ABSTRACT BISHOP, WEST MICHAEL. a Risk

Home , Elakatothrix, Genicularia, Schroederia

ABSTRACT

BISHOP, WEST MICHAEL. A Risk-based Decision Information System for Selecting an Algal Management Program. (Under the direction of Dr. Robert J. Richardson).

Freshwater resources are a vital component needed to support numerous human activities. Climate change, water use dynamics, and eutrophication have synergistically promoted nuisance and noxious algal blooms throughout the world. The increasing distribution, duration, and intensity of algal blooms has created unprecedented concerns over safety and usability of freshwater. Some of the largest blooms ever documented have occurred in the recent past. This dissertation is designed to provide a framework to assist in

1) understanding risks associated with nuisance algal infestations and 2) comparatively assessing the efficacy, costs, and collateral risks of different management approaches. The management decision framework consists of characterizing dimensions of the problem, identifying management objectives, and applying research to create a strategic management approach that incorporates risks of no action as well as risks associated with different management strategies. By considering information from the research chapters included herein, an accurate assessment of risks can be conducted in terms of the effectiveness and ecological integrity of management approaches. This research should be used to make informed decisions regarding the choice to manage nuisance algae and the selection of an appropriate management approach.

by West Michael Bishop

A dissertation submitted to the Graduate Faculty of North Carolina State University in partial fulfillment of the requirements for the degree of Doctor of Philosophy

Crop Science

Raleigh, North Carolina

2016

APPROVED BY:

______Dr. Robert J. Richardson Dr. William F. Hunt III Committee Chair

______Dr. Travis W. Gannon Dr. JoAnn M. Burkholder

DEDICATION

I dedicate this dissertation to my children, Emerson Michael and Adelynn Marie

Bishop, may they pursue their passion and never settle for pretty good.

BIOGRAPHY

West Michael Bishop grew up in a small town in Michigan. He had a unique love for the natural environment and biological sciences field. Following attainment of a Bachelor of

Science in biology from Western Michigan University and internship at University of

Wisconsin-Madison, he discovered a passion for the aquatic toxicology field. This led to

completion of a Master of Science degree at Clemson University with a focus on nuisance and noxious algal management and preservation of non-target species. A further opportunity

to study at North Carolina State University arose with a great water resource scientist, Dr.

Richardson, to further advance the science of algal management.

iii

ACKNOWLEDGMENTS

First and foremost, to my committee. Their invaluable guidance, time, and effort to

support this research is greatly commended. Thanks to Bill Culpepper and his company,

SePRO Corporation, for the unending pursuit of solutions to solve nuisance algal issues and

support advancements in research. Unimaginable thanks to Dr. Tyler Koschnick for his vision and continuing guidance; this would not have been possible if not for his actions. My

colleagues Ben Willis and Dr. Mark Heilman also provided critical support to advance this

research. Numerous other mentors and colleagues helped support this work. Lastly to my

wife, Marguerite, for her tireless support of the family through this time and my sister, Dr.

Heidi Zubeck, for being a great leader and helping to guide my path.

TABLE OF CONTENTS

LIST OF TABLES ...... viii

LIST OF FIGURES ...... xi

CHAPTER ONE: Introduction

Overview ...... 1

Eutrophication and Environmental Conditions ...... 2

Water Dynamics...... 5

Magnitude of Issue ...... 7

Regulatory Assessment ...... 8

Management Overview ...... 9

Proactive Management...... 10

Integrated Management ...... 12

Reactive Management ...... 13

Organization of Dissertation ...... 16

References ...... 18

CHAPTER TWO: Impacts of Phosphorus Inactivation on Algal Assemblage

Composition and Water Quality Characteristics in Hypereutrophic Water Resources

Abstract ...... 28

Introduction ...... 29

Materials and Methods ...... 34

Results ...... 38

Discussion ...... 42

Conclusion ...... 46

References ...... 48

CHAPTER THREE: Sensitivity of Microcystis aeruginosa Strains to Copper and

Influence of Phosphorus

Abstract ...... 79

Introduction ...... 80

Methods...... 84

Results and Discussion ...... 88

Conclusion ...... 94

References ...... 96

CHAPTER FOUR: Comparison of Partitioning and Efficacy Between Copper

Algaecide Formulations: Refining the Critical Burden Concept

Abstract ...... 117

Introduction ...... 118

Materials and Methods ...... 124

Results ...... 129

Discussion ...... 133

Conclusion ...... 142

References ...... 145

CHAPTER FIVE: Evaluation of Risks from Copper Algaecide Applications: The

Competing Ligand Hypothesis

Abstract ...... 172

Introduction ...... 173

Materials and Methods ...... 178

Results ...... 183

Discussion ...... 186

Conclusion ...... 194

References ...... 196

CHAPTER SIX: Conclusions and a Recommended Decision Information System

Framework

Background ...... 224

Management Objectives and Primary Use ...... 226

Risks Assessment: No Action ...... 227

Proactive Risks...... 234

Reactive Risks ...... 236

Risk Assessment: Management Decision Examples ...... 240

Decision Assessment ...... 242

Summary ...... 245

References ...... 247

APPENDICES ...... 261

vii

LIST OF TABLES

CHAPTER TWO: Impacts of Phosphorus Inactivation on Algal Assemblage

Composition and Water Quality Characteristics in Hypereutrophic Water Resources

Table 1. Description of site characteristics and ranges of water chemistry in the two

experimental golf course ponds ...... 59

Table 2. Description of analytical parameters and methods for measured water quality

parameters ...... 60

Table 3. Description of sediment P fractions from sequential extraction procedures ...... 61

Table 4. Statistical comparisons using a paired t-test or Wilcoxon Rank Sum test for

algae, water quality and sediment parameters between the pre-treatment and post-

treatment growing season averages (April-September) over two years. Significance

denoted with a P value < 0.05 ...... 62

CHAPTER THREE: Sensitivity of Microcystis aeruginosa Strains to Copper and

Influence of Phosphorus

Table 1. Description of the five Microcystis aeruginosa strains used in testing and growth

rates throughout exposure duration. Cell diameter was measured at experiment initiation

and conclusion. Growth rates are averages throughout experiment duration ...... 104

Table 2. Mean and ranges of water chemistry parameters for COMBO nutrient media

used throughout all replicate experimentation ...... 105

Table 3. Calculated 96-hr LC50 values for Ma chlorophyll a content and cell densities.

The range represents the 95% confidence interval for the nonlinear (polynomial,

quadratic) regression analysis. The corresponding R2 value is also given ...... 106

viii

CHAPTER FOUR: Comparison of Partitioning and Efficacy Between Copper

Algaecide Formulations: Refining the Critical Burden Concept

Table 1. Comparison of physical and chemical properties of the copper formulations

tested ...... 154

Table 2. Description of exposure-response relationships using nonlinear regression

analysis with a best fit analysis for the data. Equations and R squared values are calculated

for each treatment in 7-day L. wollei and P. varia toxicity experiments. Chlorophyll a

responses were analyzed by nonlinear regression using exponential decay, single three

parameter (y = y0+a*exp(-b*x)) equation. Filament viability was analyzed using

exponential decay, single two-parameter (y = a*exp(-b*x)) and polynomial quadratic (y =

y0+a*x+b*x^2) equations for L. wollei and P. varia, respectively ...... 155

Table 3. Calculated 7-day EC85 toxicity values for L. wollei and P. varia exposed to

copper algaecide formulations. Regression models of responses (chlorophyll a or filament

viability) versus exposures (nominal or internalized copper) were used to calculate

toxicity thresholds ...... 156

Table 4. Average bio-concentration and bio-internalization factors for copper formulation

exposures in L. wollei and P. varia toxicity experiments. Ranges represent values across

all treatment exposure concentrations ...... 157

CHAPTER FIVE: Evaluation of Risks from Copper Algaecide Applications: The

Competing Ligand Hypothesis

Table 1. Mean and ranges of water chemistry parameters for COMBO nutrient media

used throughout all replicate experimentation ...... 203

Table 2. Freshwater COMBO media constituents for culturing and testing of Scenedesmus dimorphus and Daphnia magna ...... 204

Table 3. Comparison of physical and chemical properties of the two copper formulations tested ...... 205

Table 4. Description of targeted concentrations used in toxicity testing to define the exposure-response relationship ...... 206

Table 5. Average Scenedesmus dimorphus cellular characteristics ...... 207

Table 6. Description of exposure-response relationships using nonlinear regression analysis with a three-parameter sigmoidal curve. Equations and R squared values are calculated for copper sulfate and Captain treatments in 48-hr Daphnia magna toxicity tests ...... 208

Table 7. Average percent free and total copper in solution and sorbed to algae in all treatments over the 48-hr experiment durtion ...... 209

Table 8. 48-hr NOEC, LOEC, and LC50 values in copper formulation toxicity experiments

1) without algae, 2) containing 5 x 105 cells/mL, and 3) containing 5 x 106 cells/mL .....210

LIST OF FIGURES

CHAPTER TWO: Impacts of Phosphorus Inactivation on Algal Assemblage

Composition and Water Quality Characteristics in Hypereutrophic Water Resources

Figure 1. Map of the Chockyotte irrigation pond and designation of sampling locations. 63

Figure 2. Map of the Hickory Meadows irrigation pond and designation of sampling

locations ...... 64

Figure 3. Daily mean, high, and low temperatures in Whitakers, NC throughout the study

duration...... 65

Figure 4. Total daily rainfall in Whitakers, NC throughout the study duration ...... 66

Figure 5. Secchi depth measurements in the Chockyotte irrigation pond throughout the

growing season pre-treatment (2014) and two years after treatment (2015-2016). Each

point represents the average from three sites ...... 67

Figure 6. Secchi depth measurements in the Hickory Meadows irrigation pond throughout

the growing season pre-treatment (2014) and two years after treatment (2015-2016). Each

point represents the average from three sites ...... 68

Figure 7. Total and free reactive P measurements in the Chockyotte irrigation pond

throughout the study duration. Error bars represent one standard deviation from the

mean ...... 69

Figure 8. Total and free reactive P measurements in the Hickory Meadows irrigation pond

throughout the study duration. Error bars represent one standard deviation from the

mean ...... 70

Figure 9. Total algal and cyanobacterial densities in conjunction with TN:TP ratios in the

Chockyotte irrigation pond throughout the study duration ...... 71

Figure 10. Percent abundance of algal groups throughout the study duration at the

Chockyotte irrigation pond ...... 72

Figure 11. Total algal and cyanobacterial densities in conjunction with TN:TP ratios in

the Hickory Meadows irrigation pond throughout the study duration ...... 73

Figure 12. Percent abundance of algal groups throughout the study duration at the

Hickory Meadows irrigation pond ...... 74

Figure 13. Mobile sediment P fractions in the Chockyotte irrigation pond through time.

Error bars represent one standard deviation from the mean ...... 75

Figure 14. Apatite/residual and combined mobile sediment P fractions in the Chockyotte

irrigation pond through time. Error bars represent one standard deviation from the

mean...... 76

Figure 15. Mobile sediment P fractions in the Hickory Meadows irrigation pond through

time. Error bars represent one standard deviation from the mean ...... 77

Figure 16. Apatite/residual and combined mobile sediment P fractions in the Hickory

Meadows irrigation pond through time. Error bars represent one standard deviation from

the mean ...... 78

CHAPTER THREE: Sensitivity of Microcystis aeruginosa Strains to Copper and

Influence of Phosphorus

xii

Figure 1. Mean P content per cell for five cultured Ma strains across three P growth regimes. Error bars represent one standard deviation from the mean. Different letters signify significant differences (P < 0.05) ...... 107

Figure 2. Mean chlorophyll a content per cell for five cultured Ma strains across three P growth regimes. Error bars represent one standard deviation from the mean. Different letters signify significant differences (P < 0.05) ...... 108

Figure 3. Mean chlorophyll a content of five Ma strains grown in full P media following a

96-hr exposure to copper sulfate pentahydrate. Error bars represent one standard error from the mean ...... 109

Figure 4. Mean cell densities of five Ma strains grown in full P media following a 96-hr exposure to copper sulfate pentahydrate. Error bars represent one standard error from the mean ...... 110

Figure 5. Mean chlorophyll a content of five Ma strains grown in 0.1 P media following a

96-hr copper exposure to copper sulfate pentahydrate. Error bars represent one standard error from the mean ...... 111

Figure 6. Mean cell densities of five Ma strains grown in 0.1 P media following a 96-hr exposure to copper sulfate pentahydrate. Error bars represent one standard error from the mean ...... 112

Figure 7. Mean chlorophyll a content of five Ma strains grown in 0.05 P media following a 96-hr exposure to copper sulfate pentahydrate. Error bars represent one standard error from the mean ...... 113

xiii

Figure 8. Mean cell densities of five Ma strains grown in 0.05 P media following a 96-hr

exposure to copper sulfate pentahydrate. Error bars represent one standard error from the

mean ...... 114

Figure 9. Cumulative mean 96-hr LC50 values based on chlorophyll a content for five Ma

strains grown in full, 0.1 P, and 0.05 P media following exposure to copper sulfate

pentahydrate. Error bars represent 95% confidence intervals around the mean ...... 115

Figure 10. Cumulative mean 96-hr LC50 values based on cell densities for five Ma strains

grown in full, 0.1 P, and 0.05 P media following exposure to copper sulfate pentahydrate.

Error bars represent 95% confidence intervals around the mean ...... 116

CHAPTER FOUR: Comparison of Partitioning and Efficacy Between Copper

Algaecide Formulations: Refining the Critical Burden Concept

Figure 1. Percent viable L. wollei filaments with a series of copper exposures from three

different algaecide formulations. The horizontal reference line represents an 85% decrease

compared with untreated controls following a 7-day exposure duration. Error bars

represent 95% confidence intervals around the mean. Nonlinear regression was performed

using an exponential decay, single two-parameter equation y = a*exp(-b*x) ...... 158

Figure 2. Chlorophyll a content of L. wollei with a series of copper exposures from three

different algaecide formulations. The horizontal reference line represents an 85% decrease

compared with untreated controls following a 7-day exposure duration. Error bars

represent 95% confidence intervals around the mean. Nonlinear regression was performed

using an exponential decay, single three-parameter equation y = y0+a*exp(-b*x) ...... 159

xiv

Figure 3. Percent viable P. varia filaments with a series of copper exposures from three different algaecide formulations. The horizontal reference line represents an 85% decrease compared with untreated controls following a 7-day exposure duration. Error bars represent 95% confidence intervals around the mean. Nonlinear regression was performed using the polynomial quadratic equation y = y0+a*x+b*x^2...... 160

Figure 4. Chlorophyll a content of P. varia with a series of copper exposures from three different algaecide formulations. The horizontal reference line represents an 85% decrease compared with untreated controls following a 7-day exposure duration. Error bars represent 95% confidence intervals around the mean. Nonlinear regression was performed using an exponential decay, single three parameter equation y = y0+a*exp(-b*x)...... 161

Figure 5. Adsorbed copper to L. wollei following a 7-day exposure duration with a series of copper exposures from three different algaecide formulations. Error bars represent 95% confidence intervals around the mean. Different letters represent significant differences between formulations at a specific copper exposure (α = 0.05)...... 162

Figure 6. Adsorbed copper to P. varia following a 7-day exposure duration with a series of copper exposures from three different algaecide formulations. Error bars represent 95% confidence intervals around the mean. Different letters represent significant differences between formulations at a specific copper exposure (α = 0.05)...... 163

Figure 7. Infused copper in L. wollei following a 7-day exposure duration with a series of copper exposures from three different algaecide formulations. Error bars represent 95%

confidence intervals around the mean. Different letters represent significant differences between formulations at a specific copper exposure (α = 0.05)...... 164

Figure 8. Infused copper in P. varia following a 7-day exposure duration with a series of copper exposures from three different algaecide formulations. Error bars represent 95% confidence intervals around the mean. Different letters represent significant differences between formulations at a specific copper exposure (α = 0.05)...... 165

Figure 9. Relationship between infused copper and viable L. wollei filaments following a

7-day exposure with three copper formulations. The trend line shows a linear relationship for all data (R2 = 0.920, P < 0.0001). Error bars represent 95% confidence intervals around the mean of both infused copper and response ...... 166

Figure 10. Relationship between infused copper and L. wollei chlorophyll a content following a 7-day exposure with three copper formulations. The trend line shows a linear relationship for all data (R2 = 0.935, P < 0.0001). Error bars represent 95% confidence intervals around the mean of both infused copper and response...... 167

Figure 11. Relationship between infused copper and viable P. varia filaments following a

7-day exposure with three copper formulations. The trend line shows a linear relationship for all data (R2 = 0.807, P < 0.0001). Error bars represent 95% confidence intervals around the mean of both infused copper and response ...... 168

Figure 12. Relationship between infused copper and P. varia chlorophyll a content following a 7-day exposure with three copper formulations. The trend line shows a linear

xvi

relationship for all data (R2 = 0.826, P < 0.0001). Error bars represent 95% confidence

intervals around the mean of both infused copper and response ...... 169

Figure 13. Relationship between adsorbed copper and viable L. wollei filaments following

a 7-day exposure with three copper formulations. The horizontal reference line represents

an 85% decrease in response compared with untreated controls. Error bars represent 95%

confidence intervals around the mean of both adsorbed copper and response ...... 170

Figure 14. Relationship between adsorbed copper and L. wollei chlorophyll a content

following a 7-day exposure with three copper formulations. The horizontal reference line

represents an 85% decrease in response compared with untreated controls. Error bars

represent 95% confidence intervals around the mean of both adsorbed copper and

response ...... 171

CHAPTER FIVE: Evaluation of Risks from Copper Algaecide Applications: The

Competing Ligand Hypothesis

Figure 1. Mean free and total copper concentrations in each treatment over the 48-hr

exposure duration. Error bars represent one standard deviation from the mean ...... 211

Figure 2. Mean free copper compared with total nominal copper concentrations at 2, 24,

and 48 hr after copper sulfate treatments with no algae present. Error bars represent one

standard deviation from the mean ...... 212

Figure 3. Mean free copper compared with total nominal copper concentrations at 2, 24,

and 48 hr after Captain treatments with no algae present. Error bars represent one standard

deviation from the mean...... 213

xvii

Figure 4. Mean free copper compared with total nominal copper concentrations at 2, 24, and 48 hr after copper sulfate treatments containing 5 x 105 cells/mL of algae. Error bars represent one standard deviation from the mean...... 214

Figure 5. Mean free copper compared with total nominal copper concentrations at 2, 24, and 48 hr after Captain treatments containing 5 x 105 cells/mL of algae. Error bars represent one standard deviation from the mean...... 215

Figure 6. Mean free copper compared with total nominal copper concentrations at 2, 24, and 48 hr after copper sulfate treatments containing 5 x 106 cells/mL of algae. Error bars represent one standard deviation from the mean...... 216

Figure 7. Mean free copper compared with total nominal copper concentrations at 2, 24, and 48 hr after Captain treatments containing 5 x 106 cells/mL of algae. Error bars represent one standard deviation from the mean...... 217

Figure 8. Mean Daphnia magna 48-hr LC50 values for each treatment based on total copper and free copper in exposures. Error bars represent a 95% confidence interval around the mean ...... 218

Figure 9. Mean percent mortality of Daphnia magna in copper sulfate treatments consisting of 1) no algae, 2) 5 x 105 cells/mL, and 3) 5 x 106 cells/mL. Regression equations listed for each exposure-response relationship. Error bars represent a 95% confidence interval around the mean ...... 219

Figure 10. Mean percent mortality of Daphnia magna in Captain treatments consisting of

1) no algae, 2) 5 x 105 cells/mL, and 3) 5 x 106 cells/mL. Regression equations listed for

xviii

each exposure-response relationship. Error bars represent a 95% confidence interval around the mean ...... 220

Figure 11. Mean free and sorbed copper (bars) on primary Y axis and 48-hr LC50 values

(lines) for total and free copper concentrations on the secondary Y axis for copper sulfate

(left) and Captain (right). Error bars are 95% confidence intervals around the mean ...... 221

Figure 12. Mean percent mortality of Daphnia magna compared with percent algal inhibition for 5 x 105 cells/mL exposures to copper sulfate and Captain ...... 222

Figure 13. Mean percent mortality of Daphnia magna compared with percent algal inhibition for 5 x 106 cells/mL exposures to copper sulfate and Captain ...... 223

xix

CHAPTER ONE

INTRODUCTION

OVERVIEW

Concomitant with human population growth, an increasing demand on freshwater resources arises. As freshwater resources are critical for all aspects of food development

(e.g., agriculture, livestock), drinking water, and energy production, demands are projected to continually increase dramatically (Spang 2012). Also, with an anthropocentric desire for pond and lake proximity and use (e.g., property values, recreation), the need for safe, usable water is crucial. Factors that threaten the availability and quality of freshwaters pose substantial economic and ecological risks to humans.

Excessive growths of algae can cause significant disruption of critical water resource uses including drinking, irrigation, recreation as well as be aesthetically displeasing (Paerl

1988; Wetzel 2001; Hoagland et al. 2002; Landsberg 2002; Briand et al. 2003). Algal biomass can restrict industrial water resource uses such as decreased cooling capacity or clogging of intakes in treatment systems. Algal growth dynamics or population senescence

(i.e., organic matter degradation) can create dissolved oxygen decreases and cause dead zones. Mat and scum formation and increased turbidity can impact wildlife habitat and ecological structure (Casanova et al. 1999; Burkholder and Glibert 2013). In many cases, algal densities may be minimal while still causing significant health concerns and ecosystem risks (Chorus and Bartram 1999). Cyanobacterial toxin and taste/odor compound production has become an increasing priority for water resource managers due to associated risks to

humans, domestic pets, livestock, and wildlife who depend on the water resource (WHO

1993; Falconer 1999; Carmichael et al. 2001; Zurawell et al. 2005). Since these harmful and aesthetically displeasing compounds are unabated with many typical treatment programs, and in light of continued discovery of new harmful compounds produced by algae, management strategies that proactively address noxious algae or more efficiently combat reactively are needed. By understanding the effectiveness and associated risks of algal management programs, informed decisions can be made regarding implementation to achieve water resource use objectives. Scientifically defensible research regarding implementation of management decisions in freshwater resources is critical to maintaining the integrity, functionality and uses thereof. This research framework is an integrated spectrum used to assess the ability of nutrient mitigation to control algal assemblage composition and alter sensitivities to control techniques, and to further understand use efficiency and risks associated with algaecides applied to reactively mitigate algal infestations.

EUTROPHICATION AND ENVIRONMENTAL CONDITIONS

Eutrophication has been described as a ‘wicked problem’ affecting freshwaters throughout the world, and a multi-dimensional approach will be needed to address this on a large-scale (Thornton et al. 2013). Increasing anthropogenic activities with population growth further promote nutrient inputs to freshwaters from fertilizer use, soil erosion, livestock operations and septic systems (Townsend and Howarth 2010; Paerl et al. 2016).

Nonpoint source nutrient inputs can be difficult and costly to curtail (Schindler and

Vallentyne 2008; United States Environmental Protection Agency [USEPA] 2010). The

recovery of systems following nutrient removal are variable and complex, since many factors alter this transition including species invasions, overfishing, flushing and climate change

(Duarte et al. 2008; Joehnk et al. 2008; Glibert 2010). In many scenarios, reductions in nutrient loads alone cannot be expected to cause systems to return to earlier baseline conditions (Duarte et al. 2008; Paerl 2014). Many freshwater systems may not be as responsive to changes in nutrient loads alone, due to compounded effects of climate warming on promoting nuisance blooms (Jeppesen et al. 2005; Kosten et al. 2012). Furthermore, in situ nitrogen-fixation by diazotrophic cyanobacteria can fuel the growth of potentially toxic, non-heterocystous cyanobacteria under low nitrogen conditions, that may be achieved by management (Beversdorf et al. 2013).

Deposited cyanobacterial biomass in sediments can also increase the rate of phosphorus cycling and maintain the eutrophic status (Zhao et al. 2012; Chen et al. 2014). As the climate warms, amplification of stratification duration and hypolimnetic temperature are expected to result in an increase in the severity of benthic dissolved oxygen depletion (Fang and Stefan 1997). Temperature increases in sediments also have a large impact on iron reduction and phosphate release (Robador et al. 2009; Chen et al. 2014). Stagnant waters

(decreased flushing) from increased withdrawal rates may further promote nuisance cyanobacteria as this condition provides an advantage for buoyancy mechanisms and access to legacy nutrient stores (Wilhelm and Adrian 2008).

Nutrient loads from watersheds have accumulated through long-term agricultural activity, and even if additional amendments were managed, massive continuing phosphorus loads (e.g., 500 metric tons/year into Lake Okeechobee) are expected in many systems for

decades (Reddy et al. 2011). Internal cycling of historically accumulated phosphorus, referred to as legacy phosphorus, can be a significant, ongoing source of phosphorus loading that supports deleterious water quality impacts, even if external loads could be curtailed

(Nürnberg 1997; Søndergaard et al. 2003; Schindler 2012). For example, Jarvie et al. (2013) discussed how, after decades of loading into aquatic systems, the internal fueling and subsequent accessibility (primarily by cyanobacteria) can continually fuel blooms for decades.

Reductions in anthropogenic phosphorus loads may initially result in a decline in phytoplankton biomass, but the sediment “pump” of stored phosphorus can replenish supplies in the water-column, promoting productivity (Glibert et al. 2011). Sediment phosphorus fluxes due to altered benthic biogeochemistry and feces/other metabolic wastes of animals can also provide a source for nuisance algae (Burkholder 2009). External nutrient mitigation management strategies will have more difficulty achieving results in aquatic systems due to historic accumulations (USEPA 2007), increasing temperature (Kosten et al.

2012), and dynamic hydrological conditions (Paerl et al. 2016). Additionally, restrictions on nutrient loading may be insufficient to achieve management objectives as many regulations are set above eutrophication thresholds or based on shifted baseline values (Paerl et al. 2016).

Enhanced internal phosphorus loading, more prevalent in warmer temperatures, would also strongly counteract any potential reduction in external nutrient load (Kleeberg and Dudel 1997; Søndergaard et al. 2003; Kohler et al. 2005). Acquisition of these nutrients below the thermocline are favored by nuisance cyanobacterial species with buoyancy regulation capabilities or benthic mat-forming algae (Wilhelm and Adrian 2008;

Vanderploeg et al. 2010). Nuisance algae can store nutrients from deeper waters and then rise toward the surface to fuel epilimnetic blooms (Perakis et al. 1996). Moreover, hotter climate conditions may select for more toxic strains and result in increased bloom toxicity (Davis et al. 2009; Dziallas and Grossart 2011). For example, Anabaena and Microcystis were found to produce highest levels of toxin between 18° and 25°C (Sivonen and Jones 1999). Increasing temperatures and interactions with legacy phosphorus and ongoing eutrophication from the watershed pose unprecedented challenges in restoring water resources.

WATER DYNAMICS

Increased climate warming will make it more difficult to offset cyanobacterial dominance with nutrient mitigation, especially in systems where internal accumulations are present (Jeppesen et al. 2005; Carey et al. 2008). Warmer temperatures can promote stratification initiation earlier in spring and continually later in the fall (Peeters et al. 2007), as well as a more intense thermal stratification gradient (Paerl and Huisman 2009). Diel vertical migration and tolerance of varying light conditions enable cyanobacteria to adapt during blooms, which are prevalent in warmer water temperatures (Reynolds 2006).

Maximum growth rates are attained by most cyanobacteria at temperatures above 25°C

(Robarts and Zohary 1986). At these elevated temperatures (> 24°C), cyanobacteria can more readily outcompete green algae and diatoms across many nutrient regimes (Burkholder and

Glibert 2013 and references therein). Cyanobacteria can irradiate heat to raise the water temperature further and thereby negatively affect other algae, which promotes cyanobacterial dominance in eutrophic waters.

Climate change and extreme weather patterns are expected to become more intense, such as high precipitation events, temperature variability, and droughts (International Panel on Climate Change [IPCC] 2007). These conditions will further promote nuisance algae because of their ability to ascertain episodic nutrient inputs associated with high runoff events, and to tolerate altered light, salinity, water temperature, and stratification changes

(O’Neil et al. 2012; Paerl and Paul 2012). More intense and episodic rainfall events (Paerl and Huisman 2009) may promote increased nutrient inputs from runoff due to decreased efficacy of watershed management practices (Trenberth 2005). Additionally, increased impervious surfaces (Reed et al. 2010) and more land shifted to agricultural use (Ramankutty and Foley 1999) has created more runoff than from watersheds historically. In some areas, more intense use of water may actually decrease flushing potential and concentrate nutrients.

Romo et al. (2013) found that cyanobacteria were dominant under episodic drought conditions. Some cyanobacteria have the ability to move toward water, and have other mechanisms to help prevent desiccation (e.g., mucilage, sheath), so that increased evaporation/water use creating transient water systems may promote their dominance and early colonization (Pringault and Garcia-Pichel 2004).

An intensified need for water is projected in many regions throughout the US to accommodate the growing population, primarily for domestic and public use, energy, and agriculture (Brown 1999). This need, coupled with a projected hotter, drier climate, is expected to result in less flushing of water downstream, increases in thermal gradients, and concentration of nutrients (Wetzel 2001). Droughts, more prevalent with climatic changes, impact water availability as well as nutrient forms and locations in water resources (Milly et

al. 2005). The dry conditions promote low redox in the sediments and release of ammonia and phosphate (Wetzel 2001), as well as increase nutrient levels due to decreased dilution of waste water treatment discharge (Jassby 2005). With accelerated or “actual” eutrophication in many water bodies and storage reservoirs, there is less storage capacity and increase in organic matter accumulation. These conditions promote nuisance cyanobacteria, which have significant advantages in stagnant and stratified water conditions (Reynolds and Walsby

1975; Carey et al. 2008).

MAGNITUDE OF THE ISSUE

Nuisance algal blooms are a ubiquitous problem globally (Chorus and Bartram 1999;

Anderson et al. 2002), and are increasing globally (Hallegraeff 1993; GEOHAB 2006;

Heisler et al. 2008). Nutrient levels are the second biggest impairment in US freshwater resources, with 50% of waters sampled classified as eutrophic/hypereutrophic (USEPA

2007). The US has been intensely impacted by harmful algal blooms. Microcystins have been detected in 39% of the streams (n=75) with median detected concentrations of 0.29 µg/L in the southeastern US (Loftin et al. 2016a). Cyanobacteria have been found in 98% of the

1,161 lakes and reservoirs sampled throughout the US (Loftin et al. 2016b). Toxigenic anatoxin-, cylindrospermopsin-, microcystin-, and saxitoxin-producing cyanobacteria were present in 81, 67, 95, and 79 percent of samples, respectively. Of the 1,161 lakes analyzed,

32% of lakes samples had detectable microcystin, 4% cylindrospermosins, and 7.6% saxitoxins (Loftin et al. 2016c). A drinking water survey from 45 utility companies throughout the US revealed that 80% of 677 samples analyzed contained microcystins, and

4% of the positive results were above the World Health Organization (WHO) guideline risk level discussed below (Chorus and Bartram 1999; Carmichael 2001). Cyanobacterial blooms have increased in frequency, distribution, and intensity throughout the US and are projected to continually increase (Paerl and Otten 2013).

REGULATORY ASSESSMENT

The WHO (1999) published guidelines, based on documented risk assessments for cyanobacterial toxins, suggesting 1 µg/L and 20 µg/L microcystin LR as the maximum allowable level for drinking waters and recreational waters, respectively. Microcystin LR was classified as a possible human carcinogen (group 2B) by the WHO (Grosse et al. 2006).

Only recently has the USEPA (2012) considered regulations of cyanobacterial toxins for drinking water risks, limited to only three specific toxins including microcystin-LR, anatoxin-a, and cylindrospermopsin. Thus, a single analogue (LR) of over 90 different microcystin congeners was considered (USEPA 2015a). For most toxins and analogues, there are no readily available standards or standardized methods for analysis (Paerl and Otten

2013). Microcystins and cylindrospermopsins can be altered through chlorination and typically decrease in toxicity, though harmful disinfection by-products of concern can be formed. Other toxins like anatoxin-a may not be responsive to chlorination, and are not reduced in concentration or eliminated in routine drinking water treatment processes (Merel et al. 2010). Some potable water treatment plants have voluntarily adopted monitoring assessments for a few toxins, though many have chosen not to because it is costly and not specifically mandated (Burkholder et al. 2010). This would also discount their stance of

deniability. If they do not analyze for cyanotoxins, there are no data to report to constituents.

Nevertheless, the USEPA, acknowledging sufficient published data, has at least issued health advisories for two cyanotoxins that recommend levels below 0.3 µg/L for microcystins and

0.7 µg/L for cylindrospermopsin in 10-day drinking water exposures for children less than six years old (USEPA 2015a, b). Despite these designated risk levels in drinking water and data supporting human health risks (Lun et al. 2002), few states (e.g., Minnesota, Ohio,

Oregon, Vermont) have mandated toxin guidelines and no national regulation exists.

MANAGEMENT OVERVIEW

Many freshwaters have exceeded a critical threshold level where nutrient input reductions may be futile in supporting short-term changes in lake systems and insufficient to cause and maintain long-term changes. Since legacy nutrients have accumulated in freshwaters, reducing external inputs may fail to address nutrient sources available to algae.

In particular, algae adapted to attaining internal nutrient stores have a further advantage, and are more likely to dominate assemblages, if external focus is the only approach.

Stoichiometric nutrient ratios and threshold levels are critical to understand to address eutrophication in freshwaters (Downing et al. 2001; Burkholder 2009; Glibert et al. 2011;

Carvalho et al. 2013; Fastner et al. 2015). Diversity in nuisance algae may alter how the assemblage responds to mitigation techniques; some species may be more responsive to changes in nutrient availability, whereas others are more sensitive to changes in temperature.

Therefore, management strategies need to be comprehensive in order to mitigate diverse types of nuisance algal blooms (Rigosi et al. 2014).

Management strategies are often focused on what is perceived to be the source of nuisance algal blooms (e.g., watershed, nutrient inputs) rather than the algal bloom directly.

Even if all inputs are reduced, it may take many years for the water resource to return to a pre-eutrophication state (Jarvie et al. 2013). Even oligotrophic defined systems (by current measures) are having recurring and substantial cyanobacterial blooms driven primarily by sediment nutrient acquisition and storage (Carey et al. 2008). Decreased nutrient inputs externally into these systems (low already) are not likely to change the current abundance and bloom dynamics and may further promote a long-term cyanobacterial dominated system.

This research evaluated direct strategies focused on existing, in-water problems and impacts directly to the nuisance organisms. This approach avoids predictions of responses from theoretical reductions in inputs, which often may be misleading in terms of direct correlations with nuisance algae due to unique nutrient acquisition strategies. Water resource managers may choose to monitor the bloom and put up signs to close the lake, often citing risks of the management approach for reasons not to manage. Not managing may, however, allow the algae to persist and gain a further “foothold” in the system. This could decrease the ability to effectively manage in the future, or could limit the efficacy of management approaches that may have been applicable for small-scale infestations.

PROACTIVE MANAGEMENT

Management of aquatic ecosystems must incorporate multiple levels of organization for a long-term balance. Factors that cause a specific organism or population to achieve nuisance levels or fail to be balanced within the ecosystem are often a function of an altered

environmental condition. To change the system, restore equilibrium, and suppress that same nuisance species from consistently dominating in the future, specifically tailored proactive strategies are required. Only targeting the algae in a management program may not holistically address the causative factor responsible for its presence and dominance. Proactive management is often directed at a few factors (e.g., light, temperature, nutrients).

Understanding the influence of these factors in accordance with ecosystem characteristics can allow more efficient development of proactive, strategic management approaches.

Proactive management is of particular interest due to the identification of crucial environmental factors governing presence and dominance of numerous nuisance algal blooms (Burkholder 2009). Abating the presence of blooms concomitantly decreases risks of toxin exposure as well as offsets the need for reactive solutions. Employment of new technology concomitant with in situ validation is needed to support the efficiency of novel proactive management approaches.

Nutrient mitigation from external and in situ sources as well as co-management of multiple nutrients and their interrelationship are critical factors in proactive management.

Phosphorus levels and stoichiometric nutrient ratios (e.g., N:P, Si:P) can govern algal productivity (Carpenter et al. 2008) and problematic algal presence and dominance (Rhee

1982; Tilman et al. 1982). Excessive phosphorus inputs and altered nutrient ratios have significantly increased the frequency and distribution of toxin-producing cyanobacterial blooms (Hallegraeff 1993; Watson et al. 1997). As many cyanobacteria can fix atmospheric nitrogen (Anabaena, Aphanizomenon, Lyngbya, Nostoc, etc.), lower ratios between nitrogen and phosphorus can select for these nuisance cyanobacteria (Smith 1983; Seale et al. 1987;

Paerl 1990, 1991; Ghadouani et al. 2003). Paerl et al. (2016) discusses limited techniques for immobilizing nitrogen forms in water and denitrification as the only effective in situ mitigation approach which may be limited in scope. Schindler et al. (2008) concluded that phosphorus reduction and the stoichiometric ratio of nutrients (often N:P) should be the focus of efforts to manage eutrophication in freshwaters as well as govern algal assemblage composition. Decreasing watershed phosphorus inputs can provide benefits to water quality

(Jeppesen et al. 2005), however, the internal cycling of legacy phosphorus can be a significant, ongoing source of phosphorus loading that supports deleterious water quality impacts (Jarvie et al. 2013). Solutions that decrease bio-available phosphorus and increase the N:P ratio in aquatic systems need to be assessed in terms of ability to combat the eutrophication process, mitigate accumulated phosphorus, and alter algal assemblage composition.

INTEGRATED MANAGEMENT

With increasing eutrophication in freshwaters and corresponding harmful algal blooms, nutrient mitigation (e.g., decreasing loading or availability) should be a key focus of management (Hallegraeff 1993; Heisler et al. 2008). The implications of eutrophication may go beyond simply promoting the presence and growth of noxious algae. In addition to structuring an algal assemblage, nutrient levels can affect the susceptibility of algae to reactive control techniques as algal health often parallels essential nutrient availability.

Depending on the physiological requirements of a specific algal strain, the impacts of altered nutrient regimes should differ accordingly. Increased tolerance can be observed with

sufficient nutrient availability and satiation (Rhee 1973; Twiss and Nalewajko 1992), or increased sensitivity can occur when an essential nutrient becomes limiting. Based on the greater physiological need of many cyanobacteria for phosphorus, reduced phosphorus supplies would be expected to enhance susceptibility as compared with other species, so that the efficacy of reactive management strategies increase. By evaluating the increase in sensitivity to altered phosphorus regimes, more efficient management, incorporating multiple approaches (e.g., phosphorus mitigation and copper exposures), can be implemented.

REACTIVE MANAGEMENT

Chemical amendment to directly kill or suppress an algal bloom is often a large component of reactive management. When critical water resource uses are obviated by nuisance algal species and require immediate attention, algaecides are commonly selected due to the (potential) rapid ability to control nuisance algae (Mastin et al. 2002). Numerous algaecidal active ingredients are registered by the USEPA to mitigate algal blooms in freshwaters, including diquat dibromide, endothall, acrolein, peroxide, copper and diuron.

Many of these are restricted to localized sites or have selective efficacy at approved use rates.

Copper is overwhelmingly the most widely used algaecide in the US (Oliveira-Filho et al.

2004), and copper-based algaecides have been used for more than a century (Clark 1902;

Flemming and Trevors 1989).

Copper algaecides exist in many formulations, and most are designed to increase formulation stability, decrease corrosivity, and increase efficacy toward nuisance algae

(Freedenthal et al. 1985). The amount of elemental copper applied to US water bodies for

aquatic weed and algae control is estimated at 4,218 to 5,125 metric tons annually

(9,000,000-11,000,000 pounds; USEPA 2009). Non-target toxicity studies are used to create copper product labels, as approved by USEPA Office of Pesticide Programs, to offset unreasonable risks to humans and other non-target species. The re-registration eligibility decision recently approved continued use of copper in aquatic environments factoring in application rate, frequency, and environmental risk (USEPA 2009). This information is reflected in USEPA registered copper pesticide labels to minimize adverse impacts.

Advances in copper use efficiency can decrease the amount needed to achieve desired control and alter potential risks to non-target species.

Algaecide Efficiency

Recent governmental regulations have sought to minimize discharges resulting from application of pesticides (USEPA 2011). Algal management strategies are needed that maintain the desired level of control while decreasing overall algaecide inputs. This research investigates a mechanism that accounts for the most significant component of an exposure

(i.e., the organism), and how its inherent sensitivity can be shifted. By altering the specific susceptibility of nuisance algae, the efficiency (i.e., degree of response per applied amount) of reactive management strategies can be significantly improved due to a decrease in amount of product needed, increase in selectivity, and more significant impacts to the target organism. Additionally, understanding the activity of reactive solutions allows for comparison and selection of the most efficient product to achieve desired management objectives. This research defines a method to assess the copper amount and location required to achieve control of a specific algal infestation, and compares efficiency of different copper

formulations. By understanding the affinity, uptake, and correlation of the mass of copper applied to the desired response, the most effective copper algaecide formulation can be selected and amount of reactive product required to achieve control can be specifically targeted. Based on the amount of algae present and formulation efficiency at achieving the critical burden, the reactive algaecide rate necessary for control can be calculated to minimize discharges and avoid excessive use (Fitzgerald 1964; Bishop and Rodgers 2012).

The copper formulation with the highest affinity to the algae and efficiency of control (lowest critical burden) will hypothetically result in the lowest aqueous copper exposures available to non-target species (Murray-Gulde et al. 2002). This research will provide more efficient use of copper algaecide formulations (i.e., identifying most efficacious formulation and mass/volume of product required) while decreasing cost and non-target species risks.

Algaecide Risk Evaluation

Immediate restoration of water resource uses is often required when nuisance algae proliferate. As chemical management of problematic algae is an integral component to rapid and selective remediation of risks associated with nuisance blooms, accurate data are needed to support informed decisions about algaecide use. As explained, copper algaecides have been widely used to rapidly and effectively control nuisance algal infestations and restore water resource uses (Mastin et al. 2002). Ecological concerns associated with copper-based algaecides have been raised based upon measured copper toxicity to non-target organisms.

As these products are often designed to specifically target algae, when applied correctly, non- target copper toxicity is rarely observed in field situations. Research regarding accurate risk assessment following application in real-world situations is needed to clarify the perceived

risks extracted from non-representative laboratory toxicity evaluations. As copper rapidly adheres to algae upon application in aquatic systems (Crist et al. 1990; Levy et al. 2007), and acute risks to non-target species primarily comes from dissolved fractions (Meyer et al.

1999), the bio-available copper exposure is significantly altered as compared to standardized toxicity studies (Sprague 1973). The competing ligand hypothesis assesses the influence of algae in altering the exposure of copper to non-target organisms. This will provide data on the targeted uptake rates and extents of copper from copper-based algaecides and provide environmentally relevant margins of safety associated with predicted exposures in field applications. This research will assist water resource managers in accurate risk assessments associated with common reactive algal management programs.

ORGANIZATION OF DISSERTATION

This dissertation is organized into four publishable research chapters. Each addresses and integrates different approaches for managing algae and water quality in freshwaters including: I) internal phosphorus mitigation in proactive algal management, II) influence of cyanobacterial strain and phosphorus regime in altering sensitivity to reactive management,

III) comparative efficiency of copper formulations for enhancing algaecidal impact, and IV) accurate risk assessment associated with reactive copper algaecide applications. A final chapter of conclusions summarizes research results to establish a recommended framework for the decision information system. The chapters have been formatted to align with corresponding journal requirements. The research chapters, in terms of title and targeted journal for submission, are described below:

Chapter 2: Impacts of Phosphorus Inactivation on Algal Assemblage Composition and

Water Quality Characteristics in Hypereutrophic Water Resources

o Targeted journal, Environmental Science and Pollution Research

Chapter 3: Sensitivity of Microcystis aeruginosa Strains to Copper and Influence of

Phosphorus

o Targeted journal, Journal of Aquatic Plant Management

Chapter 4: Comparison of Partitioning and Efficacy Between Copper Algaecide

Formulations: Refining the Critical Burden Concept

o Targeted journal, Archives of Environmental Contamination and Toxicology

Chapter 5: Evaluation of Risks from Copper Algaecide Applications: The Competing

Ligand Hypothesis

o Targeted journal, Ecotoxicology

Chapter 6: Conclusions and a Recommended Decision Information System Framework

REFERENCES

Anderson, DM, Glibert PM, Burkholder JM. 2002. Harmful algal blooms and eutrophication: nutrient sources, composition and consequences. Estuaries 25: 562-584.

Beversdorf, LJ, Miller TR, McMahon KD. 2013. The role of nitrogen fixation in cyanobacterial bloom toxicity in a temperate, eutrophic lake. PLoS ONE 8: e56103.

Bishop, WM, Rodgers Jr JH. 2012. Responses of Lyngbya wollei to exposures of copper- based algaecides: The critical burden concept. Archives of Environmental Contamination and Toxicology 62: 403-410.

Briand, J-F, Jacquet S, Bernard C, Humbert J-F. 2003. Health hazards for terrestrial vertebrates from toxic cyanobacteria in surface water ecosystems. Veterinary Research 34: 361-377.

Brown, TC. 1999. Past and future freshwater use in the United States: A technical document supporting the 2000 USDA Forest Service RPA Assessment. General Technical Report, RMRS-GTR-39.

Burkholder, JM, Glibert PM. 2013. Eutrophication and oligotrophication, pp. 347-371. In: Encyclopedia of Biodiversity, 2nd edition, Volume 3, by Levin S.A. (ed.). Academic Press, Boston, MA.

Burkholder JM, Frazier W, Rothenberger MB. 2010. Source Water Assessment and Control/Treatment Strategies for Harmful and Noxious Algae. In Manual M57, Algae: Source to treatment. Denver, CO.

Burkholder, JM. 2009. Harmful algal blooms. In: Likens GE (ed.) Encyclopedia of Inland Waters, vol. 1, pp. 264-285. Elsevier, Oxford, United Kingdom.

Carey, CC, Weathers KC, Cottingham KL. 2008. Gloeotrichia echinulata blooms in an oligotrophic lake: Helpful insights from eutrophic lakes. Journal of Plankton Research 30: 893-904.

Carmichael, WW, Azevedo SMFO, An JS, Molica RJR, Jochimson EM, Lau S, Rinehart KL, Shaw GR, Eaglesham GK. 2001. Human fatalities from cyanobacteria: Chemical and biological evidence for cyanotoxins. Environmental Health Perspectives 109: 663- 668.

Carmichael, WW. 2001. Assessment of blue-green algal toxins in raw and finished drinking water. AWWA Research Foundation. American Water Works Association. Denver, CO.

Carpenter, SR. 2008. Phosphorus control is critical to mitigating eutrophication. PNAS USA 105: 11039-11040.

Carvalho, L, McDonald C, de Hoyos C, Mischke U, Phillips G, Borics G, Poikane S, Skjelbred B, Lyche Solheim A; Van Wichelen J, Cardoso AC. 2013. Sustaining recreational quality of European lakes: minimizing the health risks from algal blooms through phosphorus control. Journal of Applied Ecology 50: 315-323.

Casanova, M., Buruch MD, Brock MA, Bond PM. 1999. Does toxic Microcystis aeruginosa affect aquatic plant establishment? Environmental Toxicology 14: 97-109.

Chen, M, Ye T-R, Krumholz LR, Jiang H-L. 2014. Temperature and cyanobacterial bloom biomass influence phosphorous cycling in eutrophic lake sediments. PLoS ONE 9: e93130.

Chorus, I, Bartram J. 1999. Toxic Cyanobacteria in Water: A Guide to Public Health Significance, Monitoring and Management. WHO. Chapman & Hall, London, United Kingdom.

Clark, J. 1902. On the toxic properties of some copper compounds with special reference to Bordeaux mixture. Botanical Gazette 33: 26-48.

Crist, RH, Martin JR, Guptill PW, Eslinger JM, Crist DR. 1990. Interaction of metals and protons with algae. 2. Ion-exchange in adsorption and metal displacement by protons. Environmental Science and Technology 24: 337-342.

Davis, TW, Berry DL, Boyer GL, Gobler CJ. 2009. The effects of temperature and nutrients on the growth and dynamics of toxic and non-toxic strains of Microcystis during cyanobacteria blooms. Harmful Algae 8: 715-725.

