. and Schelske, C.L. 2001. The importance of considering biological
processes when setting total maximum daily loads (TMDL) for phosphorus in
shallow lakes and reservoirs. Environmental Pollution 113:1-9.
Havens, K.E., East, T.L., Hwang, S.J., Rodusky, A.J., Sharfstein, B., and Steinman, A.D.
1999. Algal responses to experimental nutrient addition in the littoral community
of a subtropical lake. Freshwater Biology 42:329-344.
Hebblewhite, M. 2006. Linking predation risk and forage to ungulate population
dynamics. Dissertation. University of Alberta, Edmonton, Alberta, Canada.
125
Hillebrand, H. and Kahlert, M. 2001. Effect of grazing and nutrient supply on periphyton
biomass and nutrient stoichiometry in habitats of different productivity.
Limnology and Oceanography 46:1881-1898.
Hobbs, R.J. 1993. Effects of landscape fragmentation on ecosystem processes in the
Western Australian wheat belt. Biological Conservation 64:193-201.
Hoffmann, C., Ladewig, E., Claassen, N., and Jungk, A. 1994. Phosphorus uptake of
maize as affected by ammonium and nitrate nitrogen - Measurements and model
calculations. Zeitschrift für Pflanzenernährung und Bodenkunde 157:225-232.
Hoffmann, J.P. 1998. Wastewater treatment with suspended and nonsuspended algae.
Journal of Phycology 34:757-763.
Hough, A.F. and Forbes, R.D. 1943. The ecology and silvics of forests in the high
plateaus of Pennsylvania. Ecological Monographs 13:299-320.
Huberty, A. 2005. Nutrient limitation and its consequences for performance and the
homeostatic regulation of macronutrient composition in two phytophagous insects
with divergent life-history strategies. Dissertation. University of Maryland,
College Park, Maryland, USA.
Hullar, M.A. and Vestal, J.R. 1989. The effects of nutrient limitation and stream
discharge on the epilithic microbial community in an oligotrophic arctic stream.
Hydrobiologia 172:19-26.
Jansson, M. 1993. Uptake, exchange and excretion of orthophosphate in phosphate
starved Scenedesmus quadricauda and Pseudomonas K7. Limnology and
Oceanography 38:1162-1178.
126
Jaworski, N.A. 1981. Sources of nutrients and the scale of eutrophication problems in
estuaries. In: B.J. Neilson and L.E. Cronin (Ed.), Estuaries and nutrients. Humana,
Clifton, N.J., p. 83-110.
Jensén, P. and Ktönig, T. 1982. Development of regulation mechanisms for SO42-
influx
in spring wheat roots. Physiologia Plantarum 55:459-464.
Karr, J.R. 1991. Biological integrity: A long-neglected aspect of water resource
management. Ecological Applications 1:66-84.
Kaya, M.V. and Picard, G. 1995. The viability of Scenedesmus bicellularis cells
immobilized on alginate screens following nutrient starvation in air at 100%
relative humidity. Biotechnology and Bioengineering 46:459-464.
Kriest, I. and Oschlies, A. 2007. Modelling the effect of cell-size-dependent nutrient
uptake and exudation on phytoplankton size spectra. Deep-Sea Research I
54:1593-1618.
Lapointe, B.E. 1989. Macroalgal production and nutrient relations in oligotrophic areas
of Florida Bay USA. Bulletin of Marine Science 44:312-323.
Lean, D.R.S. and Pick, F.R. 1981. Photosynthetic response of lake plankton to nutrient
enrichment: A test for nutrient limitation. Limnology and Oceanography 26:1001-
1019.
LeBauer, D.S. and Treseder, K.K. 2008. Nitrogen limitation of net primary productivity
in terrestrial ecosystems is globally distributed. Ecology 82:371-379.
Li, M., Mao, L., Zhou, C., Vogelmann, J.E., and Zhu, Z. 2010. Comparing forest
fragmentation and its drivers in China and the USA with Globcover v2.2. Journal
of Environmental Management 91:2572-2580.
127
Likens, G.E., Bormann, F.H., Johnson, N.M., Fisher, D.W., and Pierce, R.S. 1970.
Effects of forest cutting and herbicide treatment on nutrient budgets in the
Hubbard Brook watershed-ecosystem. Ecological Monographs 40:23-47.
Mainstone, C.P. and Parr, W. 2002. Phosphorus in rivers—ecology and management. The
Science of the Total Environment 282–283:25-47.
Makino, W. and Cotner, J.B. 2004. Elemental stoichiometry of a heterotrophic bacterial
community in a freshwater lake: Implications for growth- and resource-dependent
variations. Aquatic Microbial Ecology 34:33-41.
Malone, T.C. and Chervin, M.B. 1979. The production and fate of phytoplankton size
fractions in the plume of the Hudson River, New York Bight. Limnology and
Oceanography 24:683-696.
Marks, J.C. and Lowe, R.L. 1993. Interactive effects of nutrient availability and light
levels on the periphyton composition of a large oligotrophic lake. Canadian
Journal of Fisheries and Aquatic Sciences 50:1270-1278.
Martin, R.A., Harms, T.K., and Grimm, N.B. 2011. Chronic N loading reduces N
retention across varying base flows in a desert river. Journal of the North
American Benthological Society 30:559-572.
Matthews, R., Hilles, M., and Pelletier, G. 2002. Determining trophic state in Lake
Whatcom, Washington (USA), a soft water lake exhibiting seasonal nitrogen
limitation. Hydrobiologia 468:107-121.
McColl, R.H.S. 1974. Self-purification of small freshwater streams: Phosphate, nitrate,
and ammonia removal. NZ Journal of Marine and Freshwater Research 8:375-
388.
128
Menge, D.N.L., Ballantyne, F., and Weitz, J.S. 2011. Dynamics of nutrient uptake
strategies: Lessons from the tortoise and the hare. Theoretical Ecology 1:163-177.
Menzel, D.W. and Corwin, N.R. 1965. The measurement of total phosphorus in seawater
based on the liberation of organically bound fractions by persulfate digestion.
Limnology and Oceanography 10:280-282.
Millie, D.F., Pigg, R.J., Fahnenstiel, G.L., and Carrick, H.J. 2009. Algal chlorophylls: A
synopsis of analytical methodologies. In: American Water Works Association,
Manual M57, Algae. American Water Works Association, Denver, Colorado
USA. p. 93-122.
Montuelle, B., Dorigo, U., Bérard, A., Volat, B., Bouchez, A., Tlili, A., Gouy, V., and
Pesce, S. 2010. The periphyton as a multimetric bioindicator for assessing the
impact of land use on rivers: an overview on the Ardières–Morcille experimental
watershed (France). Hydrobiologia 657:123-141.
Noe, G.B., Scinto, L.J., Taylor, J., Childers, D.L., and Jones, R.D. 2003. Phosphorus
cycling and partitioning in an oligotrophic Everglades wetland ecosystem: A
radioisotope tracing study. Freshwater Biology 48:1993-2008.
O'Brien, J.M. and Dodds, W.K. 2010. Saturation of NO3- uptake in prairie streams as a
function of acute and chronic N exposure. Journal of the North American
Benthological Society 29:627-635.
Oelkers, E.H. and Valsami-Jones, E. 2008. Phosphate mineral reactivity and global
sustainability. Elements 4:83-88.
Olander, L.P. and Vitousek, P.M. 2004. Biological and geochemical sinks for phosphorus
in soil from a wet tropical forest. Ecosystems 7:404-419.
129
Omernik, J.M. 1987. Ecoregions of the conterminous United States. Map Supplement
(scale 1:7,5000,000). Annals of the Association of American Geographers
77:118-125.
PADCNR (Pennsylvania Department of Conservation and Natural Resources, Bureau of
Forestry). 2010. Pennsylvania statewide forest resource assessment. Pennsylvania
Department of Conservation and Natural Resources. Harrisburg, Pennsylvania,
210 p.
Paerl, H.W. 2009. Controlling eutrophication along the freshwater–marine continuum:
Dual nutrient (N and P) reductions are essential. Estuaries and Coasts 32:593-601.
Paerl, H.W., Xu, H., McCarthy, M.J., Zhu, G., Qin, B., Li, Y., and Gardner, W.S. 2010.
Controlling harmful cyanobacterial blooms in a hypereutrophic lake (Lake Taihu,
China): The need for a dual nutrient (N & P) management strategy. Water
Research 16:2358-2365.
PASDA (Pennsylvania Spatial Data Access). 2005. Retrieved August 1, 2005.
www.pasda.psu.edu/access.
Pennell, K.D., Allen, H.L., and Jackson, W.A. 1990. Phosphorus uptake capacity of 14-
year-old loblolly pine as indicated by a 32
P root bioassay. Forest Science 36:358-
366.
Peterjohn, W.T. and Correll, D.L. 1984. Nutrient dynamics in an agricultural watershed:
Observations on the role of a riparian forest. Ecology 65:1466-1475.
130
Peterson, B.J., Wollheim, W.H., Mulholland, P.J., Webster, J.R., Meyer, J.L., Tank, J.L.,
Martí, E., Bowden, W.B., Valett, H.M., Hershey, A.E., McDowell, W.H., Dodds,
W.K., Hamilton, S.K., Gregory, S., and Morrall, D.D. 2001. Control of nitrogen
export from watersheds by headwater streams. Science 292:86-90.
Platt, T., Harrison, W.G., Irwin, B., Horne, E.P., and Gallegos, C.L. 1982. Photosynthesis
and photoadaptation of marine phytoplankton in the Arctic. Deep Sea Research
29:1159-1170.
Powell, N., Shilton, A.N., Pratt, S., and Chisti, Y. 2008. Factors influencing luxury
uptake of phosphorus by microalgae in waste stabilization ponds. Environmental
Science and Technology 42:5958-5962.
Powers, S.M., Stanley, E.H., and Lottig, N.R. 2009. Quantifying phosphorous uptake
using pulse and steady-state approaches in streams. Limnology and Oceanography
Methods 7:498-508.
Price, K.J. and Carrick, H.J. 2011. Meta-analytical approach to explain variation in
microbial phosphorus uptake rates in aquatic ecosystems. Aquatic Microbial
Ecology 65:89-102.
Puckett, L.J. 1995. Identifying the major sources of nutrient water pollution.
Environmental Science and Technology 29:408-14.
Qian, S.S., Cuffney, T.F., Alameddine, I., McMahon, G., and Reckhow, K.H. 2010. On
the application of multilevel modeling in environmental and ecological studies.
Ecology 91:355-361.
R Development Core Team. 2010. R: A language and environment for statistical
computing. R Foundation for Statistical Computing, Vienna, Austria.
131
Raghothama, K.G. 1999. Phosphate acquisition. Annual Review of Plant Physiology and
Plant Molecular Biology 50:665-693.
Rausch, C. and Bucher, M. 2002. Molecular mechanisms of phosphate transport in plants.
Planta 216:23-37.
Reddy, K.R., Kadlec, R.H., Flaig, E., and Gale, P.M. 1999. Phosphorus retention in
streams and wetlands: A review. Critical Reviews in Environmental Science and
Technology 29:83-146.
Reynolds, C.S. 1984. The ecology of freshwater phytoplankton. Cambridge University
Press, Cambridge, UK.
Rhee, G-Y. 1973. A continuous culture study of phosphate uptake, growth rate, and
polyphosphate in Scenedesmus sp. Journal of Phycology 9:495-506.
Rhee, G-Y. 1974. Phosphate uptake under nitrate limitation by Scenedesmus sp. and
ecological implications. Journal of Phycology 10:470-475.
Ribaudo, M. and Johansson, R. 2006. Water quality: Impacts of agriculture. In:
Agricultural resources and environmental indicators. United States Department of
Agriculture Economic Research Service. Washington, DC.
Rijstenbil, J.W., Dehairs, F., Ehrlich, R., and Wijnholds, J.A. 1998. Effect of the nitrogen
status on copper accumulation and pools of metal-binding peptides in the
planktonic diatom Thalassiosira pseudonana. Aquatic Toxicology 42:187-209.
Ristvey, A.G., Lea-Cox, J.D., and Ross D.S. 2007. Nitrogen and phosphorus uptake
efficiency and partitioning of container-grown azalea during spring growth.
Journal of the American Society for Horticultural Science 132:563-572.
132
Rosenberg, H. 1987. Phosphate transport in prokaryotes. In: B.P. Rosen and S. Silver
(Ed.), Ion transport in prokaryotes. Academic, p. 205-246.
Rüegg, J., Tiegs, S.D., Chaloner, D.T., Levi, P.S., Tank, J.L., and Lamberti, G.A. 2011.
Salmon subsidies alleviate nutrient limitation of benthic biofilms in southeast
Alaska streams. Canadian Journal of Fisheries and Aquatic Sciences 68:277-287.
Runcie, J.W., Ritchie, R.J., and Larkum, A.W.D. 2004. Uptake kinetics and assimilation
of phosphorus by Catenella nipae and Ulva lactuca can be used to indicate
ambient phosphate availability. Journal of Applied Phycology 16:181-194.
Sabater, S., Guasch, H., Romaní, A., and Muñoz, I. 2002. The effect of biological factors
on the efficiency of river biofilms in improving water quality. Hydrobiologia
469:149-156.
Scheffer, M., Carpenter, S., Foley, J.A., Folke, C., and Walker, B. 2001. Catastrophic
shifts in ecosystems. Nature 413:591-596.
Schindler, D.W. 1977. Evolution of phosphorus limitation in lakes. Science 195:260-262.
Schindler, D.W., Hecky, R.E., Findlay, D.L., Stainton, M.P., Parker, B.R., Paterson, M.J.,
Beaty, K.G., Lyng, M., and Kasian, S.E.M. 2008. Eutrophication of lakes cannot
be controlled by reducing nitrogen input: Results of a 37-year whole-ecosystem
experiment. Proceedings of the National Academy of Sciences USA 105:11254-
11258.
Scott, J.T. and McCarthy, M.J. 2010. Nitrogen fixation may not balance the nitrogen pool
in lakes over timescales relevant to eutrophication management. Limnology and
Oceanography 55:1265-1270.
133
Singh, S.K., Pandey, V., and Pandey, K.D. 2007. Phosphate uptake kinetics and its
regulation in N2-fixing cyanobacterium Anabaena oryzae Fritsch under salt stress.
African Journal of Biotechnology 6:2363-2368.
Smith, F.W. and Jackson, W.A. 1987. Nitrogen enhancement of phosphate transport in
roots of Zea mays L. II. Kinetic and inhibitor studies. Plant Physiology 84:1319-1324.
Smith, F.W., Mudge, S.R., Rae, A.L., and Glassop, D. 2003. Phosphate transport in
plants. Plant and Soil 248:71-83.
Steinman, A.D. and Mulholland, P.J. 2006. Phosphorus limitation, uptake, and turnover
in benthic stream algae. In: F.R. Hauer and G.A. Lamberti (Ed.), Methods in
stream ecology, 2nd
Ed., Academic Press, p. 187-212.
Sterner, R.W. and Elser, J.J. 2002. Ecological stoichiometry: The biology of elements
from molecules to the biosphere. Princeton University Press, Princeton, N.J.
Stevenson, R.J. 1990. Benthic algal community dynamics in a stream during and after a
spate. Journal of the North American Benthological Society 9:277-288.
Stevenson, R.J. 1996. The stimulation and drag of current. In: R.J. Stevenson, M.
Bothwell, and R.L. Lowe (Ed.), Algal ecology: Freshwater benthic ecosystems.
Academic Press, Inc., San Diego, p. 321-340.
Stockner, J.G. and Shortreed, K.R.S. 1978. Enhancement of autotrophic production by
nutrient addition in a coastal rainforest stream on Vancouver Island. Journal of the
Fisheries Research Board of Canada 35:28-34.
Svendsen, L.M., Kronvang, B., Kristensen, P., and Græsbøl, P. 1995. Dynamics of
phosphorus compounds in a lowland river system: Importance of retention and
non-point sources. Hydrological Processes 9:119-142.
134
Tisdale, S.L., Nelson, W.L., and Beaton, J.D. 1985. Soil fertility and fertilizers.
McMillan Publishing Co. New York.
Triska, F.J., Kennedy, V.C., Avanzino, R J., Zellweger, G.W., and Bencala, K.E. 1990. In
situ retention-transport response to nitrate loading and storm discharge in a third-
order stream. Journal of the North American Benthological Society 9:229-239.
Tyrrell, T. 1999. The relative influences of nitrogen and phosphorus on oceanic primary
production. Nature 400:525-531.
Versaw, W.K. and Harrison, M.J. 2002. A chloroplast phosphate transporter, PHT2;1,
influences allocation of phosphate within the plant and phosphate starvation
responses. Plant Cell 14:1751-1766.
Verspoor, J.J., Braun, D.C., and Reynolds, J.D. 2010. Quantitative links between Pacific
salmon and stream periphyton. Ecosystems 13:1020-1034.
Vitousek, P.M. and Howarth, R.W. 1991. Nitrogen limitation on land and in the sea: How
can it occur? Biogeochemistry 13:87-115.
Wang, W-X. and Dei, R.C.H. 2001. Metal uptake in a coastal diatom influenced by major
nutrients (N, P, and Si). Water Research 35:315-321.
Ward, B.A., Dutkiewicz, S., Barton, A.D., and Follows, M.J. 2011. Biophysical aspects
of resource acquisition and competition in algal mixotrophs. American Naturalist
178:98-112.
Webster, J.R., Newbold, J.D., Thomas, S.A., Valett, H.M., and Mulholland, P.J. 2009.
Nutrient uptake and mineralization during leaf decay in streams—a model
simulation. International Review of Hydrobiology 94:372-390.
135
Wetzel, R.G. and Lichens, G.E. 2000. Limnology analyses. 3rd
Ed., Academic Press. San
Diego, CA.
Yin, K.D., Song, X.X., Sun, J., and Wu, M.C.S. 2004. Potential P limitation leads to
excess N in the Pearl River estuarine coastal plume. Continental Shelf Research
24:1895-1907.
136
Table 3.1. A summary of geographic and biogeochemical characteristics for streams where ISES experiments were conducted in 2008. Data shown are
means (± SD) for environmental conditions during deployment and retrieval (n= 2) of each ISES to test the effects of P and N-loading on native stream
biofilms. Note: refer to Figure 3.1 for Site ID geographical locations.
Province Stream Site
ID
Temperature
(°C)
Conductivity
(µS/cm)
Chlorophyll-a
(mg/m2)
O2 (%) TP
(mg/L)
Forested
(%)
Agricultural
(%)
Plateau East Hickory 1 17.66± 1.12 74.0± 5.7 30.05± 19.94 84.40±
7.92
0.014±
0.000
98.1 1.9
Tionesta 2 20.35± 1.51 127.0± 19.8 22.20± 15.91 83.10±
8.34
0.011±
0.002
95.7 3.0
Cowanesque 3 23.95± 3.50 260.0± 18.4 62.10± 0.00 108.90±
9.76
0.052±
0.019
63.2 35.9
Tunkhannock 4 17.18± 7.30 261.5± 7.8 142.90±
164.12
99.65±
21.43
0.021±
0.008
65.9 31.0
Piedmont Red Clay 5 19.45± 0.36 528.5± 77.1 163.98± 88.92 47.05±
32.60
0.423±
0.350
30.0 52.5
Cooks 6 15.72± 1.01 363.5± 0.7 57.90± 36.56 67.20±
2.26
0.025±
0.007
60.0 38.8
Penns 7 14.03± 0.49 494.0± 17.0 84.13± 25.00 41.30±
22.63
0.022±
0.006
68.9 30.5
Spring 8 11.40± 0.08 541.5± 7.8 766.50±
667.09
41.45±
5.16
0.002±
0.003
35.1 53.4
137
Table 3.2. Regression statistics for biofilm Chl-specific P-uptake (log10(nmolP/μgChl/day)) versus P-
loading (log10(µg[PO4]/day)) determined from ISES experiments carried out in streams of varying
nutrient content in both Plateau and Piedmont provinces (n= 10 per stream).
Province Stream y-intercept Std. Error
y-intercept
Slope Std. Error
Slope
r2 p
Plateau East Hickory 1.420 0.099 -0.268 0.054 0.757 0.001
Tionesta 1.269 0.101 -0.211 0.055 0.651 0.005
Cowanesque 1.711 0.085 -0.204 0.046 0.711 0.002
Tunkhannock 1.164 0.122 -0.280 0.066 0.690 0.003
Piedmont Red Clay 1.890 0.108 -0.070 0.060 0.148 0.273
Cooks 0.830 0.040 -0.109 0.022 0.753 0.001
Penns 0.659 0.107 -0.090 0.060 0.218 0.174
Spring 0.183 0.029 -0.046 0.016 0.500 0.022
138
Table 3.3. Likelihood ratio test results from iterative model building in determining important
parameters in explaining P-uptake against P-loading. Note: degrees of freedom (DF) is difference in
the number of parameters between the full and reduced models and statistical significance is
determined as p< 0.10.
Model Reduced Model Parameters Full Model Parameters x2 DF p
1 vs. 2 Fixed intercept Fixed intercept
Random stream (intercept)
135.97 1 < 0.001
2 vs. 3 Fixed intercept
Random stream (intercept)
Fixed intercept
Random stream (intercept)
Random loading (slope)
5.58 2 0.061
3 vs. 4 Fixed intercept
Random stream (intercept)
Random loading (slope)
Fixed intercept
Random stream (intercept)
Random loading (slope)
Fixed percent Ag
0.25 1 0.616
3 vs. 5 Fixed intercept
Random stream (intercept)
Random loading (slope)
Fixed intercept
Random stream (intercept)
Random loading (slope)
Fixed interaction percent Ag
5.95 1 0.015
139
Table 3.4. Regression statistics for biofilm Chl-specific P-uptake (log10(nmolP/μgChl/day)) versus P-
loading (log10(µg[PO4]/day)) without (n= 5) or with (n= 5) simultaneous N-loading.
Stream Treatment y-intercept Std. Error
y-intercept
Slope Std. Error
Slope
r2 p
East Hickory P (without N) 1.464 0.158 -0.290 0.086 0.792 0.043
P (with N) 1.376 0.161 -0.246 0.087 0.725 0.067
Tionesta P (without N) 1.390 0.067 -0.276 0.036 0.951 0.005
P (with N) 1.148 0.197 -0.146 0.107 0.383 0.266
Cowanesque P (without N) 1.733 0.120 -0.242 0.065 0.822 0.034
P (with N) 1.690 0.131 -0.165 0.071 0.641 0.104
Tunkhannock P (without N) 1.306 0.198 -0.324 0.108 0.751 0.057
P (with N) 1.023 0.155 -0.236 0.084 0.725 0.067
Red Clay P (without N) 1.911 0.100 0.002 0.056 0.000 0.973
P (with N) 1.868 0.061 -0.143 0.034 0.856 0.024
Cooks P (without N) 0.762 0.056 -0.072 0.031 0.639 0.104
P (with N) 0.898 0.044 -0.147 0.025 0.923 0.009
Penns P (without N) 0.491 0.195 -0.003 0.110 0.000 0.983
P (with N) 0.826 0.065 -0.178 0.036 0.888 0.017
Spring P (without N) 0.142 0.038 -0.024 0.022 0.285 0.354
P (with N) 0.225 0.042 -0.068 0.024 0.735 0.063
140
Figure 3.1. Mid-Atlantic region (left) and Pennsylvania statewide map (right) showing the location of
the eight ISES experiments in both of the major physiographic provinces in Pennsylvania
(Appalachian Plateau and Piedmont). Map was created using ArcGIS 10.0 (Environmental Systems
Research Institute, Inc., Redlands, CA, USA) and displays the boundaries of the US EPA Level III
Ecoregions and county borders.
141
Figure 3.2. Scatter plot of Chl-specific P-uptake (log10(nmolP/μgChl/day)) versus P-loading
(log10(µg[PO4]/day)) for each stream tested with ISES (n= 10 per stream) fit with linear regression
and 95% confidence intervals.
142
Figure 3.3. Regression slopes (left) and intercepts (right) modeled as linear functions of percent
agriculture across eight streams sampled using ISES. Note scale difference.
143
2.01.51.00.50.0 2.01.51.00.50.0
1.6
1.2
0.8
0.4
0.0
3.02.52.01.51.0
1.6
1.2
0.8
0.4
0.0
3.02.52.01.51.0
Log N:P, Plateau
Lo
g C
hl S
p P
-Up
take
Log N:P, Piedmont
Log C:P, Plateau Log C:P, Piedmont
Figure 3.4. Scatter plot of log P-uptake (log10(nmolP/μgChl/day)) vs. log10 N:P (top) and log10 C:P
(bottom) ratios fit with linear regression at subgroups (N-load) for streams in each province. Open
square markers and dotted line regressions indicate concurrent N with P-loadings while closed circle
markers and solid line regressions indicate sole P-loadings. The vertical line at x= 1.204 for log10 N:P
plots indicates the optimal Redfield N:P ratio (16:1) and the vertical line at x= 2.025 for log10 C:P
plots indicates the optimal Redfield C:P ratio (106:1). Note scale difference between N:P and C:P
plots.
144
CHAPTER 4
QUALITATIVE EVALUATION OF SPATIO-TEMPORAL PHOSPHORUS FLUXES IN STREAM
BIOFILMS
145
4.1 ABSTRACT
Uptake and storage by biofilms facilitates the ability of streams to act as sinks for
inorganic nutrients. P fluxes over short time periods are important for examining
physiological responses to new nutrients inputs. Therefore, I sampled intact stream
biofilm assemblages seasonally and across a spatial productivity gradient and tested their
response to a new P source over the first few minutes of exposure so as to resolve
changes in initial assimilatory kinetics and integrate the singular and interactive effects of
space and time on P fluxes. Intact biofilm communities were sampled seasonally (2009 -
2010) across eight streams between two eco-regions (Appalachian Plateau and Piedmont)
in Pennsylvania. Biofilms were subjected to short-term radiotracer experiments to
estimate uptake rates. Time-series plots were made of P flux over time and data were fit
using LOESS, facilitating estimations of P-uptake and efflux rates. The data revealed
distinct breakpoints in P fluxes; specifically mean max uptake occurred at 1.65 and 2.60
minutes while max efflux occurred at 4.51 and 4.62 minutes in Piedmont and Plateau
provinces, respectively. Stream/biofilm biochemical parameters were strong predictors
of P-uptake and efflux. A MANOVA showed significant differences in P-uptake and P-
efflux by province (F2, 50= 5.91, p= 0.005) and season (F6, 100= 4.23, p= 0.001). These
results demonstrate considerable and rapid exchange processes occurring at early time
periods (i.e., < 5 minutes), the magnitude of which seems to diminish over longer periods
(i.e., 15 - 30 minutes), indicating nutrient processing as a near instantaneous
physiological, dynamic process and further suggesting that experimental time periods
scaled to hours or longer obscure such essential short-term responses. This research
146
further indicates that spatio-temporal effects, specifically productivity and seasonality,
are strong determinants of P fluxes.
4.2 INTRODUCTION
Uptake and storage by organisms growing attached to submerged substrata
(biofilms) facilitates the ability of streams to act as sinks for inorganic nutrients that can
cause eutrophication (Vollenweider 1968, Cardinale et al. 2002, Jöbgen et al. 2004).
Existing at the interface between sediments and the water column, microbial biofilms can
regulate benthic/pelagic nutrient cycling by increasing nutrient flux from the water
column into the sediments (Confer 1972, Spears et al. 2007). As such, biofilm uptake
capacity for key, bio-limiting nutrients (i.e., carbon (C), nitrogen (N), and phosphorus
(P)) is of great importance, particularly for P which has been identified as the primary
limiting nutrient in freshwater systems (Correll 1998, Karl 2000, Schindler et al. 2008).
For example, P-uptake and adsorption by biofilms attached to duckweed (Lemna gibba
L.) accounted for 31 - 71% of the total P removal in a wastewater treatment system
(Körner and Vermaat 1998); and biofilms on submerged artificial substrata have proven
effective in P removal from eutrophic lakes (Jöbgen et al. 2004). Michaelis-Menten
kinetics have been used to describe nutrient uptake into biofilms (Reuter et al. 1986),
facilitating estimations of the maximum uptake rate (Vmax) and half-saturation constant
(Km); although there are many instances where this model is not supported (e.g.,
Tarapchak and Herche 1986). While Michaelis-Menten kinetic estimations are useful,
there are however, several types of error induced by experimental protocol during
determination of Vmax and Km (Flynn 1998). Estimations of nutrient uptake parameters
147
can be confounded by both temporal and spatial factors. For instance, in a meta-analysis
of P-uptake experiments, Price and Carrick (2011) found that methodological variation in
sample time can influence P assimilation, such that sampling over shorter time scales
yields higher uptake rates. While likely more related to membrane transport kinetics,
short term processes assuredly affect assimilatory kinetics as well (Wheeler et al. 1982,
Flynn 1998), which highlights the importance of experimentally investigating short-term
P flux phenomena. Similarly, microbe physiological pre-history is crucial in determining
kinetic uptake responses (Halterman and Toetz 1984); however, little research has
compared biofilm uptake capacity across stream trophy (biogeochemical history). P
export rate often increases with land disturbance (agricultural or urban lands) and
proportion of impervious surfaces (Byron and Goldman 1989, Puckett 1995) causing
variation in P-loads (and thus nutrient history) across land-use; therefore, information on
spatial heterogeneity of biofilm uptake capacities could be beneficial to predicting
nutrient effects on stream environs.
The physical and chemical architecture of biofilms may act to enhance the uptake
and storage of nutrients in dynamic and turbulent stream ecosystems (Neu and Lawrence
1997). A polysaccharide matrix surrounding biofilm communities provides a mechanism
for entrapment and concentration of nutrients (Lock et al. 1984, Laspidou and Rittmann
2002), effectively increasing local nutrient availability compared to overlying waters
(Freeman et al. 1995). There is thus contention that uptake is a coupled process: 1)
adsorption, or the movement of P into such surface-adsorbed P pools, and 2) absorption,
or the movement of adsorbed P into intercellular pools (Short and McRoy 1984, Sañudo-
Wilhelmy et al. 2004, Yao et al. 2011). Nonetheless, uptake is not necessarily
148
unidirectional. For example, Kuenzler et al. (1979) found that microbial recycling
(efflux) of reactive phosphorus accounted for 37 - 90% of microbial uptake in the
Pamlico River Estuary. This association of P-uptake with efflux is resonated in the
higher plant literature as well (Bieleski 1973, Elliot et al. 1984). Therefore, to the extent
that surface-adsorption and recycling (efflux) of phosphate may affect uptake processes
(e.g., Yao et al. 2011), examination of P flux patterns is necessary, particularly through
short-term experiments so as to estimate the influence of initial assimilatory processes on
biofilm P dynamics (cf. Wheeler et al. 1982). Moreover, given that nutrient uptake has
shown strong relationships with nutrient (P) content of cells (Riegman and Mur 1984) as
well as microbial biomass (Riedel et al. 1996), such predictive factors might as well
explain some variation in efflux and thus warrant further correlative consideration.