Downing, JA, Watson SB, McCauley E. 2001. Predicting cyanobacteria dominance in lakes. Canadian Journal of Fisheries Aquatic Sciences 58: 1905-1908.

Duarte, CM, Conley DJ, Carstensen J, Sanchez-Camacho M. 2008. Return to Neverland: Shifting baselines affect eutrophication restoration targets. Estuaries and Coasts 32: 29-36.

Dziallas, C, Grossart H. 2011. Increasing oxygen radicals and water temperature select for toxic Microcystis sp. PLoS One 6: 25569.

Falconer, IR. 1999. An overview of problems caused by toxic blue-green algae (cyanobacteria) in drinking and recreational water. Environmental Toxicology 14: 5- 12.

Fang, X, Stefan HG. 1997. Simulated climate change effects on dissolved oxygen characteristics in ice-covered lakes. Ecological Modelling 103: 209-229.

Fastner, J, Abella S, Litt A, Morabito G, Vörös L, Pálffy K, Straile D, Kümmerlin R, Matthews D, Phillips MG, Chorus I. 2015. Combating cyanobacterial proliferation by avoiding or treating inflows with high P load-experiences from eight case studies. Aquatic Ecology 1-17.

Fitzgerald, GP. 1964. Factors in the testing and application of algicides. Applied Microbiology 12: 247-253.

Flemming, CA, Trevors JT. 1989. Copper toxicity and chemistry in the environment: A review. Water, Air and Soil Pollution 44: 141-158.

Freedenthal, CB, Meyers MD, Stoner GA. 1985. Copper based algaecides and aquatic herbicides. United States Patent number 4,505,734.

GEOHAB. 2006. Global Ecology and Oceanography of Harmful Algal Blooms, Harmful Algal Blooms in Eutrophic Systems. P. Glibert (ed.). IOC and SCOR, Paris, France and Baltimore, MD.

Ghadouani, A, Pinel-Alloul B, Prepas EE. 2003. Effects of experimentally induced cyanobacterial blooms on crustacean zooplankton communities. Freshwater Biology 48: 363-381.

Glibert, PM. 2010. Long-term changes in nutrient loading and stoichiometry and their relationships with changes in the food web and dominant pelagic fish species in the San Francisco Estuary, California. Reviews in Fisheries Science 18: 211-232.

Glibert, PM, Fullerton D, Burkholder JM, Cornwell JC, and Kana TM. 2011. Ecological stoichiometry, biogeochemical cycling, invasive species and aquatic food webs: San Francisco Estuary and comparative systems. Reviews in Fisheries Science 19: 358- 417.

Grosse, Y, Baan R, Straif K, Secretan B, Ghissassi F, Cogliano VE. 2006. Carcinogenicity of nitrate, nitrite, and cyanobacterial peptide toxins. Lancet Oncology 7: 628-629.

Hallegraeff, GM. 1993. A review of harmful algal blooms and their apparent global increase. Phycologia 32: 79-99.

Heisler, J, Glibert PM, Burkholder JM, Anderson DA, Cochlan WP, Dennison WC, Dortch Q, Gobler C, Heil CA, Humphries E. 2008. Eutrophication and harmful algal blooms: A scientific consensus. Harmful Algae 8: 3-13.

Hoagland, P, Anderson DM, Kaoru Y, White AW. 2002. The economic effects of harmful algal blooms in the United States: estimates, assessment issues, and information needs. Estuaries 25: 819-837.

IPCC (International Panel on Climate Change). 2007. Climate Change 2007: The Scientific Basis. Contribution of Working Group I to the fourth assessment report of the intergovernmental panel on climate change. Cambridge University Press, New York, NY.

Jarvie, HP, Sharpley AN, Spears B, Buda AR, May L, Kleinman PJ. 2013. Water quality remediation faces unprecedented challenges from "legacy phosphorus". Environmental Science and Technology 47: 8997-8998.

Jassby, AD. 2005. Phytoplankton regulation in a eutrophic tidal river (San Joaquin River, California). San Francisco Estuary and Watershed Science 3: Article 3.

Jeppesen, E, Søndergaard M, Jensen JP, Havens KE, Anneville O, Carvalho L, Coveney MF et al. 2005. Lake responses to reduced nutrient loading: an analysis of contemporary long-term data from 35 case studies. Freshwater Biology 50: 1747-1771.

Joehnk, KD, Huisman JEF, Sharples J, Sommeijer B, Visser PM, Stroom JM. 2008. Summer heatwaves promote blooms of harmful cyanobacteria. Global Change Biology 14: 495-512.

Kleeberg, A, Dudel GE. 1997. Changes in extent of phosphorus release in a shallow lake (Lake Groβer Müggelsee; Germany, Berlin) due to climatic factors and load. Marine Geology 139: 61-75.

Köhler, J, Hilt S, Adrian R, Nicklisch A, Kozerski HP, Walz N. 2005. Long‐ term response of a shallow, moderately flushed lake to reduced external phosphorus and nitrogen loading. Freshwater Biology 50: 1639-1650.

Kosten, S, Huszar VLM, Becares E, Costa LS, van Donk E, Hansson LA, Jeppesen E, Kruk K, Lacerot G, Mazzeo N, DeMeester L, Moss B, Lurling M, Noges T, Romo S, Scheffer M. 2012. Warmer climates boost cyanobacterial dominance in shallow lakes. Global Change Biology 18: 118-126.

Landsberg, JH. 2002. The effects of harmful algal blooms on aquatic creatures. Reviews in Fisheries Science 10: 113-390.

Levy, JL, Stauber JL, Jolley DF. 2007. Sensitivity of marine microalgae to copper: The effect of biotic factors on copper adsorption and toxicity. Science of the Total Environment 387: 141-154.

Loftin, KA, Dietze JE, Meyer MT, Graham JL, Maksimowicz MM, Toyne KD, 2016a. Total cylindrospermopsins, microcystins/nodularins, and saxitoxins data for the 2007 United States Environmental Protection Agency National Lake Assessment. US Geological Survey Data Series 929, 9 pp.

Loftin, KA, Graham JL, Hiborn ED, Lehmann SC, Meyer MT, Dietze JE, Griffith CB. 2016b. Cyanotoxins of inland lakes of the United States--Occurrence and potential recreational health risks in the EPA National Lakes Assessment 2007: Harmful Algae 56: 77-90.

Loftin, KA, Clark JM, Journey CA, Kolpin DW, Van Metre PC, Bradley PM. 2016c. Spatial and temporal variation in microcystins occurrence in wadeable streams in the southeastern USA. Environmental Toxicology and Chemistry 35: 2281-2287.

Lun, Z, Hai Y, Kun C. 2002. Relationship between microcystin in drinking water and colorectal cancer. Biomedical and Environmental Sciences 15: 166-171.

Mastin, BJ, Rodgers Jr JH, Deardorff TL. 2002. Risk evaluation of cyanobacteria-dominated algal blooms in a North Louisiana reservoir. Journal of Aquatic Ecosystem Stress and Recovery 9: 103-114.

Merel, S, Clément M, Thomas O. 2010. State of the art on cyanotoxins in water and their behaviour towards chlorine. Toxicon 55: 677-91.

Meyer, JS, Santore RC, Bobbitt JP, DeBrey LD, Boese CJ, Paquin PR, Allen HE, Bergman HL, Di Toro DM. 1999. Binding of nickel and copper to fish gills predicts toxicity when water hardness varies, but free-ion activity does not. Environmental Science and Technology 33: 913-916.

Milly, PC, Dunne KA, Vecchia AV. 2005. Global pattern of trends in streamflow and water availability in a changing climate. Nature 438: 347-350.

Murray-Gulde, CL, Heatley JE, Schwartzman AL, Rodgers JH Jr. 2002. Algicidal effectiveness of Clearigate, Cutrine-Plus, and copper sulfate and margins of safety associated with their use. Archives of Environmental Contamination and Toxicology 43: 19-27.

Nürnberg, G. 1997. Coping with Water Quality Problems due to Hypolimnetic Anoxia in Central Ontario Lakes. Water Quality Research Journal of Canada 32: 391-405.

Oliveira-Filho, EC, Lopes RM, Paumgartten FJR. 2004. Comparative study on the susceptibility of freshwater species to copper-based pesticides. Chemosphere 56: 369- 374.

O’Neil, JM, Davis TW, Burford MA, Gobler CJ. 2012. The rise of harmful cyanobacteria blooms: the potential roles of eutrophication and climate change. Harmful Algae 14: 313-334.

Paerl, HW. 1988. Nuisance phytoplankton blooms in coastal, estuarine, and inland waters. Limnology and Oceanography 33: 823-847.

Paerl, HW. 1990. Physiological ecology and regulation of N2 fixation in natural waters. Advances in Microbial Ecology 11: 305-344.

Paerl, HW, Prufert LE, Ambrose WW. 1991. Contemporaneous N2 fixation and oxygenic photosynthesis in the nonheterocystous mat-forming cyanobacterium Lyngbya aestuarii. Applied and Environmental Microbiology 57: 3086-3092.

Paerl, HW, Huisman J. 2009. Climate change: a catalyst for global expansion of harmful cyanobacterial blooms. Environmental Microbiology Reports 1: 27-37.

Paerl, HS, Paul VJ. 2012. Climate change: Links to global expansion of harmful cyanobacteria. Water Research 46: 1349-1363.

Paerl, HW, Otten TG. 2013. Harmful cyanobacterial blooms: causes, consequences, and controls. Microbial Ecology 65: 995-1010.

Paerl, HW. 2014. Mitigating harmful cyanobacterial blooms in a human- and climatically- impacted world. Life 4: 988-1012.

Paerl, HW, Gardner WS, Havens KE, Joyner AR, McCarthy MJJ, SE Newell, Qin B, Scott JT. 2016. Mitigating cyanobacterial harmful algal blooms in aquatic ecosystems impacted by climate change and anthropogenic nutrients. Harmful Algae 54: 213-222.

Peeters, F, Straile D, Lorke A, Livingstone DA. 2007. Earlier onset of the spring phytoplankton bloom in lakes of the temperate zone in a warmer climate. Global Change Biology 13: 1898-1909.

Perakis, SS, Welch EB, Jacoby JM. 1996. Sediment-to-water blue-green algal recruitment in response to alum and environmental factors. Hydrobiologia 318: 165-177.

Pringault, O, Garcia-Pichel F. 2004. Hydrotaxis of cyanobacteria in desert crusts. Microbial Ecology 47: 366-73.

Ramankutty, N, Foley JA. 1999. Estimating Historical Changes in Global Land Cover: Croplands from 1700 to 1992. Global Biogeochemical Cycles 13: 997-1027.

Reddy, KR, Newman S, Osborne TZ, White JR, Fitz HC. 2011. Phosphorous cycling in the Greater Everglades Ecosystem: Legacy phosphorous implications for management

and restoration. Critical Reviews in Environmental Science and Technology 41: 149- 186.

Reed, RE, Burkholder JM, Allen EH. 2010. Recent Developments in Online Monitoring. Technology for Surveillance of Algal Blooms, Potential Toxicity, and Physical– Chemical Structure in Rivers, Reservoirs, and Lakes. In Manual M57, Algae: Source to treatment. Denver, CO.

Reynolds, CS, Walsby AE. 1975. Water-blooms. Biological Reviews 50: 437-481.

Reynolds, CS. 2006. The ecology of phytoplankton. Cambridge University Press, Cambridge, United Kingdom.

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. 1982. Effects of environmental factors on phytoplankton growth. In: Marshall KC (ed.) Advances in Microbial Ecology, pp. 33-74. Plenum Press, New York, NY.

Rigosi, A, Carey CC, Ibelings BW, Brookes JD. 2014. The interaction between climate warming and eutrophication to promote cyanobacteria is dependent on trophic state and varies among taxa. Limnology and Oceanography 59: 99-114.

Robador, A, Bruchert V, Jørgensen BB. 2009. The impact of temperature change on the activity and community composition of sulfate-reducing bacteria in arctic versus temperate marine sediments. Environmental Microbiology 11: 1692-1703.

Robarts, RD, Zohary T. 1986. Influence of cyanobacterial hyperscum on heterotrophic activity of planktonic bacteria in a hypertrophic lake. Applied and Environmental Microbiology 51: 609-613.

Romo, S, Soria J, Fernandez F, Ouahid Y, Baron-Sola A. 2013. Water residence time and the dynamics of toxic cyanobacteria. Freshwater Biology 58: 513-522.

Schindler, DW, Hecky RE, Findlay DL, Stainton MP, Parker BR, Paterson M, Beaty KG, Lyng M, Kasian SEM. 2008. Eutrophication of lakes cannot be controlled by reducing nitrogen input: results of a 37-year whole ecosystem experiment. PNAS USA 105: 11254-11258.

Schindler, DW. 2012. The dilemma of controlling cultural eutrophication of lakes. Proceedings of the Royal Society B: Biological Sciences 279: 4322-4333.

Schindler, DW, Vallentyne JR. 2008. The Algal Bowl: Overfertilization of the World's Freshwaters and Estuaries. University of Alberta Press, Edmonton, Alberta, Canada.

Seale, DB, Boraas ME, Warren GJ. 1987. Effects of sodium and phosphate on growth of cyanobacteria. Water Research 21: 625-631.

Sivonen, K, Jones G. 1999. Cyanobacterial toxins. In: Chorus I, Bartram J (eds.) Toxic cyanobacteria in water: a guide to their public health consequences, monitoring and management. E&FN Spon, London, United Kingdom.

Smith, VH. 1983. Low nitrogen to phosphorus ratios favor dominance by blue-green algae in lake phytoplankton. Science 221: 669-671.

Søndergaard, Jensen JP, Jeppesen E. 2003. Role of sediment and internal loading of phosphorus in shallow lakes. Hydrobiologia 506-509: 135-145.

Spang, E. 2012. A thirst for power: A global analysis of water consumption for energy production. GWF Discussion Paper 1246, Global Water Forum, Canberra, Australia.

Sprague, JB. 1973. The ABC’s of pollutant bioassay using fish. Biological methods for assessment of water quality In: American Society for Testing and Materials. Philadelphia, PA pp. 6-30.

Thornton, JA, Harding WR, Dent M, Hart RC, Lin H, Rast CL, Rast W, Ryding S-O, Slawski TM. 2013. Eutrophication as a ‘wicked’ problem. Lakes and Reservoirs: Research and Management 18: 298-316.

Tilman, D, Kilham SS, and Kilham P. 1982. Phytoplankton community ecology: The role of limiting nutrients. Annual Review of Ecology and Systematics 13: 349-372.

Townsend, A, Howarth RW. 2010. Human acceleration of the global nitrogen cycle. Scientific American 302: 32-39.

Trenberth, KE. 2005. The impact of climate change and variability on heavy precipitation, floods, and droughts. In: Anderson, M.G. (Ed.), Encyclopedia of Hydrological Sciences. John Wiley and Sons, Ltd., Chichester, United Kingdom.

Twiss, MR, Nalewajko C. 1992. Influence of phosphorus nutrition on copper toxicity to three strains of Scenedesmus acutus (Chlorophyceae). Journal of Phycology 28: 291-298.

United States Environmental Protection Agency (USEPA). 2007. National Lakes Assessment: Field Operations Manual. EPA 841-B-07-004 U.S. Environmental Protection Agency, Office of Water and Office of Research and Development. Washington, DC.

United States Environmental Protection Agency (USEPA). 2009. Re-registration eligibility decision for coppers. 738-R-09-304. Washington, DC.

United States Environmental Protection Agency (USEPA). 2010. Water Quality Standards for the State of Florida’s Lakes and Flowing Waters. Vol. 75, No. 233. 40 CFR Part 131 [EPA–HQ–OW–2009–0596; FRL–9228–7] RIN 2040-AF11. Washington, DC.

United States Environmental Protection Agency (USEPA). 2011. National Pollutant Discharge Elimination System (NPDES). Pesticide General Permit (PGP) for Discharges from the Application of Pesticides. Washington, DC.

United States Environmental Protection Agency (USEPA). 2012. Cyanobacteria and Cyanotoxins: Information for Drinking Water Systems. EPA-810F11001. Office of Water. Washington, DC.

United States Environmental Protection Agency (USEPA). 2015a. Drinking Water Health Advisory for the Cyanobacterial Toxin Microcystin. Report EPA 820R15100. Office of Water. Washington, DC.

United States Environmental Protection Agency (USEPA). 2015b. Drinking Water Health Advisory for the Cyanobacterial Toxin Cylindrospermopsin. Report EPA 820R15101. Office of Water. Washington, DC.

Vanderploeg, HA, Liebig JR, Nalepa TF, Fahnenstiel GL, Pothoven SA. 2010. Dreissena and the disappearance of the spring phytoplankton bloom in Lake Michigan. Journal of Great Lakes Research 36: 50-59.

Watson, SB, McCauley E, Dowing JA. 1997. Patterns in phytoplankton taxonomic composition across temperate lakes of different nutrient status. Limnology and Oceanography 42: 487-495.

Wetzel, RG. 2001. Limnology: Lake and River Ecosystems, 3rd ed. Academic Press, San Diego, CA.

World Health Organization (WHO). 1993. Guidelines for drinking water quality, 2nd ed. Recommendations, vol. 1. World Health Organization, Geneva, Switzerland.

World Health Organization (WHO). 1999. Chapter 5: Safe Levels and Safe Practices, In: Toxic Cyanobacteria in Water. A guide to their public health consequences, monitoring and management. I. Chorus and J. Bartram, eds. Taylor and Francis. New York, NY.

Wilhelm, S, Adrian R. 2008. Impact of summer warming on the thermal characteristics of a polymictic lake and consequences for oxygen, nutrients and phytoplankton. Freshwater Biology 53: 226-237.

Zhao, G, Du J, Jia Y, Lv Y, Han G, Tian X. 2012. The importance of bacteria in promoting algal growth in eutrophic lakes with limited available phosphorus. Ecological Engineering 42: 107-111.

Zurawell, RW, Chen H, Burke JM, Prepas EE. 2005. Hepatotoxic cyanobacteria: A review of the biological importance of microcystins in freshwater environments. Journal of Toxicology and Environmental Health Part B 8: 1-37.

CHAPTER TWO

IMPACTS OF PHOSPHORUS INACTIVATION ON ALGAL ASSEMBLAGE

COMPOSITION AND WATER QUALITY CHARACTERISTICS IN

HYPERUETROHPIC WATER RESOURCES

ABSTRACT

Eutrophication has been accelerated in freshwaters resources and commonly results in prolific growths of nuisance algae or cyanobacteria. Solely reducing external nutrient inputs may fail to address sources available to algae such as accumulated legacy nutrients. More data are needed regarding legacy phosphorus mitigation and alteration of water-column nitrogen-to-phosphorus (N:P) ratios in management programs. In this research we evaluated the ability of an in situ phosphorus binding technology (Phoslock®) to alter available water- column and sediment phosphorus, and the subsequent impact on nutrient ratios and algal assemblage composition. Two golf course irrigation ponds with legacy nutrient loads and chronic cyanobacterial blooms were treated with Phoslock and monitored for two years post- treatment. Phoslock significantly (P < 0.05) decreased water-column total phosphorus levels and shifted mobile sediment phosphorus fractions (i.e., labile, reductant-soluble, organic) to residual fractions. Total N:P ratios (by mass) significantly increased and were sustained at over 30:1 in the Hickory Meadows irrigation pond and 100:1 in the Chockyotte irrigation pond throughout the study. Consequent changes in the algal assemblage included decreases in dominance and overall density of cyanobacteria as well as a shift away from scum-forming genera (e.g., Microcystis spp. and Anabaena sp.) to planktonic forms (e.g., Pseudanabaena

sp. and Planktolyngbya sp.). This research provides information on the efficacy of Phoslock to specifically bind in situ water-column and sediment phosphorus, subsequently shifting nutrient ratios and altering algal assemblage composition.

INTRODUCTION

Eutrophication is a global issue that has increased the intensity (i.e., duration, density, distribution) of harmful algal blooms (Carpenter et al. 1998; Anderson et al. 2002; Heisler et al. 2008). Point source nutrient reductions have led to positive effects in some systems in offsetting nuisance algae (Cullen and Forsberg 1988; Mallin et al. 2005; Van Nieuwenhuyse

2007). However, nonpoint nutrient sources like stormwater runoff (DeBusk et al. 2010), wildlife excretion (Nürnberg and LaZerte 2016), watershed accumulations (Foy et al. 1995;

Reddy et al. 2011; Dunne et al. 2011), and atmospheric deposition (Wetzel 2001) can account for substantial nutrient inputs and be difficult to reduce. Even if external loads could be adequately managed, legacy nutrients accumulated in bottom sediments can support elevated phytoplankton biomass and delay lake recovery (Marsden 1989; Jeppensen et al.

2005). Removal of phosphorus from external loads can promote cycling from benthic stores through the sediment ‘‘pump’’ (Glibert et al. 2011; Burkholder and Glibert 2013) and can actually increase water-column phosphorus (P) levels (Bachmann et al. 2003). Internal cycling of historically accumulated P can be a significant, ongoing source of P loading that can continue for decades (Nürnberg 1997; Søndergaard et al. 2001; Schindler 2012; Jarvie et al. 2013) and have a proportionally greater impact on nuisance algal growth (Wilhelm and

Adrian 2008). Many failures of nutrient mitigation efforts in lake recovery programs have

neglected the internal source of P (Søndergaard et al. 2001; Gulati and Van Donk 2002).

Addressing both internal P and watershed transport is needed to enable rapid recovery of eutrophic systems (Mehner et al. 2008). More research is needed regarding in situ P mitigation techniques in order to provide water resource managers scientifically defensible data to proactively address nuisance algae and correlative water quality issues.

Bio-available P levels may be readily depleted or reduced in the water-column due to algal assimilation, or may be innately low (Carey et al. 2008). Sediment accumulation can act as both a large sink and a source of P in aquatic systems mediated by release from both inorganic and organic compounds, depending on the environmental conditions (Marsden

1989). Exploitation of sediment P is an important component of cyanobacterial ecology, and sediment P can be a significant contributer to the available nutrient pool (Burkholder and

Glibert 2013). Benthic algae are very influential in transporting sediment-bound P (Bjork-

Ramberg 1985). Barbiero and Welch (1992) found high densities of cyanobacteria in P- enriched lake sediments, especially with low bio-available P in the epilimnion. The ability of some planktonic cyanobacteria to regulate buoyancy allows rapid uptake and storage of hypolimnetic P when epilimnetic P is limiting (Ganf and Oliver 1982; Kromkamp et al.

1989). Cyanobacteria often originate from resting stage akinetes in sediments, while accumulating P, and migrate from sediments to form blooms (Reynolds 1972; Perakis et al.

1996). The cells can divide multiple times using accumulated sediment P (Istvanovics et al.

1993). To efficiently combat nuisance cyanobacterial infestations in freshwaters, both water- column and sediment associated P accumulations must be addressed.

Bottom sediment plays an important role in structuring phytoplankton assemblages, in particular as P inputs accumulate (Lagus et al. 2007). Since sediment P is present in many forms, it is critical to understand the bio-available P supply in aquatic sediments in order to evaluate the potential contribution to water quality and algal assemblage composition.

Fractions of P in lake sediments depend on adsorption-desorption mechanisms, the chemisorption binding potential of the sediments, redox conditions and biological transformation activity (Koski-Vähälä and Hartikainen 2001). Cyanobacteria can use large amounts of sediment-associated P, including fractions that other algae cannot use (e.g., organically bound P). Therefore, an understanding of the forms and concentrations of available P are of particular interest in selecting away from nuisance algal types (Jansson et al. 1988; Dignum et al. 2004). Sequential P extractions are required to accurately assess bioavailability and correlative influence on lake productivity rather than consideration of only total P (Pettersson and Istvanovics 1988; Gonsiorczyk et al. 1998). Evaluation of management strategies that alter sediment composition toward less bio-available forms is critical in water resource management.

The stoichiometric ratio of nutrients, often referring to the nitrogen-to-phosphorus

(N:P) ratio, is an important influence on algal assemblages (Smith 1983; Schindler et al.

2008). Paerl et al. (2016) discussed limited techniques for immobilizing N forms in water, and denitrification was considered the only effective in situ mitigation approach that could be difficult to anthropogenically implement on a large-scale. Harris et al. (2014a) amended N to increase the N:P ratio (above 75:1), and found reduced cyanobacterial biovolume and microcystin production along with increased Secchi depth and zooplankton biomass. Orihel

et al. (2015) found that anoxic sediments provided P inputs and lowered the N:P ratio, favoring diazotrophic cyanobacterial blooms. Cyanobacterial blooms can enhance P release from bottom sediments, which can subsequently increase bloom intensity (Downing and

McCauley 1992; Nikolai and Dzialowski 2014). Chen et al. (2014) reported that cyanobacterial biomass deposited in sediment increased the rate and extent of P cycling, which maintained the eutrophic status regardless of external inputs. Management of external

N and P inputs is also important in maintaining ecosystem integrity (Gobler et al. 2016), though external sources must be considered in combination with internal management and availability of accumulated nutrients (Welch 2009). In this research, we sought to specifically decrease P availability in both the water and sediments in order to increase the stoichiometric

N:P ratio, and measured subsequent impacts on the algal assemblage.

Phoslock is an innovative technology consisting of a patented lanthanum-modified bentonite clay formulation designed to reduce bio-available P in water and sediments

(Douglass 2002). Phoslock was developed by the Commonwealth Scientific and Industrial

Research Organization (CSIRO) of Australia to promote P uptake while decreasing risks from dissolved lanthanum to aquatic biota. Initial reaction products of Phoslock with free P indicate formation of rhabdophane, a hydrated mineral of the formula LaPO4·nH2O

(Jonasson et al. 1988). Rhabdophane further transforms to a more stable mineral, monazite

(LaPO4), through time (Cetiner et al. 2005; Dithmer et al. 2015). The lanthanum-phosphate complex is highly insoluble and can form with low concentrations of reactants over a wide pH range (Diatloff et al. 1993; Hagheseresht et al. 2009). The efficacy of phosphate precipitation by lanthanum is greater than that of other salts evaluated (i.e., iron, aluminum;

Recht et al. 1970) and the solubility product constant is more tightly bound and stable under differing environmental conditions (Firsching and Brune 1991; Ross et al. 2008). Field-scale applications of Phoslock have led to significant decreases in water-column P concentrations and a corresponding decrease in cyanobacterial abundance (Robb et al. 2003). Additional research by Meis et al. (2012) and Bishop et al. (2014) showed decreases in bio-available sediment P fractions (labile, reductant-soluble), and a concomitant increase in the most tightly bound fraction (residual). As in situ P mitigating technologies are limited, and with the documented non-target species safety and specificity of binding in Phoslock, additional data are needed on the utility of Phoslock to combat eutrophication in freshwaters.

By understanding the ability of P mitigating technologies to inactivate bio-available P throughout aquatic systems and alter resultant algal productivity, more effective strategies can be implemented to proactively achieve management objectives. The overall purpose of this research was to evaluate the potential of Phoslock to improve water resource integrity in golf course ponds where P commonly causes severe impacts to water quality (e.g., 86.5% exceedance of regional P guidelines; Baris et al. 2010).

Specific objectives of this research were to:

1) Measure the change in free reactive phosphorus (FRP) and total phosphorus (TP)

concentrations in the water-column following Phoslock application to Chockyotte and

Hickory Meadows golf course irrigation ponds;

2) Measure the impact of Phoslock amendment on sediment P fractions (bio-available

and residual forms) through time;

3) Compare water quality parameters in Phoslock treated ponds pre- and post-

application; and

4) Compare algal assemblage composition and nutrient ratios in Phoslock treated ponds

pre- and post-application.

MATERIAL AND METHODS

Site Selection

Two golf course irrigation ponds with documented cyanobacterial blooms impacting irrigation functionality were selected in this study. Hickory Meadows irrigation pond is located in Whitakers, NC (36°07´16.2˝ N; 77°47´47.6˝ W), and Chockyotte irrigation pond is located in Weldon, NC (36°25´04.3˝ N; 77°37´14.3˝ W). Both ponds are designed irrigation systems with a small shallow shelf, deeper center, and minimal runoff (Figures 1 and 2).

Background water and sediment nutrient parameters were evaluated to determine Phoslock application amounts (Table 1). Phoslock was applied to each pond to target binding of the bio-available mass of P present in both the water and upper 5 cm sediment depth. Phoslock was applied at a rate of 100 kg Phoslock to 1 kg of targeted P (1:1 La:P ratio) for the P budget in the ponds (i.e., mass of P in water and sediment) from the one month pre-treatment sampling data. This ratio was based on the documented binding specificity and capacity of

Phoslock (Reitzel et al. 2013). Applications occurred in fall of 2014 and were monitored for two years post-treatment. Reference chambers consisted of in situ enclosures (3) made of acrylic (2 m x 15 cm diameter) and driven into the sediments to a depths of > 20 cm fastened by rebar. These were placed in three locations throughout each pond. They prevented water

and nutrients from entry into enclosures, while maintaining similar environmental conditions.

Pre-treatment water and sediment was also collected and maintained in ambient laboratory conditions to use as a reference. Conditions including air temperature (daily mean, maximum and minimum) and daily rainfall were collected using a 2900 ET Weather Station (Spectrum

Technologies, Inc.) set up in Whitakers, NC.

Water Quality Measurement Parameters

Water quality parameters were measured pre-treatment and periodically post- treatment, including pH, dissolved oxygen, temperature, alkalinity, chlorophyll a, Secchi depth, total nitrogen (nitrate, nitrite, total Kjeldahl nitrogen), TP, FRP, turbidity, and algal assemblage composition and cell densities of abundant taxa and groups. Grab samples were taken from 20 cm sampling depths. Three samples were randomly taken within 5 m of each sampling point per pond (n=3) and composited. Any visual macroscopic algal growths not in the sampling areas were additionally collected and analyzed (i.e., identification, biomass) if observed. All analyses were conducted according to Standard Methods for the Examination of Water and Watewater (SMEWW 2005) with documented quality assurance/quality control

(Table 2; USEPA 1978). Samples were analyzed at the SePRO Research and Technology

Campus (SRTC) accredited laboratory (International Organization for Standardization

17025; accredidation #77497) in Whitakers, NC. For algal analyses, samples were collected and immediately cooled to 4°C for transport to the SRTC. Homogenized subsamples (taken immediately following three inversions) of 50 mL were concentrated into a 1 mL Wildco

Sedgwick-Rafter counting slide (Cole-Parmer Inc.) and analyzed visually on a Zeiss

Axioskop 20 light microscope. Algae were identified to genus level following Prescott

(1970), Wehr and Sheath (2003), and Graham et al. (2016). Algae hereby refer to eukaryotic

‘mostly’ photosynthetic organisms whereas cyanobacteia are prokaryotic.

Total lanthanum was measured using inductively coupled plasma-optical emission spectrometry (Shimadzu ICPE 9000) at a wavelength of 379.478 nm, with a matrix-matched calibration curve from serial dilution of a 1,000 mg/L standard (Ricca Chemical Co., part

#PLA1KN; SMEWW 2005). Sampling frequency included the growing season pre- treatment, immediately pre-treatment, 15 days after treatment (DAT), monthly from April-

September and bi-monthly November through April for two years post-treatment.

Sediment Collection and Analysis

Sediments were collected using an Eckman Dredge (20 cm x 20 cm). The top 5 cm layer of sediment (homogenized by vigorous stirring) was analyzed based on bio-availabilty

(Cooke et al. 2005) and interactivity with Phoslock (Meis et al. 2012), for each of the three sites throughout the treated ponds. Sediment samples were placed in polyethylene centrifuge tubes, filled completely to minimize air space, and stored in darkness at < 4°C for transport to the SRTC for processing. Sequential extraction (6 g fresh weight initially per extraction) was used to determine relevant P fractions, and to compare concentrations of bio-available and residual forms of P following Phoslock amendments. Procedures incorporated methods modified from Chang and Jackson (1957), Kapanen (2008), and Meis et al. (2012), and included: (1) extraction in 1 M NH4Cl (10 mL) for 30 min to determine loosely adsorbed phosphorus (labile P); (2) extraction with 0.11 M NaHCO3/ 0.11 M Na2S2O4 for 1 h terminated by removal of supernatant, repeated for 5 min with fresh extractant to quantify the

P mainly bound to Fe-hydroxides or manganese (Mn) compounds (reductant-soluble P); 3)

extraction with 0.1 M (10mL) NaOH (17 h) to release hydroxide-exchangeable (FRP= metal oxide P; TP-FRP = organic P); and 4) digestion with 30% (v/ v) H2SO4 and 8% K2S2O4 at

121˚C for 30 min (apatite and residual P; Table 3). Sediment samples were taken during the growing season pre-treatment (2 months and 1 month), immediately pre-treatment (1 day), 15

DAT, 1 month after treatment (MAT) and bi-monthly for the first year and quarterly for the second year following Phoslock addition. Potentially mobile sediment P was defined as the summation of labile, reductant-soluble, metal oxide and organic fractions.

Statistics

A paired t-test was used to discern differences between the Phoslock treatment samples and references at each sampling event, although it should be noted that the integrity of reference enclosures was compromised after the 2 MAT sampling. When data were normally distributed with equal variances, a paired t-test was used to assess significant differences (P < 0.05) in response parameters between pre- and post-treatment sampling events and growing season averages. A pairwise Wilcoxon Rank-Sum test was used when data were non-parametric. Measurement of the three sampling points per pond were averaged for each sampling event and compared with the pre-treatment averages. Mobile sediment P fractions and the sum of mobile fractions were analyzed using one-way analysis of variance

(ANOVA), with differences from reference sediments identified with a Dunnett’s post hoc test (α = 0.05; McDonald 2014). All data were analyzed using Microsoft Excel® 2007 and

SigmaPlot® 12.5.

RESULTS

Water Chemistry and Environmental Parameters

General water chemistry (e.g., pH, alkalinity, hardness, conductivity) remained similar at all sites throughout the study (Table 1). Mean, high, and low temperatures were also similar throughout the growing seasons between study years (April-September 2014-

2016). Temperatures averaged 22.83, 23.13, and 23.11°C for 2014, 2015 and 2016, respectively (Figure 3). Total rainfall was also similar between growing seasons with 73.86,

65.94, and 74.24 cm received in 2014, 2015 and 2016, respectively (Figure 4). Total lanthanum was below detection (0.005 mg/L) in background samples from each site. In the

Chockyotte irrigation pond, total lanthanum averaged 0.118, 0.106, and 0.101 mg/L at 2, 4, and 8 weeks after treatment (WAT), respectively. In the Hickory Meadows irrigation pond, total lanthanum averaged 0.103, 0.084, and 0.049 mg/L at 2, 4, and 8 WAT, respectively.

Secchi depth increased significantly (P < 0.001) in both systems post-treatment throughout the growing season. Secchi depth in the Chockyotte irrigation pond averaged 0.83 m pre- treatment and 1.81 m and 1.62 m in post-treatment growing seasons 2015 and 2016, respectively (Figure 5). Secchi depth in the Hickory Meadows irrigation pond averaged 0.74 m pre-treatment and 1.60 m and 1.48 m in post-treatment growing seasons 2015 and 2016, respectively (Figure 6).

The Chockyotte irrigation pond had an average TP concentration of 63.6 µg/L pre- treatment, but TP never exceeded 37 µg/L in any post-treatment sampling event. Free reactive P averaged 18.8 µg/L pre-treatment and did not exceed 18 µg/L in any post- treatment sampling event (Figure 7). Hickory Meadows irrigation pond had an average TP of

603 µg/L pre-treatment and decreased to 101 µg/L at 1 MAT and only exceeded 100 µg/L two other times throughout all sampling events. Free reactive P averaged 14.4 µg/L pre- treatment and ranged from 6.0 - 22.3 µg/L in all post-treatment sampling events (Figure 8).

Free reactive P was not significantly different in post-treatment growing seasons compared with pre-treatment levels in either system. Total P significantly decreased (P < 0.05) in both sites in post-treatment growing seasons. Total P was also lower in Chockyotte at all sampling points post-treatment compared with pre-treatment levels and, with exception of the 2 WAT sampling, at all sampling events post-treatment in Hickory Meadows (Table 4).

Nitrogen was primarily found in the organically bound form included in TKN analyses. Very little N was measured in free forms of nitrate and nitrite (i.e., did not exceed

83 µg/L in either system). Total N in Hickory Meadows averaged 4.5 mg/L pre-treatment and 2.87 mg/L post-treatment. Total N in Chockyotte averaged 1.44 mg/L pre-treatment and

2.8 mg/L post-treatment. In addition to consistently low FRP levels, TN:TP ratios (by mass) were used to assess algal assemblage impact.

Algal Assemblages and Nutrient Ratios

The Chockyotte algal assemblage was dominated by cyanobacteria pre-treatment and correlated with TN:TP ratios < 33 by mass. Following Phoslock amendments, the TN:TP ratio significantly (P < 0.001) increased to > 150:1 and cyanobacterial abundance decreased, and cyanobacteria remained at < 30% of the algal assemblage through 20 MAT. In that period, the TN:TP ratio was less than 100:1 on only two sampling dates. Toward the end of the study, cyanobacteria accounted for ~40% of the algal assemblage even with TN:TP ratios above 75:1. However, overall cyanobacterial abundance post-treatment was still significantly

lower (≤ 5,000 cells/mL) on each sampling date in comparison to pre-treatment abundance

(~60,000 cells/mL; Figure 9). Pre-treatment cyanobacterial genera were diverse but were dominated by Microcystis sp. Post-treatment genera consisted mostly of Aphanocapsa sp.,

Planktolyngbya sp., and Pseudanabaena sp. Planktonic green algae were the most abundant taxa in post-treatment growing seasons, with a bloom of Helicodictyon sp. dominating in year one and multiple types of green algae in year two (Figure 10; Appendix A). The presence of Chara sp. was noted in one section of the pond beginning approximately one year after treatment (July 2015). Its growth was restricted to < 5% of the pond surface area, along the littoral shelf of the west edge. Maximum Chara sp. biomass occurred in August

2016 (276 g dry weight/ m2). Growing season average TN:TP ratios significantly increased

(P < 0.001) both years post-Phoslock application (Table 4).

Hickory Meadows was dominated by cyanobacteria pre-treatment and correlated with

TN:TP ratios < 10:1 by mass. The TN:TP ratio increased gradually after Phoslock treatment to > 30:1 by 4 MAT, and only decreased below 30 on three sampling dates throughout the rest of the study. On one sampling event in year one, and three in year two, cyanobacteria accounted for > 50% of the algal assemblage, even with TN:TP ratios above 30:1. However, overall cyanobacterial densities post-treatment were still significantly (P < 0.05) less (≤ 5,000 cells/mL) than pre-treatment levels (~30,000 cells/mL; Figure 11). Pre-treatment cyanobacterial genera were dominated by Microcystis spp. and Anabaena sp. Post-treatment genera consisted mostly of Aphanocapsa sp., Oscillatoria sp., and Pseudanabaena sp. Post- treatment algal assemblages were diverse and changed seasonally, with multiple chlorophytes and streptophytes prevalent through both growing seasons and euglenophytes in

year two (Figure 12; Appendix B). Growing season averages of TN:TP ratios significantly increased both years post-Phoslock application (Table 4).

Sediment Phosphorus

Significant decreases were measured in the combined Chockyotte mobile sediment fractions from 4 MAT through 13 MAT analyses (Figure 13). Specific fractions that comprised the significant decreases were labile, reductant-soluble, and organic, with 4, 7, and

10 out of 11 post-treatment sampling events showing significant decreases, respectively.

Metal oxide had significant increases at 2 WAT and 1 MAT sampling events. Apatite and residual sediment P levels increased significantly at 7 out of 11 post-treatment sampling events including 4, 6, 8, 10, 13, 16, and 22 MAT (Figure 14). Mobile sediment P growing season averages were significantly decreased in only the first year post-Phoslock application

(P < 0.05; Table 4).

Significant decreases were measured in the combined Hickory Meadows mobile sediment fractions from 1 MAT through 22 MAT analyses (e.g., 10 of 11 sampling events;

Figure 15). Specific fractions that comprised the significant decreases were labile, reductant- soluble, and organic, each with 10 out of 11 post-treatment sampling events showing significant decreases. Metal oxide had significant increases at 1, 4, 6, and 22 MAT sampling events. Apatite and residual sediment P levels increased significantly from 6 MAT through

22 MAT (e.g., 7 of 11 sampling events; Figure 16). Mobile sediment P growing season averages significantly decreased in both years post-Phoslock application (P < 0.05; Table 4).

DISCUSSION

Excessive P inputs and decreased N:P ratios have been linked to increased frequency and distribution of toxigenic cyanobacterial blooms (Smith 1983; Seale et al. 1987;

Ghadouani et al. 2003; Paerl 1990; Paerl 1991; Hallegraeff 1993). There are few approaches for removing in situ N in freshwaters (Paerl et al. 2016). Even if feasible, concomitant in situ

P mitigation would be needed to maintain a higher N:P ratio. Cyanobacteria can use diverse forms of N (O’neil et al. 2012), including atmospheric deposition and groundwater sources

(Paerl 1997; Paerl et al. 2002). Many cyanobacteria can also fix N2, depending on the amount of P available for algal metabolism (Stewart et al. 1971). Horne and Fogg (1970) found the

2 annual rate of N2 fixation from cyanobacteria at 0.04-0.29 g N/m in the English Lakes

District. Granhall and Lundgren (1971) reported that 40% of the N contribution to Lake

Erken in Sweden was from cyanobacterial N2 fixation, and that cyanobacteria were the highest seasonal contributors. Nitrogen deficits in cyanobacteria are not necessarily compensated for by N2 fixation (Howarth and Paerl 2008). However, subsequent N inputs from N2-fixing cyanobacteria have been shown to fuel the growth of non-heterocystous cyanobacteria (e.g., Microcystis sp.; Beversdorf et al. 2013). Although severe N limitation can limit production of N rich cyanotoxins (Van de Waal et al. 2014), lower N:P ratios have been associated with higher concentrations of the potent N-rich hepatotoxin, microcystin

(Orihel et al. 2012; Harris et al. 2014a).

The available water-column P content has been positively correlated with combined

N content in lakes, and could exacerbate eutrophication independent of external loads

(Stewart et al. 1971; Larsson et al. 2001; Sañudo-Wilhelmy et al. 2001). Co-management of

external N and P inputs is crucial in mitigating cyanobacterial blooms (Vollenweider 1976;

Chislock et al. 2014; Gobler et al. 2016). Since many freshwaters have been chronically polluted (USEPA 2007), an internal nutrient mitigation focus is also needed. Internal P depletion has initiated cell death or resting cell formation in cyanobacterial blooms (Walve and Larsson 2007), and P is required at a threshold amount to support bloom formation in water (Downing et al. 2001; Carvalho et al. 2013) or sediments (Istvanovics et al. 1993;

Carey et al. 2008). Schindler et al. (2008) concluded that P reduction and the stoichiometric ratio of nutrients (N:P) should be the focus of efforts to manage eutrophication in freshwaters. Cyanobacterial densities in freshwaters are often highly correlated with P concentration (Watson et al. 1997; Schindler et al. 2008). Sediments can be a large P source in freshwaters. Consideration of water-column nutrient concentrations alone is misleading in attempts to relate algal abundance to nutrients, due to the sediment pump of stored nutrients and the storage capability of many algae (Gilbert et al. 2011). By adding Phoslock to target both internal sediment and water-column P, improved statistics can lead to more reliable conclusions about relationships between algae and nutrient levels.

There are two main approaches to increasing the N:P ratio- adding N or removing P

(or both). Aluminum sulfate is commonly used for P mitigation but may also remove some N forms, and may maintain a decreased N:P ratio that promotes cyanobacterial dominance

(Jacoby et al. 1994; Harris et al. 2014b). Adding N may not be a sustainable approach due to rapid denitrification processes (Beutel et al. 2016), and could fuel existing cyanobacterial growth and/or toxin production (Paerl 1997; Burkholder and Glibert 2013; Chislock et al.

2014). This research evaluated Phoslock for specifically removing P and thereby shifting the

N:P ratio higher, as well as decreasing the P levels below thresholds commonly selecting for cyanobacteria (Trimbee and Prepas 1987; Downing et al. 2001). In other research, cyanobacteria were overwhelmingly dominant in artificial enclosures at TN:TP ratios (by mass) from 5:1 to 20:1 (Ghadouani et al. 2003). Seale et al. (1987) similarly found that cyanobacteria were dominant at low TN:TP ratios (by mass) < 4:1, with high P levels in Lake

Michigan waters. Conversely, an elevated TN:TP ratio of > 29:1 was consistently dominated by green algae (chlorophytes; Smith 1983). Sabour et al. (2009 and references therein) also showed decreased growth rates and growth optima at TN:TP ratios > 30:1 for many cyanobacteria. Harris et al. (2014a) reported that at a TN:TP ratio > 75:1, chlorophytes dominated the phytoplankton assemblage and cyanobacterial biovolume and toxin concentration was significantly decreased. This was accomplished through N addition in experimental mesocosms. Harris et al. (2014b) found that at a TN:TP ratio > 50:1, manipulated by combined P reduction and N addition, cyanobacterial biovolume and density were reduced, and chrysophyte and cryptophyte biovolume and density increased. We found great variability between the two ponds in N:P ratio change, but a similar effect on algal assemblage was observed, with shifts away from cyanobacterial abundance. Chockyotte had background TN:TP ratios that averaged 30:1 and Hickory Meadows had background around

7:1, and both sustained chronic cyanobacterial blooms mostly of Microcystis spp. (as well as

Anabaena sp. in Hickory Meadows). Large shifts were observed in the TN:TP ratios, but not on the same scale between sites. Hickory Meadows had an increased TN:TP ratio to 30:1 following Phoslock application, and Chockyotte increased to > 150:1 post-Phoslock

application. In general, Phoslock addition significantly increased the TN:TP ratio and decreased cyanobacterial abundance.

Following Phoslock applications, Robb et al. (2003) found a shift away from cyanobacterial blooms in the Vasse River (Australia), but less evident in the Canning River likely due to macrophyte cycling interference. Bishop et al. (2014) showed significant shifts away from Aphanizomenon sp. in Laguna Niguel lake, CA, throughout one growing season post-Phoslock application. Oosterhout and Lürling (2013) found that Phoslock caused a dose- dependent decrease in Anabaena sp. levels in short-term laboratory experiments. In this research, cyanobacterial abundance significantly decreased post-Phoslock application, although on some sampling dates (usually in late summer), cyanobacteria still comprised a substantial portion of the assemblage. Phoslock application also shifted the cyanobacterial assemblage to small, more fusiform growth types (e.g., Pseudanabaena, Planktolyngbya) as compared with larger colonial scum-formers pre-treatment. Levine and Schindler (1999) noted species shifts of cyanobacteria, with Pseudanabaena responding positively to increased N:P ratios while Anabaena greatly decreased. Chislock et al. (2014) reported that

Cylindrospermopsis raciborskii was dominant under both low and high N:P ratios that were attained by nutrient additions to ambient eutrophic pond water. Phosphorus mitigation significantly altered the dominance, composition, and overall density of cyanobacteria in these systems.

Several studies have shown effective water-column P binding by Phoslock (Ross et al. 2008; Haghseresht et al. 2009; Spears et al. 2016). However, assessment of shifts in sediment P post-Phoslock application is also important to interpret effectiveness. Meis et al.

(2012) found significant increases in residual sediment P fractions in the surficial 2 cm of sediment in Clatto Reservoir. Phoslock-bound P was not release-sensitive, with 79% found in refractory compounds after 28 days in laboratory experiments. Bishop et al. (2014) showed decreases in bio-available sediment P fractions (labile, reductant-soluble) and a concomitant increase in the most tightly bound fraction (residual). Similarly, we observed decreases in the labile and reductant-soluble fractions, and additional decreases in the organic sediment P fraction following Phoslock application. Temporary increases in the metal-oxide bound fraction were measured in both sites, as observed in other work (Bishop et al. 2014), and may be an intermediate step to formation of residual P. Dithimer et al. (2016) repoted that rhabdophane (LaPO4·nH2O) was the primary form of P ultimately sequestered by Phoslock in the sediments of 10 lakes. Rhabdophane was only recoverable following HCl extraction.

In this research, we also showed significant increases in the apatite/residual sediment-bound

P fractions which required, in part, acid digestion to release. Decreasing sediment P availability is a crucial aspect in many lake management programs (Søndergaard et al. 2003), and Phoslock can be used as an approach to reduce these internal stores.

CONCLUSION

Phoslock was able to decrease available P levels in both the water and sediments, and to increase the TN:TP ratio. Despite variability among sites post-Phoslock treatment, the types of cyanobacteria were altered and overall abundance was significantly decreased. This research provides information on the applicability of a specific in situ P mitigation technology to manage P levels, nutrient ratios, and subsequently shift the algal assemblage.

Many freshwaters have surpassed a critical threshold level where nutrient input reductions may be futile in promoting desirable short-term changes (Duarte et al. 2008). Since legacy nutrients have accumulated in the sediments, reducing external inputs may fail to address available nutrient sources. In particular, algae adapted to attaining internal nutrient stores

(e.g., cyanobacteria) have a further advantage and are more likely to dominate assemblages that depend mostly on water-column, external sources. Cyanobacteria are increasing globally, and are a threat to human and animal health (Hallegraeff 1993; Briand et al. 2003).

Multiple interactive factors govern cyanobacterial presence and dominance (Kosten et al.

2012), and this research focused on a better understanding of nutrient dynamics and alteration thereof. With increased eutrophication of water resources and chronic cyanobacterial growths, Phoslock can provide water resource managers a viable option for in situ nutrient mitigation and alteration of algal assemblage composition.

REFERENCES

Anderson DM, Glibert PM, Burkholder JM (2002) Harmful algal blooms and eutrophication: nutrient sources, composition and consequences. Estuaries 25:562-584

Bachmann RW, Hoyer MV, Fernandez C, Canfield Jr DE (2003) An alternative to proposed phosphorus TMDLs for the management of Lake Okeechobee. Lake and Reservoir Management 19:251-264

Barbiero RP, Welch EB (1992) Contribution of benthic blue-green-algal recruitment to lake populations and phosphorus translocation. Freshwater Biology 27:249-260

Baris RD, Cohen SZ, Barnes NL, Lam J, Ma Q (2010) Quantitative analysis of over 20 years of golf course monitoring studies. Environmental Toxicology and Chemistry 29:1224-1236

Beutel MW, Duvil R, Cubas FJ, Matthews DA, Wilhelm FM, Grizzard TJ, Austin D, Horne AJ, Gebremariam S (2016) A review of managed nitrate addition to enhance surface water quality. Critical Reviews in Environmental Science and Technology 46:673- 700

Beversdorf LJ, Miller TR, McMahon KD (2013) The role of nitrogen fixation in cyanobacterial bloom toxicity in a temperate, eutrophic lake. PLoS ONE 8:e56103

Bishop WM, Hyde S, McNabb T, Cormican I, Willis BE (2014) Operational evaluation of Phoslock phosphorus locking technology in Laguna Niguel Lake, CA. Water, Air, and Soil Pollution 225:1-11

Björk-Ramberg S (1985) Uptake of phosphate and inorganic nitrogen by a sediment-algal system in a subarctic lake. Freshwater Biology 15:175-183

Boström B, Andersen JM, Fleischer S, Jansson M (1988) Exchange of phosphorus across the sediment water interface. Hydrobiologia 170:229-244

Briand JF, Jacquet S, Bernard C, Humbert JF (2003) Health hazards for terrestrial vertebrates from toxic cyanobacteria in surface water ecosystems. Veterinary Research 34:361- 377

Burkholder JM (2009) Harmful algal blooms. In: Likens GE (ed.) Encyclopedia of Inland Waters, vol. 1, pp. 264-285. Elsevier, Oxford, United Kingdom

Burkholder JM, Glibert PM (2013) Eutrophication and oligotrophication, pp. 347-371. In: Encyclopedia of Biodiversity, 2nd edition, Volume 3, by Levin S.A. (ed.). Academic Press, Boston, MA

Carey CC, Weathers KC, Cottingham KL (2008) Gloeotrichia echinulata blooms in an oligotrophic lake: Helpful insights from eutrophic lakes. Journal of Plankton Research 30:893-904

Carpenter SR, Caraco NF, Correll DL, Howarth RW, Sharpley AN, Smith VH (1998) Nonpoint pollution of surface waters with phosphorus and nitrogen. Ecological Applications 8:559-568

Carvalho L, McDonald C, de Hoyos C, Mischke U, Phillips G, abor Borics G, Poikane S, Skjelbred B, Lyche Solheim A, Van Wichelen J, Cardoso AC (2013) Sustaining recreational quality of European lakes: minimizing the health risks from algal blooms through phosphorus control. Journal of Applied Ecology 50:315-323

Cetiner ZS Wood SA, Gammons CH (2005) The aqueous geochemistry of the rare earth elements. Part XIV. The solubility of rare earth element phosphates from 23 to 150 °C. Chemical Geology 217:147-169

Chang SC, Jackson ML (1957) Fractionation of soil phosphorus. Soil Science 84:133-144

Chislock MF, Sharp KL, Wilson AE (2014) Cylindrospermopsis raciborskii dominates under very low and high nitrogen-to-phosphorus ratios. Water Research 49:207-214

Chen M, Ye T-R, Krumholz LR, Jiang H-L (2014) Temperature and cyanobacterial bloom biomass influence phosphorous cycling in eutrophic lake sediments. PLoS ONE 9:e93130

Cooke GD, Welch EB, Peterson SA, Nichols SA (2005) In: Restoration and Management of Lakes and Reservoirs, third ed. Taylor & Francis Group, Boca Raton, FL

Cullen P, Forsberg C (1988) Experiences with reducing point sources of phosphorus to lakes. Hydrobiologia 170:321-336

DeBusk KM, Hunt WF, Hatch U, Sydorovych O (2010) Watershed retrofit and management evaluation for urban stormwater management systems in North Carolina. Journal of Contemporary Water Research and Education 146:64-74

Diatloff E, Asher CJ, Smith FW (1993) Use of geochem-pc to predict rare-earth element (REE) species in nutrient solutions. Plant and Soil 156:251-254

Dignum M, Hoogveld HL, Floris V, Gons HJ, Matthijs HCP, Pel R (2004) Flow cytometric detection of phosphatase activity combined with 13C-CO2 tracer-based growth rate assessment in phytoplankton populations from a shallow lake. Aquatic Microbial Ecology 37:159-169

Dithmer L, Lipton AS, Reitzel K, Warner TE, Lundberg D, Nielsen UG (2015) Characterization of phosphate sequestration by a lanthanum modified bentonite clay: A solid state NMR, EXAFS and PXRD study. Environmental Science and Technology 49:4559-4566

Dithmer L, Nielsen UG, Lürling M, Spears BM, Yasseri S, Lundberg D, Moore A, Jensen ND, Reitzel K (2016) Responses in sediment phosphorus and lanthanum concentrations and composition across 10 lakes following applications of lanthanum modified bentonite. Water Research 97:101-110

Douglass GB (2002) Remediation material and remediation process for sediments. United States Patent number 6350383

Downing JA, McCauley E (1992). The nitrogen:phosphorus relationship in lakes. Limnology and Oceanography 37:936-945

Downing JA, Watson SB, McCauley E (2001) Predicting cyanobacteria dominance in lakes. Canadian Journal of Fisheries and Aquatic Sciences 58:1905-1908

Duarte CM, Conley DJ, Carstensen J, Sanchez-Camacho M (2008) Return to Neverland: Shifting baselines affect eutrophication restoration targets. Estuaries and Coasts 32:29-36

Dunne EJ, Clark MW, Corstanje R, Reddy KR (2011) Legacy phosphorus in subtropical wetland soils: influence of dairy, improved and unimproved pasture land use. Ecological Engineering 37:1481-1491

Firsching FH, Brune SN (1991) Solubility products of the trivalent rare-earth phosphates. Journal of Chemical and Engineering Data 36:93-95

Foy RH, Smith RV, Jordan C, Lennox SD (1995) Upward trend in soluble phosphorus loadings to Lough Neagh despite phosphorus reduction at sewage treatment works. Water Research 29:1051-1063

Ganf GG, Oliver RL (1982) Vertical separation of the light and available nutrients as a factor of causing replacement of green algae by blue-green algae in the plankton of a stratified lake. Journal of Ecology 70:829-844

Ghadouani A, Pinel-Alloul B, Prepas EE (2003) Effects of experimentally induced cyanobacterial blooms on crustacean zooplankton communities. Freshwater Biology 48:363-381

Glibert PM, Fullerton D, Burkholder JM, Cornwell JC, Kana TM (2011) Ecological stoichiometry, biogeochemical cycling, invasive species and aquatic food webs: San

Francisco Estuary and comparative systems. Reviews in Fisheries Science 19:358- 417

Gobler CJ, Burkholder JM, Davis TW, Harke MJ, Johengen T, Stow CA, Van de Waal DB (2016) The dual role of nitrogen supply in controlling the growth and toxicity of cyanobacterial blooms. Harmful Algae 54:87-97

Gonsiorczyk T, Casper P, Koschel R (1998) Phosphorus binding forms in the sediment of an oligotrophic and a eutrophic hardwater lake of the Baltic lake district (Germany). Water Science and Technology 37:51-58

Graham LE, Graham JM, Wilcox LW, Cook ME (2016) Algae (Third Edition version 3.1) LJLM Press, LLC, Madison, WI

Granhall U, Lundgren A (1971) Nitrogen Fixation in Lake Erken. Limnology and Oceanography 16:711-719

Gulati RD, Van Donk E (2002) Lakes in the Netherlands, their origin, eutrophication and restoration: state-of-the-art review. Hydrobiologia 478:73-106

Hagheseresht F, Wang S, Do DD (2009) A novel lanthanum-modified bentonite, Phoslock, for phosphate removal from wastewaters. Applied Clay Science 46:369-375

Hallegraeff GM (1993) A review of harmful algal blooms and their apparent global increase. Phycologia 32:79-99

Harris TD, Wilhem FM, Graham JL, Loftin KA (2014) Experimental manipulation of TN:TP ratios suppress cyanobacterial biovolume and microcystin concentration in large-scale mesocosms. Lake and Reservoir Management 30:72-83

Harris TD, Wilhelm FM, Graham JL, Loftin K (2014) Experimental additions of aluminum sulfate and ammonium nitrate to in situ mesocosms to reduce cyanobacterial biovolume and microcystin concentration. Lake and Reservoir Management 30:84-93

Heisler J, Gilbert PM, Burkholder JM, Anderson DM, Cochlan W, Dennison WC, Dortch Q, Gobler CJ, Heil CA, Humphries E, Lewitus, R. Magnien, H.G. Marshall, K. Sellner, Stockwell DA, Stoeker DK, Suddleson M (2008) Eutrophication and harmful algal blooms: A scientific consensus. Harmful Algae 8:3-13

Hession WC, Storm DE (1999) Watershed-Level Uncertainties: Implications for Phosphorus Management and Eutrophication. Journal of Environmental Quality 29:1172-1179

Horne AJ, Fogg GE (1970) Nitrogen fixation in some English lakes. Proceedings of the Royal Society of London Series B 175:351-366

Howarth RW, Paerl HW (2008) Coastal marine eutrophication: Control of both nitrogen and phosphorus is necessary. PNAS USA 105:E103

Hupfer M, Gächter R, Giovanoli R, (1995) Transformation of phosphorus species in settling seston and during early sediment diagenesis. Aquatic Sciences 57:305-324

Istvanovics V, Petterson K, Rodrigo MA, Pierson D, Padisak J, Colom W (1993) Gloeotrichia echinulata, a colonial cyanobacterium with a unique phosphorus uptake and life strategy. Journal of Plankton Research 15:531-552

Jacoby JM, Gibbons HL, Stoops KB, Bouchard DD (1994) Response of a shallow, polymictic lake to buffered alum treatment. Lake and Reservoir Management 10:103- 112

Jansson M, Olsson H, Petterson K (1988) Phosphatases; origin, characteristics and function in lakes. Hyrobiologia 170:157-175

Jarvie HP, Sharpley AN, Spears B, Buda AR, May L, Kleinman PJ (2013) Water quality remediation faces unprecedented challenges from "legacy phosphorus." Environmental Science and Technology 47:8997-8998

Jeppesen E, Søndergaard M, Jensen JP, Havens KE, Anneville O, Carvalho L, Coveney MF, et al. (2005) Lake responses to reduced nutrient loading: an analysis of contemporary long-term data from 35 case studies. Freshwater Biology 50:1747-1771

Jonasson RG, Bancroft GM, Boatner LA (1988) Surface reactions of synthetic, 933 end- member analogues of monazite, xenotime and rhabdophane, and evolution of natural waters. Geochmica et Cosmochimica Acta 52:767-770

Kapanen G (2008) Phosphorus fractionation in lake sediments. Estonian Journal of Ecology 57:244-255

Koski-Vähälä J, Hartikainen H (2001) Assessment of the risk of phosphorus loading due to resuspended sediment. Journal of Environmental Quality 30:960-966

Kromkamp J, Van Den Heuvel A, Mur LR (1989) Phosphorus uptake and photosynthesis by phosphate-limited cultured of the cyanobacterium Microcystis aeruginosa. British Phycological Journal 24:347-355

Kosten S, Huszar VLM, Bécares E, Costa LS, Donk E, Hansson LA (2012) Warmer climates boost cyanobacterial dominance in shallow lakes. Global Change Biology 18:118-126

Lagus A, Suomela J, Helminen H, Sipura J (2007) Impacts of nutrient enrichment and sediment on phytoplankton community structure in the northern Baltic Sea. Hydrobiologia 579:351-368

Larsson U, Hajdu S, Walve J, Elmgren R (2001) Baltic Sea nitrogen fixation estimated from the summer increase in upper mixed layer total nitrogen. Limnology and Oceanography 46:811-820

Levine SN, Schindler DW (1999) Influence of nitrogen to phosphorus supply ratios and physicochemical conditions on cyanobacteria and phytoplankton species composition in the Experimental Lakes Area, Canada. Canadian Journal of Fisheries and Aquatic Sciences 56:451-466

Lukkari K, Hartikainen H, Leivuori M (2007) Fractionation of sediment phosphorus revisited. I: fractionation steps and their biogeochemical basis. Limnology and Oceanography: Methods 5:433-444

Mallin MA, McIver MR, Wells HA, Parsons DC, Johnson VL (2005) Reversal of eutrophication following sewage treatment upgrades in the New River Estuary, North Carolina. Estuaries 28:750-760

Marsden MW (1989) Lake restoration by reducing external phosphorus loading: the influence of sediment phosphorus release. Freshwater Biology 21:139-162

McDonald, JH (2014) Handbook of Biological Statistics (3rd ed.). Sparky House Publishing, Baltimore, MD

Meis S, Spears BM, Maberly SC, O’Malley MB, Perkins RG (2012) Sediment amendment with Phoslock in Clatto Reservoir (Dundee, UK): investigating changes in sediment elemental composition and phosphorus fractionation. Journal of Environmental Management 93:185-193

Mehner T, Diekmann M, Gonsiorczyk T, Kasprzak P, Koschel R, Krienitz L, Rumpf M, Schulz M, Wauer G (2008) Rapid recovery from eutrophication of a stratified lake by disruption of internal nutrient load. Ecosystems 11:1142-1156

Microsoft (2010) Microsoft Excel [computer software]. Redmond, WA

Nikolai SJ, Dzialowski AR (2014) Effects of internal phosphorus loading on nutrient limitation in a eutrophic reservoir. Limnologica 49:33-41

Nürnberg GK, LaZerte BD (2016) Trophic state decrease after lanthanum-modified bentonite (Phoslock) application to a hyper-eutrophic polymictic urban lake frequented by Canada geese (Branta canadensis). Lake and Reservoir Management 32:74-88

Nürnberg G (1997) Coping with water quality problems due to hypolimnetic anoxia in central Ontario lakes. Water Quality Research Journal of Canada 32:391-405

Oosterhout F, Lürling M (2013) Phosphorus binding clay (Phoslock) in mitigating cyanobacterial nuisance: effects on water quality variables and plankton. Hydrobiologia 710:265-277

O’Neil JM, Davis TW, Burford MA, Gobler CJ (2012) The rise of harmful cyanobacteria blooms: the potential roles of eutrophication and climate change. Harmful Algae 14:313-334

Orihel DM, Bird DF, Brylinsky M, Chen H, Donald DB, Huang DY, Giani A, Kinniburgh D, Kling H, Kotak BG, Leavitt PR, Nielsen CC, Reedyk S, Rooney RC, Watson SB, Zurawell RW, Vinebrooke RD (2012) High microcystin concentrations occur only at low nitrogen-to-phosphorus ratios in nutrient-rich Canadian lakes. Canadian Journal of Fisheries and Aquatic Sciences 69:1457-1462

Orihel DM, Schindler DW, Ballard NC, Graham MD, O’Connell DW (2015) The “nutrient pump:” Iron-poor sediments fuel low nitrogen-tophosphorus ratios and cyanobacterial blooms in polymictic lakes. Limnology and Oceanography 60:856-871

Paerl HW (1990) Physiological ecology and regulation of N2 fixation in natural waters. Advances in Microbial Ecology 11:305-344

Paerl HW, Prufert LE, Ambrose WW (1991) Contemporaneous N2 fixation and oxygenic photosynthesis in the nonheterocystous mat-forming cyanobacterium Lyngbya aestuarii. Applied and Environmental Microbiology 57:3086-3092

Paerl HW (1997) Coastal eutrophication and harmful algal blooms: Importance of atmospheric deposition and groundwater as 'new' nitrogen and other nutrient sources. Limnology and Oceanography 42:1154-1165

Paerl HW, Dennis RL, Whitall DR (2002) Atmospheric deposition of nitrogen: implications for nutrient over-enrichment of coastal waters. Estuaries 25:677-693

Paerl HW, Gardner WS, Havens KE, Joyner AR, McCarthy MJ, Newell SE, Qin B, Scott JT (2016) Mitigating cyanobacterial harmful algal blooms in aquatic ecosystems impacted by climate change and anthropogenic nutrients. Harmful Algae 54:213-222

Perakis SS, Welch EB, Jacoby JM (1996) Sediment-to-water blue-green algal recruitment in response to alum and environmental factors. Hydrobiologia 318:165-177

Pettersson K, Istvanovics V (1988) Sediment phosphorus in Lake Balaton: forms and mobility. Archives of Hydrobiology 30:25-41

Prescott GW (1970) How to know the freshwater algae. 3rd ed. McGraw Hill, Boston, MA

Psenner R, Boström B, Dinka M, Petterson K, Pucsko R, Sager M (1988) Fractionation of phosphorus in suspended matter and sediment. Archive für Hydrobiologie: Ergebnisse der Limnologie 30:98-103

Recht HL, Ghassemi M, Kleber EV (1970) Precipitation of phosphates from water and waste water using lanthanum salts. Proceedings of the 5th International Water Pollution Research. Pergamon, pp. 1-17

Reddy K, Newman RS, Osborne TZ, White JR, Fitz HC (2011) Phosphorous cycling in the greater everglades ecosystem: Legacy phosphorous implications for management and restoration. Critical Reviews in Environmental Science and Technology 41:149-186

Reitzel K, Andersen FO, Egemose S, Jensen HS (2013) Phosphate adsorption by lanthanum modified bentonite clay in fresh and brackish water. Water Research 47:2787-2796

Reynolds CS (1972) Growth, gas vacuolation and buoyancy in a natural population of a planktonic blue-green alga. Freshwater Biology 2:87-106

Robb MS, Greenop B, Goss Z, Douglas G, Adeney J (2003) Application of Phoslock, an innovative phosphorus binding clay, to two Western Australian waterways: Preliminary findings. Hydrobiologia 494:237-243

Ross G, Haghseresht F, Cloete TM (2008) The effect of pH and anoxia on the performance of Phoslock, a phosphorus binding clay. Harmful Algae 7:545-550

Sabour B, Loudiki M, Vasconcelos V (2009) Growth responses of Microcystis ichthyoblabe and Anabaena aphanizomenoides (cyanobacteria) under different nitrogen and phosphorus conditions. Chemistry and Ecology 25:337-344

Sañudo-Wilhelmy SA, Kustka AB, Gobler CJ, Hutchins DA, Yang M, Lwiza K, Burns J, Capone DG, Raven JA, Carpenter EJ (2001) Phosphorus limitation of nitrogen fixation by Trichodesmium in the central Atlantic Ocean. Nature 411:66-69

Schindler DW, Hecky RE, Findlay DL, Stainton MP, Parker BR, Paterson M, Beaty KG, Lyng M, Kasian SEM (2008) Eutrophication of lakes cannot be controlled by reducing nitrogen input: results of a 37 year whole ecosystem experiment. PNAS USA 105:11254-11258

Schindler DW (2012) The dilemma of controlling cultural eutrophication of lakes. Proceedings of the Royal Society of London: Biological Sciences 279:4322-4333

Seale DB, Boraas ME, Warren GJ (1987) Effects of sodium and phosphate on growth of cyanobacteria. Water Research 21:625-631

SigmaPlot version 12.5 (2014) Systat Software, Inc., San Jose, CA

Smith VH (1983) Low nitrogen to phosphorus ratios favor dominance by blue-green algae in lake phytoplankton. Science 221:669-671

Smith VH, Tilman GD, Nekola JC (1999) Eutrophication: Impacts of excess nutrient inputs on freshwater, marine, and terrestrial ecosystems. Environmental Pollution 100:179- 196

Søndergaard M, Jensen PJ, Jeppesen E (2001) Retention and internal loading of phosphorus in shallow, eutrophic lakes. The Scientific World Journal 1:427-442

Søndergaard M, Jensen PJ, Jeppesen E (2003) Role of sediment and internal loading of phosphorus in shallow lakes. Hydrobiologia 506-509:135-145

Spears BM, Carvalho L, Paterson DM (2007) Phosphorus partitioning in a shallow lake: implications for water quality management. Water and Environment Journal 21:43-53

Spears BM, Mackay EB, Yasseri S, Gunn IDM, Waters KE, Andrews C, Cole S, De Ville M, Kelly A, Meis S, Moore AL, Nürnberg GK, van Oosterhout F, Pitt JA, Madgwick G, Woods HJ, Lürling M (2016) A meta-analysis of water quality and aquatic macrophyte responses in 18 lakes treated with lanthanum modified bentonite (Phoslock®). Water Research 97:111-121

Standard Methods for the Examination of Water and wastewater, 21st ed. (SMEWW) (2005) American Public Health Association (APHA). Washington, DC

Stewart WDP, Alexander G (1971) Phosphorus availability and nitrogenase activity in aquatic blue-green algae. Freshwater Biology 1:389-404

Trimbee AM, Prepas EE (1987) Evaluation of total phosphorus as a predictor of the relative biomass of blue-green algae with emphasis on Alberta lakes. Canadian Journal of Fisheries and Aquatic Sciences 44:1337-1342

United States Environmental Protection Agency (USEPA) (1971) Method 130.1 Hardness, Total (mg/L as CaCO3) (Colorimetric, Automated EDTA). Cincinnati, OH

United States Environmental Protection Agency (USEPA) (1974) Method 310.2 Alkalinity (Colorimetric, Automated, Methyl Orange). Office of Research and Development, Environmental Monitoring and Support Laboratory. Cincinnati, OH

United States Environmental Protection Agency (USEPA) (1978) Phosphorous, All Forms method 365.3 manual ascorbic acid two reagent. Office of Research and Development, Environmental Monitoring and Support Laboratory. Cincinnati, OH

United States Environmental Protection Agency (USEPA) (1982) Method 150.1: pH (Electrometric). Washington, DC

United States Environmental Protection Agency (USEPA) (1983a) Method 120.1, Conductance (Specific Conductance, umhos at 25°C). Washignton, DC

United States Environmental Protection Agency (USEPA) (1993) Method 351.2, Determination of Total Kjeldahl Nitrogen by Semiautomated Colorimetry. Office of Research and Development, Environmental Monitoring and Support Laboratory. Cincinnati, OH

United States Environmental Protection Agency (USEPA) (1993) Method 353.2, Determination of Nitrate-Nitrite Nitrogen by Automated Colorimetry. Office of Research and Development, Environmental Monitoring and Support Laboratory. Cincinnati, OH

United States Environmental Protection Agency (USEPA) (1993c) Determination of turbidity by nephelometry. Revision 2.0. Office of Research and Development, Environmental Monitoring and Support Laboratory. Cincinnati, OH

United States Environmental Protection Agency (USEPA) (2010) Water Quality Standards for the State of Florida’s Lakes and Flowing Waters. Vol. 75, No. 233. 40 CFR Part 131 [EPA–HQ–OW–2009–0596; FRL–9228–7] RIN 2040–AF11. Washington, DC

Van de Waal DB, Smith VH, Declerck SAJ, Stam ECM, Elser JJ (2014) Stoichiometric regulation of phytoplankton toxins. Ecological Letters 17:736-742

Van Nieuwenhuyse EE (2007) Response of summer chlorophyll concentration to reduced total phosphorus concentration in the Rhine River (Netherlands) and the Sacramento – San Joaquin Delta (California, USA). Canadian Journal of Fisheries and Aquatic Sciences 64(11):1529-1542

Vollenweider RA (1976) Advances in defining critical loading levels for phosphorus in lake eutrophication. Memorie dell'Istituto Italiano di Idrobiologia 33:53-83

Walve J, Larsson U (2007) Blooms of Baltic Sea Aphanizomenon sp.(Cyanobacteria) collapse after internal phosphorus depletion. Aquatic Microbial Ecology 49:57-69

Watson SB, McCauley E, Dowing JA (1997) Patterns in phytoplankton taxonomic composition across temperate lakes of different nutrient status. Limnology and Oceanography 42:487-495

Wehr JD, Sheath RG (2003) Freshwater Algae of North America. Academic Press, New York, NY

Welch EB (2009) Should nitrogen be reduced to manage eutrophication if it is growth limiting? Evidence from Moses Lake. Lake and Reservoir Management 25:401-409

Wetzel RG (2001) Limnology: Lake and River Ecosystems, 3rd ed. Academic Press, San Diego, CA

Wilhelm S, Adrian R (2008) Impact of summer warming on the thermal characteristics of a polymictic lake and consequences for oxygen, nutrients and phytoplankton. Freshwater Biology 53:226-237

Table 1. Description of site characteristics and ranges of water chemistry in the two experimental golf course ponds.

Hickory Meadows Chockyotte

Size (ha) 0.27 0.38

Mean depth (m) 1.58 1.87 Max depth (m) 3.1 3.2

Perimeter (m) 321.14 245.6

Volume (m3) 4,292.5 7,178.9 pH (SU) 6.3-7.8 7-8.5 Alkalinity (mg/L as 10-22 10-28 CaCO3) Hardness (mg/L as 14-24 14-32 CaCO3)

Conductivity (µS/cm) 64-98 45-84

TN (mg/L) 1.6-6.7 1.5-3.4

Sediment OM content 18.1-24.4% 8.9-14.8% Sediment moisture 44-64% 68-82% content Sediment Bulk density 1.09-1.15 1.11-1.29 (g/cm3) Sediment TP (mg/kg) 457-725 704-1,077

TP water-column (kg) 3.61 0.29

Sediment mobile P (kg) 5.50 6.15

Total P targeted (kg) 9.11 6.44

Phoslock amended (kg) 923 644.1

Table 2. Description of analytical parameters and methods used for measured water quality parameters.

Water Sample Measurement Units Method Standard Methods Parameter Limit (MDL) SMEWW 2005: 4500; pH 0.01 SU pH meter EPA Method 150.1 mg/L as Colorimetric discrete analyzer Alkalinity 1 CaCO EPA Method 310.2 3 (Konelab Aqua 20) mg/L as Colorimetric discrete analyzer Hardness 1 CaCO EPA Method 130.2 3 (Konelab Aqua 20) Conductivity 0.1 µS/cm Calibrated meter EPA 120.1 SMEWW 2005: 4500; Dissolved Oxygen 0.1 mg/L Calibrated meter (YSI) EPA Method 360.1 Total Suspended Solids 0.1 mg/L 0.45 µm filter SMEWW 2005, 2540D Excitation/emission fluorescence, Chlorophyll a 1 µg/L EPAMethod 445 Wallac Victor2 fluorometer Colorimetric discrete analyzer Total Phosphorus 1 µg/L EPA Method 365.3 (Konelab Aqua 20) Free Reactive Colorimetric discrete analyzer 1 µg/L EPA Method 365.3 Phosphorus (Konelab Aqua 20) Total Nitrogen 0.02 mg/L Calculation Total Kjeldahl Nitrogen 0.02 mg/L Semi-automated colorimetry EPA Method 351.2 Nitrate/Nitrite 0.02 mg/L Colorimetry EPA Method 353.2 Turbidity 0.1 NTU Turbidimeter (Cole-Parmer Inc.) EPA Method 180.1

Table 3. Description of sediment P fractions from sequential extraction procedures.

Description of Bio- Phosphorus Release Phosphorus availability References Fraction Mechanisms Fraction probability Directly Desorption; available P; diffusion; Boström et al. (1988), Labile P pore water P; concentration High Hupfer et al. (1995), loosely bound gradients Spears et al. (2007) or adsorbed P

Psenner et al. (1988), P bound to Fe- Reductant- Anoxia, low Boström et al. (1988), hydroxides soluble P redox High Hupfer et al. (1995), and Mn- potential Lukkari et al. (2007), compounds Spears et al. (2007) P adsorbed to Psenner et al. (1988), Metal-oxide metal oxides Medium/ Boström et al. (1988), adsorbed P (mainly Al, Fe); pH high Hupfer et al. (1995), P exchangeable Lukkari et al. (2007) against OH- Psenner et al. (1984, 1988); Boström et al. Biodegrad- Medium/ (1988); Organic P Biogenic P ation high Hupfer et al. (1995); Lukkari et al. (2007); Meis et al. 2012 P bound to carbonates Psenner et al. (1984, Apatite and apatite Low pH; 1988), Boström et al. bound P/ minerals; Strong acid Low (1988), Residual P Refractory digestion Hupfer et al. (1995),

compounds Spears et al. (2007)

Table 4. Statistical comparisons using a paired t-test or Wilcoxon Rank Sum test for algae, water quality and sediment parameters between the pre-treatment and post-treatment growing season averages (April-September) over two years.

Significance denoted with a P value < 0.05.

Hickory Meadows Chockyotte 2014 2015 2016 2014 2015 2016 Avg Avg P Avg P Avg Avg P Avg P Secchi depth 0.738 1.60 < 0.001 1.48 < 0.001 0.825 1.81 < 0.001 1.62 < 0.001 (m) Cyanobacteria l density 39,708 1,762 < 0.05 1,859 < 0.05 63,917 1,945 < 0.001 1,864 < 0.001 (cells/mL) Total Algal Density 41,654 5,396 < 0.05 6,573 < 0.05 65,875 25,103 < 0.05 9,184 < 0.001 (cells/mL) TN:TP ratio 7.58 47.1 < 0.001 33.6 < 0.001 24.7 145.3 < 0.001 97.4 < 0.001 TP 602.66 75.27 < 0.05 99.20 < 0.05 63.63 22.72 < 0.001 28.61 < 0.001 (µg/L) FRP 14.38 15.44 0.819 9.28 0.262 18.80 9.94 0.381 10.17 0.178 (µg/L) Sediment TP 596.5 719.8 0.215 707.3 0.40 844.3 830.3 0.912 891.5 0.792 (mg/kg dw) Sediment mobile P 285.78 74.56 < 0.001 57.0 < 0.001 322.54 97.89 0.035 154.2 0.167 (mg/kg dw)

Figure 1. Map of the Chockyotte irrigation pond and designation of sampling locations.

Figure 2. Map of the Hickory Meadows irrigation pond and designation of sampling locations.

50.00

40.00

30.00

C ° 20.00

10.00 Temperature Temperature

0.00

-10.00

Mean temperature High temperature Low temperature -20.00

Figure 3. Daily mean, high, and low temperatures in Whitakers, NC throughout the study duration.

8 7 6 5 4 3 2 1

Rainfall (cm) Rainfall 0

4/1/2014

5/1/2014

6/1/2014

7/1/2014

8/1/2014

9/1/2014

1/1/2015

10/1/2014

2/1/2015

11/1/2014

3/1/2015

12/1/2014

4/1/2015

5/1/2015

6/1/2015

7/1/2015

8/1/2015

9/1/2015

10/1/2015

1/1/2016

11/1/2015

2/1/2016

12/1/2015

3/1/2016

4/1/2016

5/1/2016

6/1/2016

7/1/2016

8/1/2016 9/1/2016

Daily total rainfall

Figure 4. Total daily rainfall in Whitakers, NC throughout the study duration.

Sampling period

A M J J A S 0

0.5

1.5 Secchi (m) Secchi Depth 2

2.5

2014 2015 2016

Figure 5. Secchi depth measurements in the Chockyotte irrigation pond throughout the growing season pre-treatment (2014) and two years after treatment (2015-2016). Each point represents the average from three sites.

Sampling period

A M J J A S 0 0.2 0.4 0.6 0.8 1 1.2

Secchi (m) Secchi Depth 1.4 1.6 1.8 2

2014 2015 2016

Figure 6. Secchi depth measurements in the Hickory Meadows irrigation pond throughout the growing season pre-treatment (2014) and two years after treatment (2015-2016). Each point represents the average from three sites.

160

140 Free Reactive Phosphorus 120 Total Phosphorus 100

60 Phosphorus (µg/L) Phosphorus 40

Sampling Period

Figure 7. Total and free reactive P measurements in the Chockyotte irrigation pond throughout the study duration. Error bars represent one standard deviation from the mean.

Abbreviations (M = month; D = day; WAT = weeks after treatment; MAT = months after treatment).

1,400

1,200 Free Reactive Phosphorus 1,000 Total Phosphorus

800

600

Phosphorus (µg/L) Phosphorus 400

200

Sampling Period

Figure 8. Total and free reactive P measurements in Hickory Meadows irrigation pond throughout the study duration. Error bars represent one standard deviation from the mean.

Abbreviations (M = month; D = day; WAT = weeks after treatment; MAT = months after treatment).

80,000 300

Total Algae 70,000 Cyanobacteria 250 TN:TP 60,000

200 50,000

40,000 150 TN:TP 30,000 100

Cell Density (cells/mL) Density Cell 20,000

50 10,000

0 0

Sampling Period

Figure 9. Total algal and cyanobacterial densities in conjunction with TN:TP ratios in the

Chockyotte irrigation pond throughout the study duration. Abbreviations (M = month; D = day; WAT = weeks after treatment; MAT = months after treatment).

100%

90%

80% Chrysophyceae

70% Synurophyceae Cryptophyta 60% Raphidophyta 50% Dinophyta Euglenophyta 40% Bacillariophyta

Percent Abundance Percent 30% Chlorophyta Streptophyta 20% Cyanophyta 10%

Sampling Period

Figure 10. Percent abundance of algal groups throughout the study duration at the Chockyotte irrigation pond. Abbreviations

(M = month; D = day; WAT = weeks after treatment; MAT = months after treatment).

70,000 70

Total Algae 60,000 Cyanobacteria 60 TN:TP 50,000 50

40,000 40

30,000 30 TN:TP

20,000 20 Cell Density (cells/mL) Density Cell

10,000 10

0 0

Sampling Period

Figure 11. Total algal and cyanobacterial densities in conjunction with TN:TP ratios in the

Hickory Meadows irrigation pond throughout the study duration. Abbreviations (M = month;

D = day; WAT = weeks after treatment; MAT = months after treatment).

100%

90%

80% Chrysophyceae Synurophyceae 70% Cryptophyta Raphidophyta 60% Dinophyta Euglenophyta 50% Bacillariophyta 40% Chlorophyta Percent Abundance Percent Streptophyta 30% Cyanophyta

20%

10%

Sampling Period

Figure 12. Percent abundance of algal groups throughout the study duration at the Hickory Meadows irrigation pond.

Abbreviations (M = month; D = day; WAT = weeks after treatment; MAT = months after treatment).

500

450 labile 400 Reductant/soluble 350 Metal oxide 300 Organic 250

200 mg mg P/kg sediment DW 150 * * * * * 100

0 2 M 1 M 1 D 2 1 2 4 6 8 10 13 16 19 22 pre pre pre WAT MAT MAT MAT MAT MAT MAT MAT MAT MAT MAT

Figure 13. Mobile sediment P fractions in the Chockyotte irrigation pond through time. Error bars represent one standard deviation from the mean. Abbreviations (M = month; D = day;

WAT = weeks after treatment; MAT = months after treatment).

1,200

1,000

800

600

400 mg mg P/kg sediment DW

200

0 2 M 1 M 1 D 2 1 2 4 6 8 10 13 16 19 22 pre pre pre WAT MAT MAT MAT MAT MAT MAT MAT MAT MAT MAT

Apatite/ Residual Sediment Phosphorus Mobile Sediment Phosphorus

Figure 14. Apatite/residual and combined mobile sediment P fractions in Chockyotte irrigation pond through time. Error bars represent one standard deviation from the mean.

Abbreviations (M = month; D = day; WAT = weeks after treatment; MAT = months after treatment).

400

350 labile

Reductant/soluble 300

Metal oxide 250 Organic 200 *

150 * mg mg P/kg sediment DW * 100 * * * * * * * 50

0 2 M 1 M 1 D 2 1 2 4 6 8 10 13 16 19 22 pre pre pre WAT MAT MAT MAT MAT MAT MAT MAT MAT MAT MAT

Figure 15. Mobile sediment P fractions in the Hickory Meadows irrigation pond through time. Error bars represent one standard deviation from the mean. Abbreviations (M = month;

D = day; WAT = weeks after treatment; MAT = months after treatment).

900

800

700

600

500

400

300 mg mg P/kg sediment DW 200

100

0 2 M 1 M 1 D 2 1 2 4 6 8 10 13 16 19 22 pre pre pre WAT MAT MAT MAT MAT MAT MAT MAT MAT MAT MAT

Apatite/ Residual Sediment Phosphorus Mobile Sediment Phosphorus

Figure 16. Apatite/residual and combined mobile sediment P fractions in Hickory Meadows irrigation pond through time. Error bars represent one standard deviation from the mean.

Abbreviations (M = month; D = day; WAT = weeks after treatment; MAT = months after treatment).

CHAPTER THREE

SENSITIVITY OF MICROCYSTIS AERUGINOSA STRAINS TO COPPER AND

INFLUENCE OF PHOSPHORUS

ABSTRACT

Cyanobacterial blooms are widespread and increasingly impacting freshwater resources globally. Phosphorus (P) enrichment is often described as promoting blooms, and may also play a critical function in growth rates and stress response. Here, the impact of P levels in the growth media and cellular P quotas were assessed in terms of the susceptibility of Microcystis aeruginosa (Kutzing) Lemmerman (Ma) to a common reactive management technique, copper sulfate application. Five strains of Ma were tested under three different P regimes (1,500 µg/L = Full P; 150 µg/L= 0.1 P; 75 µg/L = 0.05 P). All Ma strains grown at lower P levels had significantly decreased (P < 0.05) P content and chlorophyll a content per cell. Ma strains grown in full and 0.1 P regimes had similar 96-hr LC50 values ranging from

0.068 to 0.139 mg Cu/L based on chlorophyll a content and from 0.047 to 0.168 mg Cu/L based on cell densities. In the 0.05 P growth regime, Ma 2386 had a significantly decreased

96-hr LC50 value (based on chlorophyll a content and cell density) than all other Ma strains, and Ma 2665 had a significantly higher 96-hr LC50 value compared with Ma strains 2386,

2388, and 2664. The relative sensitivities of strains grown in 0.05 P media to copper were

Ma 2386 > 2388 ≥ 2664 ≥ 2061 ≥ 2665. All strains had significantly decreased growth rates under 0.05 P compared with full P conditions, but only Ma 2386 had increased sensitivity to copper. This research provides insights about altered sensitivity of cyanobacteria to reduced

P supplies, toward decreasing the environmental burden of copper. By understanding responses of specific algal strains to management approaches and P levels, improved assessment of the effectiveness of management programs can be gained. Decreasing P availability can decrease the amount and need for reactive copper algaecides by altering the growth rates and carrying capacity of Ma strains and, in specific cases, by increasing the sensitivity of the cells to copper.

INTRODUCTION

Copper has been widely used to manage nuisance algal assemblages in freshwaters

(Oliveira-Filho 2004). Management practices that decrease application of copper-based algaecides are needed to promote ecological safety and comply with regulatory standards

(USEPA 2011). Methods that increase the sensitivity of a given algal strain to copper would be a valuable approach to decrease the amount of algaecide applied while also enhancing use efficiency. With increasing eutrophication in freshwaters and corresponding harmful algal blooms, nutrient mitigation (e.g., decreasing loading or availability) should be a key focus of management (Hallegraeff 1993; USEPA 2007; Heisler et al. 2008). The implications of eutrophication may go beyond simply promoting the presence and growth of noxious algae.

Previous research has shown that increased available phosphorus (P) in growth media has led to enhanced metal detoxification and decreased algal sensitivities to copper exposures (Rhee

1972, 1973; Jensen et al. 1982; Twiss and Nalewajko 1992; Guasch et al. 2004). Also, increases in cellular P content has coincided with a concomitant increase in copper tolerance, whereas P-limited cells were more susceptible (Wang and Dei 2006). By decreasing the

external bio-available P supply to algae, sensitivity to copper could be increased. In this research, the relationships between P bioavailability, cellular P content, and algal sensitivity were evaluated toward the goal of enhancing the efficiency of copper algaecide application.

The sensitivity of a noxious algal population to copper can vary widely based on water chemistry, physiological algal characteristics, and the copper formulation (Fitzgerald

1964). The innate sensitivity of an alga is considered constant if the exposure characteristics are similar, or if the exposure provides a similar dose (Bishop and Rodgers 2011). Algae deficient in P have shown increased sensitivity to copper exposure compared with the same population grown under P-replete conditions (Hall et al. 1989). This research investigates a mechanism that may account for the most significant component of an exposure (i.e., the organism), and how the inherent sensitivity can be shifted. Hypothetically, the sensitivities can be altered by limiting essential nutrients (Serra et al. 2010). Management strategies that remove bioavailable P from aqueous systems can be used to reduce cyanobacterial growth

(Oosterhout and Lürling 2013) and increase sensitivities to applied copper algaecides.

Addressing P levels may assist in indirect control of noxious algae by limiting carrying capacity and selecting for more beneficial algal assemblages (Smith 1983; Tilman et al.

1986; Seale et al. 1987). By shifting an algal population to a lower density/biomass, long- term control can be more easily maintained below action threshold levels, and a concomitant decreased amount of copper would be required to achieve a critical burden threshold (i.e., amount of copper per unit biomass needed for control; Bishop and Rodgers 2012). This research provides water resource managers with helpful information regarding the efficiency of managing noxious algal infestations by integrating P mitigation with algaecide controls.

Phosphorus and stoichiometric nutrient ratios (e.g., N:P; Si:P) are important controls on harmful cyanobacteria dominance (Rhee 1982; Tilman et al. 1982); therefore P removal may be a strategic approach to managing nuisance algal infestations (Grantz et al. 2013).

Inherent sensitivities differ among algal species and strains, so the shifts in response to P removal should also differ (Hutchinson and Stokes 1975). Differences in nutrient use, salinity/stress tolerance, morphology (Lakeman et al. 2009), and copper sensitivity (Twiss et al. 1993; Garcia-Villada et al. 2004) have been documented for different algal strains. As different algal strains require different levels of P to maintain growth (Tilman et al. 1982;

Saxton et al. 2012), P removal may impact those strains to a greater extent. Thus, increased selectivity can differentially impact noxious cyanobacteria (Downing et al. 2001).

Relationships between P levels and Cu toxicity have been measured in some laboratory

(Twiss and Nalewajko 1992; Rocha et al. 2016) and field studies (Guasch et al. 2004) on sensitive green algae and mixed periphyton.

The cyanobacterium Microcystis aeruginosa (Kutzing) Lemmerman (Ma) is widespread globally in freshwaters (Reynolds and Walsby 1975; Chorus and Bartram 1999).

Ma can be a prolific toxin producer, and can cause severe impacts to the quality of potable water, impede recreational activities, and kill livestock (Carmichael 2001). Negative ecological impacts are commonly observed during and following Ma blooms such as hypoxic zones (Paerl 1988), depressed fisheries health (Zhao et al. 2012; Liu et al. 2014), and impaired food web integrity (Lampert 1982). Ma has numerous adaptations to low-P environments, including storage (Jacobson and Halmann 1982; Whitton et al. 1991), luxury uptake (Baldia et al. 2007), use of high affinity transporters (Harke et al. 2012), alkaline

phosphatase activity (Štrojsová et al. 2005), and upregulation of multiple genes in P limiting conditions (Harke et al. 2012). Buoyancy regulation in Ma allows access to the hypolimnion to sequester P (Kromkamp et al. 1989; Jacoby and Frazer 2009), especially when epilimnetic

P is limiting (Ganf and Oliver 1982). Ma can also acquire and use diverse forms of P

(Štrojsová et al. 2005; Saxton et al. 2011). The majority of P accumulated in Ma is typically internalized (89%; Saxton et al. 2012) and stored as polyphosphate bodies which can be used to maintain growth in P-limiting conditions (Reynolds 2006; Shen and Song 2007).

Polyphosphate bodies have a high affinity for metals, and can sequester and ameliorate toxicity of Cd, Cu, Pb, Zn (Twiss and Nalewajko 1992; Verma et al. 1993), and Fe (Chaffin et al. 2011). The Ma phenotype can affect sensitivity to copper (Wu et al. 2007). More data are needed to evaluate the influence of P levels and accumulated P content on Ma susceptibility to various management practices.

Growth rates and algal health often parallel essential nutrient availability (Watson et al. 1997). By evaluating the increase in sensitivity to altered P regimes, more efficient management can be implemented. Management strategies can be created that incorporate multiple approaches (e.g., P mitigation and copper exposures) to achieve desired control. The overall goals of this research were to evaluate the responses of Ma strains to copper and to determine changes in sensitivity in response to P mitigation efforts. This research can be used to increase the efficiency of reactive algaecide programs while decreasing routine dependence on algaecide use and the overall algaecide loads applied.

The specific objectives of this research were to:

1. Measure the sensitivities of five cultured Microcystis aeruginosa strains to copper in

different P regimes;

2. Compare the sensitivities of Microcystis aeruginosa strains and assess the influence

of cellular P content;

3. Measure the physiological and toxicological responses of Ma strains following copper

exposures; and

4. Contrast differences in copper exposures required for control of Ma strains with

differing P growth regimes and cellular P quotas.

METHODS

Strains and exposure testing regimes

Five strains of Microcystis aeruginosa were attained from the University of Texas at

Austin culture collection (Table 1). Testing commenced with strains maintained in a stable, healthy culture within three months of culture initiation. They were cultured and tested in ultrafiltration purified COMBO nutrient media (Kilham et al. 1998; Table 2). Cultures were maintained in a controlled environment at 23 ± 1°C and a 16 hr light/8 hr dark photoperiod with fluorescent lighting (Spectralux T5/HO 6500K blue; 3000K red) at an intensity of 67.5

± 2.7 µmol photons/m2/s (Lewis et al. 1994). Total microcystins (following three freeze/thaw cycles in the dark) were measured in recently initiated cultures and cultures maintained for >

5 months using a Microcystins/Nodularins enzyme-linked immunosorbent assay (ELISA) Kit

(PN 520011; Abraxis, Inc. Warminster, PA 18974).

Experiments compared different initial levels of free P in the media to algal growth rates, chlorophyll a content, cellular P quota through time. Cell densities, chlorophyll a content, and P content were measured initiatially and 96 hr following exposure to copper sulfate pentahydrate (CuSO4 · 5H2O; C489-1 Fisher Scientific, Inc.) amendments, including

0.031, 0.0625, 0.125, 0.25, or 0.5 mg Cu/L in treatments (n=3; 125 mL volume in 150 mL acid-washed Erlenmeyer flasks). Controls did not receive copper amendments. Results from all experiments were pooled. Aqueous total copper concentrations were measured by taking

15 mL of exposure solution, acidifying to 2% v/v trace metal grade nitric acid then filtering

(0.22µm Whatman GF/F glass microfiber). Copper was measured using inductively coupled plasma-optical emission spectrometry (ICPE 9000; Shimadzu Corporation), including a matrix-matched calibration curve from serial dilution of a 1000 mg Cu/L standard (Fisher

Scientific, Inc. SC194; SMEWW 2005). The limit of detection for copper was 1 µg/L.