The objective of this study was to sample intact stream biofilm assemblages
seasonally and across a spatial productivity gradient and test their response to a new P
source over the first few minutes of exposure so as to resolve changes in initial
assimilatory kinetics and integrate the singular and interactive effects of space and time
on P fluxes.
4.3 METHODS
To test my hypotheses on natural, intact biofilm communities, a set of unglazed
ceramic tiles (surface area= 8.42 cm2) were secured to cement blocks and established in
eight streams of varying nutrient content and relative productivity; temporal effects were
estimated by sampling streams seasonally (Fall 2009 - Summer 2010; n= 64 experimental
units) (Appendix P). The eight streams were divided evenly between two eco-regions
149
(Appalachian Plateau and Piedmont) in the state’s Water Quality Network (see Carrick et
al. 2009). Specifically, streams located in the Plateau province were East Hickory
(41.6160, -79.3701), Tionesta (41.6019, -79.0503), Cowanesque (42.0014, -77.1278), and
Tunkhannock (41.5581, -75.8950); streams in the Piedmont province were Spring
(40.7917, -77.7980), Penns (40.8590, -77.5796), Cooks (40.5854, -75.2061), and Red
Clay (39.81667, -75.69194) (see Carrick and Price 2011 for additional stream
physicochemical information). After a 30 day incubation period, tiles with adhered
biofilms were removed from each stream and placed into 60 mL translucent
polypropylene (Nalgene) incubation jars with site-specific water. Once tiles were
returned to the laboratory, incubation jars with intact biofilms were enriched with 200 μg
of KH2PO4 to prevent potential isotope-dilution, or recycling of substrate (e.g., Harrison
and Harris 1986) and to fulfill latent requirements of extracellular potassium (K) for PO4
uptake (e.g., Weiden et al. 1967). For fine resolution into the time sensitivity of biofilm
uptake and estimations of short-term P fluxes, 1 mL of water overlying biofilms was
removed from each replicate incubation chamber using a sterile pipette at 6 - 11 time
periods ≤ 30 minutes after 0.75 mL injection of radiotracer (H333
PO4) (activity= 10
μCi/L). Water samples were then placed into labeled 20 mL scintillation vials filled with
5 mL Ecolume scintillation cocktail (ICN Pharmaceuticals, Costa Mesa, CA, USA) and
read for activity using liquid scintillation counting (LSC) (model LS 6000 IC; Beckman-
Coulter, Fullerton, CA). No obvious signs of seston were present in the incubation jars.
Scintillation counts per minute (CPM) from the LSC were converted into disintegrations
per minute (DPM) using an internal quench curve.
150
Samples from the experiments were filtered (Whatman EPM 2000 glass fiber) for
particulate phosphate (part-P) and polyphosphate (poly-P). One-half of each filter was
used for poly-P analysis and the other half for part-P analysis. Filters were analyzed
using spectrophotometry for poly-P (hot-water extraction) and part-P (persulfate
digestion) following standard methods (Fitzgerald and Nelson 1966, Eixler et al. 2005).
Additionally, for all experiments, water physicochemical conditions (e.g., temperature,
conductivity, oxygen concentration) were monitored using an YSI (Yellow Springs
Instrument) data Sonde (model 6600). All samples were also analyzed for chlorophyll-a
(Chl) content; samples were extracted in a mixture of dimethyl sulfoxide (DMSO) (CAS
67-68-5) and 90% acetone at -10°C for 1 h (Carrick et al. 1993). Chl concentrations were
determined using a Turner 10-AU fluorometer. Chl accumulation rates on artificial
substrata (mgChl/m2/day) were used as a proxy for microbial biofilm production
(Godwin and Carrick 2008).
Statistical analyses
For this fine resolution uptake analysis on intact biofilms, data were first
converted to µgP/L from DPM using a ratio of total DPM to enriched [P] (200 µgP/L)
assuming the distribution of tracer estimates the distribution of phosphate (Lean and
Nalewajko 1976, Hessen et al. 2012); this value was then multiplied by total sample
volume and normalized to Chl (µg) (final units= µgP/µgChl). Data were incorporated
into R (version 2.10.1) and analyzed using the ‘segmented’ package (piecewise
regression, Muggeo 2008) to estimate breakpoints or thresholds in uptake responses over
time. Additionally, data were fit using locally weighted regression (LOESS), a robust
nonparametric regression technique that fits linear regressions over localized subsets of
151
the data (Cleveland and Devlin1988). A LOESS smoothing parameter (λ) of 0.3 was
chosen to provide adequate local regression sensitivity to variation in the data while
limiting ‘over-fitting’ (Jacoby 2000). LOESS regression analyses are useful for
ecological time-series data (Trexler and Travis 1993) and plots have proven effective in
aiding selection of candidate breakpoints (e.g., MacNulty et al. 2009). Breakpoints are
defined here as distinct divergences in the time-series such that the slope of the trend
changes sign (Gao et al. 2010). Precise coordinates of maximum trough (uptake) and
maximum peak (efflux) were then derived from the LOESS regressions in Minitab
statistical package version 14.0 (Minitab, State College, PA). Specifically, breakpoints
were visually estimated (Ryan and Porth 2007) using the crosshair mode in Minitab to
show precise coordinates of trough/peak points on the data region of each graph. Max
uptake was determined by the difference between initial [P] and max trough over elapsed
time and efflux was determined by the difference between max trough and subsequent
peak (typically max) over elapsed time. Further, the duplicate tiles were analyzed using a
two-way analysis of variance (ANOVA) to test the effects of province, season, and the
interaction of province and season on P-uptake and P-efflux (log10(µgP/µgChl/min)) (n=
64). Additional analyses were performed using SPSS software version 19.0 (SPSS Inc.,
Chicago, IL, USA). A preliminary Kolmogorov-Smirnov test indicated that the data
were normally distributed; statistical significance levels were determined as p< 0.05.
4.4 RESULTS
Fine resolution into the time sensitivity of uptake was accomplished in an
experiment focused on intact biofilm P-uptake over the first few minutes of exposure to
152
new P. Phosphorus flux estimates from LOESS plots of P-uptake (log10(µgP/µgChl)) vs.
time (minute) conducted on intact biofilm assemblages from eight Pennsylvania streams
of varying productivity over four seasons (2009 - 2010) revealed distinct breakpoints in P
fluxes (Figures 4.1 - 4.4); specifically mean max uptake occurred at 1.65 and 2.60
minutes while max efflux occurred at 4.51 and 4.62 minutes in Piedmont and Plateau
provinces, respectively. These results show considerable and rapid exchange processes
occurring at early time periods (i.e., < 5 minutes), the magnitude of which seems to
diminish over longer periods (i.e., 15 - 30 minutes) (Table 4.1). The breakpoints
estimated using the R algorithm were largely in agreement with those from the LOESS
plots (t= 1.56, p= 0.125, df= 49). P flux estimates derived from the LOESS plots
regressed against stream/biofilm biochemical parameters showed that uptake and efflux
had significant inverse linear relationships with part-P (mgP/m2), and Chl accumulation
(mgChl/m2/d) for biofilm assemblages established on artificial substrata (tiles) (Figure
4.5, Table 4.2). Similarly uptake and efflux had significant inverse linear relationships
with part-P and Chl (mgChl/m2) for biofilm assemblages established on natural substrata
(rocks) (Figure 4.6, Table 4.2). Coefficients of determination (r2) were generally higher
for efflux than uptake; further, uptake and efflux were more strongly related to predictors
from biofilm assemblages established on natural substrata (rocks) compared to those on
artificial substrata (tiles).
Additionally, no differences were observed between biofilm assemblages from
duplicate tiles for uptake (F1,60= 0.150, p= 0.699) or efflux (F1,57= 0.010, p= 0.909).
Further, uptake and efflux were found to be strongly correlated (Pearson correlation
coefficient (r)= 0.848, p< 0.001); as such, a two-way multivariate analysis of variance
153
(MANOVA) was performed for uptake (log10(µgP/µgChl/min)) and efflux
(log10(µgP/µgChl/min)) by province and season (n= 64). The MANOVA showed
significant differences in uptake and efflux by province (Wilks's lambda= 0.809, F2, 50=
5.91, p= 0.005), season (Wilks's lambda= 0.636, F6, 100= 4.23, p= 0.001), and the
interaction of province and season (Wilks's lambda= 0.758, F6, 100= 2.48, p= 0.028).
Further two-way ANOVA analysis indicated a significant effect of province on both
uptake (F1, 51= 8.19, p= 0.006), averaging 0.039 ±0.016 (SE) (log10(µgP/µgChl/min)) in
the Piedmont and 0.099 ±0.030 (SE) (log10(µgP/µgChl/min)) in the Plateau province, and
efflux (F1, 51= 12.06, p= 0.001), averaging 0.015 ±0.007 (SE) (log10(µgP/µgChl/min)) in
the Piedmont and 0.065 ±0.018 (SE) (log10(µgP/µgChl/min)) in the Plateau province
(Table 4.3). The two-way ANOVA furthermore showed a significant effect of season on
uptake (F3, 51= 6.40, p= 0.001) and efflux (F3, 51= 5.13, p= 0.004). A Tukey post-hoc test
revealed that biofilm (log10) uptake rates in winter (x = 0.158) and spring (x = 0.099)
seasons were significantly higher compared to summer (x = 0.026) and fall (x = 0.003)
(Table 4.4). Similarly, the Tukey post-hoc test revealed that biofilm (log10) efflux rates
in winter (x = 0.094) and spring (x = 0.039) seasons were significantly higher compared to
fall (x = 0.025) and summer (x = 0.014) (Table 4.5). Finally, a strong interaction between
province and season for P-uptake was also observed (F3, 51= 3.20, p= 0.031), although
these sources showed no such interaction for P-efflux. Descriptive statistics for P fluxes
(µgP/µgChl/min) (n= 64) revealed that biofilm P-uptake (0.364 ±0.125 SE) and P-efflux
(0.194 ±0.060 SE) were greater in the Plateau province compared to P-uptake (0.126
±0.063 SE) and P-efflux (0.0404 ±0.0214 SE) in the Piedmont province (Table 4.6,
Figure 4.7).
154
4.5 DISCUSSION
Initial assimilatory processes
I have run a time-course experiment of the disappearance of P from water
overlying intact benthic biofilm assemblages to observe the pattern of uptake with time
(Harrison et al. 1989, Pérez-Lloréns and Niell 1995). This assay provided essential
insight into the short-term uptake kinetics of these important freshwater microbes. The
results of the experiments using P-uptake estimations from radiotracer loss in water
showed that active P-uptake is not unidirectional, such that there appears to be both
biofilm uptake and significant efflux of P back into the incubation medium over time.
The results appear to conform to a phosphate exchange process, whereby orthophosphate
is exchanged between the biofilm protoplasm and the surrounding medium. This
dynamic response of biofilms to new a P source further showed consistency across space
(stream) and time (season).
These findings are in accord with previous research that has found phosphorus is
not only assimilated rapidly by microbes, but exchanged between intracellular
compartments and the water (Goldberg et al. 1951, Rice 1953), given that cytoplasmic P
is readily exchangeable with external P (Cembella et al. 1982). In fact, Lean and
Nalewajko (1976) found that P exchange between organisms and their environment
might considerably exceed net uptake. Microbes in the biofilm saturated with P or
internally replete will not necessarily ‘turn off’ their uptake systems, but rather exchange
phosphate between the external medium and an internal pool of inorganic phosphorus
(Jansson 1988). I did find that there was a linear fraction of uptake < 3 minutes at most
sites, whereas afterwards a strong efflux was observed (> 5 minutes). Similarly, Harrison
155
et al. (1989) and Jansson (1993) found a two-phase uptake process with an initial rapid
(surge) uptake associated with a high-affinity system which fills an internal vacuole and a
secondary slower uptake associated with a low-affinity system which fuels metabolic
requirements. These kinetic patterns have been found as well in complex rooted seed-
producing plants (Pérez-Lloréns and Niell 1995). Taft et al. (1975), studying
phytoplankton uptake and release rates for inorganic phosphate in the Chesapeake Bay
estuary, found that P exchange in membrane transport was the principle process during
the initial phase of uptake (< 15 min), whereas net incorporation of phosphate was
dominant thereafter (> 15 min). This rapid surface sorption followed by slower trans-
membrane incorporation (Karlson 1989) supports my findings. That is, the LOESS plots
of P flux over time showed a similar rapid and considerable uptake occurrence and (after
an efflux period) a period of stabilized uptake. My time-course experimental results
expound on these findings and suggest that P-uptake into benthic biofilms may be a three
step kinetic process: rapid initial adsorption onto the biofilm matrix, efflux to conserve
membrane potential (see below), and slower incorporation. Fernández et al. (1997)
proposed a five compartment model (with components of intracellular and extracellular)
in explaining phosphorus fluxes in phytoplankton, underscoring the complexity of
microbial-nutrient dynamics.
As observed in my experiment, initial P-uptake was high, but fleeting (ca. 2
minutes). It is possible that the initial uptake observed in my experiment for the first few
minutes could be explained by two coupled processes in nutrient uptake: adsorption and
absorption (Short and McRoy 1984). Adsorption here is referring to the movement of P
into surface-adsorbed phosphorus pools, a fraction of which can be internalized and used
156
for growth (Sañudo-Wilhelmy et al. 2004). The biofilm assemblage may be well adapted
structurally for rapid uptake of P into surface areas. The heterogeneous architecture of
benthic biofilms with channel networks, interstitial spacings, and highly hydrated
matrixes (Sutherland 2001) allow them to function as living zones of transient storage
and biochemical processing (Battin et al. 2003). The chemical composition of microbial
cell surfaces typically generates a net negative charge at neutral pH (Sherbet 1978,
Dolowy 1984). Although phosphate (PO43-
) and most phosphate groups (mono- and
diesters) also have a negative charge above pH 3 (Beveridge and Murray 1980), the
biofilm matrix contains cell-surface-bound metal hydroxides and oxides (Sañudo-
Wilhelmy et al. 2004). Oxyanions such as phosphate can form (by means of ligand
exchange) surface complexes with these hydrous oxides (Sigg and Stumm 1981). Initial
flux/binding of phosphate ions by biofilms is then a largely adsorptive process, governed
by electrostatic interaction between negatively charged phosphate groups in the medium
and positively charged alkali metal hydroxides on the biofilm surface. The change from
the first phase, delineated by an initial rapid or surge uptake, to the second phase, with
decreased uptake, could be explained on the basis of a feedback inhibition (from filling
an inter-cellular pool) and/or a regulation of the phosphate efflux.
Efflux rates for biofilms in my study averaged 39 and 66% of mean uptake in the
Piedmont and Plateau provinces, respectively, and correspond well to previous research.
Harrison (1983b) found that that soluble phosphorus regeneration (efflux) by
microplankton populations ranged from 45 - 68% of uptake rates during a seasonal study
in a coastal embayment of Nova Scotia. Similarly, Wen et al. (1997) showed that about
half of the PO4 taken up was excreted in freshwater algae. Moreover, higher plant
157
literature has shown too that P influx is associated with efflux (Elliot et al. 1984, Bieleski
and Läuchli 1992). It has been observed that ammonium influx must be coupled with an
active cation efflux in order to maintain membrane potential (Parslow et al. 1984).