Method blanks were analyzed with each run to ensure no contamination by the materials used in sample preparation or analysis.

Phosphorus growth regime

The cyanobacteria were grown in three different P regimes to assess the effect of P supply on growth rates and sensitivities. Treatments of COMBO media P levels were 1,500 ±

90 µg P/L for full media (Full P), 0.1 P was 150 ± 20 µg P/L, and 75 ± 10 µg P/L represented the 0.05 P treatment level. Growth rates differed among treatments and strains. Ma densities of 5 x 106 cells/mL were used for initiation of toxicity experiments. Cyanobacteria grown in full and 0.1 P media were mostly in exponential growth phase when cell densities attained testing levels, whereas the cultures were approaching stationary phase in the 0.05 P media

(Table 1). Phenotypes and densities of Ma can alter exposure to copper (Franklin et al. 2002;

Wu et al. 2007) so we used similar densities of cells and unicellular or dividing phase cells for testing, and no large (> 10 cell) colonies were observed, as this could have altered the exposure efficiency.

Algal response parameters

Responses were measured initially and 96 hr post-treatment and compared based on the percent inhibition from untreated controls. The growth rate, µ, was calculated as follows:

Equation 1. µ = ln (Nt / No) / t

Nt and No represent final and initial cell densities respectively, and t is the exposure time after test initiation. Percent inhibition of growth (as cell density and chlorophyll a concentration) for each treatment were calculated as follows:

Equation 2. % Inhibition = [(µc - µt) / µc] * 100

µc and µt denote the mean value of response in the control and treatments, respectively.

Homogenized subsamples (10 mL) of Ma treatments were filtered (0.45 µm nitrocellulose) and used to measure chlorophyll a concentration. Each sample was placed in 5 mL of buffered acetone and sonicated to lyse cells (modified Standard Methods for the Examination of Water and Watewater [SMEWW] 2005). Chlorophyll a was measured fluorometrically using a Wallac Victor2 spectrofluorometer, by correlating with a matrix-matched standard calibration curve (10-640 µg/L; Sigma C-5753). Spectrophotometric scan (390-700 nm) was used to ensure chlorophyll a was the dominant light-harvesting pigment. Cell viability was measured by adding the mortal stain, methylene blue, to homogenized subsamples from Ma exposure vessels, targeting a final concentration of 0.02%. Counts of viable cells (not stained

blue; Corradi and Gorbi 1993), under a Zeiss Axioskop 20 light microscope, were conducted following 5 min of exposure. Cell densities were measured using an improved Neubauer

Hemocytometer (Hausser Scientific Co.) according to Standard Methods (10200 E/F,

SMEWW 2005).

Phosphorus content

Phosphorus content per cell was evaluated from 10 mL samples of homogenized cultures immediately before the experiments. Two techniques were used to confirm consistent results. First, cells were centrifuged at 4,000 rpm for 10 min and the pellet was analyzed. Second, 10 mL of homogenized sample (5 x 106 cells/mL) was gently filtered through a 0.45 µm nitrocellulose filter, rinsed with 10 mL COMBO media (- P) to remove any loosely adsorbed P, and then the cells and filtrate were analyzed separately for P. Results were similar between methods (5% variation), and the filtration technique was selected for use to account for loosely associated external P. Total P in cells and in water was analyzed following persulfate digestion (10 mL, 11 N sulfuric acid) and analysis on a discrete analyzer

(Konelab Aqua 420) according to Standard Methods (USEPA 1978). Calibration was conducted using P reference standard (Thermofisher Scientific, Inc.) of known concentration, processed in the same manner as sample materials. The P content per cell was calculated by dividing the mass of P attained by the number of cells filtered (50 ± 0.5 x 106 cells). Chaffin et al. (2011) found a positive correlation between many essential nutrients (i.e., K, Ca, Mg, S, and Fe) and P, based on cell quota. Nagai et al. (2007) reported that Fe-limited Microcystis had lower P content. All essential nutrients were supplied in excess in this experiment to

decrease any confounding factors from co-limitation, especially as found with Fe (Sterner

2008).

Statistics

A one-way analysis of variance (ANOVA) and Dunnett’s procedure were used to assess significant differences between untreated controls and P and algaecide treatments.

ANOVA and Tukey’s multiple comparison procedure (α = 0.05) were used to compare responses (growth rate, P content, chlorophyll a content, LC50 value) in different P growth regimes (McDonald 2014). Differences in sensitivities between strains and between P regimes within strains were also compared. Nonlinear regression, using a polynomial

2 quadratic equation (y = y0 + ax + bx ), best fit the data and was used to assess the exposure- response relationship and to calculate the 96-hr LC50 values of copper for each strain at each

P treatment level. A Shapiro-Wilk test was used to assess normality of the data, and all data were tested for constant variance (F-test; McDonald 2014). All data were analyzed using

Microsoft Excel (Microsoft 2010) and SigmaPlot version 12.5 (Systat Software 2014).

RESULTS AND DISCUSSION

Phosphorus and chlorophyll a content

Elevated P concentration in the media consistently led to elevated P content in Ma cells. There was a wide range in average P content of cells across all Ma strains tested (19-

238 fg P/cell), as has been reported in other works (e.g., 39-173 fg P/cell; Saxton et al. 2012).

Saxton et al. (2012) also found significant variance in Ma strains in terms of the cellular P

quota in relation to aqueous P concentration. All strains tested here responded similarly with significantly increased P content with increasing P concentration in the media (Figure 1).

A proportional increase in Ma chlorophyll a content with increasing P levels has been measured in other works, although significantly positively correlated with cell size (Chen et al. 2011). Chen et al. (2011) reported larger cell size when grown under low P conditions (4 mg/L) and smaller size at higher P concentrations (10 mg/L), much higher than the concentrations tested in this research. Cell size remained in a consistent range (Table 1) throughout the experiments. Therefore, changes in chlorophyll a and cellular P content likely resulted from physiological changes rather than cell size characteristics. The cellular chlorophyll a content is species- and strain-specific (Rhee and Gotham 1981). Here, increased chlorophyll a content was measured with increases in P concentrations. Strain Ma

2664 had similar chlorophyll a content in the 0.05 and 0.1 P regimes, although chlorophyll a content in the full P regime was significantly higher than in the 0.05 P treatment (P < 0.05).

All other Ma strains had significantly higher chlorophyll a content with each increase in aqueous P concentration (Figure 2).

Decreased P regimes in this research significantly altered the apparent carrying capacity and growth rates of all Ma strains. Growth rates in the 0.05 P media were significantly lower than in the full P media for all strains, and ranged from 0.009 d-1 in Ma

2386 to 0.036 d-1 in Ma 2664 and Ma 2061 (P < 0.05; Table 1). Growth rates in the full P and

0.1 P regimes ranged from 0.058-0.245 d-1 for all strains, and were similar to those found for two different Ma strains in previous research (0.058-0.162 d-1; Saxton et al. 2012).

Innate sensitivity of Ma strains

Since the growth rates and chlorophyll a content of Ma differed among strains and P regimes, the level of response that encompassed the 96-hr LC50 and potency slopes also differed (Figures 3-8). At the full P and 0.1 P regimes, all Ma strains tested had similar 96-hr

LC50 values. These ranged from 0.068 to 0.139 mg Cu/L based on chlorophyll a content, and

0.047 to 0.168 mg Cu/L based on cell densities (Figures 9 and 10; Table 3). In the 0.05 P growth regime, strain Ma 2386 had a significantly lower 96-hr LC50 value, based on chlorophyll a content and cell density, than all other Ma strains. Additionally, strain Ma 2665 had a significantly higher 96-hr LC50 value than Ma strains 2386, 2388, or 2664 (Figure 3;

Table 3). The relative sensitivities of strains grown in 0.05 P media to copper were Ma 2386

> 2388 ≥ 2664 ≥ 2061 ≥ 2665.

Toxicological influence of P content

The cyanobacterium Ma has multiple adaptations for acquiring P, including acquisition of diverse forms of P (Štrojsová et al. 2005; Saxton et al. 2011) and access to P supplies in lake sediments (Barbiero and Welch 1992; Jacoby and Frazer 2009). Harke and

Gobler (2013) reported that much of the Ma genome was differentially expressed under P limitation, which may trigger stress response genes that could alter susceptibility. Here, cyanobacteria were cultured in a consistent P media to minimize impacts from P storage or altered P acclimation conditions (as in Rocha et al. 2016). Under chronic decreased P levels, alterations in cellular processes have been measure including decreased protein content (Ji and Sherrell 2008), cell division processes, chlorophyll a synthesis, and photosynthetic rate

(Van Mooy et al. 2009). Low P can also promote increased cell membrane permeability due to a decrease in phospholipids, which may allow for increased penetration of copper

(Lombardi and Wangersky 1991). Although chlorophyll a content was significantly related to

P regimes across all strains, only one strain (Ma 2386) had significantly increased sensitivity to copper.

P content and copper sensitivity

Copper is often used to control cyanobacterial blooms (Hrudey et al 1999). This action can provide immediate relief, albeit temporary control (Haughey et al. 2000; Mastin et al. 2002). Understanding the amount of copper needed to efficiently control the target population is important for decreasing the copper applied and the potential long-term impacts of copper use (Jacob et al. 2016), while restoring the functionality of the water resource

(Mastin et al. 2000). Toxic modes of action of copper are mostly internal to the cell, including free radical production and inhibition of photosynthesis, respiration, enzyme function, and ATP production (Florence and Stauber 1986). Accumulated polyphosphate bodies have a high affinity for metals, and have been shown to sequester and ameliorate toxicity (Twiss and Nalewajko 1992; Verma et al. 1993; Chaffin et al. 2011). Naturally present copper-resistant cells have also been documented in field Ma populations, and they exhibited decreased fitness in the absence of copper (Garcia-Villada et al. 2004). Decreasing management dependency on copper sulfate use can reduce the likelihood of forming a less susceptible Ma population. This research revealed differences in susceptibility to copper among strains, and variability in the alteration of susceptibility based on accumulated P content. Copper detoxification by P compounds may allow for sustained blooms in elevated

P conditions (Zeng and Wang 2009), further decreasing the efficiency of copper at achieving control and increasing the continued need for reactive management. Such a situation could

further propagate naturally occurring copper-resistant cells, and increase the likelihood of copper-resistant populations (Garcia-Villada et al. 2004). Incorporating strategic P mitigation may enable increased efficacy of reactive management.

Zeng and Wang (2009) found much higher uptake of metals (Cd, Zn) with increasing ambient P concentrations in culture media, and P-replete cultures were more tolerant to these metal exposures. In this research, many strains did not exhibit increased sensitivity to copper under low P, suggesting that copper may parallel P in cellular internalization. Increased copper uptake with higher P cultures may correspond to their internal capability to detoxify copper in polyphosphate granules. In other work, cyanobacteria grown under P-replete conditions accumulated more heavy metals in polyphosphate granules than P-deplete cells

(Verma et al. 1993). Cells grown in low-P conditions may have less copper uptake, and less ability to ameliorate internal copper effects. Sustained, elevated growth rates in full-P cultures (> 0.1 d-1; except for Ma 2664) throughout the testing duration may have further altered uptake of copper. Stationary growth phase in the low-P cultures (growth rates < 0.05 d-1) likely affected uptake ability and the amount of ions taken up by the cells. Verma et al.

(1991, 1993) found less copper taken up by P-deplete cells in comparison to P-replete cells.

They hypothesized that phosphate plays an important role in regulating uptake of copper in low-phosphate conditions. In this study, cellular P quotas were inversely correlated with susceptibility to copper exposure for only one strain (Ma 2386). Increased growth rates were measured for all Ma strains in increasing P regimes, which may elicit a shorter interval between reactive treatments to maintain the designated uses of the water resource (Table 1).

Higher abundance (cells or biomass) can alter the mass of copper required to control the

population (Bishop and Rodgers 2012). By further understanding interactions among P concentration in the water, cellular P quotas and response of Ma, the feasibility of a reactive algaecide approach can be more reliably assessed. Furthermore, the amount of algaecide needed in different environmental conditions can be better predicted in order to achieve the desired level of control while decreasing the environmental burden of copper.

Future research directions

Although both nitrogen (N) and P have been indicated in supporting Ma blooms

(Chaffin et al. 2013; Monchamp et al. 2014), difficulty exists in limiting in situ N to Ma, especially since previously fixed atmospheric N can be used as a source (Beversdorf et al.

2013). Elevated N supply (Watanabe and Oishi 1985; Horst et al. 2014; Beversdorf et al.

2015) or decreased N:P ratios have increased toxin production in Ma (Orihel et al. 2012;

Harris et al. 2014). Microcystins have been implicated in hydroxyl free radical abatement

(Zilliges et al. 2011; Meissner et al. 2013), which can result from oxidized glutathione following copper exposures (Stauber and Florence 1987; Tripathi et al. 2006). Thus, N levels and N-rich cyanotoxins may affect copper toxicity, and will be specifically investigated in future work. Ma 2386 was the only strain tested that is not known to produce any microcystins, and it was the only strain that showed significantly altered sensitivity to copper depending on the P regime. Ma 2061 and Ma 2386 cultures had non-detect microcystins (<

0.1 µg/L) throughout the study. Ma 2665 cultures ranged from 12.3-20.1 µg microcystins/L,

Ma 2664 cultures ranged from 2.8-20.0 µg microcystins/L, and Ma 2388 cultures ranged from 0.6-12.9 µg microcystins/L. If toxin can ameliorate oxidative stress following algaecide exposures (e.g., copper or peroxide), selection for toxic strains may result (Burkholder and

Glibert 2006). Physiological changes in Ma strains, as governed by site-specific environmental conditions (Jacoby et al. 2000), need to be further understood to reliably predict responses of these noxious cyanobacteria to management approaches.

CONCLUSION

In this research, we assessed exposure dynamics that incorporate innate sensitivities of different Ma strains to copper sulfate, and influence of the P regime. P-deplete cells had decreased growth rates, chlorophyll a content, and P content in comparison to P-replete cells, across all Ma strains tested. Innate sensitivities to copper differed among Ma strains, and the influence of P levels also differed among strains. Only one strain (Ma 2386) had an increased sensitivity to copper in a low-P growth regime. This research describes another ramification of the eutrophication process, and further supports the need for P mitigation in water resource management. Incorporating P mitigation can allow for more strategic and efficient use of copper algaecides. By understanding the specific algal strain and P regime, management effectiveness can be better predicted.

Since environmental conditions can alter responses of Ma populations to copper application, it is important for water resource managers to consider these factors in designing an appropriate management strategy. This research can guide management recommendations to balance the ability to attain control using copper, with the benefits of P mitigation. In some situations, such as with strain Ma 2386, a hybrid approach to management (e.g., P reduction and copper algaecide application) may decrease the environmental burden of copper needed to attain control. Incorporation of nutrient mitigation may also provide the benefit of long-

term assemblage alteration, away from cyanobacterial dominance. Nuisance algal blooms are dynamic in space and time, and require multi-dimensional management approaches to attain control, especially under changing environmental conditions (e.g. temperature, CO2, nutrients, toxins) that may synergistically alter cyanobacterial response. This research provides insights on the direction and efficacy of an integrated approach to noxious algal management.

REFERENCES

Baldia SF, Evangelista AD, Aralar EV, Santiago AE. 2007. Nitrogen and phosphorus utilization in the cyanobacterium Microcystis aeruginosa isolated from Laguna de Bay, Philippines. Journal of Applied Phycology 19:607-613.

Barbiero RP, Welch EB. 1992. Contribution of benthic blue-green-algal recruitment to lake populations and phosphorus translocation. Freshwater Biology 27:249-260.

Beversdorf LJ, Miller TR, McMahon KD. 2013. The role of nitrogen fixation in cyanobacterial bloom toxicity in a temperate, eutrophic lake. PLoS ONE 8:e56103.

Beversdorf LJ, Miller TR, McMahon KD. 2015. Long-term monitoring reveals carbon- nitrogen metabolism is key to microcystin production in eutrophic lakes. Frontiers in Ecology 6:1-12.

Bishop WM, Rodgers Jr JH. 2011. Responses of Lyngbya magnifica Gardner to an algaecide exposure in the laboratory and field. Ecotoxicology and Environmental Safety 74:1832-1838.

Bishop WM, Rodgers Jr JH. 2012. Responses of Lyngbya wollei to exposures of copper- based algaecides: The critical burden concept. Archives of Environmental Contamination and Toxicology 62:403-410.

Burkholder JM, Glibert PM. 2006. Intraspecific variability: An important consideration in forming generalizations about toxigenic algal species. African Journal of Marine Science 28:177-180.

Carmichael WW. 2001. Assessment of blue-green algal toxins in raw and finished drinking water. AWWA Research Foundation. American Water Works Association, Denver, CO.

Chaffin JD, Bridgeman TB, Heckathorn SA, Mishra S. 2011. Assessment of Microcystis growth rate potential and nutrient status across a trophic gradient in western Lake Erie. Journal of Great Lakes Research 37:92-100.

Chaffin JD, Bridgeman TB, Bade DL. 2013. Nitrogen constrains the growth of late summer cyanobacterial blooms in Lake Erie. Advances in Microbiology 3:16-26.

Chen M, Li J, Dai X, Sun Y, Chen F. 2011. Effect of phosphorus and temperature on chlorophyll a contents and cell sizes of Scenedesmus obliquus and Microcystis aeruginosa. Limnology 12:187-192.

Chorus I, Bartram JE. 1999. Toxic Cyanobacteria in Water: A Guide to Public Health Significance, Monitoring and Management. E&FN Spon, London, United Kingdom.

Corradi MG, Gorbi G (1993) Chromium toxicity on two linked trophic levels. II. Morphophysiological effects on Scenedesmus acutus. Ecotoxicology and Environmental Safety 25:72-78.

Downing JA, Watson SB, McCauley E. 2001. Predicting cyanobacteria dominance in lakes. Canadian Journal of Fisheries and Aquatic Sciences 58:1905-1908.

Fitzgerald GP. 1964. Factors in the testing and application of algicides. Applied Microbiology 12:247-253.

Florence TM, Stauber JL. 1986. Toxicity of copper complexes to the marine diatom Nitzschia closterium. Aquatic Toxicology 8:11-26.

Franklin NM, Stauber JL, Apte SC, Lim RP. 2002. Effect of initial cell density on the bioavailability and toxicity of copper in microalgal bioassays. Environmental Toxicology and Chemistry 21:742-51.

Ganf GG, Oliver RL. 1982. Vertical separation of the light and available nutrients as a factor of causing replacement of green algae by blue-green algae in the plankton of a stratified lake. Journal of Ecology 70:829-844.

García-Villada L, Rico M, Altamirano M, Sánchez-Martín L, López-Rodas V, Costas E. 2004. Occurrence of copper resistant mutants in the toxic cyanobacteria Microcystis aeruginosa: characterisation and future implications in the use of copper sulphate as algaecide. Water Research 38:2207-2213.

Grantz EM, Haggard BE, Scott JT. 2013. Stoichiometric imbalance in rates of nitrogen and phosphorus retention, storage, and recycling can perpetuate nitrogen deficiency in highly productive reservoirs. Limnology and Oceanography 59:2203-2216.

Guasch H, Navarro E, Serra A, Sabater S. 2004. Phosphate limitation influences the sensitivity to copper in periphytic algae. Freshwater Biology 49:463-473.

Hall J, Healey FP, Robinson GGC. 1989. The interaction of chronic copper toxicity with nutrient limitation in two chlorophytes in batch culture. Aquatic Toxicology 14:1-14.

Hallegraeff GM. 1993. A review of harmful algal blooms and their apparent global increase. Phycologia 32:79-99.

Harke MJ, Berry DL, Ammerman JW, Gobler CH. 2012. Molecular response of the bloom- forming cyanobacterium, Microcystis aeruginosa, to phosphorus limitation. Microbial Ecology 63:188-189.

Harke MJ, Gobler CJ. 2013. Global transcriptional responses of the toxic cyanobacterium, Microcystis aeruginosa, to nitrogen stress, phosphorus stress, and growth on organic matter. PLoS ONE 8(7).

Harris TD, Wilhem FM, Graham JL, Loftin KA. 2014. Experimental manipulation of TN:TP ratios suppress cyanobacterial biovolume and microcystin concentration in large-scale mesocosms. Lake and Reservoir Management 30:72-83.

Haughey MA, Anderson MA, Whitney RD, Taylor WD, Losee RF. 2000. Forms and fate of Cu in a source drinking water reservoir following CuSO4 treatment. Water Research. 34:3440-3452.

Heisler J, Glibert PM, Burkholder JM, Anderson DA, Cochlan WP, Dennison WC, Dortch Q, Gobler C, Heil CA, Humphries E. 2008. Eutrophication and harmful algal blooms: A scientific consensus. Harmful Algae 8:3-13.

Horst GP, Sarnelle O, White JD, Hamilton SK, Kaul RB, Bressie JD. 2014. Nitrogen availability increases the toxin quota of a harmful cyanobacterium, Microcystis aeruginosa. Water Research 54:188-98.

Hrudey S, Burch S, Burch M, Drikas M, Greorgy R. 1999. Remedial measures. In: Chorus, I., Bartram, J. (Eds.), Toxic Cyanobacteria in Water. A Guide to their Public Health Consequences, Monitoring and Management. Routledge, London, United Kingdom.

Hutchinson TC, Stokes PM. 1975. Heavy metal toxicity and algal bioassays. American Society for Testing and Materials; special technical publication 573:320-343.

Jacob AP, Culver DA, Lanno RP, Voigt A. 2016. Ecological impacts of fluridone and copper sulphate in catfish aquaculture ponds. Environmental Toxicology and Chemistry 35:1183-1194.

Jacobson L, Halmann M. 1982. Polyphosphate metabolism in the blue-green alga Microcystis aeruginosa. Journal of Plankton Research 4:481-488.

Jacoby CA, Frazer TK. 2009. Eutrophication: time to adjust expectations. Science 324:723- 724.

Jacoby JM, Collier DC, Welch EB, Hardy FJ, Crayton M. 2000. Environmental factors associated with a toxic bloom of Microcystis aeruginosa. Canadian Journal of Fisheries and Aquatic Sciences 57:231-240.

Ji Y, Sherrell RM. 2008. Differential effects of phosphorus limitation on cellular metals in Chlorella and Microcystis. Limnology and Oceanography 53:1790-1804.

Jensen TE, Baxter M, Rachlin JW, Jani V. 1982. Uptake of heavy metals by Plectonema boryanum (Cyanophycea) into cellular components, especially polyphosphate bodies: An X-ray energy dispersive study. Environmental Pollution Series A: Ecological and Biological 27:119-127.

Kilham SS, Kreeger DA, Lynn SG, Goulden CE, Herrera L. 1998. COMBO: a defined freshwater culture medium for algae and zooplankton. Hydrobiologia 377:147-159.

Kromkamp J, Van Den Heuvel A, Mur LR. 1989. Phosphorus uptake and photosynthesis by phosphate-limited cultured of the cyanobacterium Microcystis aeruginosa. British Phycological Journal 24:347-355.

Lakeman MB, von Dassow P, Cattolico RA. 2009. The strain concept in phytoplankton ecology. Harmful Algae 8:746-758.

Lampert W. 1982. Further studies on the inhibitory effect of the toxic blue-green Microcystis aeruginosa on the filtering rate of zooplankton. Archives of Hydrobiology 95:207- 220.

Lewis PA, Klemm DJ, Lazorchak JM, Norberg-King TJ, Peltier TJ, Peltier WH, Heber MA. 1994. USEPA Short-term methods for estimating the chronic toxicity of effluents and receiving waters to freshwater organisms (Green alga, Selenastrum capricornutum, Growth 1003.0). 3rd ed. EPA/600.4-91/002, Cincinnati, OH.

Liu W, Qiao Q, Chen Y, Wu K, Zhang X. 2014. Microcystin-LR exposure to adult zebrafish (Danio rerio) leads to growth inhibition and immune dysfunction in F1 offspring, a parental transmission effect of toxicity. Aquatic Toxicology 155:360-367.

Lombardi AT, Wangersky P. 1991. Influence of phosphorus and silicon on lipid class production by the marine diatom Chaetoceros gracilis grown in turbidostat cage cultures. Marine Ecology Progress Series 77:39-47.

Mastin BJ, Rodgers Jr JH. 2000. Toxicity and bioavailability of copper herbicides (Clearigate, Cutrine Plus, and Copper Sulfate) to freshwater animals. Archives of Environmental Contamination and Toxicology 39:445-451.

Mastin BJ, Rodgers Jr JH, Deardorff TL. 2002. Risk evaluation of cyanobacteria-dominated algal blooms in a North Louisiana reservoir. Journal of Aquatic Ecosystem Stress and Recovery 9:103-114.

McDonald, JH. 2014. Handbook of Biological Statistics (3rd ed.). Sparky House Publishing, Baltimore, MD.

Meissner S, Fastner J, Dittmann E. 2013. Microcystin production revisited: Conjugate formation makes a major contribution. Environmental Microbiology 15:1810-1820.

Microsoft. 2010. Microsoft Excel [computer software]. Redmond, WA.

Monchamp M-E, Pick FR, Beisner BE, Maranger R. 2014. Nitrogen forms influence microcystin concentration and composition via changes in cyanobacterial community structure. PLoS ONE 9: e85573.

Nagai T, Imai A, Matsushige K, Fukushima T. 2007. Growth characteristics and growth modeling of Microcystis aeruginosa and Planktothrix agardhii under iron limitation. Limnology 8:261-270.

Oliveira-Filho EC, Lopes RM, Paumgartten FJR. 2004. Comparative study on the susceptibility of freshwater species to copper-based pesticides. Chemosphere 56:369- 374.

Oosterhout F-v, Lürling M. 2013. Phosphorus binding clay (Phoslock) in mitigating cyanobacterial nuisance: Effects on water quality variables and plankton. Hydrobiologia 710:265-277.

Orihel DM, Bird DF, Brylinsky M, Chen H, Donald DB, Huang DY, Giani A, Kinniburgh D, Kling H, Kotak BG, Leavitt PR. 2012. High microcystin concentrations occur only at low nitrogen-to-phosphorus ratios in nutrient-rich Canadian lakes. Canadian Journal of Fisheries and Aquatic Sciences 69:1457-1462.

Paerl HW. 1988. Nuisance phytoplankton blooms in coastal, estuarine, and inland waters. Limnology and Oceanography 33:823-847.

Reynolds CS. 1987. Cyanobacterial water blooms. Advances in Botanical Research 13:67- 143.

Reynolds CS. 2006. Ecology of Phytoplankton. Cambridge University Press, New York, NY.

Reynolds CS, Walsby AE. 1975. Water blooms. Biological Research 50:437e481.

Rhee G-Y. 1972. Competition between an alga and an aquatic bacterium for phosphate. Limnology and Oceanography 7:505-514.

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, Gotham IJ. 1981. The effect of environmental factors on phytoplankton growth: temperature and the interactions of temperature with nutrient limitation. Limnology and Oceanography 26:635-648.

Rhee G-Y. 1982. Effects of environmental factors on phytoplankton growth. In: Marshall KC (ed.) Advances in Microbial Ecology. Plenum Press, New York, NY.

100

Rocha GS, Lombardi AT, Melão Mda G. 2016. Influence of phosphorus on copper toxicity to Selenastrum gracile (Reinsch) Korshikov. Ecotoxicology and Environmental Safety 128:30-35.

Saxton MA, Morrow EA, Bourbonniere RA, Wilhelm SW. 2011. Glyphosate influence on phytoplankton community structure in Lake Erie. Journal of Great Lakes Research 37:683-690.

Saxton MA, Arnold RJ, Bourbonniere RA, McKay RML, Wilhelm SW. 2012. Plasticity of total and intracellular phosphorus quotas in Microcystis aeruginosa cultures and Lake Erie algal assemblages. Frontiers in Microbiology 3:1-7.

Seale DB, Boraas ME, Warren GJ. 1987. Effects of sodium and phosphate on growth of cyanobacteria. Water Research 21:625-631.

Serra A, Guasch H, Admiraal W, Van der Geest HG, Van Beusekom SAM. 2010. Influence of phosphorus on copper sensitivity of fluvial periphyton: the role of chemical, physiological and community-related factors. Ecotoxicology 19:770-780.

Shen H, Song L. 2007. Comparative studies on physiological responses to phosphorus in two phenotypes of bloom-forming Microcystis. Hydrobiologia 592:475-86.

Systat Software, Inc. 2014. SigmaPlot version 12.5. San Jose, CA.

Smith VH. 1983. Low Nitrogen to Phosphorus ratios favor dominance by blue-green algae in lake phytoplankton. Science 221:669-671.

Standard Methods for the Examination of Water and wastewater, 21st ed. (SMEWW). 2005. American Public Health Association, Washington, DC.

Stauber JL, Florence TM. 1987. Mechanism of toxicity of ionic copper and copper complexes to algae. Marine Biology 94:511-519.

Sterner RW. 2008. On the phosphorus limitation paradigm for lakes. International Review of Hydrobiology 93:433-445.

Štrojsová A, Vrba J, Nedoma J, Šimek K. 2005. Extracellular phosphatase activity of freshwater phytoplankton exposed to different in situ phosphorus concentrations. Marine and Freshwater Research 56:417-424.

Tilman D, Kilham SS, Kilham P. 1982. Phytoplankton community ecology: The role of limiting nutrients. Annual Review of Ecology and Systematics 13:349-372.

101

Tilman D, Kiesling R, Sterner R, Kilham SS, Johnson FA. 1986. Green, blue-green and diatom algae: taxonomic differences in competitive ability for phosphorus, silicon and nitrogen. Archiv für Hydrobiologie 106:473-485.

Tripathi BN, Mehta SK, Amar A, Gaur JP. 2006. Oxidative stress in Scenedesmus sp. during short- and long-term exposure to Cu+2 and Zn+2. Chemosphere 62:538-544.

Twiss MR, Nalewajko C. 1992. Influence of phosphorus nutrition on copper toxicity to three strains of Scenedesmus acutus (Chlorophyceae). Journal of Phycology 28:291-298.

Twiss MR, Welbourn PM, Schwartzel E. 1993. Laboratory selection for copper tolerance in Scenedesmus acutus (Chlorophyceae). Canadian Journal of Botany 71:333-338.

United States Environmental Protection Agency (USEPA). 1978. Method 365.3 Phosphorous, All Forms (Colorimetric, Ascorbic Acid, Two Reagent). Washington, DC.

Van Mooy BAS, Fredricks HL, Pedler BE, Dyhrman ST, Karl DM, Koblížek M, Lomas, MW, Mincer TJ, Moore LR, Moutin T, Rappé MS,Webb EA. 2009. Phytoplankton in the ocean use non-phosphorus lipids in response to phosphorus scarcity. Nature 458:69-72.

Verma SK, Singh SP, Singh RK. 1991. Nutritional control of copper uptake in the cyanobacterium Nostoc calcicola Breb. Biology of Metals 4:192-196.

Verma SK, Singh RK, Singh SP. 1993. Copper toxicity and phosphate utilization in the cyanobacterium Nostoc calcicola. Bulletin of Environmental Contamination and Toxicology 50:192-198.

Wang W-X, Dei RCH. 2006. Metal stoichiometry in predicting Cd and Cu toxicity to a freshwater green alga Chlamydomonas reinhardtii. Environmental Pollution 142:303- 312.

Watanabe MF, Oishi S. 1985. Effects of environmental factors on toxicity of a cyanobacterium (Microcystis aeruginosa) under culture conditions. Applied and Environmental Microbiology 49:1342-1344.

102

Watson SB, McCauley E, Downing JA. 1997. Patterns in phytoplankton taxonomic composition across temperate lakes of differing nutrient status. Limnology and Oceanography 42:487-495.

Whitton BA, Grainger SLJ, Hawley GRW, Simon JW. 1991. Cell-bound and extracellular phosphatase activities of cyanobacterial isolates. Microbial Ecology 21:85-98.

Wu ZX, Gan NQ, Huang Q, Song L-r. 2007. Response of Microcystis to copper stress-do phenotypes of Microcystis make a difference in stress tolerance? Environmental Pollution 147:324-330.

Zeng J, Yang L, Wang WX. 2009. Cadmium and zinc uptake and toxicity in two strains of Microcystis aeruginosa predicted by metal free ion activity and intracellular concentration. Aquatic toxicology 91:212-220.

Zhao SJ, Xie P, Li G, Chen J, Cai Y, Xiong Q, Zhao YY. 2012. The proteomic study on cellular responses of the testes of zebrafish (Danio rerio) exposed to microcystin-RR. Proteomics 12:300-312.

Zilliges Y, Kehr J-C, Meissner S, Ishida K, Mikkat S, Hagemann M, Kaplan A, Boerner T, Dittmann E. 2011. The cyanobacterial hepatotoxin microcystin binds to proteins and increases the fitness of Microcystis under oxidative stress conditions. PLoS ONE 6:e17615.

103

Table. 1. Description of the five Microcystis aeruginosa strains used in testing and growth rates throughout exposure duration. Cell diameter was measured at experiment initiation and conclusion. Growth rates are averages throughout experiment duration.

Description Ma 2386 Ma 2388 Ma 2664 Ma 2061 Ma 2665

Patterson’s 1036 Isolation number NRC-1(ss-17) s-15-b UWOCC019 UWOCC 017 AX W.P. Game Little Rideau Bruno, Rietvlei Dam, Reserve, Lake Mendota, Origin Lake, Saskatchewa, Pretoria, South Winburg, South Madison, WI Ontario, Canada Canada Africa Africa

Cell diameter 3.0-5.5 µm 3.5-6.0 µm 6.0-8.5 µm 1.0-3.0 µm 4.3-6.5 µm

Toxin producer No Possible Yes Possible Yes

Growth Rate 0.154 d-1 0.101 d-1 0.062 d-1 0.245 d-1 0.193 d-1 (Full P) Growth Rate 0.130 d-1 0.069 d-1 0.058 d-1 0.169 d-1 0.089 d-1 (0.1 P) Growth Rate 0.009 d-1 0.027 d-1 0.036 d-1 0.036 d-1 0.012 d-1 (0.05 P)

104

Table 2. Mean and ranges of water chemistry parameters for COMBO nutrient media used throughout all replicate experimentation.

Water Sample Measurement Measurement Parameter (range) (mean)

pH (SU) 7.8-8.7 8.4 Alkalinity 15.6-19.8 18.1 (mg/L as CaCO3) Hardness 38.1-45.4 42.2 (mg/L as CaCO3) Conductivity 250-440 318 (µS/cm) Dissolved Oxygen 8.1-9.2 8.5 (mg/L) Total Suspended Solids < 0.1 < 0.1 (mg/L) Chlorophyll a < 1 < 1 (µg/L) 1,410-1,590 (Full) 1,480 (Full) Total Phosphorus 130-170 (0.1P) 154 (0.1P) (µg/L) 65-85 (0.05P) 73 (0.05P) Total Kjeldahl Nitrogen 0.22-0.25 0.23 (mg/L) Nitrate & Nitrite 5-8 6.4 (mg/L) Turbidity (NTU) < 1 < 1

105

Table 3. Calculated 96-hr LC50 values for Ma chlorophyll a content and cell densities. The range represents the 95% confidence interval for the nonlinear (polynomial, quadratic) regression analysis. The corresponding R2 value is also given.

Chloroph- 95% Cell 95% yll a confidence R squared density confidence R squared 96-hr LC50 limit 96-hr LC50 limit Ma 2386 0.058- 0.074- Full 0.098 0.8658 0.112 0.9954 0.150 0.215 0.019- 0.016- 0.1P 0.068 0.7523 0.070 0.8638 0.153 0.132 0.012- 0.016- 0.05P 0.034 0.8694 0.039 0.8949 0.050 0.071 Ma 2061 0.082- 0.136- Full 0.129 0.9403 0.168 0.9887 0.210 0.210 0.123- 0.1P 0.139 0.081-.236 0.9067 0.146 0.9920 0.174 0.073- 0.05P 0.143 0.084-.310 0.8862 0.105 0.9517 0.159 Ma 2388 0.025- 0.031- Full 0.074 0.8764 0.064 0.8155 0.137 0.137 0.081- 0.014- 0.1P 0.105 0.9719 0.047 0.8157 0.141 0.152 0.076- 0.012- 0.05P 0.093 0.9854 0.061 0.8453 0.113 0.113 Ma 2664 0.074- 0.098- Full 0.121 0.8155 0.122 0.9427 0.188 0.155 0.068- 0.045- 0.1P 0.110 0.9048 0.137 0.9035 0.200 0.325 0.092- 0.041- 0.05P 0.121 0.9672 0.092 0.8740 0.163 0.189 Ma 2665 0.080- 0.062- Full 0.118 0.9409 0.096 0.9356 0.192 0.152 0.061- 0.071- 0.1P 0.139 0.8754 0.099 0.9642 0.264 0.136 0.166- 0.056- 0.05P 0.254 0.9365 0.123 0.8441 0.415 0.299

106

350 g

300

250

200 Ma 2061 fg Ma 2386 150 Ma 2388

fg P/cell fg de e ef de Ma 2664 100 cd d Ma 2665 c c b 50 ab ab a ab

0 0.05 P 0.1 P 1.0 P Growth Regimen

Figure 1. Mean P content per cell for five cultured Ma strains across three P growth regimes.

Error bars represent one standard deviation from the mean. Different letters signify significant differences (P < 0.05).

107

0.16 g 0.14 f f 0.12

0.1 f / cell /

a Ma 2061 0.08 Ma 2386 cde e Ma 2388 0.06 e e Ma 2664 d d d 0.04 Ma 2665

pg chlorophyll chlorophyll pg bc b b 0.02 a

0 0.05 P 0.1 P 1.0 P Growth Regimen

108

Figure 3. Mean chlorophyll a content of five Ma strains grown in full P media following a

96-hr exposure to copper sulfate pentahydrate. Error bars represent one standard error from the mean.

109

Figure 4. Mean cell densities of five Ma strains grown in full P media following a 96-hr exposure to copper sulfate pentahydrate. Error bars represent one standard error from the mean.

110

Figure 5. Mean chlorophyll a content of five Ma strains grown in 0.1 P media following a

96-hr copper exposure to copper sulfate pentahydrate. Error bars represent one standard error from the mean.

111

Figure 6. Mean cell densities of five Ma strains grown in 0.1 P media following a 96-hr exposure to copper sulfate pentahydrate. Error bars represent one standard error from the mean.

112

Figure 7. Mean chlorophyll a content of five Ma strains grown in 0.05 P media following a

96-hr exposure to copper sulfate pentahydrate. Error bars represent one standard error from the mean.

113

Figure 8. Mean cell densities of five Ma strains grown in 0.05 P media following a 96-hr exposure to copper sulfate pentahydrate. Error bars represent one standard error from the mean.

114

0.45

0.4

0.35

0.3

content) a a

50 0.25 hr LC hr

- 0.2 Full P 96 0.1P 0.15 0.05P

(mg (mg Cu/L v Chlorphyll 0.1

0.05

0 Ma 2386 Ma 2061 Ma 2388 Ma 2664 Ma 2665

Figure 9. Cumulative mean 96-hr LC50 values based on chlorophyll a content for five Ma strains grown in full, 0.1 P, and 0.05 P media following exposure to copper sulfate pentahydrate. Error bars represent 95% confidence intervals around the mean.

115

0.35

0.3

0.25

50 0.2

hr LC hr -

96 0.15 Full P 0.1P

(mg (mg Cu/L v Density) Cell 0.1 0.05P

0.05

0 Ma 2386 Ma 2061 Ma 2388 Ma 2664 Ma 2665

Figure 10. Cumulative mean 96-hr LC50 values based on cell densities for five Ma strains grown in full, 0.1 P, and 0.05 P media following exposure to copper sulfate pentahydrate.

Error bars represent 95% confidence intervals around the mean.

116

CHAPTER FOUR

COMPARISON OF PARTITIONING AND EFFICACY BETWEEN COPPER

ALGAECIDE FORMULATIONS: REFINING THE CRITICAL BURDEN CONCEPT

ABSTRACT

Filamentous mat-forming algae are increasingly impairing the designated uses of freshwater resources. Reactive management programs often involve application of a copper- based algaecide to decrease the standing crop of nuisance algae. Since copper algaecide formulations can differ significantly, this research outlined an advanced approach to evaluate formulation efficiency for targeted control of specific noxious algae. Two common filamentous algal species (Lyngbya wollei and Pithophora varia) were evaluated in laboratory bioassays to assess partitioned copper (internalized and adsorbed) and algal responses. Captain XTR achieved control (7-day EC85) of L. wollei with internal copper thresholds of 0.78 and 0.76 mg Cu/ g based on chlorophyll a content or filament viability, respectively. Cutrine Ultra achieved control of L. wollei based on filament viability at internalized copper of 0.85 mg Cu/ g, but did not achieve control based on chlorophyll a analyses. Internalized copper thresholds following Captain XTR exposures were similar for

P. varia, with values of 0.81 and 0.95 mg internal Cu/ g based on chlorophyll a and filament viability, respectively. The highest tested exposures to Cutrine Ultra and copper sulfate did not elicit control of P. varia, and did not reach the critical internal copper threshold with exposures tested in this research. Although some copper formulations did not attain the level of infused copper necessary to achieve control, infused copper levels did correlate

117

significantly with responses independent of formulation. The R2 values for relating infused copper to responses, among all formulations, were significant (P < 0.0001), with R2 at 0.920 and 0.935 for L. wollei based on filament viability and chlorophyll a content, respectively and R2 at 0.807 and 0.826 for P. varia based on filament viability and chlorophyll a content, respectively. The efficiency of copper formulations was assessed by calculating the bio- internalization factors. These were consistently higher in Captain XTR exposures (average

0.18), followed by Cutrine Ultra (0.14), and copper sulfate (0.09). Reactive management programs with copper should incorporate the efficiency of a copper algaecide formulation related to the internal critical burden of copper necessary to control the targeted algal species and strain. By measuring the efficiency of a specific algaecide and the corresponding amount required to achieve control of targeted algal biomass, management objectives can be achieved while decreasing the environmental loads of copper, the number of treatments, and the operational costs.

Key words: Algal management, copper algaecides, Lyngbya wollei, Pithophora varia, critical burden

INTRODUCTION

Copper-based algaecides are commonly used to control noxious algal blooms in water resources. The efficiency (affinity and transfer to toxic sites of action) of a specific application may be a function of several factors including water characteristics, copper formulation, concentration, and characteristics of the algal species and strain (Fitzgerald

1964, Fattom and Shilo 1984, Murray-Gulde et al. 2002). Copper formulations can differ in

118

terms of targeted algaecidal activity, likely correlated with partitioning of copper, in accordance with the critical burden concept (amont of sorbed copper required to achieve control of a certain mass of algae; Stauber and Florence 1987; Mastin and Rodgers 2000;

Murray-Gulde et al. 2002; Bishop and Rodgers 2012). Copper sulfate pentahydrate has been well studied (Kosalwat and Knight 1987; Nor 1987; Flemming and Trevors 1989), although more data are needed on chelated copper formulations as they may differ significantly in targeted algal control and non-target species risks (Masuda and Boyd 1993; Mastin and

Rodgers 2000). Bishop and Rodgers (2012) reported significant differences in algal response to copper formulations independent of total sorbed copper amounts. This finding further supports that the critical burden is not just dependent on the total amount of copper partitioned to the algae, but also on the amount that is introduced to the toxic sites of action

(i.e., the dose).

Several binding sites on an algal cell surface may have high affinity to positively charged species like copper. Copper adsorption to algal cell surfaces is rapid but transient.

Copper interaction may occur at physiologically active sites with substances such as carboxyl, phosphatic, or sulfhydryl groups that carry a net negative charge and potentially lead to uptake (Crist et al. 1990; Xue and Sigg 1990). Numerous sites on the algal cell surface are biologically inert such as amino and carboxyl groups of cell wall polysaccharides, proteins, and lipids (Veglio and Beolchini 1997; Kiefer et al 1997; Xue and Sigg 1990) that sorb copper independent of cell metabolism and may not contribute to uptake (Kaduková and

Virčíková 2005). Hassall (1962) and Tien et al. (2005) found that dead cells can sorb more copper than living cells, further suggesting inert adsorption mechanisms and independence

119

from metabolic activity. Ion exchange, complexation with exudutes, and microprecipitation are all other processes that may be involved in the external binding. Adsorption can occur rapidly (< 10 min), although it is mostly reversible and dependent on aqueous copper concentrations due to establishment of equilibrium conditions (Knauer et al. 1997; Veglio and Beolchini 1997).

Levy et al. (2007) suggested that internalization of copper likely is more correlated with toxicity than adsorption, due to inert external binding sites and exclusion capability.

Cho et al. (1994) and Khummongkol et al. (1982) considered active intracellular transport and passive diffusion as primary mechanisms of copper uptake. Specialized active transport ligands (e.g., transmembrane protein complexes) on the algal cell surface are often needed to uptake free metal ions, since the biological membrane is impermeable to charged or polar species (Sunda 1989). Copper sulfate, due to the dissociated free cupric ion form, may be restricted to facilitated active transport due to cell surface characteristics (Knauer et al. 1997;

Sunda 1989). Most studies assess only one form of copper and have drawn conclusions about algal tolerance (level of susceptibility) to copper, based on exposures to the free ion (e.g.,

CuCl2 or CuSO4 · 5H2O sources; Garcia-Villada et al. 2004). Organic chelating agents such as ethanolamine complexes are often used in copper algaecide formulations to enhance efficacy and improve stability of the copper ion (Strauss and Tucker 1993; Mastin and

Rodgers 2000). Stauber and Florence (1987) showed that lipid-soluble copper complexes can penetrate cell membranes through passive diffusion and, thus, are not strictly subject to active transport mechanisms. Stauber and Davies (2000) detected copper toxicity only after uptake into the cell. Key toxic mechanisms of action of copper are mostly internal to the cell,

120

and include enzyme inhibition, free radical production, and inhibition of photosynthesis, respiration, and ATP production (Florence and Stauber 1986). Covalent binding, redox reactions, and binding to intracellular proteins often incite the toxic action of internalized copper. These are mostly irreversible (Kuyucak and Volesky 1989), suggesting that the internally accumulated copper is critical for toxicity to a specific alga, and less transient than external copper. As sensitivities can differ among algal species and interactivity with copper formulations, this research provides a mechanistic approach for discerning the most effective formulation, based on copper introduction to key toxic sites of action.

Most standard algal toxicity testing protocols are used for regulatory purposes to evaluate the maximum potential for risks from a discharge effluent into receiving waters

(Organization for Economic Cooperation and Development [OECD] 1984; American Society for Testing and Materials [ASTM] 1994; United States Environmental Protection Agency

[USEPA] 1996, 2002; International Organization for Standardization [ISO] 2012). These toxicity bioassays often use sublethal responses of a known (or assumed) sensitive green algal or diatom species to assess environmental impact of contaminates introduced to an aquatic system. Significant results reported from these studies may be based on inhibited growth (algaestatic), though the toxicant may not have significantly depressed algal abundance. Despite valuable results from these culture studies, the methods may not be adequate to discern an acceptable degree of control in operational water resource management. Algal management is often conducted reactively to try to reduce a bloom of noxious algae, and biomass governs degree of control (Bishop and Rodgers 2012). Thus, growth inhibition is rarely acceptable for restoring water resource uses. Many mat-forming

121

filamentous algae often have decreased sensitivities to toxicants compared with planktonic, unicellular algae (Lembi 2000). Because of their nuisance growth structure and biomass accumulation, mat-forming species are common culprits impairing the designated uses of water resources, such as flood control, irrigation conveyance, potable supply, property values, and recreation (Hoagland et al. 2002; Landsberg 2002; Briand et al. 2003).

Filamentous algae can also cause devastating ecological impacts on food webs, decrease biodiversity, and alter biogeochemical cycling (Lembi et al. 1988; Paerl 1988; Speziale and

Dyck 1992; Turner et al. 1995; Mastin et al. 2002). Management programs are typically implemented to attain significant control (>75% mortality of initial algae; Mastin et al.

2002). This research is designed to evaluate mortality following exposure to potential reactive solutions for common nuisance mat-forming algae. This research seeks to further advance previously described methodologies (Bishop and Rodgers 2011; Bishop and

Rodgers 2012) to attain control of existing populations of nuisance algae by accurately characterizing the efficiency of algaecide formulations. By assessing the critical burden of algaecide for a specific algal strain, in conjunction with the efficiency of achieving internal toxicity thresholds, a specific algaecide formulation and corresponding amount to achieve control can be predicted more reliably.