Uptake of phosphate, as a polyatomic anion, may instigate similar effects as observed
with ammonium, such that a portion of the PO43-
incorporated into a cell must be
eliminated in order to maintain a stable electrical potential between the cellular interior
and the exterior. That said, it could be more energetically favorable for microbes to
excrete phosphate rather than activate and deactivate the entire phosphate transport
system at short intervals (see Jansson 1988). Phosphorus assimilation by Escherichia
coli, for instance, requires regulation of at least 31 genes (Wanner 1993). Efflux of P
possibly allows microbes to assimilate P and efflux excess while maintaining the same
transport systems and membrane potential. Nevertheless, physiological explanations for
high P-efflux when P-uptake and demand are also high remain uncertain (Nalewajko and
Lean 1980, Whitton 1992).
Non-linear time-courses observed in other nutrient uptake experiments have been
attributed to substrate exhaustion (e.g., Fisher et al. 1981). Moreover, Goldman et al.
(1981) suggested that substrate depletion may be important in explaining non-linear
nutrient uptake kinetics in natural microplankton populations. However, this
experimental artifact of isotope availability does not provide an adequate explanation for
my results here. According to computations, on average approximately 82% of the P
available was taken up during any of the time-course experiments. Therefore, the
evidence suggests that active efflux by the biofilm assemblages contributed to the time-
course patterns observed rather than methodological artifacts. That said, the rate at which
158
biofilm assemblages actively process (uptake/efflux) P may be somewhat disparate,
thereby differentially affecting the time-course patterns. For example, the LOESS plots
of P fluxes over time showed large variability among replicate experimental units (tiles).
Such inter-experimental variation is not uncommon (especially given the highly
heterogeneous nature of stream systems), and may be attributable to diverse taxonomic
composition, even in adjacent samples. For instance, Wen et al. (1997) found substantial
variability in the measurement of uptake kinetics between axenic freshwater algal
cultures. My study employed natural biofilm assemblages that likely differed, albeit
moderately, in community composition which may have contributed to the observed
variation in P fluxes. While biofilm samples for this experiment were preserved (1%
formalin) for future taxonomic analysis, examination was beyond the scope of the current
objectives, and thus such hypotheses remain to be tested.
My selected time-courses, with 11 sampling periods prior to 30 minutes (five
periods prior to five minutes) is in agreement with past phytoplankton research which
suggested that short-term nutrient uptake experiments be conducted in the range of
seconds to minutes to examine important physiological processes (Goldman and Glibert
1982); this indicates nutrient processing is a near instantaneous physiological dynamic
and further suggests that experimental time periods scaled to hours or longer obscure
such essential short-term responses (Harrison 1983a). Nonetheless, the literature remains
replete with rate measurements from nutrient uptake experiments of varying and often
arbitrary durations (Goldman et al. 1981). For instance, “short-term” nutrient uptake in
the literature has varied from ≤ 3 hours (Pérez-Lloréns and Niell 1995), to 45 - 120
minutes (O’Brien and Wheeler 1987), to < 1 minute (Parslow et al. 1984). This trend has
159
continued into current literature where “very fast” uptake kinetics of Synedropsis
(Division Ochrophyta, Order Fragilariales) was just recently estimated as sampling after
10 - 100 minutes (Hessen et al. 2012). My research here has shown that mean max
uptake occurred under three minutes while max efflux occurred under five minutes in
both Piedmont and Plateau provinces over four seasons. There are thus considerable P
exchange processes occurring at early time periods (i.e., < 5 minutes) in these biofilm
assemblages, the majority of which is being missed during many “short-term” nutrient
uptake experiments. Such experiments thus reflect net uptake, or the result of several
interacting and counteracting processes involved in microbial phosphate assimilation
(Jansson 1988). While these experiments are certainly valuable in their own right and
likely more applicable to total water column uptake rates, they are of limited value for
discerning the significance of initial assimilatory processes to nutrient uptake dynamics
(Wheeler et al. 1982). True “short-term” experiments and time-course incubations are
important to understand the physiological characteristics of nutrient uptake (Goldman et
al. 1981). Investigations into rapid initial uptake processes could be of ecological
importance, particularly if natural nutrient perturbations are on the time scale of minutes
(Collos 1983). In addition, initial uptake (≤ 5 minutes) was a significant linear predictor
of assimilatory (stable) P-uptake (30 minutes) (Figure 4.8; r2= 0.811, slope= 0.0495, p<
0.001), suggesting that indeed the magnitude of short-term processes affect longer term
kinetics (Dortch 1990). Moreover, Cembella et al. (1984) found that PO4 Vmax values
vary over six orders of magnitude; as kinetic parameters of microbial uptake systems can
fluctuate over experimental conditions (e.g., time), uniformity of “short-term” nutrient
uptake research may likely provide greater consistency of determined kinetic parameters
160
(Wagner and Falkner 2001). For instance, the average maximum uptake rate for benthic
biofilms was 0.126 and 0.364 µgP/µgChl/min for the Piedmont and Plateau provinces,
respectively. These values, while within the benthic range specified in Price and Carrick
(2011), are on the high end, suggesting that short-term uptake estimates yield generally
higher rates (also in agreement with their findings) and further demonstrate the effect of
experimental sample time on kinetics.
While the occurrence of non-linear nutrient disappearance over time has been
reported before in phytoplankton (Goldman et al. 1981, Wheeler et al. 1982, Parslow et
al. 1984, Cochlan and Harrison 1991), prokaryotic nannoplankton (Lehman and Sandgren
1982), seaweeds (Thomas and Harrison 1987), and submerged aquatic angiosperms
(Pérez-Lloréns and Niell 1995), much of it has focused on short-term N dynamics with
little regard for P and little research has been devoted to freshwater benthic biofilms. As
such, this work illuminates the short-term uptake potential for benthic stream
assemblages and highlights the need for time-dependent assays to ascertain community
physiological responses to added nutrients. Overall, the data demonstrate that non-
linearity in uptake is a common occurrence, detectable only in rapid time-course studies,
and that excretory phenomena are actively occurring together with uptake in a dynamic
coupling that may be undetected with longer time samplings.
Nutrient legacies
Regressions of P-uptake and P-efflux vs. part-P estimated from biofilms on
natural (rock) substrata were stronger compared to those biofilms on artificial (tile)
substrata. The biofilms on the natural substrates would comparatively have developed
for a longer period of time and thus subjected to nutrient legacies (cf. Verspoor et al.
161
2010). These experimental results indicate that while stream biofilms may act as a
reservoir for dissolved P, their uptake and efflux magnitude is rapidly reduced in
response to prolonged high P inputs (legacy), further suggesting that downstream
transport of nutrients could be enhanced in streams that are already enriched (see Chapter
3). The strong positive correlation between part-P and Chl accumulation (Pearson
correlation coefficient (r)= 0.812, p< 0.001) further suggests that as productivity
increases and biofilms accumulate P, the capacity of the population to assimilate new P
declines. Research has shown that a fraction of cellular stored P (polyphosphate A)
serves as a noncompetitive inhibitor on the uptake system (Rhee 1973). My data reflect
this finding, such that those biofilms containing high concentrations of part-P showed a
much lowered capacity to take-up new P. The LOESS plots (Figures 4.5, 4.6) further
support these data and indicate a sharp decline (threshold) in uptake/efflux magnitude as
biomass and productivity increase. Further, the breakpoints estimated by the LOESS plot
for uptake and efflux correspond along the predictor axes, suggesting that as uptake
diminishes efflux, is moderated as well.
Despite my effort for quantitative rigor in this experiment, precise internal P
exchange(s) could not be estimated given that a single radiotracer was used, and thus
explicit quantification of the [P] being effluxed from microbes within the biofilm was not
possible. I therefore utilized my chemical analysis of poly-P (intercellular phosphate
stores (Kromkamp 1987)), as an estimation of internal [P]. I then regressed poly-P
against P-efflux in order to determine the effect of internally stored P on efflux processes.
Observations with DFFITS values greater than 2√(k/n), where k is the number of
predictors (including constant), were removed from the analysis, following Belsley et al.
162
(1980). Overall, P-efflux yielded a significant negative regression with poly-P (slope= -
0.0195 (±0.004 SE), p< 0.001, r2= 0.332, Appendix Q) which suggests that as poly-P
decreases (or internal stores of P are reduced), these smaller internal P stores can be more
readily and swiftly exchanged with the external medium. Further, the higher efflux rates
occurring in biofilm assemblages with low polyphosphate content could be attributable to
elevated mobilization/utilization of stored P as an energy source by microbes in the
biofilm; some research has found evidence that P release during the utilization of
polyphosphates by benthic microorganism can contribute to enhanced P fluxes
(Sannigrahi and Ingall 2005).
For biofilms in the Plateau province, efflux comprised approximately 66% of
uptake, whereas in the Piedmont province efflux averaged only 39%. This finding is in
accord with the expected difference in electrochemical potential of cell membranes
between biofilms located in the two provinces. For instance, (log10) poly-P was
significantly (F1, 49= 14.65, p< 0.001) lower in biofilms situated in Plateau streams (x =
0.68 ±0.51 (SD) mgP/m2) compared to those in Piedmont streams (x = 1.26 ±0.56 (SD)
mgP/m2). Further, TP (mg/L) concentrations were more than 3-fold lower in the Plateau
compared with the Piedmont streams (see Carrick et al. 2009). Thus, once biofilms were
returned to the laboratory and subjected to a new elevated, yet realistic P source (33
P
carrier), P ions would face a more unfavorable electrochemical gradient for uptake in
those biofilms situated in the Plateau province, resulting in more considerable leakage or
efflux of acquired P (Elliott et al. 1984, Britto and Kronzucker 2006).
The differential in efflux rates between biofilms in the Plateau versus Piedmont
provinces might further be explained through mineral composition analyses. For
163
instance, both calcium (Ca) and zinc (Zn) have been found to influence uptake processes
in higher plants. Ca sufficiency can enhance P-uptake in mung beans (Vigna radiata)
(Tanada 1955), and Zn deficiencies can up-regulate expression of genes encoding high-
affinity P transporters in barley (Hordeum vulgare L. cv Weeah) (Huang et al. 2000).
Biofilms in the Piedmont province had significantly (F1, 57= 6.95, p= 0.011) higher Ca
concentrations, averaging 120.5 ±189.3 (SD) g/m2, compared to biofilms in the Plateau
province (x = 5.5 ±3.4 (SD) g/m2) (see PADEP 2005, Carrick et al. 2009). Moreover,
biofilms in the Piedmont province had moderately (p> 0.05) lower Zn concentrations,
averaging 0.177 ±0.141 (SD) g/m2, compared to biofilms in the Plateau province (x =
0.230 ±0.156 (SD) g/m2). Therefore, higher Ca and lower Zn concentrations in biofilms
located in the Piedmont province might facilitate greater retention of P, thereby reducing
the magnitude of efflux events.
Spatial-temporal effects
While both province and season were significant factors in the ANOVAs for P-
uptake and P-efflux, the F-ratio for province was greater, suggesting that variation
between the provinces more strongly overrides variation within each and that spatial
effects may more effectively influence differences in uptake/efflux. This is in accord
with past studies demonstrating the strong variability of uptake across stream systems
(e.g., Orr et al. 2006). For example, Ensign et al. (2006) reported over a 7-fold range (1.0
to 7.2 mm min-1
) in PO43-
retention between adjacent second- and third-order streams,
draining into the South River estuary in North Carolina. Such spatial variability in
nutrient retention has thus been linked with even moderately heterogeneous
biogeochemistry of streambeds. Certainly the Pennsylvania Plateau to Piedmont
164
transition I tested against here reflects the prominent geology of the provinces as well as
a strong gradient of environmental conditions (see Carrick et al. 2009). Nonetheless,
Martí and Sabater (1996) suggested that differences in nutrient retention efficiencies
across streams might be related to differences in nutrient availability. This was also the
case in the streams I selected here, such that TP concentrations were more than 3-fold
higher in the Piedmont compared with the Plateau streams (see Carrick et al. 2009).
Despite the strength and influence of spatial effects, season remained a significant factor
in my study and revealed interesting trends in the P-uptake and P-efflux over time. I
found that uptake and efflux were particularly elevated in winter; other research has
found as well that microbes in winter can exhibit strong assimilation capacities— PO43-
uptake was highest in the Pamlico estuary in late winter (Kuenzler et al. 1979). Further,
inorganic nutrients were likely most limiting during winter and spring in my experiment,
as suggested by stream conductivity data (e.g., Sharpe 2007), and therefore, biofilm
communities were likely more physiologically poised to respond to a new P source.
Microbial adaptive responses to altered phosphate concentrations can initiate new uptake
potentials (Aubriot et al. 2011). For instance, under nutrient stressed (limiting)
conditions, microbes are capable of inducing high-affinity uptake systems (Mann 1995)
which exhibits strong anion selectivity and contains high affinity P-binding proteins
(Dignum et al. 2005). Such systems have been found in higher plants as well during
periods of nutrient stress, such that high-affinity inorganic P-uptake systems are induced
in cluster roots of P-deficient plants (Vance et al. 2003). Further, high uptake and
retention of nutrients that I recorded in my experiment during spring has been observed
previously. Mulholland (1992) found that uptake by stream autotrophs contributes
165
significantly to nutrient retention in Walker Branch (east Tennessee, USA), during
periods of high ambient light prior to leaf emergence (e.g., spring). My spring
experimental period occurred before complete leaf-out at all sites (personal observation),
and thus biofilms were growing during conditions (high light/low leaf density) favorable
for uptake.
4.6 ACKNOWLEDGMENTS
I thank J. Carlson and J. Lynch for helpful discussions and comments on a
previous version of this manuscript. Funding for this research was provided to H.J.C. by
the Pennsylvania Department of Environmental Protection (Grant No. 4100034506).
4.7 LITERATURE CITED
Aubriot, L., Bonilla, S., and Falkner, G. 2011. Adaptive phosphate uptake behaviour of
phytoplankton to environmental phosphate fluctuations. FEMS Microbiology
Ecology 77:1-16.
Battin, T.J., Kaplan, L.A., Newbold, J.D., and Hansen, C.M.E. 2003. Contributions of
microbial biofilms to ecosystem processes in stream mesocosms. Nature 426:439-
442.
Belsley, D., Kuh, E., and Welsch, R. 1980. Regressions diagnostics: Identifying
influential data and sources of collinearity. New York: John Wiley and Sons.
Beveridge, T.J. and Murray, R.G.E. 1980. Sites of metal deposition in the cell wall of
Bacillus subtilis. Journal of Bacteriology 141:876-887.
166
Bieleski, R.L. 1973. Phosphate pools, phosphate transport, and phosphate availability.
Annual Review of Plant Physiology 24:225-252.
Bieleski, R.L. and Läuchli, A. 1992. Phosphate uptake, efflux and deficiency in the water
fern Azolla. Plant Cell and Environment 15:665-673.
Britto, D.T. and Kronzucker, H.J. 2006. Futile cycling at the plasma membrane: A
hallmark of low-affinity nutrient transport. Trends in Plant Science 11:529-534.
Byron, E.R. and Goldman, C.R. 1989. Land-use and water quality in tributary streams of
Lake Tahoe, California-Nevada. Journal of Environmental Quality 18:84-88.
Cardinale, B.J., Palmer, M.A., Swan, C.M., Brooks, S., and Poff, N.L. 2002. The
influence of substrate heterogeneity on biofilm metabolism in a stream ecosystem.
Ecology 83:412-422.
Carrick, H.J. and Price, K.J. 2011. Determining variation in TMDL reduction criteria.
Final Report, Pennsylvania Department of Environmental Protection, 84 p.
Carrick, H.J., Aldridge, F.J., and Schelske, C.L. 1993. Wind influences phytoplankton
biomass and composition in a shallow, productive lake. Limnology and
Oceanography 38:1179-1192.