Due to recent governmental regulations that mandate the action of minimizing discharges resulting from application of pesticides (USEPA 2011), algal management strategies which maintain the desired level of control while decreasing overall algaecide inputs are necessary. Furthermore, off-site movement of copper residues can increase potential impacts on non-target species. A large fraction of applied copper rapidly partitions

122

to the algae (Levy et al. 2007; Crist 1990) and can adversely affect the algae in this extracellular form (Sunda and Huntsman 1998), although toxicity is typically manifested only after transfer of copper into the cell (Stauber and Florence 1987; Stauber and Davies

2000). Calculating the critical burden of copper for an alga needs to be further refined to evaluate internalization of copper from copper-based algaecide exposures, due to increased correlation with toxicity with the multiple internal toxic sites of action. By selecting the copper formulation with the greatest affinity to the algal strain and efficiency of control, the application will result in the lowest amount of bio-available copper to non-target species.

This research will provide more efficient use of copper algaecide formulations (i.e., identifying most efficacious formulation and mass/volume of product required), thereby promoting more ecologically sound and cost-effective management (Sunda and Lewis 1978).

The goal of this research is to further correlate responses of algae to copper algaecide formulation exposures based upon internalization, external adsorption, and overall affinity of copper. By understanding the efficiency of applying a specific copper algaecide formulation, more accurate selection of an effective field use product and amount can be predicted and the margin of safety to non-target species increased.

Specific objectives of this research were to:

1. Compare copper adsorption and internalization by Pithophora varia and

Lyngbya wollei from exposures of three different copper algaecide

formulations;

2. Relate externally adsorbed and internalized copper amounts to control; and

123

3. Contrast the critical burden of copper required to control the targeted mass of

algae to copper partitioning between specific copper algaecide formulations.

MATERIALS AND METHODS

Testing and Culture Conditions

Two species of common nuisance filamentous algae were used in these experiments.

Pithophora varia Wille was collected from a reservoir at the SePRO Research and

Technology Campus (SRTC) in Whitakers, NC (36º 07ʹ 22.74ʺ N, 77º 43ʹ 40.47ʺ W).

Lyngbya wollei (Farlow ex Gomont) Speziale and Dyck was collected from Lay Lake, AL

(32º 59ʹ 19.17ʺ N, 86º 29ʹ 56.05ʺ W). Both species were identified using light microcopy

(Prescott 1970; Speziale and Dyck 1992; Wehr and Sheath 2003). The algae were cultured and tested in a controlled environment with a temperature of 23 ± 1°C and a 16-hr light/8-hr dark photoperiod with fluorescent lighting (Spectralux T5/HO 6500K blue; 3000K red) at an intensity of 67.5 ± 2.7 µmol photons/m2/s (Lewis et al. 1994). Studies were conducted at the

SRTC using well water which was similar to the site water in pH (7 ± 1.), DO (8 ± 1 mg

2 O2/L), temperature (23 ± 2°C), conductivity (130-350 μS/cm ), alkalinity 80-110 (mg/L as

CaCO3), and hardness (90-120 mg/L as CaCO3). All analyses were measured according to

Standard Methods (American Public Health Association [APHA] 2005).

Experimental Design

The mass of P. varia and L. wollei in each treatment was held constant at 0.1g ±

0.01g wet weight, and exposed to a series of aqueous copper concentrations targeting background, 1.25, 2.5, 5, 7.5, and 10 mg Cu/ g algae as Captain® XTR, Cutrine®-Ultra, and

124

copper sulfate pentahydrate in 500 mL flasks containing 250 mL water in seven-day (7-day) toxicity tests. Filaments were cleaned of debris and rinsed with deionized water to remove most other algae and organic material (Middleton and Frost 2014). The percent wet weight and standard deviation from the mean were 83.9 ± 3.0% for P. varia and 76.7 ± 3.1% for L wollei. Stock algaecide solutions were prepared within 4 hr of test initiation, and serial dilutions were used to obtain treatment copper concentrations. Characteristics of copper formulations are outlined in Table 1 (CuSO4 · 5H2O [Fisher Scientific, Inc.; C489-1];

Captain® XTR [SePRO Corporation, Carmel, IN]; Cutrine®-Ultra [Applied Biochemists, Inc.,

Germnatown, WI]). Six replicates of each exposure concentration were tested for each algaecide along with six untreated controls. All treatments and untreated controls were gently swirled twice daily by hand. Three replicates were randomly selected for response analyses and three were used for copper measurements. Replicated experiments (i.e., samples treated similarly as described above) were also conducted using field samples taken over a two-year period from the same field sites. This was done to evaluate the temporal consistency of laboratory results and rule out potential strain alterations due to laboratory culture conditions.

Furthermore, this reinforces field exposure relevance derived from laboratory results

(Lakeman et al. 2009). Responses and copper partitioning did not significantly differ between corresponding experiments; therefore, the data were pooled for analysis.

Algal Response: Filament Viability

Percent filament viability was measured by taking 0.05 g of P. varia or L. wollei from three replicates of each treatment and the untreated control, and immersing the biomass in

5mL of 0.1% methylene blue for approximately 10 min. As a mortal stain, cells that allow

125

entry of methylene blue are considered dead (Corradi and Gorbi 1993). One hundred filaments per replicate were randomly selected and examined, and the percent viability was determined by counting the proportion of filaments that were not internally stained bright blue.

Algal Response: Chlorophyll a Content

Algal chlorophyll a was measured on five initial samples and three replicates of each treatment and untreated controls. Due to the robust structure of P. varia and L. wollei, chlorophyll analysis was modified from standard methods by freezing the 0.05 g sample (-

12°C) for a minimum of 24 h. Each sample was placed in 10 mL buffered acetone and sonicated to lyse cells (modified from APHA 2005). Chlorophyll a was measured fluorometrically using a Wallac Victor2 spectrofluorometer by correlating with a matrix- matched standard calibration curve (10-640 µg/L; Sigma C-5753). To ensure that chlorophyll a was the most representative response pigment, an absorbance spectra scan was completed on extracted samples. Since Lyngbya can contain phycocyanin pigments that may become dominant through chromatic adaption, this step allowed us to ensure that other pigments accounted for < 10% of the total absorbance.

Copper Sorption and Critical Burden Measurements

The amount of copper absorbed by and adsorbed to the algae was measured in five initial samples to determine background levels and (n=3) for each treatment replicate and untreated controls. Adsorbed copper was measured by rinsing the P. varia and L. wollei with

10 mL of 2 mM ethylenediaminetetraacetic acid (EDTA) for 10 min to remove adhered metal ions from the cell surface. Filament structure was observed microscopically to ensure it was

126

still intact after EDTA extraction, as EDTA does not penetrate the cell. This also was consistently an effective mechanism to differentiate adsorbed versus intracellular copper

(Knauer et al. 1997). Internalized copper was measured by digesting the rinsed algae in 2 mL of 70% trace-metal-grade nitric acid (HNO3; Fisher Scientific, Inc. A509), 2 mL of 30%

H2O2, and 6 mL NanoPure™ water, sonicating for 30 seconds to disperse filaments, and then heating at 180°C until the solution appeared clear (Tripathi et al. 2006; USEPA 1996).

Aqueous total copper concentrations were measured by taking 15 mL of exposure solution, acidifying to 2% trace-metal-grade nitric acid, and then filtering (0.22 µm pore size

Whatman GF/F glass microfiber filter). A copper mass balance was conducted through the summation of sorbed fractions and measuring copper remaining in the water. Since total flask acidification provided similar results as subsamples, and this fulfilled the mass balance assessment, we concluded that there was minimal interaction of copper with the borosilicate flasks. Copper was measured using inductively coupled plasma-optical emission spectrometry (ICPE 9000; Shimadzu Corporation) with a matrix-matched calibration curve from serial dilution of a 1000 mg/L copper standard (Fisher Scientific, Inc. SC194; APHA

2005). The limit of detection for copper was 1 µg/L (0.1 µg Cu/ g algae). Method blanks were analyzed with each run to ensure that there was no contamination by the materials used in sample preparation or analysis.

The lowest exposure of copper at which control was achieved (i.e., 85% reduction in chlorophyll a concentration or filament viability compared with untreated controls) for each algaecide was identified as the critical burden in units of mg copper per gram of algae

(aqueous exposure and partitioned fractions). The 7-day, 85% effect concentrations (EC85)

127

were determined based on regression models for both filament viability and chlorophyll a content. These values were selected because they represent a desired level of control of an existing nuisance algal population, and provided a significant decrease in response variables compared with initial and control values (α = 0.05). Effect concentrations (when applicable) and 95% confidence intervals were calculated by nonlinear regression on both nominal and infused copper concentrations for each formulation. Chlorophyll a responses were best fit with regression using exponential decay, single three-parameter (f = y0+a*exp(-b*x)) equation. Filament viability was analyzed using exponential decay, single two-parameter (f = a*exp(-b*x)) and polynomial quadratic (f = y0+a*x+b*x^2) equations for L. wollei and P. varia, respectively.

Copper Algaecide Formulation Efficiency

The algal bio-concentration factor was used to compare affinities of different copper formulations to each strain, calculated by the equation:

Equation 1. BcF = (copper adsorbed + copper internalized) / copper amended to water.

The copper formulation efficiency was further compared based upon the proportion of copper internalized, excluding adsorbed copper, versus the total applied. This algal bio- internalization factor was calculated by the equation:

Equation 2. BiF = copper internalized / copper amended to water.

Correlation analysis of algal bio-internalization factors with the critical burden measurements was conducted to assess the lowest rate of copper (from most efficient formulation) that should be applied to attain the critical burden.

Statistical Analysis

128

A one-way analysis of variance (ANOVA) was used to discern differences in response parameters (chlorophyll a concentrations and filament viability) between treatments and untreated controls for each experiment, with differences identified through multiple range testing (Tukey’s test; α = 0.05). Exposure-response relationships were evaluated for each strain exposed to each algaecide, and EC85 values were calculated by regression analysis using chlorophyll a and filament viability as the response parameters. Critical burden measurements, bio-concentration and bio-absorption factors were compared among copper algaecide formulations using ANOVA (McDonald 2014). A Shapiro-Wilk test was used to assess if the data were normally distributed, and all data were tested for constant variance (F- test). All data were analyzed using Microsoft Excel (Microsoft 2010) and SigmaPlot version

12.5 (Systat Software 2014).

RESULTS

Response Parameters

Similar trends were observed between responses in both L. wollei and P. varia toxicity experiments. Exposure-response relationships showed a decrease in chlorophyll a content and filament viability with increasing copper exposures with all formulations.

Captain XTR had a significantly more potent impact on L. wollei filament viability at exposures ≥ 2.5 mg copper per gram of algae (mg Cu/ g) compared with the other formulations. Additionally, chlorophyll a content significantly decreased at the 5, 7.5, and 10 mg Cu/ g exposures compared to both other formulations. Captain XTR decreased both response parameters below designated control levels (7-day EC85) at the 5, 7.5, and 10 mg

129

Cu/ g exposures for both response parameters. Cutrine Ultra also exceeded the control threshold at the 10 mg Cu/ g exposure based on filament viability as well as the 7.5 and 10 mg/ g exposures for chlorophyll a content. Copper sulfate did not achieve control of L. wollei in any of the exposures tested (Table 2; Figures 1 and 2).

Based on chlorophyll a content, Captain XTR was more potent to P. varia than copper sulfate at exposures ≥ 5 mg Cu/ g and at the 10 mg Cu/ g concentration compared with Cutrine Ultra. Captain XTR significantly decreased filament viability at exposures of ≥

2.5 mg Cu/ g compared with copper sulfate and ≥ 5 mg Cu/ g compared with Cutrine Ultra.

Captain XTR exceeded the control threshold for both responses at the 7.5 and 10 mg Cu/ g exposures. Cutrine Ultra exceeded the control threshold with only the 10 mg/ g exposure based on chlorophyll a content. Copper sulfate did not achieve control of P. varia in any of the exposures tested (Figures 3 and 4).

Copper Partitioning

Adsorbed copper increased with increasing exposure concentrations among all formulations tested, although adsorbed copper alone did not correlate with designated algal control. Copper sulfate treatment resulted in significantly more copper adsorbed to L. wollei in the 1.25 and 2.5 mg Cu/ g exposures compared with Captain XTR and Cutrine Ultra and additionally more than Cutrine Ultra at the 5 and 7.5 mg Cu/ g exposures (Figure 5). As mentioned, however, copper sulfate did not achieve control of L. wollei at any of the exposures tested, whereas Captain XTR and Cutrine Ultra did achieve control based on filament viability. Captain XTR and copper sulfate treatment resulted in similar adsorbed copper with P. varia at the 1.25, 2.5, and 5 mg Cu/ g exposures. Copper sulfate treatment led

130

to more adsorbed copper at the 2.5 and 5 mg Cu/ g than Cutrine Ultra, although significantly less copper adsorption than both Captain XTR and Cutrine Ultra at 7.5 and 10 mg Cu/ g exposures (Figure 6).

Infused copper differed between formulations at corresponding copper exposures.

Captain XTR application led to significantly more infused copper in L. wollei than did copper sulfate at all exposures tested, and at concentrations ≥ 2.5 mg Cu/ g, in comparison with Cutrine Ultra treatment (Figure 7). Captain XTR also led to significantly more infused copper in P. varia than copper sulfate at the 2.5, 5, and 10 mg Cu/ g exposures, and more infused copper than Cutrine Ultra at the 7.5 and 10 mg Cu/ g exposures (P < 0.05; Table 3;

Figure 8).

A significant linear trend was observed between infused copper amounts and responses (chlorophyll a content and filament viability) with both P. varia and L. wollei (P <

0.0001). The R2 values for relating infused copper to responses, among all formulations, were 0.920 and 0.935 for L. wollei filament viability and chlorophyll a content, respectively

(Figures 9 and 10). The R2 values relating infused copper to responses, amongst all formulations, were 0.807 and 0.826 for P. varia filament viability and chlorophyll a content, respectively (Figures 11 and 12). Although some copper formulations did not attain the level of infused copper necessary to achieve control, infused copper amounts did elicit a similar response independent of formulation. Since control was achieved when the internal copper exceeded a specific threshold regardless of the formulation tested, this demonstrates that the internal fraction is most critical and correlative to predicting copper toxicity. Adsorbed copper varied widely between formulations and algal species, although there still was a

131

significant relationship between adsorbed copper and chlorophyll a content and filament viability control parameters (P < 0.05). Figures 13 and 14 show relationships between adsorbed copper and response parameters for L. wollei.

Average bio-concentration factors and ranges between exposures were similar with

Captain XTR and copper sulfate, although both averaged higher than with Cutrine Ultra.

Despite similar partitioning in accordance with the BcF, however, copper sulfate treatment did not achieve the designated level of control with either alga tested. Bio-internalization factors were consistently higher in Captain XTR exposures (average 0.18) followed by

Cutrine Ultra (0.14) and copper sulfate (0.09; Table 4). Partitioning of copper to the target alga in BcF assessments is important in assessing affinity, although the BiF is more critical for assessment of formulation efficiency as pertains to achieving control.

Critical Burden Thresholds, Copper Exposure, and Infusion

Significant differences were found between the exposure amounts of specific copper formulations that were required to achieve the internal critical threshold of copper for control. Captain XTR achieved control (7-day EC85) of L. wollei at 3.89 and 4.3 mg Cu/ g exposure for chlorophyll a and filament viability, respectively. Cutrine Ultra did not achieve control based on chlorophyll a results, and copper sulfate did not attain control with either chlorophyll a of filament viability results. Cutrine Ultra achieved control based on filament viability analyses, but at a very high exposure concentration of 8.56 mg Cu/ g, which was significantly higher than the exposures required for Captain XTR (P < 0.05; Table 3).

The threshold of infused copper necessary to elicit the critical burden (when achieved) was similar between response parameters and also between formulations. Captain

132

XTR achieved control of L. wollei with internal copper thresholds of 0.78 and 0.76 mg Cu/ g for chlorophyll a and filament viability, respectively. The regression model indicated that control of L. wollei was not achieved with Cutrine Ultra treatment at the highest exposure tested based on chlorophyll a responses, though this algaecide did achieve control at 0.85 mg internal Cu/ g based on filament viability analyses. Copper sulfate did not achieve control with the response parameters measured; the highest exposure only resulted in 0.55 mg internal Cu/ g (Table 3).

The copper exposure required to control P. varia in relation to L. wollei with Captain

XTR was not significantly different for the chlorophyll a content response, but significantly more Captain XTR was needed to elicit control based on filament viability. Yet, the infused copper amounts that elicited control of P. varia were similar to amounts required to achieve control of L wollei. Infused copper thresholds were also similar between response parameters of P. varia, with values of 0.81 and 0.95 mg internal Cu/ g for chlorophyll a and filament viability, respectively. The highest exposures tested of Cutrine Ultra and copper sulfate did not elicit control of P. varia, and only resulted in maximum infused copper concentrations of

0.53 and 0.74 mg Cu/ g, respectively (Table 3).

DISCUSSION

Copper Adsorption and Exclusion Mechanisms

The algae tested in this research, L. wollei and P. varia, possess thick barriers for regulating uptake (e.g. mucilage; thick cell wall); they also have a low surface area-to- volume ratio, and previousely have been described as inherently less responsive to copper

133

(Pearlmutter and Lembi 1980; Speziale and Dyck 1992; Lembi 2000). Bishop and Rodgers

(2012) found partitioned copper amounts alone lacked a strong correlation with responses of

Lyngbya wollei, supporting the present research as infused copper was more correlative to control rather than total sorbed copper.

Brown et al. (1988) reported that reduced cell membrane permeability was responsible for metal tolerance in Amphora coffeaeformis. Yan and Pan (2002) identified that the exclusion capability of the green alga Closterium lunula resulted in less copper bioaccumulation and provided tolerance to copper. Twiss et al. (1993) also reported that copper exclusion was the primary mechanism of acquired copper tolerance in a copper tolerant strain of Scenedesmus acutus. With higher adsorptive affinity at the cell surface, less copper was transferred to internal sites of action and, thus, represents a primary tolerance mechanism to copper. Significant copper binding was found in the cell wall and exopolymer sheath of the cyanobacteria Anabaena spp. and Microcystis sp. (Yee et al. 2004; Tien et al.

2005). Li et al. (2001) observed a strong affinity of copper for mucilaginous sheaths in cyanobacteria. Tien et al. (2005) found that the mucilaginous strands of Anabaena spiroides bound 2.4 mg Cu/ g dry weight and the mucilage accounted for 20-40% of adsorbed copper.

Function of the mucilage may be to decrease metal toxicity in polluted environments, and may impact efficacy of applied copper (Reynolds 2007).

Variation in surface sugars and their configuration can explain some adsorption differences in many algal groups including cyanobacteria (Tien 2002; De Philippis and

Vicenzini 1998; De Phillipis et al. 2001, 2003). Extracellular fine particulate matter and bacteria in aquatic systems may also interfere with copper uptake and contribute to adsorbed

134

fractions of copper (Tien 1999). Iron and manganese metal complexes adsorbed to the algal cell surface sorb ionic copper and reduce toxicity. Yet, these same external complexes were unable to sorb chelated copper, and could not protect the algae from lipid-soluble copper complexes (Stauber and Florence 1985a, b). The toxicity of chelated copper may not be as ameliorated by these external copper binding mechanisms, since it is already bound to chelated components. The ability to move across the external barrier due to chelation properties may promote the attainment of the internal critical threshold at lower exposures and, consequently, significantly increase toxicity to the algae.

Copper Internalization and Effects

Many factors can contribute to amelioration of toxicity from externally sorbed metals and binding is often reversible (Yan and Pan 2002), whereas intracellular copper is not readily removed from the cell due to its irreversible interactivity with numerous cell processes (Kuyucak and Volesky 1989). Some internal detoxification mechanisms for copper have been described, such as active efflux (Levy et al. 2007), sequestration in storage compartments (Soldo et al. 2005), or phytochelatin inactivation (Le Faucheur et al. 2005,

2006). These appear to reduce toxicity with relatively low amounts of copper or levels to which the algae have been acclimated, as opposed to pulse exposures of high introduced concentrations (e.g., algaecide application) which significantly alter the exposure proportion

(Sunda and Huntsman 1998). The ability to alter the internal toxicity of copper is predicted to differ depending on the initial form of copper, and most detoxification strategies only address the free cupric ion (Levy et al. 2007). The copper thresholds identified in this research did not appear to be impacted by internal defense mechanisms, as the degree of control was

135

proportional to internalized copper regardless of formulation. Further research regarding internal detoxification mechanisms for various copper species will allow for more informed decisions on copper use and resistance potential.

Copper exclusion may be the primary mechanism of tolerance to copper sulfate exposure. Butler et al. (1980) reported that a non-tolerant strain of Chlorella vulgaris accumulated over seven-fold more copper than a tolerant strain. Foster (1977) measured a similar accumulated copper threshold amount to elicit response with tolerant and sensitive strains of Chlorella vulgaris, though a higher (four-fold) aqueous copper concentration was required to achieve response in the tolerant strain. Quigg et al. (2006) reported that the copper-sensitive cyanobacterium Synechococcus sp. accumulated up to three-fold more copper than eukaryotic algae such as Tetraselmis sp. A similar threshold of infused copper elicited toxicity to the algae tested in this research, although much higher aqueous exposures of some formulations were required to achieve that internal accumulation. These findings are in close agreement for another metal with Zeng et al. (2009), who reported that inhibition of many cyanobacterial strains was closely related to intra-Zn concentrations and not aqueous concentrations.

Copper Formulation Versus Internalization

Copper sulfate has to be actively transported through the cell membrane or bound to an appropriate carrier in order to enter the cell and elicit a response on the internal toxic sites of action (Stauber and Florence 1987; Blust et al. 1987). This active transport decreases the

+2 area available for copper ions (Cu from CuSO4) to pass through the cell membrane (Knauer et al. 1997). Since active internalization is dependent on metabolic function, environmental

136

conditions such as temperature and light may inhibit this route of copper uptake and decrease algal responses (Cho et al. 1994; Genter 1996). With the algae tested in this research, copper as copper sulfate was primarily adsorbed and had only limited internal concentrations.

Consequently, copper sulfate did not elicit the same degree of control as chelated copper formulations.

Chelation of copper can occur naturally in aquatic systems (Stumm and Morgan

1970) and often decreases the toxicity of the cupric ion (Erickson et al. 1970; Gnassia-barelli et al. 1978). Numerous eukaryotic and prokaryotic algae produce organic copper complexing agents in stationary growth phase (McKnight and Morel 1980; Koukal et al. 2007). Some planktonic algal species secrete organic compounds in the presence of excess copper to ameliorate toxicity (Davey et al. 1973; Croot et al. 2000). Butler (1980) reported that organic complexes produced by Chlorella sp. may decrease toxicity and accumulation of copper when bound to the cupric ion. Fogg and Westlake (1955) and Gouvêa et al. (2005) found that peptides and other exudates produced by cyanobacteria can decrease toxic effects of the cupric ion and detoxify contaminated water. Chelated copper, already in a complexed state, would be expected to be less reactive to cellular exudates and natural organic matter complexes than the free cupric ion, thus promoting toxicity independent of defense mechanisms.

Interaction of copper with some organic complexes can therefore increase copper toxicity (Sutton et al. 1970; McIntyre and Gueguen 2013). Numerous algaecides intentionally utilize organic chelating agents (e.g., monoethanolamine, triethanolamine, ethylenediamine) to enhance efficacy and improve stability of the copper ion (Strauss and Tucker 1993; Mastin

137

and Rodgers 2000). Chelated copper algaecides differ in the type and amounts of chelators as well as the ratio of chelator(s) to metal ions. The stability of the formulation and the degree of interaction of the organic chelating agent with the specific metal ion are important aspects of chelated formulation potency (Freedenthal et al. 1985). In general, these formulations have more routes of entry, other than active transport alone, to achieve internal copper thresholds

(Stauber and Florence 1987). Despite differences in formulations, the chelated copper tested in this research elicited greater internalization of copper and had consistently increased efficacy compared with similar exposures of copper sulfate. These findings support results of prior works wherein enhanced efficacy was measured with chelated formulations than with copper sulfate (Bishop and Rodgers 2012; Calomeni et al. 2014; Iwinski et al. 2016).

The internal fraction of copper was primarily correlated with toxicity in both algal species tested, and high external adsorptive processes did not elicit the same degree of toxicity. The documented copper exclusion capabilities and lack of associated impacts further reinforce the need to measure internalized copper to establish the toxicity correlation.

Specifically, understanding the designed algaecide copper formulations and their introduction to internalized toxic sites of action is crucial for identifying a specific amount of algaecide needed to attain control. Since species/strain sensitivities differ (Twiss et al. 1993), and copper can adversely affect cells when bound externally (Sunda and Huntsman 1998), this concept needs to be further evaluated with other species and strains.

Field Translation

In this study, treatment with the chelated copper Captain XTR led to consistently more copper internalized at a corresponding exposure than copper sulfate or the other

138

chelated formulation. Copper formulations, regardless of chelation designation, do not elicit the same degree of control of various nuisance algae. By understanding the relationship between infusion proportion and the applied copper amounts of a specific formulation, a more reliable prediction can be made regarding field application efficacy. Incorporating the formulation efficiency along with the critical internal threshold of copper needed to control the target alga, a specific (and usually lower) amount of an algaecide formulation can be selected to control the designated algal biomass at a field site.

This can be represented mathematically using the equation:

Equation 3. Cuctl = (CBinf · Bm) / BiF where Cuctl is the mass of copper from a specific algaecide formulation required to achieve control of the target algal biomass (Bm) in a field application. This equation also incorporates the infused critical burden threshold of the target alga (CBinf) in terms of mass of copper from the specific formulation necessary to achieve control, and the bio- internalization factor (BiF) or amount of copper that is internalized from a corresponding exposure of a specific copper algaecide formulation.

By acting on internal mechanisms of action and using more efficient copper amounts to achieve the infusion threshold, control may be achieved without adding high enough copper levels to impact the short-term cellular membrane integrity (Sunda and Huntsman

1998). This is an important point for control of cyanobacterial blooms. Toxin release has been documented following extreme copper sulfate exposures that appeared to have lysed the cells (Kenefick et al. 1993; Touchette et al. 2008). Cells can be controlled by copper using different formulations that may not primarily act externally and significantly alter cell

139

membrane integrity. Selection of the most efficient copper formulation and exposure to achieve control, based on internal mechanisms of action, has been shown to decrease risks of leaking intracellular toxin contents into the water (Iwinski et al. 2016).

There was a strong correlation in this study between infused copper and responses across copper formulations. The exposure required to achieve that internal threshold differed significantly between formulations. Hypothetically, the aqueous exposure of each copper algaecide could be increased to a point to drive the internalization amount required to attain the designated level of control. This level was not achieved in the exposures tested, which were representative of typical field use patterns and viable algal management programs

(Bishop et al. 2015).

Bio-concentration and Bio-internalization

The bio-concentration factor has utility for assessing risks to non-target species, as it is inversely proportional to the dissolved copper remaining in the water (Bishop et al. in prep.), although it is insufficient to assess the applicability of a specific algaecide formulation for controlling the target algae. If the nuisance algae are not sufficiently controlled (i.e., restoration of water resource uses), it may spur future applications and increased copper application rates. The resulting, overall net increase of copper loading to the aquatic system may pose greater risks to non-target species through time (Jacob et al. 2016).

Sorption and desorption kinetics may alter the ultimate bio-concentration factor, primarily from adsorbed forms of copper (Knauer et al. 1997). Some adsorbed copper may be in a transitional state to facilitate the internalization and, therefore, less available for cellular desorption (Gonzalez-Davila et al. 1995). Since this study analyzed adsorbed copper at only

140

one point in time, the data may underestimate the amount that could be desorbed and cause non-target risks. Future studies will be needed to characterize time-sensitive sorption kinetics among copper fractions, and how formulation affects binding capacity and intensity. The external fraction likely will be much more transient than the infused copper (Knauer et al.

1997; Kaduková and Virčíková 2005), and formulations that increase copper infusion can attain control with a lower applied copper amount.

The total bio-concentration factors in this research were similar between the most effective and least effective copper formulations tested, although the bio-internalization was substantially different. This relationship reinforces the need to promote consideration of internalized copper for algal control. Copper internalized in algal cells is not likely to be released back to the water-column (Kaduková and Virčíková 2005), and tends to ultimately transition to more tightly bound states through time depending on available ligands (Liu et al.

2006; Jones et al. 2008). Increased copper adsorption with some copper sulfate treatments created a similar bio-concentration as chelated copper, though did not achieve control (at concentrations tested) in these experiments. Formulations of copper-based algaecides differ and formulation advancements are designed to enhance the affinity and efficacy for targeted algal management. These results support the premise that chelated formulations of copper algaecides require lower concentrations of applied copper to achieve an effective internal threshold amount for algal control.

141

CONCLUSION

This paper describes a methodology to determine an effective reactive management approach for controlling nuisance mat-forming algae with copper-based algaecides. Copper algaecide formulations differed significantly in terms of copper partitioning and consequent algaecidal efficacy. Internalized copper was critical in attaining control, independent of the amount of adsorbed copper. The internal threshold of copper required to control the nuisance algae was similar across formulations and species, although the aqueous exposures required to achieve that threshold differed significantly. The defined bio-internalization factor was used as a measure of formulation efficiency at attaining the infused critical burden threshold.

Predicting responses of a nuisance algal population to a designed program must incorporate the formulation efficiency, especially the internal threshold for the targeted nuisance algae, in order to accurately calculate the least amount of copper that can attain the desired level of control. A predictable amount of copper can be selected to attain control of a specific target nuisance algal biomass by coupling formulation efficiency for a specific field scenario with the internal threshold required for control. This research can assist water resource managers in selecting the formulation that contributes the most to the internal fraction while decreasing the copper that is desorbed or that remains in the water for potential off-site movement and non-target risks. Multiple applications of copper algaecides may be required to achieve the critical burden when there is: 1) high algae biomass, 2) low water volume, or 3) algae with low sensitivity to the algaecide. This research allows water resource managers to conceptualize an effective management program based on specific site characteristics.

142

Copper algaecide formulations differ significantly with respect to copper partitioning and consequent algaecidal efficacy (Stauber and Florence 1987; Mastin and Rodgers 2000;

Murray-Gulde et al. 2002; Bishop and Rodgers 2011; Bishop and Rodgers 2012). Numerous algal characteristics (e.g., exclusion mechanisms, biomass/density) and water chemistries can inhibit the internalization of copper, and this research incorporated those factors to predict an effective algal management program. Recent design of copper formulations may improve internalization of copper and attainment of the infused critical burden threshold. Algal sensitivity assessments need to account for different forms of copper in order to effectively apply copper to manage noxious algal species. Studies assessing a single form of copper, especially the cupric ion, can overestimate the amount of copper required to negatively impact the algae, and can also exaggerate risks to non-target species. Chelated copper algaecides should be highly considered in reactive algal management programs where copper is selected for use. Based on site and algal characteristics, chelated copper may allow for decreased environmental loading of copper (e.g., increased efficiency, less applications), lower potential for non-target risks (Closson and Paul 2014), and enhanced control of noxious algal species and strains.

This research further refined the critical burden concept, and showed that the infused copper amount per unit mass of algae is critical to achieving control. By measuring the internal critical burden of the targeted algal strain, and aligning that information with efficiency of a formulation, a specific algaecide and corresponding amount can be predicted to achieve control of the targeted algal biomass. This research can be used to: 1) select the most efficient copper formulation for a specific algal infestation, 2) determine the predicted

143

amount of product needed based on formulation efficiency and infused critical burden thresholds, and 3) evaluate potential non-target species risks/ offsite movement based on partitioning efficiency. A management program that incorporates formulation efficiency related to the internal threshold for control can achieve management objectives while decreasing the environmental loads of copper, the number of treatments, and the operational costs.

144

REFERENCES

American Public Health Association (APHA) (2005) Standard methods for the examination of water and wastewater. 21st ed. Washington, DC

American Society for Testing and Materials (ASTM) (1994) Standard Guide for Conducting Static 96h Toxicity Tests with Microalgae. Annual Book of ASTM Standards. ASTM E1218-1290. Philadelphia, PA

Applied Biochemists (2015) Cutrine Ultra Algaecide Safety Data Sheet. Applied Biochemists Inc. Germantown, WI 53022

Berberoglu H, Jay J, Pilon L (2008) Effect of nutrient media on photobiological hydrogen production by Anabaena variabilis ATCC 29413. International Journal of Hydrogen Energy 33:1172-1184

Bishop WM, Rodgers Jr JH (2011) Responses of Lyngbya magnifica Gardner to an algaecide exposure in the laboratory and field. Ecotoxicology and Environmental Safety 74:1832-1838

Bishop WM, Rodgers Jr JH (2012) Responses of Lyngbya wollei to exposures of copper- based algaecides: The critical burden concept. Archives of Environmental Contamination and Toxicology 62:403-410

Bishop WM, Willis BE, Horton CT (2015) Affinity and Efficacy of copper following an algicide exposure: Application of the critical burden concept for Lyngbya wollei control in Lay Lake, AL. Environmental Management 55:983-990

Blust R, Bernaerts F, Van der Linden A, Thoeye C (1987) The influence of aqueous copper chemistry on the uptake and toxicity of copper in Artemia. Artemia Research and its Applications 1:311-323

Briand JF, Jacquet S, Bernard C, Humbert JF (2003) Health hazards for terrestrial vertebrates from toxic cyanobacteria in surface water ecosystems. Veterinary Research 34:361- 377

Butler M, Haskew AEJ, Young MM (1980) Copper tolerance in the green alga, Chlorella vulgaris. Plant, Cell and Environment 3:119-126

Brown LN, Robinson MG, Hall BD (1988) Mechanisms of copper tolerance in Amphora coffeaeformis internal and external binding. Marine Biology 97:581-586

145

Calomeni A, Rodgers Jr. JH, Kinley CM (2014) Responses of Planktothrix agardhii and Pseudokirchneriella subcapitata to copper sulfate (CuSO4·5H2O) and a chelated copper compound (Cutrine Ultra). Water, Air, and Soil Pollution 225:2231

Cho DY, Lee ST, Park SW, Chung AS (1994) Studies on the biosorption of heavy metals onto Chlorella vulgaris. Journal of Environmental Science and Health, Part A 29:389-409

Closson KR, Paul EA (2014) Comparison of the toxicity of two chelated copper algaecides and copper sulfate to non-target fish. Bulletin of Environmental Contamination and Toxicology 93:660-665

Corradi MG, Gorbi G (1993) Chromium toxicity on two linked trophic levels. II. Morphophysiological effects on Scenedesmus acutus. Ecotoxicology and Environmental Safety 25:72-78

Crist RH, Martin JR, Guptill PW, Eslinger JM, Crist DR (1990) Interaction of metals and protons with algae. 2. Ion exchange in adsorption and metal displacement by protons. Environmental Science and Technology 24:337-342

Croot PL, Moffett JW, Brand LE (2000) Production of extracellular Cu complexing ligands by eucaryotic phytoplankton in response to Cu stress. Limnology and Oceanography 45:619-627

Davey EW, Morgan MJ, Erickson JJ (1973) A biological measurement of copper complexation capacity of seawater. Limnologica 18:993-997

De Philippis R, Paperi R, Sili C, Vincenzini M (2003) Assessment of the metal removal capability of two capsulated cyanobacteria, Cyanospira capsulata and Nostoc PCC7936. Journal of Applied Phycology 15:155-161

De Philippis R, Sili C, Paperi R, Vincenzini M (2001) Exopolysaccharide-producing cyanobacteria and their possible exploitation: A review. Journal of Applied Phycology 13:293-299

De Philippis R, Vicenzini M (1998) Exocellular polysaccharides from cyanobacteria and their possible applications. FEMS Microbiology Reviews 22:151-175

De Schamphelaere KAC, Stauber JL, Wilde KL, Markich SJ, Brown PL, Franklin NM, Creighton NM, Janssen CR (2005) Toward a biotic ligand model for freshwater green algae: Surface-bound and internal copper are better predictors of toxicity than free Cu+2 ion activity when pH is varied. Environmental Science and Technology 39:2067-2072

146

Erickson SJ, Lackie N, Maloney TE (1970) A screening technique for estimating copper toxicity to estuarine phytoplankton. Journal of the Water Pollution Control Federation 42:270-278

Fattom A, Shilo M (1984) Hydrophobicity as an adhesion mechanism of benthic cyanobacteria. Applied and Environmental Microbiology 1:135-143

Fisher Scientific (2008) Copper (II) sulfate pentahydrate Material Safety Data Sheet. Fisher Scientific Inc. Fair Lawn, NJ

Fitzgerald GP (1964) Factors in the testing and application of algicides. Applied Microbiology 12:247-253

Flemming CA, Trevors J (1989) Copper toxicity and chemistry in the environment: a review. Water, Air and Soil Pollution 44:141-158

Florence TM, Stauber JL (1986) Toxicity of copper complexes to the marine diatom Nitzschia closterium. Aquatic Toxicology 8:11-26

Fogg GE, Westlake DF (1955) The importance of extracellular products of algae in freshwater. Verhandlungen des Internationalen Verein Limnologie 12:219-232

Foster PL (1977) Copper exclusion as a mechanism of heavy metal tolerance in a green alga. Nature (London) 269:322-323

Freedenthal CB, Meyers MD, Stoner GA (1985) Copper based algaecides and aquatic herbicides. United States Patent number 4,505,734

García-Villada L, Rico M, Altamirano M, Sánchez-Martín L, López-Rodas V, Costas E (2004) Occurrence of copper resistant mutants in the toxic cyanobacteria Microcystis aeruginosa: characterisation and future implications in the use of copper sulphate as algaecide. Water Research 38:2207-2213

Genter RB (1996) Ecotoxicology of inorganic chemical stress to algae. In: Stevenson RJ, Bothwell ML, Lowe RL, editors. Algal ecology. Freshwater benthic ecosystems. Academic Press, San Diego, CA

Gnassia-Barelli M, Romeo M, Laumond F, Pesando D (1978) Experimental studies of the relationship between natural copper complexes and their toxicity to phytoplankton. Marine Biology 47:15-19

Gonzalez-Davila M, Santana-Casiano JM, Perez-Pena J, Millero FJ (1995) Binding of Cu(II) to the surface and exudates of the alga Dunaliella tertiolecta in seawater. Environmental Science and Technology 29:289-301

147

Gouvêa SP, Vieira AA, Lombardi AT (2005) Copper and cadmium complexation by high molecular weight materials of dominant microalgae and of water from a eutrophic reservoir. Chemosphere 60:1332-1339

Hassall KA (1962) A specific effect of copper on the respiration of Chlorella vulgaris. Nature 193:90

Hoagland P, Anderson DM, Kaoru Y, White AW (2002) The economic effects of harmful algal blooms in the United States: estimates, assessment issues, and information needs. Estuaries 25:819-837

International Organization for Standardization (ISO). 2012. Water quality – Freshwater algal growth inhibition test with unicellular green algae. ISO 8692:2012

Iwinski K J, Calomeni AJ, Geer T D, Rodgers JH (2016) Cellular and aqueous microcystin- LR following laboratory exposures of Microcystis aeruginosa to copper algaecides. Chemosphere 147:74-81

Jacob AP, Culver DA, Lanno RP, Voigt A. 2016. Ecological impacts of fluridone and copper sulphate in catfish aquaculture ponds. Environmental Toxicology and Chemistry 35:1183-1194.

Jones R, Hassan S, Rodgers Jr JH (2008) Influence of contact duration on sediment associated copper fractionation and bioavailability. Ecotoxicology and Environmental Safety 71:104-116

Kaduková J, Virčíková E (2005) Comparison of differences between copper bioaccumulation and biosorption. Environment International 31:227-232

Kamrin MA (1997) Pesticide profiles: Toxicity, environmental impact and fate. Lewis Publishers, Boca Raton, FL

Kenefick SL, Hrudey SE, Peterson HG, Prepas EE (1993) Toxin release from Microcystis aeruginosa after chemical treatment. Water, Science and Technology 27:433-440

Khummongkol D, Canterford GS, Fryer C (1982) Accumulation of heavy metals in unicellular algae. Biotechnology and Bioengineering 14:2643-2660

Kiefer E, Sigg L, Schosseler P (1997) Chemical and spectroscopic characterization of algae surfaces. Environmental Science and Technology 31:759-764

Knauer K, Behra R, Sigg L (1997) Adsorption and uptake of copper by the green alga Scenedesmus subspicatus (Chlorophyta). Journal of Phycology 33:596-601

148

Kosalwat P, Knight AW (1987) Acute toxicity of aqueous and substrate-bound copper to the midge, Chironomus decorus. Archives of Environmental Contamination and Toxicology 16:275-282

Koukal B, Rossé P, Reinhardt A, Ferrari B, Wilkinson KJ, Jean-LucLoizeau J, Dominik J (2007) Effect of Pseudokirchneriella subcapitata (Chlorophyceae) exudates on metal toxicity and colloid aggregation. Water Research 41:63-70

Kuyucak N, Volesky B (1989) The mechanism of cobalt biosorption. Biotechnology and Bioengineering 33:823-31

Lakeman MB, von Dassow P, Cattolico RA (2009) The strain concept in phytoplankton ecology. Harmful Algae 8:746-758

Landsberg JH (2002) The effects of harmful algal blooms on aquatic creatures. Reviews in Fisheries Science 10:113-390

Le Faucheur SV, Behra R, Sigg L (2005) Thiol and metal contents in periphyton exposed to elevated copper and zinc concentrations: a field and microcosm study. Environmental Science and Technology 39:8099-8107

Le Faucheur SV, Schildknecht F, Behra R, Sigg L (2006) Thiols in Scenedesmus vacuolatus upon exposure to metals and metalloids. Aquatic Toxicology 80:355-361

Lembi CA, O’Neal SW, Spencer DF (1988) Algae as weeds: economic impact, ecology, and management alternatives. In: C. A. Lembi and J. R. Waaland, eds. Algae and Human Affairs, Cambridge University Press, Cambridge, United Kingdom

Lembi CA (2000) Relative tolerance of mat-forming algae to copper. Journal of Aquatic Plant Management 38:68-70

Li P, Liu Z, Xu R (2001) Chemical characterization of the released polysaccharide from the cyanobaterium Aphanothece halophytica GR02. Journal of Applied Phycology 13:71- 77

Liu R, Zhao D, Barnett M (2006) Fate and transport of copper applied in channel catfish ponds. Water, Air, and Soil Pollution 176:139-162

Levy JL, Stauber JL, Jolley DF (2007) Sensitivity of marine microalgae to copper: The effect of biotic factors on copper adsorption and toxicity. Science of the Total Environment 387:141-154

Lewis PA, Klemm DJ, Lazorchak JM, Norberg-King TJ, Peltier TJ, Peltier WH, Heber WA (1994) USEPA short-term methods for estimating the chronic toxicity of effluents

149

and receiving waters to freshwater organisms (Green alga, Selenastrum capricornutum, Growth 1003.0). 3rd ed. EPA/600.4-91/002, Cincinnati, OH

Ma M, Zhu W, Wang Z, Witkamp GJ (2003) Accumulation, assimilation and growth inhibition of copper on freshwater alga (Scenedesmus subspicatus 86.81 SAG) in the presence of EDTA and fulvic acid. Aquatic Toxicology 63:221-228

Mastin BJ, Rodgers Jr JH (2000) Toxicity and bioavailability of copper herbicides (Clearigate, Cutrine Plus, and Copper Sulfate) to freshwater animals. Archives of Environmental Contamination and Toxicology 39:445-451

Mastin BJ, Rodgers Jr JH, Deardorff TL (2002) Risk evaluation of cyanobacteria-dominated algal blooms in a North Louisiana reservoir. Journal of Aquatic Ecosystem Stress and Recovery 9:103-114

Masuda K, Boyd CE (1993) Comparative evaluation of the solubility and algal toxicity of copper sulfate and chelated copper. Aquaculture 117:287-302

McDonald, JH (2014) Handbook of Biological Statistics (3rd ed.). Sparky House Publishing, Baltimore, MD

McKnight DM, Morel FMM (1980) Copper complexation of siderophores from filamentous blue green algae. Limnology and Oceanography 25:62-71

McIntyre AM, Gueguen C (2013) Binding interactions of algal derived dissolved organic matter with metal ions. Chemosphere 90:620-626

Microsoft. 2010. Microsoft Excel [computer software]. Redmond, WA

Middleton CM, Frost PC (2014) Stoichiometric and growth responses of a freshwater filamentous green alga to varying nutrient supplies: Slow and steady wins the race. Freshwater Biology 59:2225-2234

Murray-Gulde CL, Heatley JE, Schwartzman AL, Rodgers Jr. JH (2002) Algicidal effectivenessof Clearigate, Cutrine-Plus, and Copper Sulfate and margins of safety associated with their use. Archives of Environmental Contamination and Toxicology 43:19-27

Nor YM (1987) Ecotoxicity of copper to aquatic biota: A review. Environmental Research 43:274-282

Organization for Economic Cooperation and Development (OECD). 1984. Guideline for Testing Chemicals No. 201. Alga, Growth Inhibition Test. Paris, France

150

Paerl HW (1988) Nuisance phytoplankton blooms in coastal, estuarine, and inland waters. Limnology and Oceanography 33:823-847

Pearlmutter NL, Lembi CA (1980) Structure and Composition of Pithophora oedogonia (Chlorophyta) Cell Walls. Journal of Phycology 16:602-616

Prescott GW (1970) How to know the freshwater algae. 3rd ed. McGraw Hill, Boston MA

Quigg A, Reinfelder JR, Fisher N (2006) Copper uptake kinetics in diverse marine phytoplankton. Limnology and Oceanography 51:893-899

Reynolds CS (2007) Variability in the provision and function of mucilage in phytoplankton: facultative responses to the environment. Hydrobiologia 578:37-45

SePRO Corporation (2014) Captain XTR Algaecide Product Label. SePRO Corporation Carmel, IN 46032

SePRO Corporation (2015) Captain XTR Algaecide Safety Data Sheet. SePRO Corporation Carmel, IN 46032

SigmaPlot version 12.5 (2014) Systat Software, Inc., San Jose, CA

Soldo D, Hari R, Sigg L, Behra R (2005) Tolerance of Oocystis nephrocytioides to copper: Intracellular distribution and extracellular complexation of copper. Aquatic Toxicology 71:307-317

Sorrosa JM, Satoh M, Shiraiwa Y (2005) Low temperature stimulates cell enlargement and intracellular calcification of coccolithophores. Marine Biotechnology 7:128-133

Speziale BJ, Dyck LA (1992) Comparative taxonomy of Lyngbya wollei Comb. Nov. (Cyanobacteria). Journal of Phycology 28:693-706

Stauber JL, Davies CM (2000) Use and limitations of microbial bioassays for assessing copper bioavailability in the aquatic environment. Environmental Reviews 8:255-301

Stauber JL, Florence TM (1985a) The influence of iron on copper toxicity to the marine diatom, Nitzschia closterium (Ehrenberg) W. Smith. Aquatic Toxicology 6:297-305

Stauber JL, Florence TM (1985b) Interactions of copper and manganese: a mechanism by which manganese alleviates copper toxicity to the marine diatom, Nitzschia closterium (Ehrenberg) W. Smith. Aquatic Toxicology 7:241-254

Stauber JL, Florence TM (1987) Mechanism of toxicity of ionic copper and copper complexes to algae. Marine Biology 94:511-519

151

Straus DL, Tucker CS (1993) Acute toxicity of copper sulfate and chelated copper to channel catfish Ictalarus punctatus. Journal of the World Aquaculture Society 24:390-395

Stumm W, Morgan JJ (1970) Aquatic Chemistry. John Wiley and Sons, New York, NY

Sunda WG (1989) Trace metal interactions with marine phytoplankton. Biological Oceanography 6:411-442

Sunda W, Guillard RRL (1976) The relationship between cupric ion activity and the toxicity of copper to phytoplankton. Journal of Marine Research 34:511-529

Sunda WG, Huntsman SA (1998) Processes regulating cellular metal accumulation and physiological effects: Phytoplankton as model systems. Science of the Total Environment 219:165-181

Sunda WG, Lewis JAM (1978) Effect of complexation by natural organic ligands on the toxicity of copper to a unicellular alga, Monochrysis lutheri. Limnology and Oceanography 23:870-876

Sutton D L, Weldon LW, Blackburn RD (1970) Effect of diquat on uptake of copper in aquatic plants. Weed Science 18:703-707

Tien C-J (1999) Phytoplankton Mucilage – Quantification, Characterisation and Determination of Metal-binding Activity, PhD Thesis, Manchester University, United Kingdom

Tien C-J (2002) Biosorption of metal ions by freshwater algae with different surface characteristics. Process Biochemistry 38:605-613

Tien CJ, Sigee DC, White KN (2005) Copper adsorption kinetics of cultured algal cells and freshwater phytoplankton with emphasis on cell surface characteristics. Journal of Applied Phycology 17:379-89

Touchette BW, Edwards CT, Alexander J (2008) A comparison of cyanotoxin release following bloom treatments with copper sulfate or sodium carbonate peroxyhydrate. In: H.K. Hudnell (ed.) Cyanobacterial Harmful Algal Blooms: State of the Science and Research Needs. Springer, New York, NY

Tripathi BN, Mehta SK, Amar A, Gaur JP (2006) Oxidative stress in Scenedesmus sp. during short- and long-term exposure to Cu+2 and Zn+2. Chemosphere 62:538-544

Turner MA, Robinson GGC, Townsend BE, Hann BJ, Amaral JA (1995) Ecological effects of blooms of filamentous green algae in the littoral zone of an acid lake. Canadian Journal of Fisheries and Aquatic Science 52:2264-2275

152

Twiss MR, Welbourn PM, Schwartzel E (1993) Laboratory selection for copper tolerance in Scenedesmus acutus (Chlorophyceae). Canadian Journal of Botany 71:333-338

United States Environmental Protection Agency (USEPA) (1996) Microwave assisted acid digestion of siliceous and organically based matrices, method 3052. EPA SW-846, Ch. 3.2, USEPA. Washington, DC

United States Environmental Protection Agency (USEPA) (1996) Ecological Effect Test Guidelines. OPPTS 850.5400. Algal Toxicity, Tiers I and II. Washington, DC

United States Environmental Protection Agency (USEPA) (2002) Short-term methods for estimating the chronic toxicity of effluents and receiving waters to freshwater organisms (Green Alga, Selenastrum capricornutum, Growth 1003.0). 4th ed. EPA- 821-R-02-012. Washington, DC

United States Environmental Protection Agency (USEPA) (2011) National Pollutant Discharge Elimination System (NPDES). Pesticide General Permit (PGP) for Discharges from the Application of Pesticides. Washington, DC

Veglio F, Beolchini F (1997) Removal of metals by biosorption: a review. Hydrometallurgy 44:301-316

Wehr JD, Sheath RG (2003) Freshwater Algae of North America. Academic Press, New York, NY

Xue HB, Stumm W, Sigg L (1988) The binding of heavy metals to algal surfaces. Water Research 22:917-926

Xue HB, Sigg L (1990) Binding of Cu(II) to algae in a metal buffer. Water Research 24:1129-1136

Yan H, Pan G (2002) Toxicity and bioaccumulation of copper in three green microalgal species. Chemosphere 49:471-476

Yee N, Benning LG, Phoenix VR, Ferris FG (2004). Characterization of metal-cyanobacteria sorption reactions: A combined macroscopic and infrared spectroscopic investigation. Environmental Science and Technology 38:775-782

Zeng J, Yang L, Wang WX (2009) Cadmium and zinc uptake and toxicity in two strains of Microcystis aeruginosa predicted by metal free ion activity and intracellular concentration. Aquatic Toxicology 91:212-220

153

Table 1. Comparison of physical and chemical properties of the copper formulations tested.