Carrick, H.J., Scanlan, A., and Wagner, R. 2009. Use of periphyton to estimate TMDL
end-points. Final Report, Pennsylvania Department of Environmental Protection,
57 p.
Cembella, A.D., Anita, N.J., and Harrison, P.J. 1984. The utilization of inorganic and
organic phosphorous compounds as nutrients by eukaryotic microalgae: A
multidisciplinary perspective: Part 1. CRC Critical Reviews in Microbiology
10:317-391.
167
Cleveland, W.S. and Devlin, S.J. 1988. Locally-weighted regression: An approach to
regression analysis by local fitting. Journal of the American Statistical
Association 83:596-610.
Cochlan, W.P. and Harrison, P.J. 1991. Uptake of nitrate, ammonium and urea by
nitrogen-starved cultures of Micromonas pusilla (Prasinophyceae): Transient
responses. Journal of Phycology 27:673-679.
Collos, Y. 1983. Transient situations in nitrate assimilation by marine diatoms. 4. Non-
linear phenomena and the estimation of the maximum uptake rate. Journal of
Plankton Research 5:677-691.
Confer, J.L. 1972. Interrelations among plankton, attached algae, and the phosphorus
cycle in artificial open systems. Ecological Monographs 42:1-23.
Correll, D.L. 1998. The role of phosphorus in the eutrophication of receiving waters: A
review. Journal of Environmental Quality 27:261-266.
Dignum, M., Matthijs, H.C.P., Pel, R., Lannbroek, H.J., and Mur, L.C. 2005. Nutrient
limitation of freshwater cyanobacteria. In: J. Huisman, H.C.P. Matthijs, and P.M.
Visser (Ed.), Harmful cyanobacteria. Springer, Dordrecht, The Netherlands, p. 65-
68.
Dolowy, K. 1984. Bioelectrochemistry of cell surfaces. Progress in Surface Science
15:245-368.
Dortch, Q. 1990. The interaction between ammonium and nitrate uptake in
phytoplankton. Marine Ecology Progress Series 61:183-201.
168
Eixler, S., Selig, U., and Karsten, U. 2005. Extraction and detection methods for
polyphosphate storage in autotrophic planktonic organisms. Hydrobiologia
533:135-143.
Elliott, G.C., Lynch, J., and Läuchli, A. 1984. Influx and efflux of P in roots of intact
maize plants. Plant Physiology 76:336-341.
Ensign, S.H., McMillan, S.K., Thompson, S.P., and Piehler, M.E. 2006. Nitrogen and
phosphorus attenuation within the stream network of a coastal, agricultural
watershed. Journal of Environmental Quality 35:1237-1247.
Fernández, J.A., Clavero, V., Villalobos, J.A., and Niell, F.X. 1997. A non-linear model
of phosphorous flux in the phytoplankton of a temperate eutrophic reservoir.
Hydrobiologia 344:205-214.
Fisher, T.R., Carlson, P.R., and Barber, R.T. 1981. Some problems in the interpretation
of ammonium uptake kinetics. Marine Biology Letters 2:33-44.
Fitzgerald, G.P. and Nelson, T.C. 1966. Extraction and enzymatic analyses for limiting or
surplus phosphorus in algae. Journal of Phycology 2:32-37.
Flynn, K.J. 1998. Estimation of kinetic parameters for the transport of nitrate and
ammonium into marine phytoplankton. Marine Ecology Progress Series 169:13-
28.
Freeman, C., Chapman, P.J., Gilman, K., Lock, M.A., Reynolds, B., and Wheater, H.S.
1995. Ion exchange mechanisms and the entrapment of nutrients by river
biofilms. Hydrobiologia 297:61-65.
169
Gao, B., McKee, K., Gargiulo, O., and Graham, W. 2011. Statistical evaluation of
hydrologic data in northeastern Florida and southern Georgia. Final Report St.
Johns River Water Management District Report SJ2010-SP11.
Godwin, C.M. and Carrick, H.J. 2008. Spatio-temporal variation of periphyton biomass
and accumulation in a temperate spring-fed stream. Aquatic Ecology 42:583-595.
Goldman, J.C. and Glibert, P.M. 1982. Comparative rapid ammonium uptake by four
species of marine phytoplankton. Limnology and Oceanography 27:814-827.
Goldman, J.C., Taylor, C.D., and Glibert, P.M. 1981. Nonlinear time-course uptake of
carbon and ammonium by marine phytoplankton. Marine Ecology Progress Series
6:137-148.
Halterman, S.G. and Toetz, D.W. 1984. Kinetics of nitrate uptake by freshwater algae.
Hydrobiology 114:209-214.
Harrison, W.G. 1983a. The time-course of uptake of inorganic and organic nitrogen
compounds by phytoplankton from the Eastern Canadian Arctic: A comparison
with temperate and tropical populations. Limnology and Oceanography 28:1231-
1237.
Harrison, W.G. 1983b. Uptake and recycling of soluble reactive phosphorus by marine
microplankton. Marine Ecology Progress Series 10:127-135.
Harrison, W.G. and Harris, L.R. 1986. Isotope-dilution and its effects on measurements
of nitrogen and phosphorus uptake by oceanic microplankton. Marine Ecology
Progress Series 27:253-261.
170
Hessen, D.O., Frigstad, H., Færøvig, P.J., Wojewodzic, M.W., and Leu, E. 2012. UV
radiation and its effects on P-uptake in arctic diatoms. Journal of Experimental
Marine Biology and Ecology 411:45-51.
Huang, C., Barker, S.J., Langridge, P., Smith, F.W., and Graham, R.D. 2000. Zinc
deficiency up-regulates expression of high-affinity phosphate transporter genes in
both phosphate-sufficient and -deficient barley roots. Plant Physiology 124:415-
422.
Jacoby, W.G. 2000. Loess: A nonparametric, graphical tool for depicting relationships
between variables. Electoral Studies 19:577-613.
Jansson, M. 1988. Phosphate uptake and utilization by bacteria and algae. Hydrobiologia
170:25-28.
Jansson, M. 1993. Uptake, exchange and excretion of orthophosphate in phosphate-
starved Scenedesmus quadrieauda and Pseudomonas K7. Limnology and
Oceanography 38:1162-1178.
Jöbgen, A.M., Palm, A., and Melkonian, M. 2004. Phosphorus removal from eutrophic
lakes using periphyton on submerged artificial substrata. Hydrobiologia 528:123-
142.
Karl, D. 2000. Phosphorus, the staff of life. Nature 406:31-33.
Karlson, B. 1989. Seasonal phosphate uptake by size-fractionated plankton in the
Skagerrak. Journal of Experimental Marine Biology and Ecology 127:141-154.
Körner, S. and Vermaat, J.E. 1998. The relative importance of Lemna gibba L., bacteria
and algae for the nitrogen and phosphorus removal in duckweed-covered
domestic wastewater. Water Research 32:3651-3661.
171
Kromkamp, J. 1987. Formation and functional significance of storage products in
cyanobacteria. New Zealand Journal of Marine and Freshwater Research 21:457-
465.
Kuenzler, E.J., Stanley, D.W., and Koenings, J.P. 1979. Nutrient kinetics of
phytoplankton in the Pamlico River, N.C. North Carolina Water Resources
Research Institute Report No. 139, 163 p.
Laspidou, C.S. and Rittmann, B.E. 2002. A unified theory for extracellular polymeric
substances, soluble microbial products, and active and inert biomass. Water
Research 36:2711-2720.
Lean, D.R.S. and Nalewajko, C. 1976. Phosphate exchange and organic phosphorus
excretion by freshwater algae. Journal of the Fisheries Research Board of Canada
33:1312-1323.
Lehman, J.T. and Sandgren, C.D. 1982. Phosphorus dynamics of the procaryotic
nannoplankton in a Michigan lake. Limnology and Oceanography 27:828-838.
Lock, M.A., Wallace, R.R., Costerton, J.W., Ventullo, R.M., and Charlton, S.E. 1984.
River epilithon (biofilm): Towards a structural functional model. Oikos 42:10-22.
MacNulty, D.R., Smith, D.W., Mech, L.D., and Eberly, L.E. 2009. Body size and
predatory performance in wolves: Is bigger better? Journal of Animal Ecology
78:532-539.
Mann, N.H. 1995. How do cells express nutrient limitation at the molecular level? In: I.
Joint (Ed.), Molecular ecology of aquatic microbes. NATO ASI Series, Vol. G 38.
Springer-Verlag, Berlin, Germany.
172
Martí, E. and Sabater, F. 1996. High variability in temporal and spatial nutrient retention
in Mediterranean streams. Ecology 77:854-869.
Muggeo, V.M.R. 2008. Segmented: An R package to fit regression models with broken-
line relationships. R News 8:20-25.
Mulholland, P.J. 1992. Regulation of nutrient concentrations in a temperate forest stream:
roles of upland, riparian, and instream processes. Limnology and Oceanography
37:1512-1526.
Nalewajko, C. and Lean, D.R.S. 1980. Phosphorus. In: I. Morris (Ed.), The physiological
ecology of phytoplankton. Blackwell Scientific Publication, Oxford, p. 235-258.
Neu, T.R. and Lawrence, J.R. 1997. Development and structure of microbial biofilms in
river water studied by confocal laser scanning microscopy. FEMS Microbiology
Ecology 24:11-25.
O’Brien, M.G. and Wheeler, P.A. 1987. Short term uptake of nutrients by Enteromorpha
prolifera (Chlorophyceae). Journal of Phycology 23:547-556.
Orr, C.H., Rogers, K.L., and Stanley, E.H. 2006. Channel morphology and P uptake
following removal of a small dam. Journal of the North American Benthological
Society 25:556-568.
PADEP (Pennsylvania Department of Environmental Protection). 2005. Instream
comprehensive evaluation surveys. Document ID: 391-3200-001. Bureau of
Water Standards and Facility Regulation. Harrisburg, Pennsylvania, 64 p.
Parslow, J.S., Harrison, P.J., and Thompson, P.A. 1984. Development of rapid
ammonium uptake during starvation of batch and chemostat cultures of the
marine diatom Thalassiosira pseudonana. Marine Biology 83:43-50.
173
Pérez-Lloréns, J.L. and Niell, F.X. 1995. Short-term phosphate uptake kinetics in Zostera
noltii Hornem: A comparison between excised leaves and sediment-rooted plants.
Hydrobiologia 297:17-27.
Price, K.J. and Carrick, H.J. 2011. Meta-analytical approach to explain variation in
microbial phosphorus uptake rates in aquatic ecosystems. Aquatic Microbial
Ecology 65:89-102.
Puckett, L.J. 1995. Identifying the major sources of nutrient water pollution.
Environmental Science and Technology 29:408-14.
Reuter, J.E., Stanford, L.L., and Goldman, C.R. 1986. Inorganic nitrogen uptake by
epilithic periphyton in a N-deficient lake. Limnology and Oceanography 31:149-
160.
Riedel, G.F., Sanders, J.G., and Gilmour, C.C. 1996. Uptake, transformation, and impact
of selenium in freshwater phytoplankton and bacterioplankton communities.
Aquatic Microbial Ecology 11:43-51.
Riegman, R. and Mur, L.R. 1984. Regulation of phosphate uptake kinetics in Oscillatoria
agardhii. Archives of Microbiology 139:28-32.
Ryan, S.E. and Porth, L.S. 2007. A tutorial on the piecewise regression approach applied
to bedload transport data. General Technical Report RMRS-GTR-189. United
States Department of Agriculture, Forest Service, Rocky Mountain Research
Station, Fort Collins, CO, USA, 41 p.
Sannigrahi, P. and Ingall, E. 2005. Polyphosphate as a source of enhanced P fluxes in
marine sediments overlain by anoxic waters: Evidence from 31
P NMR.
Geochemical Transactions 6:52-59.
174
Sañudo-Wilhelmy, S.A., Tovar-Sanchez, A., Fu, F-X, Capone, D.G., Carpenter, E.J., and
Hutchins, D.A. 2004. The impact of surface-adsorbed phosphorus on
phytoplankton Redfield stoichiometry. Nature 432:897-901.
Schindler, D.W., Hecky, R.E., Findlay, D.L., Stainton, M.P., Parker, B.R., Paterson, M.J.,
Beaty, K.G., Lyng, M., and Kaslan, S.E.M. 2008. Eutrophication of lakes cannot
be controlled by reducing nitrogen input: Results of a 37-year whole-ecosystem
experiment. Proceedings of the National Academy of Sciences 105:11254-11258.
Sharpe, A. 2007. Report on the investigation of low dissolved oxygen levels in the
Annapolis River estuary. Annapolis Estuary Dissolved Oxygen Report, 17 p.
Sherbet, G.V. 1978. The biophysical characterization of the cell surface. Academic Press,
New York, N.Y.
Short, F.T. and McRoy, C.P. 1984. Nitrogen uptake by leaves and roots of the seagrass
Zostera marina L. Botanica Marina 17:547-555.
Sigg, L. and Stumm, W. 1981. The interactions of anions and weak acids with the
hydrous goethite (α-FeOOH) surface. Colloids and Surfaces 2:101-107.
Spears, B.M., Carvalho, L., Perkins, R., Kirika, A., and Paterson, D.M. 2007. Sediment
phosphorus cycling in a large shallow lake: Spatio-temporal variation in
phosphorus pools and release. Hydrobiologia 584:37-48.
Sutherland, I. 2001. Biofilm exopolysaccharides: A strong and sticky framework.
Microbiology 147:3-9.
Taft, J.L., Taylor, W.R., and McCarthy, J.J. 1975. Uptake and release of phosphorus by
phytoplankton in the Chesapeake Bay estuary, USA. Marine Biology 33:21-32.
175
Tanada, T. 1955. Effects of ultraviolet radiation and calcium and their interaction on salt
absorption by excised mung bean roots. Plant Physiology 30:221-225.
Tarapchak, S.J. and Herche, L.R. 1986. Phosphate uptake by microorganisms in lake
water: Deviation from simple Michaelis-Menten kinetics. Canadian Journal of
Fisheries and Aquatic Sciences 43:319-328.
Thomas, T.E. and Harrison, P.J. 1987. Rapid ammonium uptake and nitrogen interactions
in five intertidal seaweeds grown under field conditions. Journal of Experimental
Marine Biology and Ecology 107:1-8.
Trexler, J.C. and Travis, J. 1993. Nontraditional regression analysis. Ecology 74:1629-
1637.
Vance, C.P., Udhe-Stone, C., and Allan, D.L. 2003. Phosphorus acquisition and use:
Critical adaptations by plants for securing a nonrenewable resource. New
Phytologist 157:423-447.
Verspoor, J.J., Braun, D.C., and Reynolds, J.D. 2010. Quantitative links between Pacific
salmon and stream periphyton. Ecosystems 13:1020-1034.
Vollenweider, R.A. 1968. Scientific fundamentals of the eutrophication of lakes and
flowing waters, with particular reference to nitrogen and phosphorus as factors in
eutrophication. Technical report DAS/CSI/68.27. Environmental Directorate,
Organization for Economic Cooperation and Development (OECD), Paris, 154 p.
Wagner, F. and Falkner, G. 2001. Phosphate limitation. In: L.C. Raim and J.P. Gaur
(Ed.), Algal adaptation to environmental stresses-physiological, biochemical and
molecular mechanisms. Springer-Verlag, New York, p. 65-110.
176
Wanner, B.L. 1993. Gene regulation by phosphate in enteric bacteria. Journal of Cellular
Biochemistry 51:47-54.
Weiden, P.L., Epstein, W., and Schultz, S.G. 1967. Cation transport in Escherichia coli.
VII. Potassium requirement for phosphate uptake. Journal of General Physiology
50:1641-1659.