Copper sulfate Captain® XTR a, b, c Cutrine® Ultra c, d pentahydrate c, e

Manufacturer SePRO Corporation Lonza Group Ltd. Fisher Scientific 7758-99-8 (CAS No.) C489-1 (Catalogue EPA Reg. No. 8959- EPA Reg. No. 8959- Identification number) 53 53 249.68 formula weight Copper-ethanolamine Copper-ethanolamine Active Ingredient CuSO ∙5H O; Cupric complexes (SP9000 complexes (D- 4 2 (formulation) ion surfactants) limonene) % Active 28.2 27.8 25.4 Ingredient Triethanolamine Triethanolamine- complex- 9.6% 20-30% Chelator amount NA Monoethanolamine Ethanolamine- complex- 13.3% 18-28% Appearance Blue viscous liquid Blue viscous liquid Blue crystalline solid

Water Solubility Miscible Miscible 32 g/ 100 g (20°C) 3.5-4 Concentrate pH 10-10.5 10.2-10.3 (5% aqueous (SU) solution) Specific Gravity 1.2 1.23 NA (g/cm3) a SePRO 2014 b SePRO 2015 c Kamrin 1997 d Applied Biochemists 2015 e Fisher Scientific 2008

154

Table 2. Description of exposure-response relationships using nonlinear regression analysis with a best fit analysis for the data. Equations and R squared values are calculated for each treatment in 7-day L. wollei and P. varia toxicity experiments.

Chlorophyll a responses were analyzed by nonlinear regression using exponential decay, single three-parameter (y = y0+a*exp(-b*x)) equation. Filament viability was analyzed using exponential decay, single two-parameter (y = a*exp(-b*x)) and polynomial quadratic (y = y0+a*x+b*x^2) equations for L. wollei and P. varia, respectively.

Filament Viability Chlorophyll a R squared R squared Treatment Regression Equation Regression Equation value value (y=) (y=) Lyngbya wollei y = 140.8+1,760*exp(- Captain XTR y = 97.81*exp(-0.443*x) 0.9608 0.9940 0.637*x) y = 335.6+1,567*exp(- Cutrine Ultra y = 97.28*exp(-0.225*x) 0.9868 0.9753 0.85*x) y = 613.4+1,282*exp(- Copper sulfate y = 99.23*exp(-0.128*x) 0.9570 0.9493 0.8826*x) Pithophora varia y = 97.95- y = 59.54+2,078*exp(- Captain XTR 0.9956 0.9974 15.28*x+0.5892*x^2 0.358*x) y = 97.48-8.045*x- y = 314.8+1,772*exp(- Cutrine Ultra 0.9934 0.9628 0.0003*x^2 0.376*x) y = 97.57-5.344*x- y = 183.7+2,008*exp(- Copper sulfate 0.9914 0.9914 0.0272*x^2 0.203*x)

155

Table 3. Calculated 7-day EC85 toxicity values for L. wollei and P. varia exposed to copper algaecide formulations.

Regression models of responses (chlorophyll a or filament viability) versus exposures (nominal or internalized copper) were used to calculate toxicity thresholds.

EC EC 85 EC EC 85 Chlorophyll 95% 85 95% 85 95% Filament 95% Chlorophyll Filament a Confid- Confid- Confid- viability Confid- Treatment a viability (mg Cu/ g ence ence ence (mg Infused ence (mg Infused (mg Cu/ g algae) Interval Interval Interval Cu / g algae) Interval Cu/ g algae) algae)

Lyngbya wollei Captain 2.91 - 0.44 - 3.21 - 0.57 - 3.89 0.78 4.36 0.76 XTR 7.52 1.26 5.58 0.94 Cutrine 7.34 - 0.80 - > 10 NA > 0.87 NA 8.56 0.85 Ultra 9.79 0.92 Copper > 10 NA > 0.55 NA > 10 NA > 0.55 NA sulfate Pithophora varia Captain 4.98 - 6.87 - 0.68 - 5.80 0.81 0.61 - 1.2 7.84 0.95 XTR 6.92 9.32 1.21 Cutrine > 10 NA > 0.53 NA > 10 NA > 0.53 NA Ultra Copper > 10 NA > 0.74 NA > 10 NA > 0.74 NA sulfate

156

Table 4. Average bio-concentration and bio-internalization factors for copper formulation exposures in L. wollei and P. varia toxicity experiments. Ranges represent values across all treatment exposure concentrations.

Bio-Concentration Bio-Internalization Bio-Concentration Bio-Internalization Treatment Factor Factor Factor Factor Lyngbya wollei Lyngbya wollei Pithophora varia Pithophora varia

0.460 0.180 0.519 0.156 Captain XTR (0.379 - 0.559) (0.126 - 0.235) (0.459 - 0.600) (0.109 - 0.241)

0.378 0.144 0.373 0.107 Cutrine Ultra (0.268 - 0.469) (0.087 - 0.193) (0.349 - 0.407) (0.053 - 0.165)

0.457 0.090 0.407 0.096 Copper sulfate (0.307 - 0.662) (0.056 - 0.121) (0.292 - 0.502) (0.065 - 0.148)

157

Figure 1. Percent viable L. wollei filaments with a series of copper exposures from three different algaecide formulations. The horizontal reference line represents an 85% decrease compared with untreated controls following a 7-day exposure duration. Error bars represent

95% confidence intervals around the mean. Nonlinear regression performed using exponential decay, single two-parameter equation y = a*exp(-b*x).

158

Figure 2. Chlorophyll a content of L. wollei with a series of copper exposures from three different algaecide formulations. The horizontal reference line represents an 85% decrease compared with untreated controls following a 7-day exposure duration. Error bars represent

95% confidence intervals around the mean. Nonlinear regression performed using exponential decay, single three-parameter equation y = y0+a*exp(-b*x).

159

95% confidence intervals around the mean. Nonlinear regression performed using polynomial quadratic equation y = y0+a*x+b*x^2.

160

95% confidence intervals around the mean. Nonlinear regression performed using exponential decay, single three-parameter equation y = y0+a*exp(-b*x).

161

3.5 ab

3 a

a 2.5

ab b 2 ab a b Captian XTR 1.5 b Cutrine Ultra

a Copper sulfate 1 b

a b Adsorbed copper (mg/ g) (mg/ Adsorbed copper 0.5 b b a a a 0 Control 1.25 2.5 5 7.5 10 Treatment (mg Cu/ g algae)

Figure 5. Mean adsorbed copper to L. wollei following a 7-day exposure duration with a series of copper exposures from three different algaecide formulations. Error bars represent

95% confidence intervals around the mean. Different letters represent significant differences between formulations at a specific copper exposure (α = 0.05).

162

4.5 a

4 a b 3.5

3 b c 2.5 c a a Captian XTR 2 Cutrine Ultra Copper sulfate 1.5 b

a 1 ab

Adsorbed copper (mg/ g) (mg/ Adsorbed copper b a a 0.5 b a a a 0 Control 1.25 2.5 5 7.5 10 Treatment (mg Cu/ g algae)

Figure 6. Mean adsorbed copper to P. varia following a 7-day exposure duration with a series of copper exposures from three different algaecide formulations. Error bars represent

95% confidence intervals around the mean. Different letters represent significant differences between formulations at a specific copper exposure (α = 0.05).

163

1.4 a

1.2 a a

1 b b b 0.8 Captian XTR a c Cutrine Ultra 0.6 b c c Copper sulfate

0.4 c

Infused g) (mg/ copper Infused a a b 0.2 a a a 0 Control 1.25 2.5 5 7.5 10 Treatment (mg Cu/ g algae)

Figure 7. Mean infused copper in L. wollei following a 7-day exposure duration with a series of copper exposures from three different algaecide formulations. Error bars represent 95% confidence intervals around the mean. Different letters represent significant differences between formulations at a specific copper exposure (α = 0.05).

164

1.4 a

1.2

1 a ab ab b 0.8 a b c Captian XTR a Cutrine Ultra 0.6 b ab Copper sulfate a b 0.4

Infused (mg/g) copper Infused a a

0.2

a a a 0 Control 1.25 2.5 5 7.5 10 Treatment (mg Cu/ g algae)

Figure 8. Mean infused copper in P. varia following a 7-day exposure duration with a series of copper exposures from three different algaecide formulations. Error bars represent 95% confidence intervals around the mean. Different letters represent significant differences between formulations at a specific copper exposure (α = 0.05).

165

Figure 9. Relationship between infused copper and viable L. wollei filaments following a 7- day exposure with three copper formulations. The trend line shows a linear relationship for all data (R2 = 0.920, P < 0.0001). Error bars represent 95% confidence intervals around the mean of both infused copper and response.

166

167

Figure 11. Relationship between infused copper and viable P. varia filaments following a 7- day exposure with three copper formulations. The trend line shows a linear relationship for all data (R2 = 0.807, P < 0.0001). Error bars represent 95% confidence intervals around the mean of both infused copper and response.

168

Figure 12. Relationship between infused copper and P. varia chlorophyll a content following a 7-day exposure with three copper formulations. The trend line shows a linear relationship for all data (R2 = 0.826, P < 0.0001). Error bars represent 95% confidence intervals around the mean of both infused copper and response.

169

Figure 13. Relationship between adsorbed copper and viable L. wollei filaments following a

7-day exposure with three copper formulations. The horizontal reference line represents an

85% decrease in response compared with untreated controls. Error bars represent 95% confidence intervals around the mean of both adsorbed copper and response.

170

Figure 14. Relationship between adsorbed copper and L. wollei chlorophyll a content following a 7-day exposure with three copper formulations. The horizontal reference line represents an 85% decrease in response compared with untreated controls. Error bars represent 95% confidence intervals around the mean of both adsorbed copper and response.

171

CHAPTER FIVE

EVALUATION OF RISKS FROM COPPER ALGAECIDE APPLICATIONS: THE

COMPETING LIGAND HYPOTHESIS

ABSTRACT

Copper algaecides are routinely applied to target noxious algal blooms in freshwaters.

Standard toxicity testing data with copper suggest that typical application concentrations can cause deleterious acute impacts to non-target organisms. Few studies have assessed the impact of specific binding of copper to target ligands, and the subsequent alteration of risks to non-targets species. Copper formulations also differ with the goal of enhancing the efficacy of control of the target algal species/strain. This research evaluated the influence of living algae as a component of the predicted environmental exposure with copper-based algaecide formulations, in order to inform accurate risk assessment from more realistic field exposures. Significant shifts in Daphnia magna 48-hr LC50 values were found when algae were present in exposures along with a copper salt or ethanolamine-chelated copper formulation (Captain). Copper sulfate 48-hr LC50 values shifted from 75.3 to 317.8 and 517.8 mg Cu/L whereas Captain increased from 353.8 to 414.2 and 588.5 mg Cu/L in no algae, 5 x

105 and 5 x 106 cells/mL algae treatments, respectively. Larger shifts were measured with copper sulfate exposures, although Captain was less toxic to Daphnia magna in all corresponding treatments based on total copper amended. The amount of free copper required to elicit toxicity thresholds significantly increased in copper sulfate exposures with increasing algal cell densities, whereas a decrease in free copper levels elicited toxicity in

172

Captain exposures. Captain was more effective at controlling Scenedesmus dimorphus at most concentrations, and control was inversely proportional to toxicity to D. magna. Overall incorporating target, competing ligands in toxicity testing is important for accurate risk assessment, and copper formulation can significantly alter algaecidal efficacy as well as risks to non-target organisms.

Key words: Risk assessment, algae, algaecides, Daphnia magna, copper

INTRODUCTION

Application of copper-based algaecides and herbicides is a common practice in managing surface freshwater resources (Fitzgerald 1964, Fattom and Shilo 1984). Concern of risks to non-target species is ever-present following application. Despite the typical use of applied copper concentrations in excess of many published lethal concentrations for freshwater organisms, copper toxicity is rarely observed in situ. Studies to date have over- estimated environmentally relevant exposure components and created a toxicity model that is inaccurate, and not validated in the field (Ma et al. 1999; Levy et al. 2007). The United States

Environmental Protection Agency (USEPA) registered copper-based algaecides are typically designed to enhance uptake by and toxicity to the algae, concomitantly decreasing non-target species exposure (Sunda and Lewis 1978). Research regarding refined laboratory toxicity bioassays is needed to account for the primary target ligands (i.e., the targeted algae), and to improve understanding of risks to non-target species (e.g., fish and invertebrates). Data on management method specificity for the target species and resultant impacts to other organisms is a strategic approach for more accurate assessment of effective noxious algal

173

response (Jewet et al. 2008). As chemical management of problematic algae is an integral component to rapid and selective risk remediation associated with nuisance blooms (Mastin et al. 2002), more accurate, realistic data are needed to support informed decisions about algaecide use. The goal of this research was to accurately assess the efficacy and non-target species risks from environmentally relevant exposures to copper-based algaecide applications.

The Biotic Ligand Model (BLM) is used to assess copper toxicity based on a specific binding site (e.g., fish gill). The BLM has shown that dissolved organic carbon (DOC) complexation can decrease copper availability to the target ligand (Santore et al. 2001;

USEPA 2003). This research used living algae as the targeted biotic ligand to incorporate into the exposure model for non-target species. By designing algaecides to have enhanced affinity and algal uptake (i.e., copper efficiency and algaecidal efficacy), the amount of copper partitioned to the targeted ligands would increase while copper availability to non- target species would decrease. The proportional significance of target organism uptake to non-target organism risks provides a way to accurately evaluate these products and may support their justifiable use in aquatic systems to control the targeted biotic ligands. The lethal dose of copper (i.e., the threshold copper concentration at the toxic sites of action) to non-target organisms (e.g., fish and invertebrates) primarily requires a dissolved aqueous form (e.g., gill-bound toxicity). The actual exposure and subsequent dose is altered in real- world applications (Playle et al. 1993; Meyer et al. 1999). Effectiveness of an algaecide application depends on achieving or exceeding the critical burden of copper at the toxic sites of action for a specific target alga (Bishop and Rodgers 2012). Because copper algaecides are

174

applied to target a specific nuisance algal infestation (i.e., treat the algae, not the water), and a large fraction rapidly partitions to the algae (Levy et al. 2007; Crist et al. 1990), the dissolved copper concentration and resulting non-target species exposure may never achieve the lethal threshold amount. Copper has an affinity to bind to target ligands/organic matter, so the bioavailable concentrations and exposure duration are less with viable algae present.

Copper concentrations in laboratory animal toxicity studies do not typically contain measurable amounts of algae or other particulate organic matter, and copper concentrations remain relatively constant throughout the exposure duration (Sprague 1973; Kim et al. 1999).

These studies are designed to show a maximum potential for risk, but they are not representative of field scenarios. A risk assessment of applied copper from laboratory exposures needs to translate to a field site in order to support realistic risk management decisions in the field (Cairns 1986). Without accounting for algaecide uptake, which is a significant factor altering exposure to non-target organisms, laboratory results lack accuracy for translation to field situations. To improve risk assessments, data are needed to characterize the ability of algae to ameliorate copper toxicity (decrease concentration and duration of exposure) to non-target organisms following copper-based algaecide application.

Typically, algaecides are strategically applied directly to an algal bloom or water resource, with some algae present, to restore water resource uses (e.g., drinking, irrigation, recreation). The goal of an algaecide formulation and application is to enhance exposure to the algae in order to achieve control. To accurately assess risks to in situ aquatic biota from copper algaecide applications, data are needed for the affinity and binding specificity to algae and the remaining amount of copper available to non-target organisms. As many of these

175

products (copper algaecide formulations) are efficiently designed to rapidly partition to the target organism, the dose of copper on, and internalized in, algae accounts for a significant fraction of the applied copper. The dissolved copper half-life decreases substantially with increasing ligands (i.e., organic biomass), and subsequent non-target animal exposure- response curves shift because the amount of copper available to interact with the toxic sites of action decreases below the toxicity threshold (Jackson 1978; Playle et al. 1993). Most previous studies that have considered this problem have focused on artificial humic acids

(Ma et al. 1999) or dissolved organic matter from natural sources (Parent et al. 1996; Kim et al. 1999). More research is needed on the impact of living organic matter on the exposure modification of copper to non-target organisms. This is especially critical in exposure scenarios where high biomass of live organisms, such as an algal bloom, is intentionally targeted by an algaecide. Terry and Stone (2002) reported that the majority of copper removal occured within 24 hr after amending with algae, with further decrease to over 99% of applied copper after 36 hr. An inversely proportional influence of algaecidal impacts has been measured with increasing algal densities, subsequently lowering the mass of copper/mass of algae exposure (Franklin et al. 2002; Bishop and Rodgers 2012). Limited studies have assessed the form of a specific toxicant in terms of both the algaecidal impact and the subsequent risks to non-target organisms. Aspects of a field application should be considered in risk assessment evaluations with copper-based algaecides in order to improve environmental relevance (Blanck et al. 1984; Cairns and Pratt 1989). The goal of applying these products is solely to control the targeted nuisance algal species while minimizing residual off-site movement and risks to non-target species. For enhanced exposure and

176

algaecide delivery to target algae, technologies for efficient application (trailing hoses, targeted surface spray) and safety techniques can be used; for example, to allow refugia by treating from shoreline out, treating only a portion of water body, and using a precise amount of product to control algae while avoiding waste from application of excess amounts. For accurate risk assessment of copper algaecide applications, predictions of expected responses to exposures need to be confirmed. The competing ligand hypothesis incorporates copper binding to target ligands and allows assessment of risks based on the available copper in an exposure to a toxic site of action on non-target organisms. This information will allow regulatory personnel and stakeholders to make informed decisions based on accurate risk assessment about predicted impacts from copper algaecide exposures in a field application, which differs markedly from typical standard laboratory toxicity experiments.

By understanding copper affinity and uptake kinetics, more accurate assessment can be made of the ability of targeted biotic ligands present (algae) in a field algaecide application to ameliorate direct copper toxicity risks to non-target organisms. Factors influencing the impacts from copper algaecides include: 1) the intrinsic character and sensitivity of the algae (e.g., structure, biovolume, surface area, binding capacity; Fitzgerald

1964; Fattom and Shilo 1984; Speziale et al. 1991; Dyck 1994), 2) environmental characteristics of the water resource (e.g., pH, hardness, alkalinity, conductivity, temperature;

Murray-Gulde et al. 2002), and 3) the specific copper algaecide formulation (e.g., chelation, chemistry; Mastin and Rodgers 2000). Chelated copper algaecides can have increased toxicity to target algal species compared with ionic copper formulations due to their diffusion properties, chelator toxicity (Stauber and Florence 1987; Sunda 1989; Masuda and Boyd

177

1993), and potential formation of free hydroxyl radicals (Florence et al. 1983). Additionally, chelated copper has shown decreased toxicity toward non-target species due to the shift away from the ionic copper form that affects gill site functionality (Strauss and Tucker 1993;

Mastin and Rodgers 2000; Closson and Paul 2014). Understanding copper formulation efficiency and potency is critical in designing ecologically sound management programs.

The goal of this research was to more accurately predict risks associated with typical use patterns of copper algaecides. This research, coupled with critical burden analyses, will allow applicators to make more informed decisions about the copper concentration required to achieve control of specific algal infestations while also including an effective margin of safety for non-target organisms. Specific objectives of this research were to:

1. Measure the concentration of copper that partitions to the algae and remains

dissolved through time after copper algaecide application, including two different

formulations;

2. Compare exposure-response relationships with Daphnia magna exposed to two

copper algaecide formulations with different densities of algae present; and

3. Correlate the copper exposure (concentration, form, and duration) to Daphnia

magna toxicity thresholds and algal response.

MATERIALS AND METHODS

Experimental design

Daphnia magna Straus were attained from Aquatic Research Organisms Inc.

(Hampton, NH 03842) and maintained in a healthy, stable culture for one month prior to test

178

initiation. D. magna neonates (age < 24 hr old) were used in toxicity experiments. All were cultured and tested at 23 ± 2°C under a 16 hr light/8 hr dark photoperiod illuminated by fluorescent lighting (Spectralux T5/HO 6500K blue; 3000K red) at an intensity of 67.5 ± 2.7

µmol photons/m2/s. The D. magna populations were exposed to a range of concentrations of copper in 48-hr toxicity experiments (Lewis et al. 1994; USEPA 1996b). Culturing and testing was conducted in COMBO freshwater media (Table 1), which included all ions needed for healthy culture (Kilham et al. 1998; Table 2). Stock solutions, 1,000 mg/L copper, of two different copper formulations (CuSO4 · 5H2O [Fisher Scientific, Inc.; C489-1] and

Captain® [SePRO Corporation, Carmel, IN]; Table 3), were prepared within 4 hr of test initiation. Nominal copper treatment concentrations were achieved by amending appropriate amounts of stock solutions to COMBO media (background copper concentration < 1 µg/L).

Neonates of D. magna used in testing were randomized and fed the chlorophyte

Scenesdesmus dimorphus Kutzing mixed with yeast-cerophyl-trout chow (YCT; Aquatic

Research Organisms Inc., Hampton, NH 03842) at 0.1% v/v prior to test initiation. Food was not added throughout testing, except for the algae present in the corresponding exposures. D. magna exposures containing algae were prepared by first growing algae in COMBO media and amending a concentrated amount of algae in the exponential growth phase to achieve the targeted initial density. Exposure-response relationships were evaluated 1) without algae present using two different copper algaecide formulations, tested separately; and 2) with algal amendments to achieve 5.0 x 105 and 5.0 x 106 cells/mL of S. dimorphus. Copper test concentrations used in the experiments differed among treatments and were adjusted to encompass toxicity thresholds (48-hr no observed effect concentration [NOEC], lowest

179

observed effect concentration [LOEC] and lethal concentration for 50% of test organisms

[LC50] values). A range of 8 to 12 different concentrations was used to provide a defined exposure-response curve, and to include at least three test concentrations below and above the 48-hr LC50 value (Table 4). D. magna mortality was measured by visual appearance at 24 and 48 hr after treatment (HAT) based on movement response to stimuli and structural integrity. Untreated controls were included in all experiments, and encompassed similar exposure conditions without copper amendments. Each treatment consisted of three replicates, each with ten D. magna neonates in a 500 mL exposure solution.

Copper measurement analysis

Subsamples of water (with algae when applicable) were analyzed initially and at 2,

24, and 48 HAT for total and dissolved copper concentrations. Treatment vessels were covered and gently inverted 180 degrees to homogenize algae and water prior to sampling.

To measure free copper concentrations, 10 mL samples were filtered through a 0.22 µm glass fiber filter (Phenomenox, Inc., AF0-2203-52) then centrifuged for 10 min at 6,000 rpm. Total copper was measured in a 3 mL sample of the treatment water (with homogenized algae when present) that was 1) digested in 1 mL of 10% trace-metal-grade nitric acid (Fisher

Scientific, Inc. A509), and 2) further digested in 1 mL of 30% H2O2 (Fisher Scientific, Inc.

H325; Tripathi et al. 2006; USEPA 1996). Partitioned or sorbed copper was calculated as the difference between total and free copper in the subsamples. Method blanks and standard additions for sample preparation were used to assess interference and percent recovery, respectively. Copper was measured using inductively coupled plasma-optical emission spectrometry (Shimadzu ICPE 9000) with a matrix-matched calibration curve from serial

180

dilution of a 1,000 mg/L copper standard (Fisher Scientific, Inc. SC194; APHA 2005). The limit of detection for copper was 1 µg/L. The amount of partitioned copper per cell (ng/cell) was calculated by dividing the mass of partitioned copper measured in the 10 mL of filtered algae by the cell/mL algal density times 10. Algal surface area was compared with partitioned copper amounts and calculated by the external area of the coenobium

(Scenedesmums dimorphus colony) based upon the prolate ellipsoid equation per cell subtracting adjoined cell areas. The surface area equation applied was based on He and Tebo

(1998) and calculated as SA = 2πb2+2πa2b2/ sqroot(a2-b2) · arcsin (sqroot(a2-b2)/a). Where a

= length and b = width, assuming a > b and b = depth (Table 5).

Algal responses

Algal response parameters were evaluated at 96 hr to assess algaecidal efficacy and correlation with partitioned copper. Chlorophyll a concentrations were used as a representative test endpoint. Homogenized subsamples (5 mL) of S. dimorphus treatments were filtered (0.45 µm nitrocellulose), placed in 5 mL of buffered acetone, and sonicated to lyse the cells (modified SMEWW 2005). Chlorophyll a was measured fluorometrically using a Wallac Victor2 spectrofluorometer, by correlating with a matrix-matched standard calibration curve (10-640 µg/L; Sigma C-5753). The growth rate, µ, was calculated as follows:

Equation 1: µ = ln (Nt / No) / t where Nt and No represent final and initial cell densities respectively, and t is the exposure time after test initiation. Percent inhibition of growth rate for each treatment was calculated as follows:

181

Equation 2: % inhibition = [(µc-µt) / µc] * 100 where µc and µt denote the mean value of growth rate in the control and treatments, respectively.

Non-target species impacts

Daphnia magna 48-hr NOEC, LOEC, and LC50 values were calculated in water containing no measurable concentrations of organic matter/algae. The shift in toxicity threshold values with treatments containing different algal densities was compared for each algaecide exposure. Free copper concentrations were measured through time between clean

(algal-free) water and algal amended treatments. A best fit assessment was used to correlate responses of D. magna to the total and free copper exposure.

Statistics

Nonlinear regression with a three-parameter sigmoidal curve was fit to the data and used to compare Daphnia magna exposure-response relationships and surface area copper partitioning (Table 6). A one-way analysis of variance (ANOVA) was used to compare differences in EC50 values between exposures with and without algae. An ANOVA and

Dunnett’s procedure was used to discern significant differences between controls and algaecide treatments for determining toxicity threshold levels (α = 0.05; McDonald 2014). A

Shapiro-Wilk test was used to assess normality of the data, and all data were tested for constant variance (F-test). Data were analyzed using Microsoft Excel (Microsoft 2010) and

SigmaPlot version 12.5 (Systat Software 2014).

182

RESULTS

Copper exposures

Total copper concentrations were between 95-103% of targeted nominal concentrations throughout all experiments. Thus, all result designations with total copper are presented with their nominal value. Method blanks consistently had non-detect copper concentrations, and the percent recovery from standard additions ranged from 94-101%. The operationally defined free copper in the no algae exposures comprised 93.6% and 96.4% of the total nominal copper concentrations for copper sulfate and Captain treatments, respectively. Free copper in exposures containing algae was significantly decreased (P <

0.05) in all treatments and averaged 57.4% and 76.94% of total in copper sulfate and Captain treatments containing 5 x 105 cells/mL, respectively. Treatments with 5 x 106 cells/mL further decreased free copper levels to 35.25% and 41.57% of total for copper sulfate and

Captain treatments, respectively (Table 7; Figure 1). Free copper levels remained similar in the no-algae exposures throughout the experiment but tended to decrease in concentration from 2 to 48 hr in exposures containing algae. This decrease was more pronounced in the highest (5 x 106 cells/mL) algal exposures, indicating that a longer interaction time was needed to bind available ligands. Ligands may have been closer to saturation in lower and no algae treatments, as the majority of binding appeared to occur over the first two hr (Figures

2-7).

Non-target toxicity

The amount of total copper required to elicit 48-hr LC50 values for both copper formulations significantly increased in both the 5 x 105 and 5 x 106 algal exposures (Table 8;

183

Figure 8). This shift was more pronounced with copper sulfate than Captain. A 4.22 and 6.88 factor increase in 48-hr LC50 values was measured in total copper as copper sulfate with low

(5 x 105) and high (5 x 106) amounts of algae added, respectively. In contrast, a 1.17 and 1.66 factor increase was measured for total copper as Captain with low and high amounts of algae added, respectively. Copper sulfate 48-hr LC50 values increased from 75.3 to 317.8 and 517.8 mg Cu/L whereas, 48-hr LC50 values for Captain treatments increased from 353.8 to 414.2 and 588.5 mg Cu/L in no algae, 5 x 105 and 5 x 106 cells/mL treatments, respectively (Table

8; Figures 8-10). Despite this difference in the shift in toxicity values, Captain was significantly (P < 0.05) less toxic to D. magna in all corresponding treatments based on total copper amended.

Free copper in the Captain exposures with algae was not representative of toxicity.

LC50 values, based on average free copper levels, significantly decreased (P < 0.05) from the no algae exposures to both the Captain treatments containing algae, although the LC50 values did not differ in the low- and high-algae treatments. This finding was not expected, as copper sulfate required an increased level of free copper in treatments containing algae to elicit the

LC50 (Figure 11). The data suggest that a fraction of copper not accounted for in the dissolved concentration was able to elicit toxicity. This complex was primarily formed with chelated copper rather than ionic copper sulfate. The dissolved copper levels in copper sulfate exposures with algae were required to be significantly higher to elicit the same degree of toxicity as predicted from levels measured in the no-algae exposures.

184

Copper sorption

The percentage of copper sorption to algae increased by a factor of 1.6 in copper sulfate and 2.7 in Captain (significant P < 0.05), with increasing algal density treatments.

This increase was not proportional to the increase in algal densities (10 x). The amount of copper (ng) bound per cell in the 5 x 105 exposures was 0.14 and 0.084 and in the 5 x 106 exposures was 0.026 and 0.024 in copper sulfate and Captain treatments, respectively. The data were not significantly different for the two formulations, and the formulations had very similar binding per cell at the highest algal density. The amount bound per cell or per unit surface area was significantly lower in each formulation in the high density (5 x 106) algal treatment (Table 7).

Target algal response

The impact of copper on the algae was not proportional to the sorbed copper, and differed between formulations. The percent growth inhibition compared with controls was more pronounced with Captain exposures than with copper sulfate, and was significantly different at all concentrations in the 5 x 106 density treatments. This algal influence is important to assess with simultaneous exposure of D. magna, in order to attain control while also gaining insights about understanding non-target risks. In the copper sulfate 5 x 105 exposures, at the concentration eliciting the D. magna LC50, there was a 28% algal inhibition.

In contrast, with Captain exposures, a 57% algal inhibition occurred when the LC50 concentration for D. magna was attained. Likely due to a depressed copper/cell ratio, algal inhibition was further decreased at LC50 values in the high-density algal treatments to 25% and 50% in the copper sulfate and Captain treatments, respectively (Figures 12-13).

185

DISCUSSION

Copper exposures

Previous studies have given the dissipation half-life for dissolved copper at 10 days

(USEPA 2005), reflecting the amount of time for 50% transfer of copper out of the water- column, and is used as a predictor of aqueous toxicity. Murray-Gulde et al. (2002) found half-lives to differ based on copper formulation and water chemistry within a range of 2.6-5.7 days in natural waters. Chelated copper formulations had a longer half-life in soft waters, with minimal algae, whereas the half-life for copper sulfate was lower, suggesting more rapid interaction with inorganic components in the water. Masuda and Boyd (1993) reported that chelated copper was more potent to algae and had a longer half-life than copper sulfate with most sediment types (except for 5% organic matter and clay). The chelated copper half-life was even more pronounced in control waters with no sediment. The majority of copper removal occurred in the first three days, and was substantially less with no sediment present.

Liu et al. (2006) showed rapid association of copper, from copper sulfate applications, with suspended sediment in the water (< 10 min). Most of the copper (> 99%) was transferred to sediment in two days. In this study, copper sorption occurred rapidly when algae were present, with an average of more than 50% bound in the high-density exposures at 48 hr after treatment. Such exposure duration differences need to be accounted for in environmental risk assessments.

Forms of copper interacting with gill-active sites (Na, K-atpase pump) are responsible for most toxicity, and increase potency toward non-target organisms (Pagenkopf 1983; Playle et al. 1993; Meyer et al. 1999). The cupric ion is generally considered to be the most toxic

186

species (Howarth and Sprague 1978, Campbell 1995), although other inorganic copper

+ complexes including CuOH (Pagenkopf et al. 1974), Cu(OH)2 (Chakoumakos et al. 1979;

2+ De Schamphelaere and Janssen 2002), Cu2(OH)2 , and CuCO3 can significantly contribute to toxicity (Howarth and Sprague 1978; Chakoumakos et al. 1979). In elevated pH environments, there can be increased uptake of inorganic copper complexes, imparting more toxicity to non-target organisms (De Schamphelaere and Janssen 2002). Masuda and Boyd

(1993) reported that triethanolamine copper complexes from chelated copper algaecides were more prevalent than the cupric ion at pH higher than 7.1. Copper may also be associated with

DOC complexes from algae, and low-affinity organic ligands may impart toxicity to the gill site if the binding affinity to the complex is lower than the surface interaction at the active site (Playle et al. 1993b; MacRae et al. 1999a). The dissolved forms of copper in aqueous media with and without algae likely differ, although numerous inorganic forms (e.g., media interaction or higher pH through photosynthetic activity) can alter toxic effects (De

Schamphelaere and Janssen 2002). Dissolved copper was measured in this research to account for inorganic, chelated, and low-affinity DOC forms of copper that may readily interact with non-target organism gill binding sites. Several forms of copper can elicit toxicity, including chelated copper, which were accounted for in the exposure analysis as representative on a comparative basis for copper formulations.

Algal sorption

Kim et al. (1999) showed increased interaction time of copper with dissolved organic matter (24 hr) resulted in more bound copper and decreased bioavailability. Conversely, living algal biosorption of copper is rapid, mostly occurring within 10 min (Crist et al. 1990).

187

Living algae may have a much greater proportional influence on the concentration exposure time for non-target organisms from copper applications. Both dead (Tien et al. 2005) and living algae can bind and remove copper; for example, Terry and Stone (2002) found that viable Scenedesmus removed more copper than dead cells over a three-day period, and that a high level of copper, 15 mg Cu/L, was required to render the algae non-viable. Our maximum tested concentration was 0.75 mg Cu/L, which is less than previously reported toxic concentrations, so responses were measured on growth inhibition. Based on visual cell viability, algal cellular structure maintained integrity over the experiment duration and the impact of sorption from dead cells was assumed to have been negligible.

Borgmann and Carlton (1984) showed that free cupric ion activity differed in various surface and prepared media water sources at similar toxicity endpoints for Daphnia magna.

Similarly, as in this study, however, waters containing algae (used as a food source) had decreased toxicity compared with primarily inorganic nutrients. Copper was more readily adsorbed by Daphnia magna from the dissolved phase, and accumulation decreased with increasing available food sources such as algae. The contribution of any ingested copper associated with algal cells to toxicity was considered negligible, based on previous reports that the majority of ingested copper was excreted and was not directly related to acute toxicity (Schuytema et al. 1984; Zhao et al. 2009), as opposed to other chemistries that may have enhanced toxicity from dietary uptake such as atrazine (Lampert et al. 1989). Increased

Daphnia health from algal availability is not likely responsible for the significant shifts in toxicity. Prior studies have shown that copper sorption (e.g., to food particles) overwhelmed

188

the mediation of toxic effects (Jeon et al. 2010), and that food type did not alter the acute toxicity of copper to D. magna (Winner et al. 1977).

Smaller cells can sorb more copper than larger cells on a per biovolume basis due to the larger relative surface area for binding (Yan and Pan 2002). Nevertheless, total copper sorption does not necessarily relate to toxicity. Butler et al. (1980) and Foster (1977) described copper exclusion from the cell interior as the primary mechanism of tolerance.

Thus, sorption alone may not adequately indicate potential toxicity. Bishop and Rodgers

(2012) also found that total sorbed copper was not representative of toxicity between copper formulations. In this research, we maintained a similar cell and colony size and evaluated the impact of increasing total sorption sites by altering cell densities in exposures.

Yan and Pan (2002) reported 67% to 95% removal of copper when applied at low concentrations, although less than 50% was removed when copper concentrations were elevated in comparison to cell densities. In this study, with a higher ratio of algae to copper, increased sorption was measured along with a decrease in the concentration of copper remaining in aqueous media. Franklin et al. (2000) and Fitzgerald (1964) reinforced the efficacy of an algaecide application as being proportional to the initial total algal cell density, suggesting that there is a threshold amount of copper per cell that is required for control. It is important in risk assessment to understand the interaction of copper with available algal surface area, and how this alters the subsequent exposure to other desirable species, as present in typical field scenarios. There is innate variability in binding among strains of the same species and among species (Twiss and Nalewajko 1992). Therefore, research will be needed to assess other commonly managed nuisance species/strains.

189

In this study, there was a major effect of increasing available binding ligands on the

LC50 values, although not linearly proportional to the surface area of cells. Significant differences in copper formulation toxicity were also found. Chelated copper, with and without algae, was less toxic to non-target organisms, and inversely proportional to toxicity to target ligands. This finding suggests that the more effective the algaecide formulation, the lower non-target risks could be observed. The free copper fraction in the chelated copper exposures required to elicit toxicity decreased, however, with increasing cell density. This finding was unexpected, and the opposite of what occurred in copper sulfate exposures where an increase in the free (and total) copper was needed to elicit LC50 values with higher algal densities. These data suggest that an interaction with algal exudates or small organic matter components may ameliorate the free ion toxicity of copper sulfate, although may increase toxicity of free copper from chelated formulations to D. magna. More research is needed to discern the specific copper species responsible for eliciting toxicity. Other researches have also suggested that toxicity can be elicited by multiple copper forms, rather than only by free or dissolved copper (Playle et al. 1993a).

Toxicity value and copper formulation

Daphnia is the most sensitive genus of aquatic invertebrates to copper among those which are routinely used in toxicity testing (USEPA 2005; Johnson et al. 2008). Muyssen and

Janssen (2007) estimated the copper NOEC and LOEC for Daphnia juveniles as 75 µg/L and

90 µg/L, respectively, whereas the NOEC and LOEC for Daphnia adults exceeded 130 µg/L.

In this study, Daphnia magna neonates were used to obtain a conservative estimate of risk to non-target organisms. A range of toxicity values has been reported based on the water quality

190

of the test media and the formulation. With 19 copper-amended European natural surface waters, 48-hr EC50 values for D. magna varied between 35.2 and 792 µg/L dissolved copper

(De Schamphelaere et al. 2002). Johnson et al. (2008) reported a similar response of D. magna to a different chelated copper and copper sulfate in prepared water with 96-hr LC50 values of 4.6 and 5 µg/L, respectively. From this research, D. magna 48-hr LC50 values for copper sulfate and Captain were 75.3 and 353.8 µg Cu/L, respectively, with no organic matter present. Murray-Gulde et al. (2002) showed that an ethanolamine-chelated copper algaecide was less toxic to Ceriodaphnia dubia than copper sulfate, with 96-hr LC50 values of 91.7 and 60.0 µg Cu/L, respectively. Mastin and Rodgers (2000) reported that 48-hr LC50 values with D. magna differed depending on the copper formulation, with copper sulfate at

18.9 and ethanolamine-chelated copper at 11.3 µg Cu/L though chelation with surfactant at

29.4 µg Cu/L.

Water chemistry can have a substantial impact on copper availability and forms.

Typically, alkalinity and pH may be responsible for shifts in copper forms and bioavailability, whereas little impact has been reported from higher nutrient levels (Perez et al. 2005). Influence of pH on toxicity may occurr, primarily due to lack of interference of hydrogen ions with copper binding sites at elevated pH levels. The pH may have affected the copper species in this research, for example by maintaining the chelated forms as the

(elevated pH) ethanolamine complexes. The pH levels were similar between treatments and

Meador (1991) reported that there was no significant effect on 48-hr toxicity despite potential copper species transformation. Meador (1991) considered DOC to be the primary governing factor in copper toxicity. This is also supported by findings of increased free copper

191

concentration needed with copper sulfate to achieve LC50 values when algae or other organic matter are present (Ma et al. 1999; Levy et al. 2007). The pH may also be an important factor in altering the copper species, although these may impart similar toxicity. More likely is the interaction of free copper species with hydroxide or carbonate species to render copper sulfate proportionally less toxic (Chakoumakos et al. 1979). In this study, laboratory prepared water with soft and low alkalinity designation (i.e., greater non-target risks) was used to assess interactivity with living algal cells.

There were significantly higher shifts in 48-hr LC50 values when algae were present in exposures, which further increased at higher cell densities. The shift in toxicity was much more pronounced when the copper source was copper sulfate rather than chelated copper.

This finding may suggest increased interactivity of the free cupric ion with aqueous inorganic and organic components, making it less available to other biota. In the high-density algal treatments, similar sorption occurred in the two formulations. Captain was significantly more potent to the algae at higher-density exposures than copper sulfate, suggesting that sorption did not account for toxicity and that chelated copper may introduce more internal copper to the cells thereby imparting more toxicity. In previous research, DOC derived from algal and daphnid blooms in a microcosm reduced copper toxicity to Daphnia sp. (Meador 1991) and when copper exceeded the binding capacity of the organic matter, there were dramatic increases in toxicity to non-target organisms. In this research, the amount of copper sorbed increased at higher algal densities, and also increased through time. The binding capacity apparently was not oversaturated, and increased cells or algal biomass would have continually decreased toxicity to non-target organisms. In other studies, toxicity testing with

192

C. dubia showed a LC50 shift from 20.1 to 119.2 µg Cu/L when allowed to equilibrate with dissolved organic matter for 24 hr (Kim et al. 1999). Here, a significant shift in toxicity was observed in simultaneous exposures of D. magna with living algae, as often present in natural waters. This further suggests that the living algae may be more influential than dissolved organic constituents in altering toxicty.

Some prior studies have described a correlation between dissolved or free ion copper levels and toxicity (Sunda and Guillard 1976; Meador 1991; Ma et al. 1999). The premise that dissolved copper mostly controls the LC50 was not supported for chelated copper in this research (Figure 11). Instead, the free copper form in chelated copper exposures may have interacted with inorganic or organic components in the water to increase toxicity (i.e., the free copper LC50 values decreased with increasing algal densities). Significantly more free copper (as copper sulfate) was needed in the presence of algae to elicit the LC50 for D. magna. Many algae produce small organic exudates that may render copper ions from copper sulfate less bioavailable as toxic agents (McKnight and Morel 1980; Xue et al. 1988;

Lombardi and Vieira 2000; Koukal et al. 2007). These observations support the premise that dissolved and free copper concentrations are not strictly representative of toxicity (USEPA

2005), and that toxicity is also altered depending on the copper species initially present.

Risk assessment

With the relatively recent designation of algaecides as point source discharges to aquatic systems, more appropriate risk assessment is needed to assess environmental impacts

(USEPA 2011). Real-world scenarios are needed to accurately predict risks associated with copper algaecide exposures. Standard toxicity tests may overestimate the impact of the

193

environment exposure associated with toxicant designed for a specific purpose. Algaecides are designed to be applied to an algal infestation. This research showed that copper concentration and exposure duration were significantly different when algae were present, and that algae significantly shifted the amount of copper needed to attain toxicity thresholds for the non-target species, D. magna. The LC50 values shifted to > 500 µg/L in the highest algal density exposure, and in accordance with other label mandates, suggests that risk would be further reduced. Even with maximum labeled use rates of copper (1 mg Cu/L), much higher than what is routinely used, the restriction to treat 1/3 to 1/2 of the water volume would effectively decrease a whole water volume concentration (assuming homogenous) to

500 µg/L. Therefore, toxicity values that incorporate algae will shift risk quotient levels

(estimated environmental concentration / LC50), as used in USEPA (2004) evaluations of product environmental risk, to levels not at high concern designations for acute exposures

(i.e., > 0.5).

CONCLUSION

The competing ligand hypothesis concept provides a testing framework to characterize risks from more relevant environmental exposures. It also accounts for interactivity of the toxicant with organism exudates and the potential for increased toxicity.

More research is needed to assess copper compounds created in environmental exposures with differing initial copper species. Both in a large perspective on potential concentrations and on the chemical/biological level with variability in biotic and abiotic factors in receiving waters.

194

This research assists water resource managers in understanding potential risks from copper-based algaecide applications in field situations. With increasing regulatory demands and scrutiny on pesticide use, these data will justify the efficiency of different algaecide formulations and predicted risks to non-target species. Typical non-target studies do not account for a critical exposure component. This research showed that algae can account for substantial binding ligands which can alter non-target species risks to copper exposures, and should be considered in toxicity testing with copper-based algaecides. The efficiency of copper formulations differed, concomitant to alterations in non-target toxicity. The copper formulation that decreases risks to non-target organisms while enhancing the degree of control to algae in receiving waters, should be considered in risk-based decisions regarding copper use.

195

REFERENCES

Bishop WM, Rodgers Jr JH (2012) Responses of Lyngbya wollei to exposures of copper- based algaecides: The critical burden concept. Archives of Environmental Contamination and Toxicology 62:403-410

Bishop WM, Johnson BM, Rodgers Jr JH (2014) Comparative responses of target and non- target species to exposures of a copper-based algaecide. Journal of Aquatic Plant Management 52:65-70

Blanck H, Wallin G, Engberg SA (1984) Species dependent variation in algal sensitivity to chemical compounds. Ecotoxicology and Environmental Safety 8:339-351

Borgmann U, Charlton CC (1984) Copper complexation and toxicity to Daphnia in natural waters. Journal of Great Lakes Research 10:393-398

Butler M, Haskew AEJ, Young MM (1980) Copper tolerance in the green alga, Chlorella vulgaris. Plant, Cell, and Environment 3:119-126

Cairns J (1986) The myth of the most sensitive species. BioScience 36:670-672

Cairns J, Pratt JR (1989) The scientific basis of bioassays. Hydrobiologia 188-189:5-20

Campbell PGC (1995) Interactions between trace metals and aquatic organisms: a critique of the free-ion activity model. In: Tessier, A., Turner, D.R. (Eds.), Metal Speciation and Bioavailability in Aquatic Systems. John Wiley and Sons, Chichester, United Kingdom

Chakoumakos C, Russo RC, Thurston RV (1979) Toxicity of copper to cutthroat trout (Salmo clarki) under different conditions of alkalinity, pH, and hardness. Environmental Science and Technology 13:213-219

Closson KR, Paul EA (2014) Comparison of the toxicity of two chelated copper algaecides and copper sulfate to non-target fish. Bulletin of Environmental Contamination and Toxicology 93:660-665

Crist RH, Martin JR, Guptill PW, Eslinger JM, Crist DR (1990) Interaction of metals and protons with algae. 2. Ion-exchange in adsorption and metal displacement by protons. Environmental Science and Technology 24:337-342

De Schamphelaere KAC, Heijerick DG, Janssen CR (2002) Refinement and field validation of a biotic ligand model predicting acute copper toxicity to Daphnia magna. Comparative Biochemistry and Physiology Part C 133:243-258

196

De Schamphelaere KAC, Janssen CR (2002) A biotic ligand model predicting acute copper toxicity to Daphnia magna: the effects of calcium, magnesium, sodium, potassium and pH. Environmental Science and Technology 36:48-54

Dyck LA (1994) Creation of management strategies that are compatible with the autecology of Lyngbya. Lake and Reservoir Management 9:71

Fattom A, Shilo M (1984) Hydrophobicity as an adhesion mechanism of benthic cyanobacteria. Applied and Environmental Microbiology 1:135-143

Fisher Scientific (2008) Copper (II) sulfate pentahydrate Material Safety Data Sheet. Fisher Scientific Inc. Fair Lawn, NJ

Fitzgerald GP (1964) Factors in the testing and application of algicides. Applied Microbiology 12:247-253

Florence TM, Lumsden BG, Fardy JJ (1983) Evaluation of some physico-chemical techniques for the determination of the fraction of dissolved copper toxic to the marine diatom Nitzschia closterium. Analytica Chimica Acta 151:281-295

Foster PL (1977) Copper exclusion as a mechanism of heavy metal tolerance in a green alga. Nature (London) 269:322-323

Franklin NM, Stauber JL, Apte SC, Lim RP (2002) Effect of initial cell density on the bioavailability and toxicity of copper in microalgal bioassays. Environmental Toxicology and Chemistry 21:742-751

He LM, Tebo BM (1998) Surface charge properties of Cu(II) adsorption by spores of the marine Bacillus sp. strain SG-1. Applied and Environmental Microbiology 64:1123- 1129

Howarth RS, Sprague JB (1978) Copper lethality to rainbow trout in waters of various hardness and pH. Water Research 12:455-462

Jackson TA (1978) The biogeochemistry of heavy metals in polluted lakes and streams at Flin Flon, Canada, and a proposed method for limiting heavy metal pollution of natural waters. Environmental Geology 2:173-189

Jeon J, Ra JS, Lee SH, Lee MJ, Seung HY, Kim SD (2010) Role of food and clay particles in toxicity of copper and diazinon using Daphnia magna. Ecotoxicology and Environmental Safety 73:400-406

Jewett EB, Lopez CB, Dortch Q, Etheridge SM, Backer LC (2008) Harmful Algal Bloom Management and Response: Assessment and Plan. Interagency Working Group on

197

Harmful Algal Blooms, Hypoxia, and Human Health of the Joint Subcommittee on Ocean Science and Technology. Washington, DC

Johnson BJ, Chao MM, Tedrow OR, McQueen AD, Rodgers Jr. JH (2008) Responses of Lepomis macrochirus, Pimephales promelas, Hyalella azteca, Ceriodaphnia dubia, and Daphnia magna to exposures of Algimycin® PWF and copper sulfate pentahydrate. Journal of Aquatic Plant Management 46:176-183

Kilham SS, Kreeger DA, Lynn SG, Goulden CE, Herrera L (1998) COMBO: a defined freshwater culture medium for algae and zooplankton. Hydrobiologia 377:147-159

Kim SD, Ma H, Allen HE, Cha DK (1999) Influence of dissolved organic matter on the toxicity of copper to Ceriodaphnia dubia: Effect of complexation kinetics. Environmental Toxicology and Chemistry 18:2433-2437

Levy JL, Stauber JL, Jolley DF (2007) Sensitivity of marine microalgae to copper: The effect of biotic factors on copper adsorption and toxicity. Science of the Total Environment 387:141-154

Lewis PA, Klemm DJ, Lazorchak JM, Norberg-King TJ, Peltier TJ, Peltier WH, Heber MA (1994) Short-term methods for estimating the chronic toxicity of effluents and receiving waters to freshwater organisms. 3rd ed. EPA/600.4-91/002, Cincinnati, OH

Liu R, Zhao D, Barnett M (2006) Fate and transport of copper applied in channel catfish ponds. Water, Air, and Soil Pollution 176:139-162

Lombardi AT, Vieira AAH (2000) Copper complexation by Cyanophyta and Chlorophyta exudates. Phycologia 39:118-125

Lürling M, De Lange HJ, Van Donk E (1997) Changes in food quality of the green alga Scenedesmus induced by Daphnia info chemicals: Biochemical composition and morphology. Freshwater Biology 38:619-628

Ma H, Kim SD, Cha D, Allen HE (1999) Effect of Kinetics of Complexation by humic acid on toxicity of copper to Ceriodaphnia dubia. Environmental Toxicology and Chemistry 18:828-837

MacRae RK, Smith DE, Swoboda-Colberg N, Meyer JS, Bergman HL (1999a) The copper binding affinity of rainbow trout (Oncorhynchus mykiss) and brook trout (Salvelinus fontinalis) gills: implications for assessing bioavailable metals. Environmental Toxicology and Chemistry 18:1180-1189

198

Mastin BJ, Rodgers Jr JH (2000) Toxicity and bioavailability of copper herbicides (Clearigate, Cutrine Plus, and Copper sulfate) to freshwater animals. Archives of Environmental Contamination and Toxicology 39:445-451

Mastin BJ, Rodgers Jr JH, Deardorff TL (2002) Risk evaluation of cyanobacteria-dominated algal blooms in a North Louisiana reservoir. Journal of Aquatic Ecosystem Stress and Recovery 9:103-114

McDonald, JH (2014) Handbook of Biological Statistics (3rd ed.). Sparky House Publishing, Baltimore, MD

McKnight DM, Morel FMM (1980) Copper complexation of siderophores from filamentous blue green algae. Limnology and Oceanography 25:62-71

Meador JP (1991) The interaction of pH, dissolved organic carbon, and total copper in the determination of ionic copper and toxicity. Aquatic Toxicology (Amst.) 19:13-32

Meyer JS, Santore RC, Bobbitt JP, DeBrey LD, Boese CJ, Paquin PR, Allen HE, Bergman HL, DiToro DM (1999) Binding of nickel and copper to fish gills predicts toxicity when water hardness varies, but free-ion activity does not. Environmental Science and Technology 33:913-6

Microsoft (2010) Microsoft Excel [computer software]. Redmond, WA

Murray-Gulde CL, Heatley JE, Schwartzman AL, Rodgers Jr. JH (2002) Algicidal effectiveness of Clearigate, Cutrine-Plus, and copper sulfate and margins of safety associated with their use. Archives of Environmental Contamination and Toxicology 43:19-27

Muyssen BTA, Janssen CR (2007) Age and exposure duration as a factor influencing Cu and Zn toxicity toward Daphnia magna. Ecotoxicology and Environmental Safety 68:436-442

Pagenkopf GK (1983) Gill surface interaction model for trace-metal toxicity to fishes: Role of complexation, pH, and water hardness. Environmental Science and Technology 17:342-347

Pagenkopf GK, Russo RC, Thurston RV (1974) Effect of complexation on toxicity of copper to fishes. Journal of the Fisheries Research Board of Canada 31:462-465

Parent L, Twiss MR, Campbell PGC (1996) Influences of natural dissolved organic matter on the interaction of aluminum with the microalga Chlorella: A test of the free-ion model of trace metal toxicity. Environmental Science and Technology 30:1713-1720

199

Perez P, Estevez-Blanco P, Beiras R, Fernandez E (2005) Effect of copper on the photochemical efficiency, growth and chlorophyll a biomass of natural phytoplankton assemblages. Environmental Toxicology and Chemistry 25:137-143

Playle RC, Dixon DG, Burnison K (1993a) Copper and cadmium binding to fish gills: Modification by dissolved organic carbon and synthetic ligands. Canadian Journal of Fisheries and Aquatic Sciences 50:2667-2677

Playle RC, Dixon DG, Burnison K (1993b) Copper and cadmium binding to fish gills: Estimates of metal-gill stability constants and modelling of metal accumulation. Canadian Journal of Fisheries and Aquatic Sciences 50:2678-2687

Santore RC, Di Toro DM, Paquin PR, Allen HE Meyer JS (2001) A biotic ligand model of the acute toxicity of metals. II. Application to acute copper toxicity in freshwater fish and daphnia. Environmental Toxicology and Chemistry 20:2397-2402

Schuytema GS, Nelson PO, Malueg KW, Nebeker AV, Krawczyk DF, Ratliff AK, Gakstatter JH (1984) Toxicity of cadmium in water and sediment slurries to Daphnia magna. Environmental Toxicology and Chemistry 3:293-308

SePRO Corporation (2014) Captain Algaecide Product Label. SePRO Corporation Carmel, IN 46032

SePRO Corporation (2015) Captain Algaecide Safety Data Sheet. SePRO Corporation Carmel, IN 46032

SigmaPlot version 12.5 (2014) Systat Software, Inc., San Jose, CA

Speziale BJ, Turner EG, Dyck LA (1991) Physiological characteristics of vertically-stratified Lyngbya wollei mats. Lake and Reservoir Management 7:107-114

Sprague JB (1973) The ABC’s of pollutant bioassay using fish. Biological methods for assessment of water quality In: American Society for Testing and Materials. Philadelphia, PA

Standard Methods for the Examination of Water and Wastewater, 21st ed. (SMEWW) (2005) American Public Health Association, Washington, DC

Stauber JL, Florence TM (1987) Mechanism of toxicity of ionic copper and copper complexes to algae. Marine Biology 94:511-519

Straus DL, Tucker CS (1993) Acute toxicity of copper sulfate and chelated copper to channel catfish Ictalarus punctatus. Journal of the World Aquaculture Society 24:390-395

200

Sunda WG, Guillard RRL (1976) The relationship between cupric ion activity and the toxicity of copper to phytoplankton. Journal of Marine Research 34:511-529

Sunda WG, Lewis JAM (1978) Effect of complexation by natural organic ligands on the toxicity of copper to a unicellular alga, Monochrysis lutheri. Limnology and Oceanography 23:870-876

Sunda WG (1989) Trace metal interactions with marine phytoplankton. Biological Oceanography 6:411-442

Terry PA, Stone W (2002) Biosoprtion of cadmium and copper contaminated water by Scenedesmus abundans. Chemosphere 47:249-255

Tien C-J, Sigee DC, White KN (2005) Copper adsorption kinetics of cultured algal cells and freshwater phytoplankton with emphasis on cell surface characteristics. Journal of Applied Phycology 17:379-389

Twiss MR, Nalewajko C (1992) Influence of phosphorus nutrition on copper toxicity to three strains of Scenedesmus acutus (Chlorophyceae). Journal of Phycology 28:291-298

Twiss MR, Welbourn PM, Schwartzel E (1993) Laboratory selection for copper tolerance in Scenedesmus acutus (Chlorophyceae). Canadian Journal of Botany 71:333-338

United States Environmental Protection Agency (USEPA) (1986) Guidance for reregistration of pesticide products containing copper sulfate. Fact sheet no 100. Office of Pesticide Programs. Washington, DC

United States Environmental Protection Agency (USEPA) (1996b) Ecological Effects Test Guidelines. OPPTS 850.1010 Aquatic Invertebrate Acute Toxicity Test, Freshwater Daphnids. Prevention, Pesticides, and Toxic Substances (7101). EPA 712-C-96-114. Washington, DC

United States Environmental Protection Agency (USEPA) (2002) Methods for Measuring Acute Toxicity of Effluents and Receiving Waters to Freshwater and Marine Organisms (5th Edition). EPA-821-R-02-012, Washington, DC

United States Environmental Protection Agency (USEPA) (2003) The Biotic Ligand Model: Technical Support Document for Its Application to the Evaluation of Water Quality Criteria for Copper, U.S. Environmental Protection Agency. Washington, DC

United States Environmental Protection Agency (USEPA) (2004) Overview of the Ecological Risk Assessment Process in the Office of Pesticide Programs, U.S. Environmental Protection Agency: Endangered and Threatened Species Effects Determinations. Office of Prevention, Pesticide, and Toxic Substances. Washington, DC

201

United States Environmental Protection Agency (USEPA) (2005) Ecological Risk Assessment for Copper. Office of Prevention, Pesticides and Toxic Substances Washington, DC

Winner RW, Keeling T, Yeager R, Farrell MP (1977) Effect of food type on the acute and chronic toxicity of copper to Daphnia magna. Freshwater Biology 7:343-349

Xue HB, Stumm W, Sigg L (1988) The binding of heavy metals to algal surfaces. Water Research 22:917-926

Yan H, Pan G (2002) Toxicity and bioaccumulation of copper in three green microalgal species. Chemosphere 49:471-476

Zhao C-M, Fan W-H, Wang W-X (2009) Aqueous and dietary copper uptake and elimination in Daphnia magna determined by the 67CU radiotracer. Environmental Toxicology and Chemistry 28:2360-2366

202

Table 1. Mean and ranges of water chemistry parameters for COMBO nutrient media used throughout all replicate experimentation

Measurement Measurement Parameter (range) (mean)

pH (SU) 7.8-8.7 8.4

Alkalinity 15.6-19.8 18.1 (mg/L as CaCO3) Hardness 38.1-45.4 42.2 (mg/L as CaCO3)

Conductivity 250-440 318 (µS/cm) Dissolved Oxygen 8.1-9.2 8.5 (mg/L) Total Suspended Solids < 0.1 < 0.1 (mg/L) Chlorophyll a < 1 < 1 (µg/L)

Total Phosphorus 1,300-1,500 µg/L

Free Reactive 1,000-1,400 µg/L Phosphorus Total Kjeldahl Nitrogen 0.22-0.25 0.23 (mg/L) Nitrate & Nitrite 5-8 6.4 (mg/L)

Turbidity (NTU) < 1 < 1

203

Table 2. Freshwater COMBO media constituents for culturing and testing of Scenedesmus dimorphus and Daphnia magna.

Component Final media Elemental (cation) mg/L µmol/L mg/L µmol/L

CaCl2•2H2O 36.76 250 10 250

MgCl2•7H2O 36.97 150 3.91 150

K2HPO4 8.71 50 3.91 100

NaNO3 85.01 1000 23.0 1000

NaHCO3 12.60 150 3.44 150

Na2SiO3•9H2O 28.42 100 4.6 200

H3BO3 24.00 388 4.6 426 Algal Trace Elements

Na2EDTA•2H2O 4.36 11.7 3.4 (EDTA) 11.7 (EDTA)

FeCl3•H2O 1.00 3.7 0.21 3.7

MnCl2 •4H2O 0.18 0.9 0.05 0.9

CuSO4•5H2O 0.001 0.004 0.00025 0.004

ZnSO4•7H2O 0.022 0.08 0.005 0.08

CoCl2•6H2O 0.012 0.05 0.003 0.05

Na2MoO4•2H2O 0.022 0.09 0.0086 (Mo) 0.09 (Mo)

H2SeO3 0.0016 0.012 0.001 0.012 Na3VO4 0.0018 0.01 0.0005 (Va) 0.01 (Va) Animal Trace Elements LiCl 0.31 7.3 0.05 7.3 RbCl 0.07 0.6 0.05 0.6

SrCl2•6H2O 0.15 0.57 0.05 0.57 NaBr 0.016 0.16 0.0125 (Br) 0.16 (Br) KI 0.0033 0.02 0.0025 (I) 0.02 (I) Vitamins B12 0.00055 0.0004 NA NA Biotin 0.0005 0.002 NA NA Thiamin 0.1 0.3 NA NA

204

Table 3. Comparison of physical and chemical properties of the two copper formulations tested.

Copper sulfate pentahydrate a, b Captain® c, d

Manufacturer Fisher Scientific SePRO Corporation 7758-99-8 (CAS No.) 67690-9 (EPA reg. Identification C489-1 (Catalogue number) number) 249.68 formula weight Copper-ethanolamine Active Ingredient Elemental copper (Cu+2) complexes % Active Ingredient 25 28.2

Appearance Blue crystalline solid Blue viscous liquid

Water Solubility 32 g/100 g (20°C) Miscible Concentrate pH 3.5-4 (5% aqueous solution) 10-10.5 (SU) Specific Gravity NA 1.2 (g/cm3) a Fisher Scientific 2008 b USEPA 1986 c SePRO 2014 d SePRO 2015

205

Table 4. Description of targeted concentrations used in toxicity testing to define the exposure-response relationship.

Targeted Initial Copper Range of Measured Total Concentrations as Copper Water Sample Parameter Copper as Percent of Measured Endpoints Sulfate or Captain Nominal (µg Cu/ L)

0, 10, 25, 50, 75, 100, 150, 48-hr NOEC, LOEC, and Copper sulfate no algae 98-102 200, 250 LC50 values

0, 50, 75, 100, 150, 200, 250, 48-hr NOEC, LOEC, and Captain no algae 96-101 300, 350, 400, 450, 500 LC50 values

Copper sulfate 5 x 105 0, 150, 200, 250, 300, 350, 48-hr NOEC, LOEC, and 96-100 cells/mL 400, 450, 500, 600 LC50 values

0, 150, 200, 250, 300, 350, 48-hr NOEC, LOEC, and Captain 5 x 105 cells/mL 95-98 400, 450, 500, 600 LC50 values 0, 150, 200, 250, 300, 350, Copper sulfate 5 x 106 48-hr NOEC, LOEC, and 400, 450, 500, 550, 600, 700, 97-103 cells/mL LC50 values 750 0, 150, 200, 250, 300, 350, 48-hr NOEC, LOEC, and Captain 5 x 106 cells/mL 400, 450, 500, 550, 600, 700, 95-99 LC values 750 50

206

Table 5. Average Scenedesmus dimorphus cellular characteristics.

Cell SA Cell volume Bio-volume Cells/ Coenobium Treatment (µm2) (µm3) (mm3/ mL) coenobium SA (µm2) Copper sulfate no ND ND ND ND ND algae Captain no ND ND ND ND ND algae Copper sulfate 5 x 85.34 47.88 23.94 5.1 240 105 cells/mL Captain 5 x 93.57 52.36 26.18 5.3 291 105 cells/mL Copper sulfate 5 x 91.77 49.42 247.1 5.25 281 106 cells/mL Captain 5 x 88.54 54.86 274.3 5.4 302 106 cells/mL

Abbreviations: ND = non-detect; SA = surface area

207

Table 6. Description of exposure-response relationships using nonlinear regression analysis with a three parameter sigmoidal curve. Equations and R squared values are calculated for copper sulfate and Captain treatments in 48-hr Daphnia magna toxicity tests.

Total Copper Free Copper R R Regression Regression Treatment squared squared Equation Equation value value (y=) (y=) Copper sulfate 81.45/(1+exp(-(x- 81.42/(1+exp(-(x- 0.9912 0.9908 no algae 58.63)/37.53)) 55.08)/35.56))

87.76/(1+exp(-(x- 87.62/(1+exp(-(x- Captain no algae 0.9300 0.9322 346.93)/25.96)) 340.83)/25.83))

Copper sulfate 5 101.55/(1+exp(-(x- 97.93/(1+exp(-(x- 0.9952 0.9768 x 105 cells/mL 319.08)/48.99)) 169.67)/32.93))

Captain 5 x 105 105.35/(1+exp(-(x- 95.40/(1+exp(-(x- 0.9907 0.9690 cells/mL 421.62)/71.48)) 298.10)/40.17))

Copper sulfate 5 114.69/(1+exp(-(x- 122.05/(1+exp(-(x- 0.9864 0.9534 x 106 cells/mL 546.35)/116.12)) 251.55)/81.97))

Captain 5 x 106 129.66/(1+exp(-(x- 105.13/(1+exp(-(x- 0.9819 0.9721 cells/mL 642.33)/114.88)) 298.49)/67.68))

208

Table 7. Average percent free and total copper in solution and sorbed to algae in all treatments over the 48 hr experiment duration.

% sorbed to ng copper pg copper % free copper Treatment algae bound/ cell /um2 coenobia (STD) (STD) (STD) (STD)

Copper sulfate NA 93.60 (1.74) NA NA no algae

Captain no NA 96.4 (3.52) NA NA algae Copper sulfate 5 x 105 36.2 (11.40) 57.4 (9.50) 0.14 (0.035) 2.98 (0.75) cells/mL Captain 5 x 105 19.46 (6.67) 76.94 (5.55) 0.084 (0.042) 1.53 (0.78) cells/mL Copper sulfate 5 x 106 58.35 (12.13) 35.25 (10.11) 0.026 (0.0085) 0.493 (0.159) cells/mL Captain 5 x 106 54.82 (9.02) 41.57 (7.51) 0.024 (0.008) 0.432 (0.147) cells/mL

Abbreviations: NA = not applicable; STD = standard deviation

209

Table 8. 48-hr NOEC, LOEC, and LC50 values in copper formulation toxicity experiments 1) without algae, 2) containing 5 x

105 cells/mL, and 3) containing 5 x 106 cells/mL. a

48-hr 48-hr 48-hr 48-hr 48-hr LC 95% 48-hr LC 95% NOEC NOEC LOEC LOEC 50 50 Treatment (total µg confidence (free µg confidence (total µg (free µg (total µg (free µg Cu/ L) interval Cu/ L) interval Cu/ L) Cu/ L) Cu/ L) Cu/ L) Copper sulfate no 50 46.8 75 70.6 75.3 59.5-99.1 73.6 63.3-87.1 algae Captain no 300 297.6 350 342.1 353.8 340.1-369.4 351.1 341.8-361.7 algae Copper sulfate 5 x 200 100.4 250 113.7 317.8 309.3-325.9 171.3 154.3-188.1 105 cells/mL Captain 5 x 105 250 210.5 300 257.2 414.2 403.3-425.6 302.2 286.1-323.0 cells/mL Copper sulfate 5 x 250 71 300 82.6 517.8 500.3-536.4 221.6 194.6-252.9 106 cells/mL Captain 5 x 106 300 100.1 350 120.1 588.5 568.3-609.5 272.0 258.1-287.2 cells/mL a Abbreviations: NOEC = no observed effect concentration; LOEC = lowest observed effect concentration; LC50 = lethal concentration needed to kill 50% of test organisms.

210

120

100

40 Sorbed Copper Free Copper

20 Percent copper applied of Percent

0 Copper Captain no Copper Captain Copper Captain sulfate no algae sulfate 5x10^5 sulfate 5x10^6 algae 5x10^5 cells/mL 5x10^6 cells/mL cells/mL cells/mL

Treatment

Figure 1. Mean free and total copper concentrations in each treatment over the 48-hr exposure duration. Error bars represent one standard deviation from the mean.

211

300

250

200

150

100 Free Cu/L) (µg Free Copper

0 0 10 25 50 75 100 150 200 250

Total Copper (µg Cu/L)

Copper sulfate 2 hr free Copper sulfate 24 hr free Copper sulfate 48 hr free

Figure 2. Mean free copper compared with total nominal copper concentrations at 2, 24, and

48 hr after copper sulfate treatments with no algae present. Error bars represent one standard deviation from the mean.

212

500

450

400

350

300

250

200

150 Free Cu/L) (µg Free Copper

100

0 0 50 75 100 150 200 250 300 350 400 450 500

Total Copper (µg Cu/L)

Captain 2 hr free Captain 24 hr free Captain 48 hr free

Figure 3. Mean free copper compared with total nominal copper concentrations at 2, 24, and

48 hr after Captain treatments with no algae present. Error bars represent one standard deviation from the mean.

213

500

450

400

350

300

250

200

150 Free Cu/L) (µg Free Copper 100

0 0 150 200 250 300 350 400 450 500 600 Total Copper (µg Cu/L)

Copper sulfate 2 hr free Copper sulfate 24 hr free Copper sulfate 48 hr free

Figure 4. Mean free copper compared with total nominal copper concentrations at 2, 24, and

48 hr after copper sulfate treatments containing 5 x 105 cells/mL of algae. Error bars represent one standard deviation from the mean.

214

600

550

500

450

400

350

300

250

200

150 Free Cu/L) (µg Free Copper 100

0 0 150 200 250 300 350 400 450 500 600 Total Copper (µg Cu/L)

Captain 2 hr free Captain 24 hr free Captain 48 hr free

Figure 5. Mean free copper compared with total nominal copper concentrations at 2, 24, and

48 hr after Captain treatments containing 5 x 105 cells/mL of algae. Error bars represent one standard deviation from the mean.

215

550

500

450

400

350

300

250

200

150 Free Cu/L) (µg Free Copper 100

0 0 150 200 250 300 350 400 450 500 550 600 700 750 Total Copper (µg Cu/L) Copper sulfate 2 hr free Copper sulfate 24 hr free Copper sulfate 48 hr free

Figure 6. Mean free copper compared with total nominal copper concentrations at 2, 24, and

48 hr after copper sulfate treatments containing 5 x 106 cells/mL of algae. Error bars represent one standard deviation from the mean.

216

600

550

500

450

400

350

300

250

200

150 Free Cu/L) (µg Free Copper 100

0 0 150 200 250 300 350 400 450 500 550 600 700 750 Total Copper (µg Cu/L) Captain 2 hr free Captain 24 hr free Captain 48 hr free

Figure 7. Mean free copper compared with total nominal copper concentrations at 2, 24, and

48 hr after Captain treatments containing 5 x 106 cells/mL of algae. Error bars represent one standard deviation from the mean.

217

700

Total Copper 600

Free Copper 500

400

(µg Cu/L) (µg 50

300

hr hr LC -

48 200

100

0 Copper Captain no Copper Captain Copper Captain sulfate no algae sulfate 5x10^5 sulfate 5x10^6 algae 5x10^5 cells/mL 5x10^6 cells/mL cells/mL cells/mL

Treatment

Figure 8. Mean Daphnia magna 48-hr LC50 values for each treatment based on total copper and free copper in exposures. Error bars represent a 95% confidence interval around the mean.

218

219

Figure 10. Mean percent mortality of Daphnia magna in Captain treatments consisting of 1) no algae, 2) 5 x 105 cells/mL, and 3) 5 x 106 cells/mL. Regression equations listed for each exposure-response relationship. Error bars represent a 95% confidence interval around the mean.

220

120 700 120 700

600 100 600 100

500 500 80 80 400 400 60 60 300 300

40 µg L µg Copper/ µg L µg Copper/ 40

200 200 Percent copper applied of Percent 20 Free copper applied of Percent 100 20 100 Free Copper Copper Sorbed Sorbed 0 0 copper 0 0 Copper Copper Copper copper LC50 Captain no Captain Captain sulfate no sulfate sulfate LC50 Total algae 5x10^5 5x10^6 algae 5x10^5 5x10^6 Total LC50 cells/mL cells/mL cells/mL cells/mL LC50 Free Treatment Free Treatment

Figure 11. Mean free and sorbed copper (bars) on primary Y axis and 48-hr LC50 values (lines) for total and free copper concentrations on the secondary Y axis for copper sulfate (left) and Captain (right). Error bars are 95% confidence intervals around the mean.

221

Captain non-target survival Copper sulfate non-target survival Captain algal response Copper sulfate algal response

100 100

90 90

80 80

70 70

60 60

50 50

Survival (%) Survival 40 40

30 30 Algal Inhibition Inhibition (%) Algal Daphnia Daphnia 20 20

10 10

0 0 250 300 350 400 450 500 600

Total Copper Concentration (µg/L)

Figure 12. Mean percent mortality of Daphnia magna compared with percent algal inhibition for 5 x 105 cells/mL exposures to copper sulfate and Captain.

222

Captain non-target survival Copper sulfate non-target survival Captain algal response Copper sulfate algal response

100 100

90 90

80 80

70 70

60 60

50 50

Survival (%) Survival 40 40

30 30 Algal Inhibition Inhibition (%) Algal

Daphnia Daphnia 20 20

10 10

0 0 350 400 450 500 550 600 700 750 Total Copper Concentration (µg/L)

Figure 13. Mean percent mortality of Daphnia magna compared with percent algal inhibition for 5 x 106 cells/mL exposures to copper sulfate and Captain.

223

CHAPTER SIX

CONCLUSIONS AND A RECOMMENDED DECISION INFORMATION SYSTEM

FRAMEWORK

BACKGROUND

Anthropogenic activities are changing global environmental conditions at a faster rate than has been documented in over 10,000 years (Miller 2002). Such changes often select for the most adaptable organisms (Schopf 2000a). Algal and cyanobacterial blooms are increasing in intensity and frequency throughout the world (Hallegraeff 1993), and some of the largest blooms ever documented have occurred in the recent past (Paerl and Otten 2013).

Anthropogenically induced climate change and nutrient mobilization are key factors promoting the expansion of nuisance and harmful algal blooms (Anderson et al. 2002;

Heisler et al. 2008; Paerl and Paul 2012; Taranu et al. 2015). Atmospheric alteration in greenhouse gases may exacerbate ozone depletion, ultraviolet radiation, and temperature increases which, in turn, favor toxic cyanobacteria (Miller 2002; Solomon et al. 2009).

Climate extremes such as heat waves and extensive droughts may also select for bloom forming cyanobacteria due to rapid recruitment capability and stress adaptations (Paerl and

Huisman 2009; Kosten et al. 2012). All of these factors interact synergistically to make harmful algal blooms an unprecedented threat to freshwater resources (Paerl and Scott 2010).

When the ecological ramifications of noxious algal blooms become known or the economic impacts are realized, there will be more concerted management efforts. There are many potential management options to consider, but risks associated with their

224

implementation are often cited to dismiss them from consideration. Ironically, the risks that spurred the need for investigating management options often are neglected. Yet, failure to address noxious algal blooms can accelerate their dominance, spread, and negative impacts to water resources. Noxious algae can even enhance the in situ eutrophication process and engineer a system that promotes their continued dominance (Beversdorf et al. 2013; Chen et al. 2014). Freshwaters have been plagued by extreme, chronic nutrient enrichment (USEPA

2007), and are at a point where ‘in lake’ intervention is a critical component of the approach needed to control algal blooms (Jarvie et al. 2013). When water resource managers realize this critical turning point, they likely will seek a scientific approach to justify management.

This dissertation provides a risk-based framework incorporating key research arenas for selecting an algal management program. It balances the ‘no action’ approach with efficiency and risks associated with taking action. The overall goal is to promote informed decisions regarding the need for, and direction of, management.

Many management actions can have negative impacts on water resources, and often come with a cost. Due to this cost and risks of impact following implementation, many resource managers choose not to take action to combat noxious algal infestations. However, the cost and ecological impacts from no management can outweigh those associated with management programs (Mastin et al. 2002; Getsinger et al. 2014). This chapter summarizes the risk-based management approach, wherein risks associated with different management approaches (e.g., proactive, reactive) are assessed in comparison to risks associated with no management. Incorporating information from the research chapters included herein, a more accurate assessment of risks is presented in terms of effectiveness and ecological integrity of

225

management approaches. This research is contributed toward the goal of making improved, informed decisions about whether to manage, and about selection of an appropriate management approach.

MANAGEMENT OBJECTIVES AND PRIMARY USE

Prioritizing the primary uses of the water resource often assists in clarifying the direction of management and acceptability of associated risks. Many water resources are multi-use, though they often were originally established for a few specific purposes. For example, some reservoirs were established solely to be a potable water source, cooling water for industry, or hydropower production. Though there may be public access to support recreational activities like fishing, these may not be prioritized in a management program.

Therefore, risks to fisheries health may be ranked below that of preservation of the primary function. Property values are often a key influencing value to a water resource, and although impact to these may correlate with other functional uses, they may not have been the original purpose of the water resource. Some desired uses may be contradictory and further require prioritization (Wagner and Oglesby 1984). Management approaches may also have restrictions for the designated water uses such as some herbicides/algaecides in potable water or waters destined for irrigation. The primary and secondary management objectives can influence the management approach and prioritize criticality and timing of management effectiveness versus collateral risks.

226

RISK ASSESSMENT: NO ACTION

The risks of not managing nuisance algal infestations could be great and need to be weighed in assessing a decision to manage. For example, allowing a toxic algal bloom to continue promotes wildlife and human exposures, even if the lake is closed (Miller et al.

2010; Banack et al. 2015; Woller-Skar et al. 2015). Also, a chronic toxin exposure is possible and acute exposure is increasingly likely by not managing and allowing a toxic bloom to continue unabated (Chorus and Bartram 1999). In the past few decades, blooms have been documented to last longer and achieve extreme densities (Robarts and Zohary 1986). With elevated water temperatures and increased growing seasons, bloom intensity and duration is predicted to increase (Paerl and Huisman 2008) as well as related toxicity (Davis et al. 2009;

Dziallas and Grossart 2011). Water resource managers may eventually be held liable for both wildlife and human health risks associated with designated water uses. Multiple negative impacts to the water resource and interconnected environment, as discussed below, can be realized if algal blooms are not mitigated.

Economic and functional value

Accumulated biomass can impede flood control which, in turn, can impact property values (Thunberg et al. 1992). Additionally, clogging of intake structures for hydropower or decreasing cooling capacity can be ramifications of algal growths for industrial uses

(Getsinger et al. 2014). Recreational activities are also commonly impeded by weed and toxic algal infestations (Bergstrom et al. 1996). Some algal mats can harbor fecal bacteria or toxic gram-positive bacteria that can pose health risks to humans and wildlife (Byappanahalli and Whitman 2009; Chun et al. 2013; Vijayavel et al. 2013). Cyanobacteria may also

227

decrease the quality of irrigation water or directly produce toxins that can negatively influence irrigated agricultural crops or turf (Legrand et al. 2003; Babica et al. 2006; Bhadha et al. 2014). Direct impacts of toxic algae on fish can also have major ramifications for the value of the water resource. Toxic algal blooms have caused mass fish mortalities throughout the US (34 million documented fish mortalities in Texas from Prymnesium parvum valued at over $13 million US; Southard et al. 2010). Catfish aquaculture is also severely impacted with $15-23 million US in annual production costs required to manage off flavor (Hanson

2006b). Through many pathways, eutrophication has degraded ecosystem goods and services with related costs (very) conservatively estimated at over $2 billion annually within the US alone (Carpenter et al. 1998; Dodds et al. 2009).

Toxins

Numerous toxins have been described from cyanobacteria in freshwaters that can cause health risks to humans, domestic pets, livestock, and wildlife associated with the water resource (Turner et al. 1990; Falconer 1999; WHO 1999; Carmichael 2001). There are some commonly studied and well described cyanobacterial toxins or cyanotoxins (e.g., microcystin

LR, cylindrospermopsin, anatoxin-a), and many variants are being described as well as entirely new compounds and classes of toxins (Namikoshi 1994, 1995; Sano et al. 2001).

Unidentified and undescribed compounds that can have extremely potent impacts to test organisms are also commonly reported, and there are no standard analytical methodologies for their characterization or measurment (Scott 1991; Skulburg 1992, 1994; Onodera et al.

1997). Novel cyanotoxin groups have recently been discovered as well, such as lyngbyaureidamides, anabaenopeptins (Grach-Progrebinsky and Carmeli 2008; Zi et al.

228

2012), nodulopeptins (Schumacher et al. 2012), jamaicamides (Edwards et al. 2004), and aeruginosins (Ishida et al. 2009). These have new and unique impacts compared with the historically known toxins. Anas et al. (2016) reported that aeruginosins decreased clotting factors in bleeding humans, and promoted blood complications and disorders. Due to the continually new information on cyanotoxins, and in light of the potential chronic (decadal) exposure and accumulation potential, cyanotoxin science may surely unveil just how menacing toxins can be and have been.

Despite advancements in monitoring technologies (Reed et al. 2010), limited adoption of these technologies and the typically heterogeneous distribution of blooms in time and space commonly impede detection. Impacts may be difficult to manage or avoid due to costly potable water chemical treatment methods or inefficiencies of toxin removal (Burkholder et al. 2010). Analytical detection can be complex and costly to conduct, and the lack of standardized testing regulations poses challenges to the assessment of risks from cyanobacterial blooms. Since toxin production may be intermittent and change within a population over the bloom duration (Beversdorf et al. 2013; Gobler et al. 2016), source control of the potential producers should be considered. Managing algae before cyanotoxin levels are sufficient to cause concern (even if toxin was released) would be ideal. Allowing an increase in population numbers could produce higher toxin levels (described or unknown), and increase the risk of natural release from the cells (White et al. 2005; Lehman et al. 2013).

A ‘guilty until proven innocent approach’ to managing cyanobacteria may be appropriate and perhaps the best way to ensure safety of the water resource (Codd et al. 1997; Carmichael

2001; Otten and Paerl 2015).

229

By treating effectively, cyanotoxin production is depressed and total toxin levels, used in risk assessment by the WHO (1999), also decrease. The leaky cell hypothesis may shift toxins from in the cell to aqueous phase (Kenefick et al. 1993), though total toxin typically is diluted (rapidly in open water) and degraded (e.g., half-lives of 48 hr - 2 weeks;

Lahti et al. 1997; Jones and Orr 1994; Edwards et al. 2008). Also, with strategic algaecide formulations and rates, control can be achieved with minimal cyanotoxin release (Iwinski et al. 2016). Regulatory guidelines are typically based on total toxin concentration (in cell and in water), and risks to humans and wildlife are better encompassed by total exposure due to the ability of algae (and associated toxin) to concentrate physically through buoyancy and wind action. Such physical concentration can lead to localized toxin concentrations more than 1,000 fold higher than water-column concentrations, especially in high-use areas for humans and wildlife (e.g., shorelines and beach areas; Chorus and Bartram 1999). If an effective reactive control measure was implemented, dead cyanobacterial cells would not be able to adjust buoyancy (nor produce toxin), thus offsetting the highly concentrated scums and other potential “hot spots” for acute exposure to cyanotoxins (Chorus and Bartram 1999), and/or continued chronic exposure.

Ecological implications

In moderation, some nutrient enrichment can actually promote beneficial phytoplankton and bottom-up food chain effects (Burkholder and Glibert 2013 and references therein). Fish biomass and size typically increases in early stages of eutrophication, and may be viewed as a beneficial outcome. Piscivore-dominated, long- length food chains have also been described as more resilient to phosphorus (P) loading, as

230

the P is incorporated more rapidly and completely into the biota (Burkholder and Glibert

2013). Jeppeson et al. (2005) found that following external P reductions, there was less overall fish biomass although the percentage of piscivores, typically characteristic of less eutrophic waters, increased in 80% of the lakes. However, long-term eutrophication often results in profound and irreversible impacts on fisheries (Burkholder and Glibert 2013). Loss of sensitive species is commonly the first sign of this process in freshwaters (Schindler

1987). Large-scale alterations in habitat and major shifts in aquatic food webs result in sustained loss of biodiversity as a long-term impact (Burkholder et al. 2008). Cultural eutrophication, water use dynamics, and temperature extremes can promote stratification in aquatic systems. “Rough fish” (e.g., mostly undesirable to anglers) tend to dominate the benthic zone in highly eutrophic water due to the ability to tolerate low oxygen (Burkholder and Glibert 2013). This can allow for increased sediment suspension and nutrient mobilization to support nuisance algal blooms. Cyanobacteria may dominate in these scenarios due to their ability to grow well in lower light levels and to acquire nutrients from the lower water-column (Ganf and Oliver 1982; Carey et al. 2008). Furthermore, cyanobacteria can take advantage of various forms of nutrients and episodic inputs, such as are contributed from fish (Reinersten et al. 1986; Schaus and Vanni 2000).

Cyanobacteria commonly dominate the plankton, and sometimes the periphyton, of eutrophic waters (Burkholder 2009). Their growth characteristics (colony size, mat- formation, mucilage) and toxin production discourages most zooplankton grazing (Webster and Peters 1978; Lampert 1981, 1987; Reinikainen et al. 1995). Beneficial zooplankton such as Daphnia can selectively graze on green algae over cyanobacteria (Mitra and Flynn 2006,

231

2006a; Wang et al. 2010). Some cyanotoxins negatively affect juvenile fish and thereby promote the growth of herbivorous zooplankton (Pavagadhi et al. 2013). If zooplankton do consume toxic cyanobacteria, the toxins can bio-accumulate, resulting in increased bio- magnification up the food web (Watanabe et al. 1992). Decreased grazer size and subsequent grazing pressure has been observed due to loss of protective habitat (e.g., beneficial macrophytes) related to nutrient enrichment. Microcystins act as endocrine disruptors in fish, altering the activities of enzymes that are critical to steroid hormone synthesis and metabolism (Rogers et al. 2011). Physiological stress and damaged gonad tissue (lesions, cell apoptosis, and testicular ultrastructure alteration) in fish have been documented following microcystin LR exposure (Trinchet et al. 2011; Zhao et al. 2012; Qiao et al. 2013).

Accidental introduction of invasive species can also alter ecosystem balance and select for noxious algae. Invasive species such as zebra mussels (Dreissena polymorpha) may preferentially graze on beneficial algae and cycle nutrients to fuel noxious algal growth

(Vanderploeg et al. 2001). Other intentionally introduced species such as triploid grass carp are widely used for submersed vegetation control (> 45 states in the US; Mitchell and Kelly

2006), although they do not readily consume most noxious cyanobacteria (Kasinak et al.

2015), and they can also promote through nutrient cycling and sediment mobilization

(Scheffer et al. 2001). Anthropogenically mediated spread of non-native and invasive species, coupled with or promoted by increased eutrophication, has stimulated noxious algal blooms with severe ramifications for the ecosystems.

Human health implications

232

Cyanobacteria produce an array of toxins and also numerous other bioactive metabolites (compounds that can affect other organisms) whose impacts are yet unknown.

Despite the large body of literature on cyanotoxins, there is still much to be understood about exposures to these toxins. These toxins can accumulate in biota over decades, and as discussed below, new types of exposures are being revealed that are correlative with increasing bloom prevalence. Severe ramifications on human health have been documented including neurotoxicity, kidney/liver impairment, gastrointestinal disorders (Bell and Codd

1994), and multiple types of cancer (Hernandez et al. 2009). Understanding the potential routes of cyanobacterial toxin exposure is critical to improve assessment of long-term health risks. Many toxins have been shown to accumulate in key organs (e.g., kidney, liver, brain) and manifestation of effects may result after years of exposure (Banack et al. 2015). Thus, additive accumulation from all potential routes of exposure should be considered.

Cyanobacterial blooms have occurred in many drinking source waters, and commonly have caused severe taste-and-odor issues, though toxins have more recently gained public awareness (USEPA 2012). Microcystins in drinking waters have been associated with hepatic tumors and colorectal cancer (Ueno et al. 1996; Lun et al. 2002). Human exposure can occur from many routes, such as inhalation of aerosolized toxins, ingestion/consumption of contaminated water/algal cells, and recreational skin contact with cyanobacterial infestations

(Backer et al. 2008, 2010; Wood and Dietrich 2011; Stommel et al. 2013). Strong correlations have been discerned between chronic cyanotoxin exposure and neurodegenerative diseases, such as amyotrophic lateral sclerosis (ALS) and Alzheimer’s

(Bradley and Mash 2009), and human mortalities also have been documented (Carmichael et

233

al. 2001) under extreme exposure scenarios. For example, Banack et al. (2015) found a 25- fold increase in ALS in people bordering Lake Mascoma in Enfield, NH, that contained chronic cyanobacterial blooms producing the neurotoxic amino acid β-N-methylamino-L- alanine (BMAA). Acknowledging the accumulative health risks associated with toxin exposures as well as their increased impacts to young children (USEPA 2015a, b) should be important components in the decision to manage, and to the management approach.

PROACTIVE RISKS

Overview

Some of the greatest risks of proactive mitigation include costs and efficacy. Since infestations may be caused by numerous factors, it becomes more difficult to predict the time to, and the degree of, influence of singlet factors that can be managed. Also, as the system may have been altered in other ways, only focusing on one approach may not fully resolve the problem. Nutrient mitigation can be influential in multiple ways in regards to algal blooms (Heisler et al. 2008). Decreased nutrient inputs have resulted in concomitant decreases in algae in some scenarios (Mallin et al. 2005), although in others, it has taken decades for any positive effects of decreased nutrient inputs to be observed (Jarvie et al.

2013). Alternatively, the algal blooms can worsen due to in situ environmental alterations

(Bachmann et al. 2003; Glibert et al. 2011; Orihel et al. 2015). It is often difficult to adequetley mitigate the nutrients entering a system, and the legacy accumulation can easily exceed a tipping point for internal fueling of algal blooms (Duarte et al. 2008). Thus, predicting the capability and time to noticeably improve water quality are more difficult

234

when depending solely upon proactive nutrient management approaches. Predictions of efficacy involving cyanobacteria also differs with species and strain, and depends on environmental factors (Rigosi et al. 2014). For example, temperature increases can alter cyanobacterial species responsiveness to nutrient reductions (Kosten et al. 2012). Watershed management can also be very costly in attempts to remove large amounts of phosphorus and nitrogen. Due to large increases in nutrient loads, even high removal efficiencies and percent nutrient reductions may not be sufficient to offset growth rates of algae in receiving waters

(Paerl et al. 2016). Alterations in climatic factors including extreme rainfalls and/ or prolonged droughts may also decrease the efficiencies of stormwater management practices

(Trenberth 2005; Paerl 2014). Non-point source nutrient introductions can be difficult to identify and even more difficult to manage (Smith et al. 2014), and can contribute significantly to overall inputs (Wetzel 2001). Proactive management should be highly considered so as to prevent establishment of blooms and offset associated risks, although a realistic understanding of effectiveness is also needed.

Phoslock fate and effects

There are risks and indirect effects associated with most in situ nutrient mitigation strategies. The potential impacts of Phoslock are summarized here due to Phoslock application in this work (Chapter 2). Phoslock is designed to limit free lanthanum release to water due to the unique integration into a bentonite clay matrix (Douglass 2002). This provides a limited potential for exposure to free lanthanum in the water-column. Toxicity studies on lanthanum only using the free ion (LaCl3 source) often overestimate risks as compared with Phoslock directly. Upon application to water, lanthanum associated with the

235

clay in Phoslock preferentially and rapidly binds with phosphate (PO4), forming a highly stable mineral called rhabdophane (Jonasson et al. 1988). Rhabdophane further transforms to a more stable mineral, monazite (LaPO4), through time (Cetiner et al. 2005; Dithmer et al.

2015). This is considered the ultimate fate of Phoslock due to the extreme extraction methodologies required to disassociate (Reitzel et al. 2013). Toxicity studies assessing

Phoslock impacts to sentinel zooplankton revealed levels in excess of study concentrations were needed to elicit effects. These ranged from 50 to 50,000 mg Pholsock/L (Ecotox 2008;

Lurling and Tolman 2010). Similarly, both fresh and saltwater benthic invertebrates as well as fish did not show signs of toxicity at the highest levels of Phoslock tested (≥ 450 mg/L) in third-party studies (Clearwater and Hickey 2004; Watson-Leung 2009; Reitzel et al. 2013).

Since these levels are in excess of typical field applications, no risks are projected to aquatic biota, nor have any issues been documented from field applications in the United States.

Lanthanum from Phoslock can accumulate short-term in aquatic biota in the liver and hepatopancreas tissues, not in the muscle, though is depurated through excretion (Landman et al. 2007). No significant exposure to lanthanum is predicted to humans following Phoslock applications, and lanthanum is actually a component of prescribed medicine to reduce blood phosphate levels.