Wen, Y.H., Vézina, A., and Peters, R.H. 1997. Allometric scaling of compartmental
fluxes of phosphorus in freshwater algae. Limnology and Oceanography 42:45-
56.
Wheeler, P.A., Glibert, P.M., and McCarthy, J.J. 1982. Ammonium uptake and
incorporation by Chesapeake Bay phytoplankton: Short term uptake kinetics.
Limnology and Oceanography 27:1113-1128.
Whitton, B.A. 1992. Diversity, ecology, and taxonomy of the cyanobacteria. In: N.H.
Mann and N.G. Carr (Ed.), Photosynthetic prokaryotes. Plenum Press, New York,
p. 1-51.
Yao, B., Xi, B., Hu, C., Huo, S., Su, J., and Liu, H. 2011. A model and experimental
study of phosphate uptake kinetics in algae: Considering surface adsorption and
P-stress. Journal of Environmental Sciences 23:189-198.
177
Table 4.1. P-uptake and P-efflux estimates from LOESS plots of P flux (log10(µgP/µgChl)) vs. time
(minute) and breakpoint estimates from the ‘segmented’ package in R for experiments conducted on
intact biofilm assemblages from eight Pennsylvania streams of varying productivity over four
seasons 2009 - 2010. Note * indicates failure of LOESS to detect distinct divergences in the time-
series or inability of the R algorithm to converge on a breakpoint.
Season Province Stream Time Max
Uptake
(min)
Time Max
Efflux
(min)
Segmented
Breakpoint
(min ±SE)
Uptake Rate
(µgP/µgChl/min)
Efflux Rate
(µgP/µgChl/min)
Fall Plateau Cowanesque 5.15 10.06 1.61 ±2.09 0.0021 0.0015
Fall Plateau East Hickory 2.04 5.17 * 0.1259 0.1131
Fall Plateau Tionesta 2.14 5.04 2.44 ±4.36 -0.1024 0.1780
Fall Plateau Tunkhannock 1.21 1.88 2.26 ±2.58 0.0089 0.0036
Fall Piedmont Cooks 1.91 9.96 * 0.0142 0.0032
Fall Piedmont Penns 0.90 2.02 5.29 ±1.53 0.0052 0.0063
Fall Piedmont Red Clay 2.04 5.17 * 0.0191 0.0094
Fall Piedmont Spring 2.10 9.93 * 0.0138 0.0046
Winter Plateau Cowanesque 0.99 2.05 6.19 ±2.11 0.0495 0.0876
Winter Plateau East Hickory * * * * *
Winter Plateau Tionesta 5.14 10.05 2.88 ±1.31 0.3851 0.1275
Winter Plateau Tunkhannock 1.94 5.14 * 0.9763 0.5189
Winter Piedmont Cooks 2.15 5.14 6.38 ±2.80 1.7616 0.2480
Winter Piedmont Penns 0.99 2.27 6.97 ±9.03 0.0049 0.0037
Winter Piedmont Red Clay 2.05 5.26 1.26 ±1.10 0.0278 0.0080
Winter Piedmont Spring 0.87 2.15 * 0.0606 0.0404
Spring Plateau Cowanesque 4.94 * 4.91 ±6.92 0.0044 *
Spring Plateau East Hickory 2.28 4.74 1.60 ±0.60 0.5901 0.1937
Spring Plateau Tionesta 1.04 1.93 * 1.9093 0.7379
Spring Plateau Tunkhannock 1.00 2.12 * 0.0181 0.0063
Spring Piedmont Cooks 4.84 9.53 4.22 ±6.35 0.0283 0.0080
Spring Piedmont Penns 0.92 2.04 2.57 ±2.66 0.0285 0.0065
Spring Piedmont Red Clay 0.94 4.96 * 0.1663 0.0333
Spring Piedmont Spring 1.00 1.23 5.71 ±10.8 0.0121 0.0101
Summer Plateau Cowanesque 0.99 2.27 2.44 ±0.46 0.0417 0.0058
Summer Plateau East Hickory 3.97 5.04 5.69 ±2.67 0.0622 0.0497
Summer Plateau Tionesta 2.15 4.07 4.01 ±1.92 0.0355 0.0133
Summer Plateau Tunkhannock 3.86 5.14 5.48 ±2.53 0.0392 0.0214
Summer Piedmont Cooks 1.08 2.36 2.11 ±1.30 0.0877 0.0404
Summer Piedmont Penns 1.63 3.98 1.01 ±0.07 0.0933 0.0340
Summer Piedmont Red Clay 2.05 4.19 1.52 ±0.67 0.0078 0.0022
Summer Piedmont Spring 0.87 1.94 1.15 ±0.70 0.0940 0.0252
178
Table 4.2. Linear regression statistics for (log10) P-uptake (µgP/µgChl/min) and (log10) P-efflux
(µgP/µgChl/min) vs. (log10) part-P (mgP/m2) and (log10) Chl accumulation (mgChl/m
2/d) for biofilm
assemblages established on tiles, and (log10) part-P (mgP/m2) and (log10) Chl (mgChl/m
2) for biofilm
assemblages established on natural substrata (rocks) from eight Pennsylvania streams of varying
productivity over four seasons 2009 - 2010.
Response Predictor y-
intercept
Std.
Error y-
intercept
Slope Std.
Error
Slope
r2 p
(log10)
Uptake
(log10) Part-P
(tile) 0.183 0.046 -0.097 0.032 0.254 0.005
(log10) Chl
Accumulation
(tile)
0.117 0.027 -0.144 0.046 0.261 0.004
(log10)
Efflux
(log10) Part-P
(tile) 0.066 0.011 -0.034 0.008 0.443 0.000
(log10) Chl
Accumulation
(tile)
0.041 0.007 -0.047 0.011 0.399 0.000
(log10)
Uptake
(log10) Part-P
(rock) 0.202 0.038 -0.088 0.020 0.407 0.000
(log10) Chl
(rock) 0.245 0.050 -0.095 0.023 0.371 0.000
(log10)
Efflux
(log10) Part-P
(rock) 0.072 0.009 -0.030 0.005 0.586 0.000
(log10) Chl
(rock) 0.072 0.014 -0.026 0.006 0.376 0.000
179
Table 4.3. Two-way ANOVA results for P-uptake (log10(µgP/µgChl/min)) and P-efflux
(log10(µgP/µgChl/min)) by season and province for replicate measurements on intact biofilms across
eight Pennsylvania streams of varying productivity over four seasons 2009 - 2010.
Source Dependent
Variable
Type III Sum
of Squares
df Mean
Square
F p
Province (log10) Uptake 0.102 1 0.102 8.185 0.006
(log10) Efflux 0.051 1 0.051 12.060 0.001
Season (log10) Uptake 0.240 3 0.080 6.399 0.001
(log10) Efflux 0.065 3 0.022 5.132 0.004
Province * Season (log10) Uptake 0.120 3 0.040 3.202 0.031
(log10) Efflux 0.031 3 0.010 2.444 0.075
Error (log10) Uptake 0.637 51 0.012
(log10) Efflux 0.217 51 0.004
Corrected Total (log10) Uptake 1.050 58
(log10) Efflux 0.341 58
180
Table 4.4. Homogenous subsets based on Tukey's HSD post-hoc test for (log10) P-uptake by season.
Means for groups in homogeneous subsets are displayed.
Season N Subset
1 2
Fall 15 0.0027
Summer 16 0.0260
Spring 16 0.0991 0.0991
Winter 12 0.1583
Sig. 0.106 0.487
Table 4.5. Homogenous subsets based on Tukey's HSD post-hoc test for (log10) P-efflux by season.
Means for groups in homogeneous subsets are displayed.
Season N Subset
1 2
Summer 16 0.0137
Fall 15 0.0248
Spring 16 0.0388 0.0388
Winter 12 0.0944
Sig. 0.727 0.111
181
Table 4.6. Descriptive statistics of P-uptake and P-efflux (µgP/µgChl/min) for biofilms on duplicate
tiles in eight streams situated across two geologic provinces over four seasons (n= 64).
Province Season N Mean Std. Error Minimum Maximum
Plateau Fall Uptake 8 0.0058 0.0421 -0.19 0.25
Efflux 8 0.1189 0.0663 0.00 0.54
Winter Uptake 6 0.8587 0.3795 0.02 2.58
Efflux 6 0.4636 0.2140 0.00 1.30
Spring Uptake 8 0.6628 0.3036 0.01 2.13
Efflux 8 0.2019 0.1057 0.00 0.82
Summer Uptake 8 0.0528 0.0112 0.01 0.11
Efflux 8 0.0330 0.0089 0.01 0.07
Piedmont
Fall
Uptake 8 0.0124 0.0027 0.00 0.02
Efflux 8 0.0055 0.0015 0.00 0.01
Winter
Uptake 8 0.3612 0.2392 0.01 1.88
Efflux 8 0.0962 0.0796 0.00 0.65
Spring
Uptake 8 0.0559 0.0276 0.01 0.24
Efflux 8 0.0185 0.0072 0.00 0.05
Summer
Uptake 8 0.0726 0.0222 0.01 0.18
Efflux 8 0.0315 0.0105 0.00 0.07
182
Table 4.7. Pearson correlation matrix of all LOESS derived uptake estimations and biochemical parameters tested in eight Pennsylvania streams over
four seasons (n= 32). Note: ** and * indicates correlation is significant at the 0.01 and 0.05 level (2-tailed), respectively.
Variable Max
Uptake
Time
Max
Uptake
Max
Efflux
Time
Max
Efflux
(log10)
Uptake
Rate
(log10)
Efflux
Rate
(log10)
Poly-P
(log10)
Part-P
(log10)
Conductivity
(log10) Chl
Accumulation
Max Uptake Pearson 1 0.220 0.996** 0.227 0.813** 0.752** -0.350 -0.617** -0.508** -0.588**
Sig. (2-tailed) 0.235 0.000 0.228 0.000 0.000 0.086 0.000 0.004 0.001
Time Max Uptake Pearson 1 0.253 0.776** -0.046 -0.043 -0.417* -0.248 -0.329 -0.309
Sig. (2-tailed) 0.178 0.000 0.806 0.823 0.038 0.179 0.071 0.091
Max Efflux Pearson 1 0.215 0.817** 0.784** -0.356 -0.627** -0.520** -0.592**
Sig. (2-tailed) 0.255 0.000 0.000 0.088 0.000 0.003 0.001
Time Max Efflux Pearson 1 -0.023 -0.068 -0.222 -0.210 -0.172 -0.299
Sig. (2-tailed) 0.902 0.722 0.297 0.265 0.364 0.108
(log10) Uptake Rate Pearson 1 0.866** -0.178 -0.529** -0.375* -0.537**
Sig. (2-tailed) 0.000 0.395 0.002 0.038 0.002
(log10) Efflux Rate Pearson 1 -0.474* -0.598** -0.552** -0.558**
Sig. (2-tailed) 0.019 0.000 0.002 0.001
(log10) Poly-P Pearson 1 0.738** 0.637** 0.675**
Sig. (2-tailed) 0.000 0.001 0.000
(log10) Part-P Pearson 1 0.581** 0.812**
Sig. (2-tailed) 0.001 0.000
(log10) Conductivity Pearson 1 0.590**
Sig. (2-tailed) 0.000
(log10) Chl
Accumulation
Pearson 1
Sig. (2-tailed)
183
Time (min)
(lo
g1
0)
µg
P/
µg
Ch
l
30150
0.30
0.25
0.20
0.15
30150
0.08
0.07
0.06
0.05
0.04
30150
0.045
0.040
0.035
0.030
0.025
30150
0.14
0.12
0.10
0.08
0.06
30150
0.010
0.008
0.006
0.004
0.002
30150
0.08
0.07
0.06
0.05
0.04
30150
0.10
0.08
0.06
0.04
0.02
30150
0.08
0.06
0.04
0.02
0.00
East Hickory Tionesta C owanesque Tunkhannock
Red C lay C ooks Penns Spring
Figure 4.1. Scatter plot of P flux (log10(µgP/µgChl)) vs. time (minute) conducted on intact biofilm
assemblages from eight Pennsylvania streams of varying productivity in Summer 2010. The lines
show locally weighted scatter plot smoothing (LOESS) curve fitting. Time samplings were conducted
at 0, 1, 2, 4, 5, 8, 10, 12, 16, 25, 30 minutes. Note scale differences for streams.
184
Time (min)
(lo
g1
0)
µg
P/
µg
Ch
l
30150
0.60
0.55
0.50
0.45
0.40
30150
0.85
0.80
0.75
0.70
0.65
30150
0.025
0.020
0.015
0.010
0.005
30150
0.016
0.014
0.012
0.010
0.008
30150
0.14
0.12
0.10
0.08
0.06
30150
0.22
0.20
0.18
0.16
0.14
30150
0.04
0.03
0.02
0.01
30150
0.0100
0.0075
0.0050
0.0025
0.0000
East Hickory Tionesta C owanesque Tunkhannock
Red C lay C ooks Penns Spring
Figure 4.2. Scatter plot of P flux (log10(µgP/µgChl)) vs. time (minute) conducted on intact biofilm
assemblages from eight Pennsylvania streams of varying productivity in Spring 2010. The lines show
locally weighted scatter plot smoothing (LOESS) curve fitting. Time samplings were conducted at 0,
1, 2, 5, 10, 30 minutes. Note scale differences for streams.
185
Time (min)
(lo
g1
0)
µg
P/
µg
Ch
l
353025
0.50
0.25
0.00
-0.25
-0.50
30150
1.00
0.95
0.90
0.85
30150
0.24
0.21
0.18
0.15
0.12
30150
0.95
0.90
0.85
0.80
0.75
30150
0.07
0.06
0.05
0.04
0.03
30150
1.0
0.9
0.8
30150
0.010
0.009
0.008
0.007
0.006
30150
0.060
0.045
0.030
0.015
0.000
East Hickory Tionesta C owanesque Tunkhannock
Red C lay C ooks Penns Spring
Figure 4.3. Scatter plot of P flux (log10(µgP/µgChl)) vs. time (minute) conducted on intact biofilm
assemblages from eight Pennsylvania streams of varying productivity in Winter 2010. The lines show
locally weighted scatter plot smoothing (LOESS) curve fitting. Time samplings were conducted at 0,
1, 2, 5, 10, 30 minutes. Note scale differences for streams; also, samples for East Hickory were
irretrievable due to ice-coverage.
186
Time (min)
(lo
g1
0)
µg
P/
µg
Ch
l
30150
0.40
0.35
0.30
30150
0.6
0.5
0.4
0.3
0.2
30150
0.0200
0.0175
0.0150
0.0125
0.0100
30150
0.0240
0.0225
0.0210
0.0195
0.0180
30150
0.05
0.04
0.03
0.02
0.01
30150
0.025
0.020
0.015
0.010
0.005
30150
0.0150
0.0125
0.0100
0.0075
0.0050
30150
0.025
0.020
0.015
0.010
0.005
East Hickory Tionesta C owanesque Tunkhannock
Red C lay C ooks Penns Spring
Figure 4.4. Scatter plot of P flux (log10(µgP/µgChl)) vs. time (minute) conducted on intact biofilm
assemblages from eight Pennsylvania streams of varying productivity in Fall 2009. The lines show
locally weighted scatter plot smoothing (LOESS) curve fitting. Time samplings were conducted at 0,
1, 2, 5, 10, 30 minutes. Note scale differences for streams.
187
(lo
g1
0)
µg
P/
µg
Ch
l/m
in
1.000.750.500.250.00
0.4
0.3
0.2
0.1
0.0
2.01.51.00.50.0
0.100
0.075
0.050
0.025
0.000
(log)Uptake*(log)Particulate-P (log)Uptake*(log)Chl Accumulation
(log)Efflux*(log)Particulate-P (log)Efflux*(log)Chl Accumulation
Figure 4.5. Scatter plot with linear regression (solid-line) and LOESS (dashed-line) of (log10) P-
uptake (µgP/µgChl/min) (top) and (log10) P-efflux (µgP/µgChl/min) (bottom) vs. (log10) part-P
(mgP/m2) (left) and (log10) Chl accumulation (mgChl/m
2/d) (right) for intact biofilm assemblages
established on artificial substrata (tiles) from eight Pennsylvania streams of varying productivity
over four seasons 2009 - 2010. Note scale differences for P-uptake vs. P-efflux.
188
(lo
g1
0)
µg
P/
µg
Ch
l/m
in
3.02.52.01.51.0
0.60
0.45
0.30
0.15
0.00
3210
0.10
0.05
0.00
-0.05
(log)Uptake*(log)Particulate-P (log)Uptake*(log)Chl
(log)Efflux*(log)Particulate-P (log)Efflux*(log)Chl
Figure 4.6. Scatter plot with linear regression (solid-line) and LOESS (dashed-line) of (log10) P-
uptake (µgP/µgChl/min) (top) and (log10) P-efflux (µgP/µgChl/min) (bottom) vs. (log10) part-P
(mgP/m2) (left) and (log10) Chl (mgChl/m
2) (right) for biofilm assemblages established on natural
substrata (rocks) from eight Pennsylvania streams of varying productivity over four seasons 2009 -
2010. Note scale differences for P-uptake vs. P-efflux.
189
Province
(lo
g1
0(µ
gP
/µ
gC
hl/
min
))
P lateauP iedmont P lateauP iedmont
0.60
0.45
0.30
0.15
0.00
P lateauP iedmont
0.60
0.45
0.30
0.15
0.00
P lateauP iedmont
(log) Uptake, F all (log) Uptake, Winter (log) Uptake, Spring (log) Uptake, Summer
(log) Efflux, F all (log) Efflux, Winter (log) Efflux, Spring (log) Efflux, Summer
Figure 4.7. Box plot of (log10) P-uptake (top) and (log10) P-efflux (bottom) (µgP/µgChl/min) across
eight streams situated in two geological provinces in Pennsylvania over four seasons (n= 64).
Rectangular boxes represent the middle 50% (interquartile range (IQR)) of the data, the top and
bottom of the boxes represent the third (Q3) and first (Q1) quartiles, respectively, lines (“whiskers”)
extending to either side indicate the general extent of the data (upper limit (Q3+1.5(IQR)) and lower
limit (Q1–1.5(IQR))), lines within the boxes designate the median value, and asterisks indicate the
outliers (data lying outside the defined “whisker” bounds).
190
(log10) Initial Uptake
(lo
g1
0)
Assim
ilato
ry U
pta
ke
0.60.50.40.30.20.10.0
0.035
0.030
0.025
0.020
0.015
0.010
0.005
0.000
Figure 4.8. Scatter plot with linear regression of (log10) assimilatory (stable) P-uptake (30 minutes)
vs. (log10) initial P-uptake (≤5 minutes) (µgP/µgChl/min) for intact biofilm assemblages established
on artificial substrata (tiles) from eight Pennsylvania streams of varying productivity over four
seasons 2009 - 2010 (n= 64). Regression statistics: F= 248.44, p< 0.001, r2= 0.811.
191
CONCLUSION
Lotic systems play a critical role in Earth’s hydrologic cycle, transportation of
nutrients and organic matter, and maintenance of biological diversity (Allan and Flecker
1993). Moreover, based on average global value of ecosystem services, lake and river
systems are among the most important biomes in the world (cf. Costanza et al. 1997).
Obviously, any potential determent to the proper functioning of such systems (e.g.,
anthropogenic nutrient loading) could have serious environmental and economic impacts.
This dissertation provides insight into the efficiency of stream ecosystems in removing
nutrients through investigating the uptake capacity of benthic biofilms, which constitute
the dominant form of microbial life in most aquatic ecosystems (Hall-Stoodley et al.
2004). The overall objective of this four chapter work was to conduct a series of
empirical and experimental research endeavors examining the P-uptake potential of
benthic biofilms by building upon preceding chapters’ discoveries and continually
refining hypotheses in an effort to help identify and understand sources of variation in
uptake rates (Figure C.1).
In Chapter 1, I performed a synthesis and analysis of peer-reviewed aquatic P-
uptake rates which helped in developing a model explaining variation in nutrient uptake
rates among aquatic microbes. This research highlighted the significant difference
between planktonic and benthic P-uptake rates and suggests that the lower affinity for P
by benthic microbes could be attributed to their adnate growth forms, which can create
boundary layers separating cells from ambient P. Further, I found that uptake rates
measured in experiments on cultured microbes were generally higher compared with wild
samples; however, this trend was not significant and suggests that cultured microbes in
192
these studies represented a reasonable surrogate for natural microbes. Such an extension
may be useful for studies where in situ experimentation is not feasible. Lastly, this
research showed that experiments sampling microbes over shorter times yielded over 3-
fold higher P-uptake rates and appear to represent more accurate estimates of gross
uptake, a concept explored further in Chapter 4.
The results from Chapter 1 suggested that differences in experimental conditions
may be driving some of the substantial variation observed in uptake rates. Chapter 2
sought to build on these findings by verifying the effects of physical disturbance (a
common biofilm experimental processing technique) on biofilm viability and P-uptake
potential. The experiments showed no difference in P-uptake rates between (physically)
scraped (x = 0.77 ±0.11 (SE) μgP/μgChl/d) and intact (x = 0.91 ±0.17 (SE) μgP/μgChl/d)
biofilms (t= 0.69, p= 0.492, df= 33). Further, microbial physiology was not depressed by
physical disturbance. These data thus lend confidence to the numerous experiments that
investigate benthic microbial physiologic responses post-disturbance and highlight the
potential of uptake following common physical disturbances that occur in turbulent
(stream) environments.
Given the findings in Chapter 2, specifically that physical abrasion (disturbance)
of the biofilm does not negatively affect uptake capacity or physiological condition, I
utilized such sampling methods in Chapter 3 to facilitate estimations of multiple
biochemical parameters (e.g., biomass, areal nutrient concentrations) from a single
sample that would otherwise be impractical to measure from an intact radioactive sample.
Overall, results from Chapter 3 experiments showed that benthic stream biofilm uptake
declined linearly with increasing experimental P-loadings across all streams. Results
193
further showed a pattern of decreasing uptake efficiency with higher point source nutrient
enrichment and stream productivity. These experimental results indicate that while
stream biofilms may act as a reservoir for dissolved P, their uptake efficiency is rapidly
reduced in response to prolonged high P inputs (legacy). My data also showed that N
enrichment provided biofilms a greater ability to respond to a given P-load, suggesting
that N may become a controlling (limiting) resource particularly when P is in high
concentrations. While N enrichments did produce a complex biofilm P-uptake
phenomenon, influenced by biogeochemistry (province) and cellular stoichiometric
ratios, my experiments provide direct, cause-effect results that underscore the need to
control loads of both P and N in streams throughout Pennsylvania (and likely the greater
Mid-Atlantic region) and present a much needed mechanistic understanding of microbial
resource availability - resource demand (cf. Davidson and Howarth 2007).
Expounding upon the findings in Chapter 1, specifically that microbes sampled
over shorter times (i.e., < 10 minutes) yielded considerably higher P-uptake rates, results
in Chapter 4 showed substantial and rapid uptake/exchange processes occurring at early
time periods (i.e., < 5 minutes), the magnitude of which seemed to diminish over longer
periods (i.e., 15 - 30 minutes). Qualitative evidence found here suggests that many
aquatic microbial kinetic uptake studies with initial samples ≥ 5 minutes (e.g., Wolfe and
Lind 2010, Yao et al. 2011) may fail to measure initial assimilatory kinetics –essential in
generating longer term uptake rates (see Figure 4.8)– and that many contemporary “short-
term” phosphate uptake experiments are generating rates that reflect net uptake (net result
of counteracting uptake/efflux events) rather than gross. Additionally, research from
Chapter 4 supported earlier results from Chapter 3 regarding the influence of nutrient
194
legacy (between provinces) on biofilm P-uptake rates. Here I found that while both
province and season were significant factors in the ANOVAs for P-uptake and P-efflux,
the F-ratio for province was greater, suggesting that spatial effects may more effectively
influence differences in uptake/efflux. Certainly the Pennsylvania Plateau to Piedmont
transition I tested against here reflects the prominent geology of the provinces as well as
a strong gradient of environmental conditions and nutrient availability.
Overall, my dissertation research provides greater understanding of stream
nutrient dynamics through these empirical and experimental approaches aimed at
examining biofilm community uptake rates. Future research might focus on microbial
taxonomic composition (see section 4.5) and diversity indices to provide further insights
into nutrient uptake - availability relationships (see Baker et al. 2009).
LITERATURE CITED
Allan, J.D. and Flecker, A.S. 1993. Biodiversity conservation in running waters:
Identifying the major factors that affect destruction of riverine species and
ecosystems. BioScience 43:32-43.
Baker, M.A., de Guzman, G., and Ostermiller, J.D. 2009. Differences in nitrate uptake
among benthic algal assemblages in a mountain stream. Journal of the North
American Benthological Society 28:24-33.
Costanza, R., d’Arge, R., deGroot, R., Farber, S., Grasso, M., Hannon, B., Limburg, K.,
Naeem, S., O’Neill, R.V., Paruelo, J., Raskin, R.G., Sutton, P., and van den Belt,
M. 1997. The value of the world’s ecosystem services and natural capital. Nature
387:253-260.
195
Davidson, E.A. and Howarth, R.W. 2007. Environmental science: Nutrients in synergy.
Nature 449:1000-1001.
Hall-Stoodley, L., Costerton, J.W., and Stoodley, P. 2004. Bacterial biofilms: From the
natural environment to infectious diseases. Nature Reviews Microbiology 2:95-
108.
Wolfe, J.E. III and Lind, O.T. 2010. Phosphorus uptake and turnover by periphyton in the
presence of suspended clays. Limnology 11:31-37.
Yao, B., Xi, B., Hu, C., Huo, S., Su, J., and Liu, H. 2011. A model and experimental
study of phosphate uptake kinetics in algae: Considering surface adsorption and
P-stress. Journal of Environmental Sciences 23:189-198.
196
Figure C.1. Schematic of my four chapter dissertation research illustrating the refinement and
concentration of hypotheses through the successive applications of preceding research findings.
197
Appendix A. Summary table of experiments testing the effects of physical disturbance on biofilm
phosphorus uptake, displaying stream geographic coordinates and sampling date(s) for each
experiment (Chapter 2).
Experiment Stream Name Latitude Longitude Sampling Date(s)
Uptake by intact vs.
scraped biofilms
Spring Creek 40.7786
40.8222
-77.7696
-77.8369
10/17/2008
6/3/2009
6/11/2009
Effect of increasing levels
of disturbance
Spring Creek 40.7786 -77.7696 5/11/2009
Estimates of M-M
parameters
Spring Creek 40.7786 -77.7696 8/5/2009
Estimates of abiotic
sorption
Spring Creek 40.7786
40.8222
-77.7696
-77.8369
6/17/2009
198
Appendix B. Descriptive statistics for site, treatment, and experiment (date) factors for intact vs.
scraped stream biofilm P-uptake rates (μgP/μgChl/d) (Chapter 2). Note: SE Mean is the standard
error of the mean and CV is the coefficient of variation.
Site Treatment Experiment (date) n Mean SE Mean CV
Upstream Intact 10/17/08 4 0.336 0.030 17.627
Upstream Scrape 10/17/08 4 0.749 0.045 12.118
Downstream Intact 10/17/08 4 1.715 0.156 18.239
Downstream Scrape 10/17/08 4 1.329 0.077 11.593
Upstream Intact 6/3/09 3 0.250 0.011 7.639
Upstream Scrape 6/3/09 3 0.115 0.017 26.039
Downstream Intact 6/3/09 3 1.963 0.088 7.758
Downstream Scrape 6/3/09 3 0.674 0.116 29.789
Upstream Intact 6/11/09 3 0.153 0.055 62.379
Upstream Scrape 6/11/09 3 0.197 0.018 15.819
Downstream Intact 6/11/09 3 0.988 0.222 38.886
Downstream Scrape 6/11/09 3 1.386 0.108 13.535
199
Appendix C. Genera list from increasing levels of disturbance experiment (Chapter 2). Cells
identified under 100x magnification. Physiognomic (growth morphology) classifications of these
genera follow Graham and Vinebrooke (1998), Wellnitz and Ward (2000), and Passy (2007).
Genera Growth
Morphology
Disturbance Treatment
Low Low Low Medium Medium Medium High High High
Melosira Filamentous 1101 187 113 1419 693 793 678 406 1018
Closterium Stalked 22 5 2 10 5 12 6 7 66
Cosmarium Stalked 46 4 6 52 6 76 24 26 48
Cymbella Stalked 5 4 0 6 2 1 6 4 6
Diatoma Stalked 0 0 0 0 0 0 0 0 0
Gomphonema Stalked 46 9 7 142 40 206 108 71 54
Nitzschia Stalked 0 0 0 0 0 1 1 0 0
Surirella Stalked 0 0 0 0 0 0 0 0 0
Synedra Stalked 1 3 0 0 0 0 0 0 0
Achnanthes Prostrate 0 0 0 0 0 0 0 0 0
Amphora Prostrate 0 0 0 0 0 0 0 2 0
Cocconeis Prostrate 17 9 0 24 14 10 2 15 33
Cyclotella Prostrate 11 2 0 3 4 3 4 1 6
Eunotia Prostrate 0 0 0 0 0 0 0 0 0
Frustulia Prostrate 10 8 2 2 7 5 7 12 14
Gyrosigma Prostrate 0 0 0 0 0 0 1 0 0
Navicula Prostrate 0 3 0 0 0 0 2 2 12
Rhoicosphenia Prostrate 0 0 0 0 0 0 0 1 0
Total - 1259 234 130 1658 771 1107 839 547 1257
200
Appendix D. Genera list from increasing levels of disturbance experiment (Chapter 2). Cells
identified under 400x magnification. Physiognomic (growth morphology) classifications of these
genera follow Graham and Vinebrooke (1998), Wellnitz and Ward (2000), and Passy (2007).
Genera Growth
Morphology
Disturbance Treatment
Low Low Low Medium Medium Medium High High High
Melosira Filamentous 29 12 15 57 18 12 25 15 24
Closterium Stalked 2 0 0 1 0 0 0 0 2
Cosmarium Stalked 3 0 0 0 0 4 0 0 0
Cymbella Stalked 0 0 0 1 0 0 3 7 3
Diatoma Stalked 0 0 0 0 0 0 3 0 2
Gomphonema Stalked 2 1 1 3 1 2 9 1 5
Nitzschia Stalked 0 0 0 1 1 0 1 0 0
Surirella Stalked 0 0 0 0 1 0 0 0 0
Synedra Stalked 0 1 0 0 0 0 0 0 0
Achnanthes Prostrate 0 0 0 0 0 0 4 10 8
Amphora Prostrate 0 0 0 0 0 1 2 1 4
Cocconeis Prostrate 1 0 0 1 0 0 4 3 4
Cyclotella Prostrate 0 0 0 2 1 0 1 2 1
Eunotia Prostrate 0 0 0 0 0 0 1 0 0
Frustulia Prostrate 1 1 0 0 0 0 0 0 1
Gyrosigma Prostrate 0 0 0 0 0 0 0 0 0
Navicula Prostrate 1 1 0 0 1 2 1 11 8
Rhoicosphenia Prostrate 0 1 0 0 0 2 3 1 0
Total - 39 17 16 66 23 23 57 51 62
201
Fra
cti
on
of
To
tal
HighMediumLow
1.0
0.8
0.6
0.4
0.2
0.0
Growth Form
Filamentous
Stalked
Prostrate
Appendix E. Fraction (of total) of biofilm growth forms (examined at 400x) recovered from each of
the three disturbance treatments applied to resident biofilms growing on natural substrates in Spring
Creek (Chapter 2).
202
[P] (µgP/L)
P U
pta
ke (
µg
P/µ
gC
hl/d
)
0 50 100 150 200 2500.0
0.5
1.0
1.5
Michaelis-Menten
Best-fit values
Vmax
Km
Std. Error
Vmax
Km
95% Confidence Intervals
Vmax
Km
Goodness of Fit
Degrees of Freedom
R square
2.247
231.4
0.7662
140.9
-1.050 to 5.544
0.0 to 837.8
2
0.9893
Appendix F. Michaelis-Menten plot and parameter estimates of P-uptake vs. P concentration for
scraped biofilm assemblages estimated over a 60 minute time period (Chapter 2). Data were fitted by
nonlinear regression according to the Michaelis-Menten equation using GraphPad Prism 5
(GraphPad Prism Software Inc., San Diego, CA, USA).
203
[P] (µgP/L)
P U
pta
ke (
µg
P/µ
gC
hl/d
)
0 200 400 6000.0
0.5
1.0
1.5
2.0
Michaelis-Menten
Best-fit values
Vmax
Km
Std. Error
Vmax
Km
95% Confidence Intervals
Vmax
Km
Goodness of Fit
Degrees of Freedom
R square
2.332
293.9
0.6299
173.9
0.3271 to 4.336
0.0 to 847.4
3
0.9757
Appendix G. Michaelis-Menten plot and parameter estimates of P-uptake vs. P concentration for
intact biofilm assemblages estimated over a 60 minute time period (Chapter 2). Data were fitted by
nonlinear regression according to the Michaelis-Menten equation using GraphPad Prism 5
(GraphPad Prism Software Inc., San Diego, CA, USA).
204
[P] (µgP/L)
P U
pta
ke (
µg
P/µ
gC
hl/d
)
0 50 100 150 200 2500.0
0.5
1.0
1.5
2.0
2.5
Michaelis-Menten
Best-fit values
Vmax
Km
Std. Error
Vmax
Km
95% Confidence Intervals
Vmax
Km
Goodness of Fit
Degrees of Freedom
R square
16.64
1457
1.400
139.4
10.62 to 22.67
857.5 to 2057
2
1.000
Appendix H. Michaelis-Menten plot and parameter estimates of P-uptake vs. P concentration for
scraped biofilm assemblages estimated over short (i.e., 5 - 12 minutes) time periods (Chapter 2). Data
were fitted by nonlinear regression according to the Michaelis-Menten equation using GraphPad
Prism 5 (GraphPad Prism Software Inc., San Diego, CA, USA).
205
[P] (µgP/L)
P U
pta
ke (
µg
P/µ
gC
hl/d
)
0 10 20 300.0
0.5
1.0
1.5
Michaelis-Menten
Best-fit values
Vmax
Km
Std. Error
Vmax
Km
95% Confidence Intervals
Vmax
Km
Goodness of Fit
Degrees of Freedom
R square
2.301
23.37
4.685
85.24
-57.23 to 61.83
0.0 to 1106
1
0.6448
Appendix I. Michaelis-Menten plot and parameter estimates of P-uptake vs. P concentration for
intact biofilm assemblages estimated over short (i.e., 5 - 12 minutes) time periods (Chapter 2). Data
were fitted by nonlinear regression according to the Michaelis-Menten equation using GraphPad
Prism 5 (GraphPad Prism Software Inc., San Diego, CA, USA).
206
[P] (µgP/L)
P U
pta
ke (
µg
P/µ
gC
hl/d
)
0 50 100 150 200 2500.0
0.5
1.0
1.5
Michaelis-Menten
Best-fit values
Vmax
Km
Std. Error
Vmax
Km
95% Confidence Intervals
Vmax
Km
Goodness of Fit
Degrees of Freedom
R square
8.422
1237
16.61
2831
-63.05 to 79.90
0.0 to 13419
2
0.9934
Appendix J. Michaelis-Menten plot and parameter estimates of P-uptake vs. P concentration for
scraped biofilm assemblages estimated over long (i.e., 30 - 60 minutes) time periods (Chapter 2). Data
were fitted by nonlinear regression according to the Michaelis-Menten equation using GraphPad
Prism 5 (GraphPad Prism Software Inc., San Diego, CA, USA).
207
[P] (µgP/L)
P U
pta
ke (
µg
P/µ
gC
hl/d
)
0 50 100 150 200 2500.0
0.1
0.2
0.3
Michaelis-Menten
Best-fit values
Vmax
Km
Std. Error
Vmax
Km
95% Confidence Intervals
Vmax
Km
Goodness of Fit
Degrees of Freedom
R square
1.106
653.4
1.251
960.9
-4.276 to 6.489
0.0 to 4788
2
0.9904
Appendix K. Michaelis-Menten plot and parameter estimates of P-uptake vs. P concentration for
intact biofilm assemblages estimated over long (i.e., 30 - 60 minutes) time periods (Chapter 2). Data
were fitted by nonlinear regression according to the Michaelis-Menten equation using GraphPad
Prism 5 (GraphPad Prism Software Inc., San Diego, CA, USA).
208
Appendix L. Pennsylvania statewide map depicting the locations of the 47 streams (within distinct
physiographic provinces) sampled during a previous study (see Carrick et al. 2009) from which eight
streams (numbered in diamonds) were selected as a representative subset to deploy the in situ
enrichment system (ISES) in order to evaluate the effects of P- and N-loading on native stream
biofilm P-uptake capacity (Chapter 3).
209
Appendix M. Summary table of in situ enrichment experiments (ISES) conducted in 2008 (Chapter
3). All streams are part of the Pennsylvania Department of Environmental Protection’s Water
Quality Network (WQN). The summary includes the county where each stream is located, WQN
number, Ecoregion, geographic coordinates, and sampling dates for the deployment and retrieval of
each ISES experiment.
County Stream Name WQN Ecoregion Latitude Longitude Sampling Dates
Warren East Hickory Creek 877 Appalachian
Plateau
41.6160 -79.3701 8/22/2008
9/05/2008
Forest Tionesta Creek 830 Appalachian
Plateau
41.6019 -79.0503 8/22/2008
9/05/2008
Cumberland Cowanesque River 320 Appalachian
Plateau
42.0014 -77.1278 8/21/2008
9/03/2008
Wyoming Tunkhannock Creek 317 Appalachian
Plateau
41.5581 -75.8950 8/21/2008
9/03/2008
Chester Red Clay Creek 150 Lower
Piedmont
39.8167 -75.6919 8/11/2008
8/27/2008
Bucks Cooks Creek 187 Lower
Piedmont
40.5854 -75.2061 8/11/2008
8/27/2008
Synder Penns Creek 229 Upper
Piedmont
40.8590 -77.5796 8/12/2008
8/29/2008
Centre Spring Creek 415 Upper
Piedmont
40.7917 -77.7980 8/12/2008
8/29/2008
210
Areal C (g/m2)
Are
al C
hl (g
/m
2)
35302520151050
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
Appendix N. Scatter plot of areal chlorophyll-a (Chl) (g/m2) vs. areal carbon (C) (g/m
2) estimated
from biofilms recovered from ISES experiments in both Plateau and Piedmont provinces (n= 80)
(Chapter 3). Pearson correlation coefficient (r)= 0.768, p< 0.001, r2= 0.590.
211
Appendix O. Total alkalinity measured in eight Pennsylvania streams of varying productivity over four seasons 2009 - 2010 (Chapter 4). Alkalinity was
determined by colorimetric titration (phenolphthalein and mixed indicator of bromocresol green-methyl red) to pH 4.6 with 0.02 N H2SO4 following
Wetzel and Likens (2000).
Stream Province Season Date Hydroxide
(mg CaCO3/L)
Carbonate
(mg CaCO3/L)
Bicarbonate
(mg CaCO3/L)
Total alkalinity
(mg CaCO3/L)
East Hickory Plateau Fall 11/11/2009 0 0 16 16
East Hickory Plateau Winter 1/13/2010 0 0 10 10
East Hickory Plateau Spring 5/14/2010 0 0 4 4
East Hickory Plateau Summer 9/15/2010 0 0 30 30
Tionesta Plateau Fall 11/11/2009 0 0 22 22
Tionesta Plateau Winter 1/13/2010 0 0 14 14
Tionesta Plateau Spring 5/14/2010 0 10 10 20
Tionesta Plateau Summer 9/15/2010 0 0 40 40
Cowanesque Plateau Fall 11/18/2009 0 0 68 68
Cowanesque Plateau Winter 1/18/2010 0 0 66 66
Cowanesque Plateau Spring 5/17/2010 0 24 21 45
Cowanesque Plateau Summer 9/17/2010 0 20 50 70
Tunkhannock Plateau Fall 11/18/2009 0 0 48 48
Tunkhannock Plateau Winter 1/18/2010 0 0 42 42
Tunkhannock Plateau Spring 5/17/2010 0 0 36 36
Tunkhannock Plateau Summer 9/17/2010 0 20 32 52
Red Clay Piedmont Fall 11/15/2009 0 0 92 92
Red Clay Piedmont Winter 1/15/2010 0 0 84 84
Red Clay Piedmont Spring 5/12/2010 0 36 3 39
Red Clay Piedmont Summer 9/20/2010 20 80 0 100
Cooks Piedmont Fall 11/18/2009 0 0 122 122
Cooks Piedmont Winter 1/15/2010 0 0 108 108
Cooks Piedmont Spring 5/12/2010 10 20 0 30
Cooks Piedmont Summer 9/20/2010 0 20 90 110
Penns Piedmont Fall 11/13/2009 0 0 188 188
Penns Piedmont Winter 1/16/2010 0 0 162 162
212
Penns Piedmont Spring 5/15/2010 0 36 9 45
Penns Piedmont Summer 9/18/2010 0 0 170 170
Spring Piedmont Fall 11/13/2009 0 0 204 204
Spring Piedmont Winter 1/16/2010 0 0 160 160
Spring Piedmont Spring 5/15/2010 0 20 100 120
Spring Piedmont Summer 9/18/2010 0 0 240 240
213
Appendix P. Summary table of sampling dates for the deployment and retrieval of each spatio-
temporal experiment (Chapter 4). Additional stream information (i.e., county, Ecoregion, geographic
coordinates) is located within Appendix M.
Plateau Piedmont
Stream Season Sampling Dates Stream Season Sampling Dates
East Hickory Fall 10/14/2009
11/11/2009
Red Clay Fall 10/9/2009
11/15/2009
Winter 12/16/2009
1/13/2010
Winter 12/18/2009
1/15/2010
Spring 4/16/2010
5/14/2010
Spring 4/14/2010
5/12/2010
Summer 8/11/2010
9/15/2010
Summer 8/16/2010
9/20/2010
Tionesta Fall 10/14/2009
11/11/2009
Cooks Fall 10/9/2009
11/15/2009
Winter 12/16/2009
1/13/2010
Winter 12/18/2009
1/15/2010
Spring 4/16/2010
5/14/2010
Spring 4/14/2010
5/12/2010
Summer 8/11/2010
9/15/2010
Summer 8/16/2010
9/20/2010
Cowanesque Fall 10/12/2009
11/18/2009
Penns Fall 10/10/2009
11/13/2009
Winter 12/21/2009
1/18/2010
Winter 12/19/2009
1/16/2010
Spring 4/19/2010
5/17/2010
Spring 4/17/2010
5/15/2010
Summer 8/13/2010
9/17/2010
Summer 8/14/2010
9/18/2010
Tunkhannock Fall 10/12/2009
11/18/2009
Spring Fall 10/10/2009
11/13/2009
Winter 12/21/2009
1/18/2010
Winter 12/19/2009
1/16/2010
Spring 4/19/2010
5/17/2010
Spring 4/17/2010
5/15/2010
Summer 8/13/2010
9/17/2010
Summer 8/14/2010
9/18/2010
214
Poly-P ((log10)mgP/m2)
P-e
fflu
x (
(lo
g1
0)
µg
P/
µg
Ch
l/m
in)
2.01.51.00.50.0
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0.00
Appendix Q. Scatter plot with linear regression of (log10) P-efflux (µgP/µgChl/min) vs. (log10) poly-P
(mgP/m2) for intact biofilm assemblages established on artificial substrata (tiles) from eight
Pennsylvania streams of varying productivity over four seasons 2009 - 2010 (Chapter 4).
KEITH J. PRICE 125 Cardinal Dr. (484) 238-8730
Conshohocken, PA 19428 [email protected]
EDUCATION
Ph.D. Wildlife and Fisheries Science, The Pennsylvania State University,
University Park, PA, 2012
o Dissertation: “Phosphorus Uptake by Stream Benthic Biofilms: Empirical
and Experimental Approaches to Explaining Variation”
M.S. Biology, West Chester University, West Chester, PA, 2005
o Thesis: “Nutrient Limitation of Periphyton Biomass in Westtown Lake”
B.S. Biology, Philadelphia University, Philadelphia, PA, 2002
o Minor: Economics
EXPERIENCE
The Pennsylvania State University, Dept. of Ecosystem Science and Management
o Teaching Assistant- Vertebrates Laboratory (WFS 301): Fall 2011
o Research Assistant: Spring 2007 - 2011
o Teaching Assistant- Limnology (WFS 435): Fall 2007 - 2010
o Teaching Assistant- Limnological Methods (WFS 436): Fall 2010
PEER-REVIEWED PUBLICATIONS
Price, K. J. and H. J. Carrick (in prep) Effects of nutrient loading on phosphorus
uptake by biofilms situated along a stream productivity gradient.
Carrick, H. J., K. L. Dananay, R. A. Eckert, and K. J. Price (2012) Decomposition
during autumn foliage leaf-fall in wetlands situated along a biogeochemical
gradient in Pennsylvania, USA. Journal of Freshwater Ecology 27:1-17 (Editor’s
Choice Article).
Price, K. J. and H. J. Carrick (2011) Meta-analytical approach to explain variation
in microbial phosphorus uptake rates in aquatic ecosystems. Aquatic Microbial
Ecology 65:89-102.
TECHNICAL PUBLICATIONS
Carrick, H. J. and K. J. Price (2011) Determining Variation in TMDL Reduction
Criteria. Final Report, Pennsylvania Department of Environmental Protection,
Harrisburg, PA, 84 p.
Carrick, H. J., R. A. Eckert, M. K. May, and K. J. Price (2011) Changes in
Biofilm Stoichiometry and Diatom Taxonomic Composition in Response to
Ecosystem-Level, Experimental Enrichment with P. Final Report, Pennsylvania
Department of Environmental Protection, Harrisburg, PA, 23 p.
Carrick, H. J., K. J. Price, M. K. May, and J. M. Regan (2009) Developing
Numeric Criteria to Guide Nutrient Controls for Streams in Pennsylvania. U.S.
Geological Survey State Water Resources Research Institute Program Report
2009PA103B, 5 p.
Price, K. J. (2005) Nutrient Limitation of Periphyton Biomass in Westtown Lake.
Masters Thesis, West Chester University, 140 p.