REACTIVE RISKS

Overview

If proactive management options are not feasible or insufficient to mitigate noxious algal blooms, reactive management should be considered. Resource managers typically

236

evaluate all practical and legal management options other than use of algaecides such as mechanical harvesting, physical mechanisms (e.g., dye, sonication, flushing), and biological controls (e.g., grass carp, viruses; Codd et al. 2005). These strategies may not resolve the problem, or they may cause other economic ramifications, environmental effects, and collateral damages (Dyck 1994; Scheffer et al. 2001). When critical water resources require immediate intervention, often strategic application of an algaecide provides the most broad- spectrum efficacy and the most rapid, temporary improvement (Mastin et al. 2002). In this research, copper, in the form of USEPA-registered algaecides, was specifically assessed due to its widespread use, increased regulatory scrutiny, and relatively low potential environmental impacts (USEPA 2009). Weighing risks of copper use versus no management provides a valuable assessment in the decision process.

Copper overview

Copper is commonly found in aquatic ecosystems, at 0.2 to 30 µg/L in the water and

0.8-50 mg/kg in bottom sediments (USEPA 1984; Flemming and Trevors 1989). The maximum contaminant level allowed in US drinking waters is set at 1.3 mg/L (USEPA

2007). Copper is a naturally occurring element and an essential micronutrient for biota. It is also a structural element of many proteins and a co-factor in many enzymes, such as plastocyanin which is involved in electron transport (Yruela 2005). Copper is important for animals for its use in hemocyanin (a copper protein) as the primary oxygen carrier in most mollusks and arthropods (Freedman et al. 1976). Copper has many functions in humans, as well, such as a cofactor for blood oxygen transfer, brain development, immune response, etc.

(Tapiero et al. 2003). Above threshold levels, however, many copper forms can elicit

237

sublethal or toxic impacts to aquatic organisms. Daphnia magna, for example, has a 96-hr

LC50 of 10 µg/L in laboratory prepared water (USEPA 1980). Salmonids are also very sensitive to copper and exposures can impair olfaction and normal migration patterns and cause avoidance at levels as low as 0.7 µg/L dissolved copper (Hansen et al. 1999). Due to rapid algal biomass die offs, copper can indirectly cause or contribute to dissolved oxygen

“sags” that can cause disease and death of aquatic biota. Numerous exposure components including water chemistry, duration of exposure and frequency of application can all impact risks of copper associated with algaecide applications.

The formulation of copper, in particular, is important for eliciting both targeted algaecidal activity and decreasing non-target risks. This research revealed that chelated copper, in general, has greater efficacy in controlling common noxious algae (Chapter 4) in comparison to copper sulfate. Closson and Paul (2014) also found that chelated copper was safer for fish in acute toxicity testing. In this study, relevant field exposure components (i.e., algae) were incorporated, and all copper algaecides tested had significantly reduced non- target species toxicity (Chapter 5). USEPA (2009) recently approved continued use of copper in aquatic environments with labeling that incorporates numerous study findings to prevent unreasonable risks to humans and other non-target species. No restrictions on recreational activities, potable water source or irrigation are present on copper algaecide labels. Risks associated with copper use (i.e., amount needed over time, acute and chronic risks [differs by formulation], and indirect risks) all need to be considered in selection of this approach.

Additionally, these risks need to be weighed against achieving management objectives and the risks associated with ineffective or no management.

238

Copper fate from aquatic pesticide applications

Following an algaecide treatment, applied copper moves from the water-column to algae and ultimately to bottom sediments in an aquatic system (Gallagher et al. 2005). The transport of copper to sediments is mediated by sorption to suspended particulates and precipitation via interaction with carbonate and sulfide (Flemming and Trevors 1989). Liu et al. (2006) measured rapid copper sorption to suspended sediment particles within 2 hours after copper applications with 99% of applied copper transferring to bottom sediments in 2 days and less than 0.01% found in fish tissues. Movement of copper as an algaecide to sediments is facilitated by algal biomass. Copper algaecides rapidly sorb (~15 min) to numerous binding sites on algae (e.g., carboxylic, sulfhydryl, phosphate groups, transport proteins; Crist et al. 1990), and the movement of copper applied as an algaecide has been monitored with settling algae following treatment (Button et al. 1977). This is consistent with

Jones et al.’s (2008) finding that the majority of copper bound in treated sediments short- term after application was associated with the oxidizable fraction comprised primarily of algal biomass.

Long-term accumulations

Copper that partitions to sediments following a pesticide application is transformed to more stable forms such as the minerals malachite and chalcocite (Liu et al. 2006). Jones et al.

(2008) found that the bioavailability of sediment-sorbed copper continues to transform to more stable and biologically unreactive forms following transport to sediments, which further decreases risks to benthic biota. The sediment copper concentration that may cause adverse effects to non-target organisms can differ widely (orders of magnitude), depending on the

239

sediment characteristics (Willis et al. 2013; Huggett et al. 1999). Acid-volatile sulfides (Di

Toro et al. 1990), pH (Burton 1991), cation exchange capacity (Chapman et al. 1998), and organic matter content and type (Besser et al. 2003; Milani et al. 2003) are important sediment characteristics that influence the bioavailability of copper. Studies assessing long- term copper algaecide programs have not shown any increase in sediment copper toxicity

(Han et al. 2001; Gallagher et al. 2005; Iwinski et al. 2016).

RISK ASSESSMENT: MANAGEMENT DECISION EXAMPLES

With related macroscale algal invasions, collateral damages of management are readily assessed and acknowledged. In these systems an understanding of issues that are projected to occur from invasions are weighed against short-term risks of management, regardless of how extreme. An example of this includes caulerpa (Caulerpa taxifolia (M.

Vahl) C. Agardh) invasions wherein liquid chlorine was injected to attain control of the target pest (Carlton 2001). Caulerpa was eliminated and virtually all living organisms present in the vicinity were also killed. This intense ‘scorched earth’ application was considered to be warranted based on examples of the trajectory of other invasions. Complete habitat destruction and exclusion of native species are common attributes of caulerpa invasions

(Meinesz 1999). Successful re-establishment of many native species after eradication of the invader also supported the management decision (Anderson 2004).

Another example concerns the invasive aquatic weed, hydrilla (Hydrilla verticillata

(L. f.) Royle). Acknowledging the voracious, long-term ramifications of the plant and its ability to spread throughout aquatic systems (Langeland 1996), the California Department of

240

Food and Agriculture (CDFA) eradication program employs extreme measures for control. In one example, the Shasta Project, multiple ponds identified with hydrilla were chemically treated with herbicide and completely filled in with soil (CDFA 2003). This approach inherently sacrificed all of the biota in the aquatic system and terminated the entire water resource, because of the potential negative impacts that could be elicited by unchecked invasion.

A terrestrial invasive with a similar ‘no mercy’ management approach is cogongrass

(Imperata cylindrica (L.) Beauv.), which is a significant invader throughout the southeast

US. The primary control mechanism for this plant is routine treatment with a combination of non-selective herbicides that often kill every other plant that may be present in or near the treated cogongrass stands (Miller and Enloe 2009; McClure and Johnson 2010). This

‘scorched earth’ program undoubtedly also impacts related soil fauna and habitat suitability for wildlife. These localized impacts of eradication and collateral damage are tactically deemed acceptable compared to the wide-scale, more intense habitat destruction and native species overtake that has been realized in unmanaged scenarios.

Not only should the system in consideration for management be assessed in a management decision, but all the neighboring systems that may be at an increased risk of infestation should be included in the assessment. All of these examples included the foresight to understand the devastation that could occur in the infested site as well as the likely spread that would ensue without management. The eradication management decision, despite known non-target risks, was deemed overwhelmingly acceptable in these scenarios. The management decision needs to incorporate multi-dimensional components in space and time

241

in order to weigh short-term risks in localized areas with long-term risks to the interconnected ecological system.

DECISION ASSESSMENT

In terms of freshwater algae and cyanobacteria, the ramifications of infestations have not been fully understood. Though a large body of research was summarized that outlines alarming concerns associated with infestations, more research is needed on long-term effects.

Only a fraction of the issues related to cyanobacteria likely been uncovered due to the complex nature of (many) toxins, exposure routes, and chronic (decadal) accumulation potential to elicit impacts as well as interactive effects with other compounds (Banack et al.

2015). This research sought to similarly apply risk-based approaches toward the common, noxious algal blooms rampantly afflicting freshwater resources. As discussed, cyanobacterial infestations are increasing in freshwater and obviate every critical use including economic ramifications and habitat and native species decimation. Beyond caulerpa invasions, these also provide a direct threat to human health regardless of contact with the water. The argument posed in this research strives to reassess how management is viewed regarding noxious algal invasions in freshwaters. Due to the similar and far advanced devastation cyanobacterial blooms can cause, aggressive management with a goal of eradication should also likely be sought after, in spite of potential known collateral damages. Since collateral management risks were assessed and deemed less important in other invasions, why not also dismiss collateral risks as less important with this more nuisance and widespread invader

(i.e., cyanobacterial blooms)?

242

Management of nuisance algae and cyanobacteria can be ongoing and intense, and should be readily justifiable as well as ecologically and economically beneficial. Only considering the potential risks of management fails to accurately assess overall risks to the system. Risks of management approaches may be exaggerated and not realized in typical field applications (Chapter 5). Therefore, this work proposes a new paradigm shift wherein risks from noxious algae are exposed (i.e. communicated to stakeholders), and accurate risks of management disclosed, in order to ensure water resource managers have appropriate information needed to make informed decisions.

Copper has been banned in many water resources for use in algal control (Netherland et al. 2005). Dismissed as a toxic heavy metal that accumulates and is toxic to fish and invertebrates, often this algaecide is not considered in algal management programs. This research confirms that some copper formulations can be very effective for controlling nuisance algae (Chapter 4), and also showed that typical algaecide applications may not result in the high-level of non-target risks that has been described (Chapter 5). Copper is a micronutrient allowed in drinking water up to 1.3 mg/L (USEPA 2007). However, the

USEPA published health advisories for cyanotoxins of 0.3 µg/L for microcystins and 0.7

µg/L for cylindrospermopsin in 10-day drinking water exposures for children less than six years old (USEPA 2015a, b). These toxins should also be marked as toxic chemicals and regulated more strictly, especially as many are indirectly spurred by anthropogenic activities.

If other mitigation techniques are not feasible, in a risk-based analysis, the decision should be relatively simple to use copper if it can offset cyanobacteria and associated toxins, especially if human health risk designations are a use of the water resource. Parallel examples of the

243

caulerpa invasion management would suggest, in addition, that the potential collateral environmental risks of copper (not as extreme as often cited) would be much lower than from complete chlorination, and also lower than the no action environmental and human health ramifications of cyanotoxins.

This research does not endorse copper use but, rather, assesses accurate risks associated with copper use. In many sites, cyanobacterial blooms are unabated, and resource managers often cite potential collateral impacts of possible management strategies. The risks associated with algaecide or copper use should be weighed with the impending and increasing impacts of a harmful cyanobacterial outbreak. The caulerpa management approach weighed intense management versus the complete devastation of habitat and native species in an unchecked invasion. There are many examples of these and other issues that have been documented following cyanobacterial outbreaks. Concomitant management that seeks the same eradication endpoint should be highly considered even if it causes similar non-target devastation as chlorine use. Advanced monitoring programs have revealed the intensity and increased devastation cyanobacterial blooms can cause (Bullerjahn et al. 2016); though relatively few examples exist of large-scale management success. With the immense value of freshwaters to human activities, management should be readily considered and implemented.

Weighing non-target risks is critical and applied programs should strive to minimize, although many management approaches have gotten unfairly assessed and documented impacts pale to that of chlorination. Assessing risks of no management is perhaps of greater importance, since as the degree of infestation intensifies, water resource uses become deteriorated further. Advanced infestations may require increased need for management and

244

cost to manage if not addressed early. Also, number and effectiveness of management options decreases as blooms further intensify. Therefore, the risks to the water body and potential long-term remediation costs need to be considered. Short- and long-term impacts of both the infestation as well as the remediation should be considered. Shift in food web structure, widespread anoxia under mats, and toxins may all be factors of unchecked cyanobacterial and other nuisance blooms that destroy native species. With this understanding, most legal approaches to management (including copper) seem to not possess the same risks. Algal blooms are diverse and dynamic and will no doubt take a multidimensional approach to manage. Alignment of management objectives with assessments of efficacy and risks of proactive and reactive management is needed to resolve this growing problem.

SUMMARY

By understanding risks associated with action and no action options, water resource managers can make informed decisions regarding the need for, and the comparative risks of, management. With the increasing acknowledgement and devastation harmful algal blooms evoke on freshwater resources, management implementation will undoubtedly increase. This dissertation seeks to promote scientifically defensible and informed decisions regarding direction of the management decision. The ramifications of allowing nuisance and noxious algae to persist in freshwaters will increase concomitant with anthropogenic demands.

Actualization of economic and ecological risks associated with infestations will increase pressure to manage. Selection of a management program is intricate and diverse, though this

245

work presents an initial framework to assist in elucidating efficacy, costs, and collateral risks of management in context with the decision not to manage.

246

REFERENCES

Anas, ARJ, Nakajima A, Naruse C, Tone M, Asukabe H, Harada K-i. 2016. Determination of FVIIa-sTF inhibitors in toxic Microcystis cyanobacteria by LC-MS Technique. Marine Drugs 14: 7.

Anderson, DM, Glibert PM, Burkholder JM. 2002. Harmful algal blooms and eutrophication: Nutrient sources, composition and consequences. Estuaries 25: 562-584.

Anderson, DM. 2004. Prevention, control, and mitigation of harmful algal blooms: multiple approaches to HAB management. Harmful Algae Management and Mitigation 123- 130.

Babica, P, Blaha L, Marsalek B. 2006. Exploring the Natural Role of Microcystins-A Review of Effects on Photoautotrophic Organisms. Journal of Phycology 42: 9-20.

Bachmann, RW, Hoyer MV, Fernandez C, Canfield Jr DE. 2003. An alternative to proposed phosphorus TMDLs for the management of Lake Okeechobee. Lake and Reservoir Management 19: 251-264.

Backer, LC, Carmichael W, Kirkpatrick B, Williams C, Irvin M, Zhou Y, et al. 2008. Recreational exposure to low concentrations of microcystins during an algal bloom in a small lake. Marine Drugs 6: 389-406.

Backer, LC, McNeel SV, Barber T, Kirkpatrick B, Williams C, Irvin M, et al. 2010. Recreational exposure to microcystins during algal blooms in two California lakes. Toxicon 55: 909-921.

Banack, SA, Caller T, Henegan P, Haney J, Murby A, Metcalf JS, Powell J, Cox PA, Stommel E. 2015. Detection of cyanotoxins, β-N-methylamino-L-alanine and microcystins, from a lake surrounded by cases of amyotrophic lateral sclerosis. Toxins (Basel) 7: 322-36.

Bell, SG, Codd GA. 1994. Cyanobacterial toxins and human health. Reviews in Medical Microbiology 5: 217-277.

Bergstrom, JC, Teasley RJ, Cordell HK, Souter R, English DBK. 1996. Effects of reservoir aquatic plant management on recreational expenditures and regional economic activity. Journal of Agriculture and Applied Economic 28: 409-422.

Besser, JM, Brumbaugh WG, May TW, Ingersoll CG. 2003. Effects of organic amendments on the toxicity and bioavailability of cadium and copper in spiked formulated sediments. Environmental Toxicology and Chemistry 22: 805-815.

247

Beversdorf, LJ, Miller TR, McMahon KD. 2013. The role of nitrogen fixation in cyanobacterial bloom toxicity in a temperate, eutrophic lake. PLoS ONE 8: e56103.

Bhadha, JH, Lang TA, Alvarez OM, Giurcanu MC, Johnson JV, Odero DC, Daroub SH. 2014. Allelopathic effects of Pistia stratiotes (Araceae) and Lyngbya wollei Farlow ex Gomont (Oscillariaceae) on seed germination and root growth. Sustainable Agriculture Research 3: 121-130.

Bradley, WG, Mash DC. 2009. Beyond Guam: the cyanobacteria/ BMAA hypothesis of the cause of ALS and other neurodegenerative diseases. Amyotrophic Lateral Sclerosis 10: 7-20.

Bullerjahn, GS, McKay RM, Davis TW, Baker DB, Boyer GL, D’Anglada LV, Doucette GJ, Ho, JC, Irwin EG, Kling CL, Kudela RM. 2016. Global solutions to regional problems: Collecting global expertise to address the problem of harmful cyanobacterial blooms. A Lake Erie case study. Harmful Algae 54: 223-238.

Burkholder, JM, Glibert PM. 2013. Eutrophication and oligotrophication, pp. 347-371. In: Encyclopedia of Biodiversity, 2nd edition, Volume 3, by Levin S.A. (ed.). Academic Press, Boston, MA.

Burkholder JM, Frazier W, Rothenberger MB. 2010. Source Water Assessment and Control/Treatment Strategies for Harmful and Noxious Algae. In Manual M57, Algae: Source to treatment. Denver, CO.

Burkholder, JM. 2009. Harmful algal blooms. In: Likens GE (ed.) Encyclopedia of Inland Waters, vol. 1, pp. 264-285. Elsevier, Oxford, United Kingdom.

Burkholder, JM, Glibert PM, Skelton HM. 2008. Mixotrophy, a major mode of nutrition for harmful algal species in eutrophic waters. Harmful Algae 8: 77-93.

Burton, GA. 1991. Assessing the toxicity of freshwater sediments. Environmental Toxicology and Chemistry 10: 1585-1627.

Button, KS, Hostetter HP, Mair DM. 1977. Copper dispersal in a water-supply reservoir. Water Research 11: 539-544.

Byappanahalli, MN, Whitman RL. 2009. Clostridium botulinum type E occurs and grows in the alga Cladophora glomerata. Canadian Journal of Fisheries and Aquatic Sciences 66: 879-882.

Carpenter, SR, Bolgrien D, Lathrop RC, Stow CA, Reed T, Wilson MA. 1998. Ecological and economic analysis of lake eutrophication by nonpoint pollution. Australian Journal of Ecology 23: 68-79.

248

California Department of Food and Agriculture (CDFA). 2003. The California Department of Food and Agriculture Hydrilla Eradication Program. Annual Progress Report.

Carey, CC, Weathers KC, Cottingham KL. 2008. Gloeotrichia echinulata blooms in an oligotrophic lake: Helpful insights from eutrophic lakes. Journal of Plankton Research 30: 893-904.

Carlton, JT. 2001. Introduced species in U.S. coastal waters: Environmental impacts and management priorities. Pew Oceans Commission, Arlington VA.

Carmichael, WW. 2001. Health effects of toxin-producing cyanobacteria: The CyanoHABs. Human and Ecological Risk Assessment 7: 1393-1407.

Cetiner, ZS, Wood SA, Gammons CH. 2005. The aqueous geochemistry of the rare earth elements. Part XIV. The solubility of rare earth element phosphates from 23 to 150 °C. Chemical Geology 217: 147-169.

Chen, M, Ye T-R, Krumholz LR, Jiang H-L. 2014. Temperature and cyanobacterial bloom biomass influence phosphorous cycling in eutrophic lake sediments. PLoS ONE 9: e93130.

Chorus, I, Bartram J. 1999. Toxic cyanobacteria in water: A guide to public health significance, monitoring and management. WHO. Chapman & Hall, London, United Kingdom.

Chun, CL, Ochsner U, Byappanahalli MN, Whitman RL, Tepp WH, Lin G, Johnson EA, Peller J, Sadowsky MJ. 2013. Association of toxin-producing Clostridium botulinum with the macroalga Cladophora in the Great Lakes. Environmental Science and Technology 47: 2587-2594.

Clearwater, SJ, Hickey CW. 2004. Ecotoxicity testing of Phoslock on sediment-dwelling aquatic biota and rainbow trout. NIWA (National Institute of Water & Atmospheric Research). Prepared for Ecowise Environmental Pty Ltd. NIWA Client Report No. AUS2004-004, June 2004.

Closson, KR, Paul EA. 2014. Comparison of the toxicity of two chelated copper algaecides and copper sulfate to non-target fish. Bulletin of Environmental Contamination and Toxicology 93: 660-665.

249

Codd, GA, Morrison LF, Metcalf JS. 2005. Cyanobacterial toxins: Risk management for health protection. Toxicology and Applied Pharmacology 203: 264-272.

Codd, GA, Ward CJ, Bell SG. 1997. Cyanobacterial toxins: Occurrence, modes of action, health effects and exposure routes. In Applied Toxicology: Approaches Through Basic Science (pp. 399-410). Springer, Berlin Heidelberg, Germany.

Dithmer, L, Lipton AS, Reitzel K, Warner TE, Lundberg D, Nielsen UG. 2015. Characterization of phosphate sequestration by a lanthanum modified bentonite clay: A solid state NMR, EXAFS and PXRD study. Environmental Science and Technology 49: 4559-4566.

Di Toro, DM, Mahony JJ, Hansen DJ, Scott KJ, Hicks MB, Mayr SM, Redmend MS. 1990. Toxicity of cadmium in sediments: The role of acid volatile sulfide. Environmental Toxicology and Chemistry 9: 1487-1502.

Dodds, WK, Bouska WW, Eitzmann JL, Pilger TJ, Pitts KL, et al. (2009) Eutrophication of US freshwaters: analysis of potential economic damages. Environmental Science and Technology 43: 12-19.

Douglass, GB. 2002. Remediation material and remediation process for sediments. United States Patent number 6350383.

Duarte, CM, Conley DJ, Carstensen J, and Sanchez-Camacho M. 2008. Return to Neverland: Shifting baselines affect eutrophication restoration targets. Estuaries and Coasts 32: 29-36.

Dyck, LA. 1994. Creation of management strategies that are compatible with the autecology of Lyngbya. Lake and Reservoir Management 9: 71.

Dziallas, C, Grossart H. 2011. Increasing oxygen radicals and water temperature select for toxic Microcystis sp. PLoS One 6: 25569

ECOTOX. 2008. Toxicity assessment of Granulated Phoslock to the Cladoceran Ceriodaphnia dubia. Report prepared for Phoslock Water Solutions Ltd. ECOTOX Services Australasia. July 2008.

250

Edwards, DJ, Marquez BL, Nogle LM, McPhail K, Goeger DE, Roberts MA, Gerwick WH. 2004 Structure and biosynthesis of the jamaicamides, new mixed polyketide-peptide neurotoxins from the marine cyanobacterium Lyngbya majuscula. Chemistry and Biology 11: 817-833.

Edwards, C, Graham D, Fowler N, Lawton LA. 2008. Biodegradation of microcystins and nodularin in freshwaters. Chemosphere 73: 1315-1321.

Falconer, IR. 1999. An overview of problems caused by toxic blue-green algae (cyanobacteria) in drinking and recreational water. Environmental Toxicology 14: 5- 12.

Flemming, CA, Trevors JT. 1989. Copper toxicity and chemistry in the environment: A review. Water, Air, and Soil Pollution 44: 143-158.

Freedman, TB, Loehr JS, Loehr TM. 1976. A resonance Raman study of the copper protein, hemocyanin. New evidence for the structure of the oxygen-binding site. Journal of the American Chemical Society 98: 2809-2815.

Gallagher, J, Duke B, Rodgers Jr JH. 2005. Responses of Hyalella azteca and Ceriodaphnia dubia to reservoir sediments following chelated copper herbicide applications. Journal of Aquatic Plant Management 43: 95-99.

Ganf, GG, Oliver RL. 1982. Vertical separation of the light and available nutrients as a factor of causing replacement of Green algae by Blue-green algae in the plankton of a stratified lake. Journal of Ecology 70: 829-844.

Getsinger, KD, Dibble ED, Rodgers Jr JH, Spencer DF. 2014. Benefits of Controlling Nuisance Aquatic Plants and Algae in the United States. CAST Commentary QTA2014-1, Council on Agricultural Science and Technology, Ames, IA.

Gobler, CJ, Burkholder JM, Davis TW, Harke MJ, Johengen T, Stow CA, Van de Waal DB. 2016. The dual role of nitrogen supply in controlling the growth and toxicity of cyanobacterial blooms. Harmful Algae 54: 87-97.

Grach-Progrebinsky, O, Carmeli S. 2008. Three novel anabaenopeptins from the cyanobacterium Anabaena sp. Tetrahedron 64: 10233-10238.

251

Han, FX, Hargreaves JA, Knigery WL, Huggett DB, Schlenk DK. 2001. Accumulation, distribution, and toxicity of copper in sediments of catfish ponds receiving periodic copper sulfate applications. Journal of Environmental Quality 30: 912-919.

Hallegraeff, GM. 1993. A review of harmful algal blooms and their apparent global increase. Phycologia 32: 79-99.

Hansen, JA, Rose JD, Jenkins RA, Gerow KG, Bergman HL. 1999. Chinook salmon (Oncorhynchus tshawytscha) and rainbow trout (Oncorhynchus mykiss) exposed to copper: neurophysiological and histological effects on the olfactory system. Environmental Toxicology and Chemistry 18: 1979-1991.

Hanson, TR. 2006b. Economic impact of off-flavor to the U.S. catfish industry. In: Schrader, K. and Rimando, A. (eds) Off-flavors in Aquaculture. American Chemical Society, Washington, DC.

Heisler, J, Glibert PM, Burkholder JM, Anderson DA, Cochlan WP, Dennison WC, Dortch Q, Gobler C, Heil CA, Humphries E. 2008. Eutrophication and harmful algal blooms: a scientific consensus. Harmful Algae 8: 3-13.

Hernandez, JM, Lopez-Rodas V, Costas E. 2009. Microcystins from tap water could be a risk factor for liver and colorectal cancer: A risk intensified by global change. Medical Hypotheses 72: 539-540.

Huggett, DB, Gillespie WB, Rodgers Jr JH. 1999. Copper Bioavailability in Steilacoom Lake Sediments. Archives of Environmental Contamination and Toxicology 36: 120-123.

Ishida, K, Welker M, Christiansen G, Cadel-Six S, Bouchier C, Dittman E, Hertweck C, Tandeau de Marsac N. 2009. Plasticity and evolution of aeruginosin biosynthesis in cyanobacteria. Applied and Environmental Microbiology 75: 2017-2026.

Iwinski, KJ, Calomeni AJ, Geer TD, Rodgers JH. 2016. Cellular and aqueous microcystin- LR following laboratory exposures of Microcystis aeruginosa to copper algaecides. Chemosphere 147: 74-81.

Iwinski, KJ, McQueen AD, Kinley CM, Calomeni AJ, Geer TD, Rodgers Jr JH. 2016. Sediment Copper Concentrations, in Situ Benthic Invertebrate Abundance, and Sediment Toxicity: Comparison f Treated and Untreated coves in a Southern Reservoir. Water, Air, and Soil Pollution 227: 85.

252

Jonasson, RG, Bancroft GM, Boatner LA. 1988. Surface reactions of synthetic, 933 end- member analogues of monazite, xenotime and rhabdophane, and evolution of natural waters. Geochmica et Cosmochimica Acta 52: 767-770.

Jones, GJ, Orr PT. 1994. Release and degradation of microcystin following algicide treatment of a Microcystis aeruginosa bloom in a recreational lake, as determined by HPLC and protein phosphatase inhibition assay. Water Research 28: 871-876.

Jones, R, Hassan S, Rodgers JH. 2008. Influence of contact duration on sediment associated copper fractionation and bioavailability. Ecotoxicology and Environmental Safety 71: 104-116.

Kasinak, J-M E, Bishop CJ, Wright RA, Wilson AE. 2015. Grass carp do not consume the nuisance benthic cyanobacterium, Lyngbya wollei. Journal of Aquatic Plant Management 53: 74-80.

Kenefick, SL, Hrudey SE, Peterson HG, Prepas EE. 1993. Toxin release from Microcystis aeruginosa after chemical treatment. Water Science and Technology 27: 433-440.

Lahti, K, Rapala J, Färdig M, Niemelä M, Sivonen K. 1997. Persistence of cyanobacterial hepatotoxin, microcystin-LR in particulate material and dissolved in lake water. Water Research 31: 1005-1012.

Lampert, W. 1981. Inhibitory and toxic effects of bluegreen algae on Daphnia. International Review of Hydrobiology 66: 285-298.

Lampert, W. 1987. Laboratory studies on zooplankton-cyanobacteria interactions. New Zealand Journal of Marine and Freshwater Research 21: 483-490.

Landman, M, Brijs J, Glover C, Ling N. 2007. Lake Okareka and Tikitapu Fish Health Monitoring, 2007. Scion Report. October 2007.

Langeland, KA. 1996. Hydrilla verticillata (L.F.) Royle (Hydrocharitaceae), “The Perfect Aquatic Weed.” Castanea 61: 296-304.

253

Legrand, C, Rengefors K, Fistarol GO, Graneli E. 2003. Allelopathy in phytoplankton- biochemical, ecological and evolutionary aspects. Phycologia 42: 406-19.

Lehman, PW, Marr K, Boyer GL, Acuna S, Teh SJ. 2013. Long-term trends and causal factors associated with Microcystis abundance and toxicity in San Francisco estuary and implications for climate change impacts. Hydrobiologia 718: 141-158.

Liu, R, Zhao D, Barnett M. 2006. Fate and transport of copper applied in channel catfish ponds. Water, Air, and Soil Pollution 176: 139-162.

Lun, Z, Hai Y, Kun C. 2002. Relationship between microcystin in drinking water and colorectal cancer. Biomedical and Environmental Sciences 15: 166-171.

Lurling, M, Tolman Y. 2010. Effects of lanthanum and lanthanum-modified clay on growth, survival and reproduction of Daphnia magna. Water Research 44: 309-319.

Mallin, MA, McIver MR, Wells HA, Parsons DC, Johnson VL. 2005. Reversal of eutrophication following sewage treatment upgrades in the New River Estuary, North Carolina. Estuaries 28: 750-760.

Mastin, BJ, Rodgers Jr JH, Deardorff TL. 2002. Risk evaluation of cyanobacteria-dominated algal blooms in a North Louisiana reservoir. Journal of Aquatic Ecosystem Stress and Recovery 9: 103-114.

McClure, M, Johnson J. 2010. Cogongrass Eradication Strategies. Georgia Forestry Commission, Atlanta, GA

Meinesz, A. 1999. Killer Algae. The University of Chicago Press. 360pp. ISBN 0-226- 51922-8

Milani, D, Reynoldson TB, Borgmann U, Kolasa JK. 2003. The relative sensitivity of four benthic invertebrates to metal spiked-sediment exposures and application to contaminated field sediment. Environmental Toxicology and Chemistry 22: 845-854.

Miller Jr, GT. 2002. Living in the Environment: Principles, connections, and solutions 12th Edition. Wadsworth Publishing. Belmont, CA.

Miller, J, Enloe SF. 2009. Alabama Forestry Commission’s cogongrass control recommendations. Auburn. AL.

Miller, MA, Kudela RP, Mekebri A, Crane D, Oates SC, Tinker MT, Staedler M, Miller WA, Toy-Choutka S, Dominik C, Hardin D, Langlois G, Murray M, Ward K, Jessup DA. 2010. Evidence for a novel marine harmful algal bloom: Cyanotoxin (microcystin) transfer from land to sea otters. PLoS ONE 5: e12576.

254

Mitchell, AJ, Kelly AM. 2006. The public sector role in the establishment of grass carp in the United States. Fisheries 31: 113-121.

Mitra, A, Flynn KJ. 2006. Promotion of harmful algal blooms by zooplankton predatory activity. Biology Letters 2: 194-197.

Mitra, A, Flynn KJ. 2006a. Accounting for variation in prey selectivity by zooplankton. Ecological Modelling 199: 82-92.

Namikoshi, M, Choi BW, Sakai R, Sun F, Rinehart KL, Carmichael WW, Evans WR., Cruz P, Munro MHG, Blunt JW. 1994. New nodularins, A general method for structure assignment. Journal of Organic Chemistry 59: 2349-2357.

Namikoshi, M, Sun F, Choi BW, Rinehart KL, Carmichael WW, Evans WR, Beasley VR. 1995 Seven more microcystins from Homer lake cells: application of the general method for structure assignment of peptides containing,-dehydroamino acid unit(s). Journal of Organic Chemistry 60: 3671-3679.

Netherland, MD, Getsinger KD, Stubbs DR. 2005. Aquatic plant management: Invasive species and chemical control. Outlooks on Pest Management 16: 100-104.

Onodera, H, Satake M, Oshima Y, Yasumoto T, Carmichael WW. 1997b New saxitoxin analogues from the freshwater filamentous cyanobacterium Lyngbya wollei. Natural Toxins 5: 146-51.

Orihel, DM, Schindler DW, Ballard NC, Graham MD, O’Connell DW. 2015. The “nutrient pump:” Iron-poor sediments fuel low nitrogen-tophosphorus ratios and cyanobacterial blooms in polymictic lakes. Limnology and Oceanography 60: 856-871.

Otten, TG, Paerl HW. 2015. Health effects of toxic cyanobacteria in US drinking and recreational waters: Our current understanding and proposed direction. Current Environmental Health Reports 2: 75-84.

Paerl, HW, Huisman J. 2008. Blooms like it hot. Science 320: 57-58.

Paerl, HW, Huisman J. 2009. Climate change: a catalyst for global expansion of harmful cyanobacterial blooms. Environmental Microbiology Reports 1: 27-37.

Paerl, HW, Scott JT. 2010. Throwing fuel on the fire: synergistic effects of excessive nitrogen inputs and global warming on harmful algal blooms. Environmental Science and Technology 44: 7756-8.

Paerl, HS, Paul VJ. 2012. Climate change: Links to global expansion of harmful cyanobacteria. Water Research 46: 1349-1363.

255

Paerl, HW, Otten TG. 2013. Harmful cyanobacterial blooms: Causes, consequences, and controls. Microbial Ecology 65: 995-1010.

Paerl, HW. 2014. Mitigating harmful cyanobacterial blooms in a human- and climatically- impacted world. Life 4: 988-1012.

Pavagadhi, S, Natera S, Roessner U, Balasubramanian R. 2013. Insights into lipidomic perturbations in zebrafish tissues upon exposure to microcystin-LR and microcystin- RR. Environmental Science and Technology 47: 14376-14384.

Qiao, Q, Liu WJ, Wu K, Song T, Hu J, Huang X, Wen J, Chen L, Zhang XZ. 2013. Female zebrafish (Danio rerio) are more vulnerable than males to microcystin-LR exposure, without exhibiting estrogenic effects. Aquatic Toxicology 142: 272-282.

Reed, RE, Burkholder JM, Allen EH. 2010. Recent Developments in Online Monitoring. Technology for Surveillance of Algal Blooms, Potential Toxicity, and Physical- Chemical Structure in Rivers, Reservoirs, and Lakes. In Manual M57, Algae: Source to Treatment. Denver, CO, AWWA Chapter 1.

Reinertsen, H, Jensen A, Langeland A, Olsen Y. 1986. Algal competition for phosphorus; the influence of zooplankton and fish. Canadian Journal of Fisheries and Aquatic Sciences 43: 1135-1141.

Reinikainen, M., Kiviranta J, Ulvi V, Niku-Paavola ML. 1995. Acute toxic effects of a novel cyanobacterial toxin on the crustaceans Artemia salina and Daphnia pulex. Archives of Hydrobiology 133: 61-69.

Reitzel, K, Andersen FO, Egemose S, Jensen HS. 2013. Phosphate adsorption by lanthanum modified bentonite clay in fresh and brackish water. Water Research 47: 2787-2796.

Reitzel, K, Lotter S, Dubke M, Egemose S, Jensen HS, Andersen FØ. 2013. Effects of Phoslock treatment and chironomids on the exchange of nutrients between sediment and water. Hydrobiologia 703: 189-202.

Robarts, RD, Zohary T. 1986. Influence of cyanobacterial hyperscum on heterotrophic activity of planktonic bacteria in a hypertrophic lake. Applied and Environmental Microbiology 51: 609-613.

256

Rogers, ED, Henry TB, Twiner MJ, Gouffon JS, McPherson JT, Boyer GL, Sayler GS, Wilhelm SW. 2011. Global gene expression profiling in larval zebrafish exposed to microcystin-lr and Microcystis reveals endocrine disrupting effects of cyanobacteria. Environmental Science and Technology 45: 1962-1969.

Sano, T, Usui T, Ueda K, Osada H, Kaya K. 2001. Isolation of new protein phosphatase inhibitors from two cyanobacterial species, Planktothrix spp. Journal of Natural Proucts 64: 1052-1055.

Schaus, MH and Vanni VJ. 2000. Effects of gizzard shad on phytoplankton and nutrient dynamics: role of sediment feeding and fish size. Ecology 81: 1701-1719.

Scheffer, M, Carpenter S, Foley JA, Folke C, Walker B. 2001. Catastrophic shifts in ecosystems. Nature 413: 591-596.

Schindler, DW. 1987. Detecting ecosystem responses to anthropogenic stress. Canadian Journal of Fisheries and Aquatic Sciences 44: 6-25.

Schindler, DW, Vallentyne JR. 2008. The algal bowl: Overfertilization of the world's freshwaters and estuaries. University of Alberta Press, Edmonton, Alberta.

Schopf, JW. 2000a. The fossil record: Tracing the roots of the cyanobacterial lineage, in B. A. Whitton and M. Potts, eds., The ecology of cyanobacteria. Kluwer Academic Publishers, Dordrecht, The Netherlands.

Schumacher, M, Wilson N, Tabudravu JN, Edwards C, Lawton LA, Motti C, Wright AD, Diederich M, Jaspars M. 2012. New nodulopeptins from Nodularia spumigena KAC 66. Tetrahedron 68: 1622-1628.

Scott, WE. 1991. Occurrence and significance of toxic cyanobacteria in southern Africa. Water Science and Technology 23: 175-180.

Skulberg, OM, Carmichael WW, Anderson RA, Matsunaga S, Moore RE, Skulberg R. 1992 Investigations of a neurotoxic Oscillatorialean strain (cyanophyceae) and its toxin. Isolation and characterization of homoanatoxin-a. Environmental Toxicology and Chemistry 11: 321-329.

Skulberg, OM, Underdal B, Utkilen H. 1994. Toxic waterblooms with cyanophytes in Norway - current knowledge. Algological Studies 75: 279-289.

Smith, VH, Dodds WK, Havens KE, Engsgrom DR, Paerl HW, Moss B, Likens GE. 2014. Comment: Cultural eutrophication of natural lakes in the United States is real and widespread. Limnology and Oceanography 59: 2217-2225.

257

Solomon, S, Plattner G-K, Knutti R, Friedlingstein P. 2009. Irreversible climate change due to carbon dioxide emissions. PNAS 106: 170-1709.

Southard, GM, Fries LT, Barkoh A. 2010. Prymnesium parvum: The Texas experience. Journal of the American Water Resources Association 46: 14-23.

Stommel, EW, Field NC, Caller TA. 2013. Aerosolization of cyanobacteria as a risk factor for Amyotrophic Lateral Sclerosis. Medical Hypotheses 80: 142-145.

Taranu, ZE, Eaves IG, Leavitt PR, Bunting L, Buchaca T, Catalan J, Domaizon I, Guilizzoni P, Lami A, McGowan S, Moorhouse H, Morabito G, Pick FR, Stevenson MA, Thompson PL, Vinebrooke RD. 2015. Acceleration of cyanobacterial dominance in north temperate subarctic lakes during the Anthropocene. Ecology Letters 2015: 1-10.

Tapiero, H, Townsend DM, Tew KD. 2003. Trace elements in human physiology and pathology. Copper. Biomedicine and Pharmacotherapy 57: 386-398.

Trinchet, I, Djediat C, Huet H, Dao SP, Edery M. 2011. Pathological modifications following sub chronic exposure of medaka fish (Oryzias latipes) to microcystin-LR. Reproductive Toxicology 32: 329-340.

Thunberg, EM, Pearson CN, Milon JW. 1992. Residential flood control benefits of aquatic plant control. Journal of Aquatic Plant Management 30: 66-70.

Turner, PC, Gammie AJ, Hollinrake K, Codd GA. 1990. Pneumonia associated with contact with cyanobacteria. British Medical Journal 300: 1440-1441.

Ueno, Y, Nagata S, Tsutsumi T, Hasegawa A, Watanabe MF, Park HD, Chen GC, Yu SH. 1996. Detection of microcystins, a blue-green algal hepatotoxin, in drinking water sampled in Haimen and Fusui, endemic areas of primary liver cancer in China, by highly sensitive immunoassay. Carcinogenesis 17: 1317-1321.

United States Environmental Protection Agency (USEPA). 1980. Ambient water quality criteria for copper. Report EPA 440/5-80-036. Washington, DC.

United States Environmental Protection Agency (USEPA). 1984. Ambient water quality criteria for copper. 440/5/84/031 EPA 40 CFR Parts 141 and 142. Washington, DC.

United States Environmental Protection Agency (USEPA). 2007. National primary drinking water regulations for lead and copper: short-term regulatory revisions and clarifications, 40 CFR Parts 141 and 142. Washington, DC.

258

United States Environmental Protection Agency (USEPA). 2009. Reregistration Eligibility Decision for Coppers. Report EPA 738-R-09-304. Washington, DC.

United States Environmental Protection Agency (USEPA). 2012. Cyanobacteria and Cyanotoxins: Information for Drinking Water Systems. Office of Water. EPA- 810F11001. Washington, DC.

United States Environmental Protection Agency (USEPA). 2015a. Drinking water health advisory for the cyanobacterial toxin microcystin. Report EPA 820R15100. Washington, DC.

United States Environmental Protection Agency (USEPA). 2015b. Drinking water health advisory for the cyanobacterial toxin cylindrospermopsin. Report EPA 820R15101. Washington, DC.

Vanderploeg, HA, Liebig JR, Carmichael WW, Agy MA, Johengen TH, Fahnenstiel GL, Nalepa TF. 2001. Zebra mussel (Dreissena polymorpha) selective filtration promoted toxic Microcystis blooms in Saginaw Bay (Lake Huron) and Lake Erie. Canadian Journal of Fisheries and Aquatic Sciences 58: 1208-1221.

Vijayavel, K, Sadowsky MJ, Ferguson JA, Kashian DR. 2013. The establishment of the nuisance cyanobacteria Lyngbya wollei in Lake St. Clair and its potential to harbor fecal indicator bacteria. Journal of Great Lakes Research 39: 560-568.

Wagner, KJ, Oglesby RT. 1984. Incompatibility of common lake management objectives. Lake and Reservoir Management 1: 97-100.

Wang, X, Qin B, Gaoand G, Paerl HW. 2010. Nutrient enrichment and selective predation by zooplankton promote Microcystis (Cyanobacteria) bloom formation. Journal of Plankton Research 32: 457-470.

Watanabe, MM, Kaya K, Takamura N. 1992 Fate of the toxic cyclic heptapeptides, the microcystins, from blooms of Microcystis (cyanobacteria) in a hypertrophic lake. J. Phycol. 28: 761-767.

Watson-Leung, T. 2009. Phoslock toxicity testing with three sediment dwelling organisms (Hyalella azteca, Hexagenia spp. and Chrinonomus dilutes) and two water dwelling organisms (Rainbow Trout and Daphnia magna). Aquatic Toxicology Unit, Ontario Ministry of the Environment, Ontario, Canada.

Webster, KE, Peters RH. 1978. Some size-dependent inhibitions of larger cladoceran filterers in filamentous suspensions. Limnology and Oceanography 23: 1238-1245.

Wetzel, RG. 2001. Limnology: Lake and River Ecosystems, 3rd ed. Academic Press, San Diego, CA.

259

White, SH, Duivenvoorden L, Fabbro LD. 2005. A decision-making framework for ecological impacts associated with the accumulation of cyanotoxins (cylindrospermopsin and microcystin). Lakes and Reservoirs: Research and Management 10: 25-37.

Willis, BE, Alley BL, Rodgers Jr JH. 2013. Bioavailability and analytical measurement of copper residuals in sediments. Water, Air, and Soil Pollution 224: 1423.

Woller-Skar, MM, Jones DN, Luttenton MR, Russell AL. 2015. Microcystin detected in little brown bats (Myotis lucifugus). The American Midland Naturalist 174: 331-334.

Wood, SA and DR Dietrich. 2011. Quantitative assessment of aerosolized cyanobacterial toxins at two New Zealand lakes. Journal of Environmental Monitoring 13: 1617- 1624.

World Health Organization (WHO). 1999. Toxic Cyanobacteria in Water: A guide to the public health consequences, monitoring and management. E&FN Spon, London, United Kingdom.

Yruela, I. 2005. Copper in plants. Brazilian Journal of Plant Physiology 17: 145-156.

Zhao, SJ, Xie P, Li GY, Chen J, Cai Y, Xiong Q, Zhao YY. 2012. The proteomic study on cellular responses of the testes of zebrafish (Danio rerio) exposed to microcystin-RR. Proteomics 12: 300-312.

Zi, J, Lantvit DD, Swanson SM, Orjala J. 2012. Lyngbyaureidamides A and B, two anabaenopeptins from the cultured freshwater cyanobacterium Lyngbya sp. (SAG 36.91). Phytochemistry 74: 173-177.

260

APPENDICES

261

Appendix A.

List of algal taxa in the Chockyotte irrigation pond and average density (cells/mL) or biomass (Chara: g dry weight/m2) from the three sampling points at each sampling event. Chockyotte Algae Sampling Period 1 day 2 M pre 1 M pre 2 WAT 1 MAT 2 MAT 4 MAT 6 MAT 7 MAT 8 MAT pre Group Genus 8.9.14 9.16.14 10.29.14 11.28.14 12.18.14 2.20.15 4.12.15 5.15.15 6.16.15 10.15.14 Streptophyta (Desmids) Chara Closterium 22 8 9 25 15 76 53 33 7 Genicularia 55 3 198 55 55 18 4 141 16 532 Cosmarium 4 8 9 6 9 Pleurotaenium Staurastrum 76 555 54 78 145 319 333 542 Peudstaurastrum Chlorophyta (Green algae) Coelastrum 67 412 32 444 139 4 16 235 4533 129 Elakatothrix 33 34 78 2,350 227 66 44 1,101 1239 352 Eudorina 55 56 Gloeocystis Helicodictyon 1,210 22,090 24,994 Oedogonium 18 4,110 156 Oocystis 954 856 912 2,352 4,555 3,400 3,190 844 543 123 Schroederia Scenedesmus 350 Selenastrum Sphaerocystis

262

Chockyotte Algae Sampling Period 1 day 2 M pre 1 M pre 2 WAT 1 MAT 2 MAT 4 MAT 6 MAT 7 MAT 8 MAT pre Group Genus 8.9.14 9.16.14 10.29.14 11.28.14 12.18.14 2.20.15 4.12.15 5.15.15 6.16.15 10.15.14

Bacillariophyta (Diatoms) Achnanthes 4 23 12 42 123 431 743 352 372 222 Amphipleura Aulacoseira Cymbella 9 4 123 35 4 55 4 Diatoma Navicula 45 22 9 2 23 75 125 68 33 Stauroneis Synedra 55 23 44 45 12 44 54 561 449 1,266 Cyanophyta (Blue-green algae) Anabaena 402 189 316 373 90 36 7 Aphanocapsa 1,614 436 3,202 788 54 30 3 132 23 22 Microcystis 61,745 59,543 45,117 Nodularia 18 Oscillatoria 29 59 89 11 6 15 6 Pseudanabaena 8,364 6,456 3,332 103 368 49 375 1,350 1,459 3,432 Planktolyngbya 433 472 45 18 80 114 1,222 3,422 6,990

263

Chockyotte Algae Sampling Period 1 day 2 M pre 1 M pre 2 WAT 1 MAT 2 MAT 4 MAT 6 MAT 7 MAT 8 MAT pre Group Genus 8.9.14 9.16.14 10.29.14 11.28.14 12.18.14 2.20.15 4.12.15 5.15.15 6.16.15 10.15.14 Euglenophyta (Euglenoids) Euglena Phacus Trachelmonas 25 34 77 410 765 685 847 5,542 4,411 2,119 Raphidophyta Gonyostomum 8 13 3 9 4 8 Cryptophyta Cryptomonas 56 107 1,381 2,260 524 444 234 540 201 3 Synurophyceae Synura 47 110 363 1,333 341 262 Mallamonas 7 4 24 60 103 262 202 Chrysophyceae Chrysococcus Dinobryon 6 300 270 206 605 1,072 259 40 Dinophyta Ceratium 19 144 18 15 41 145 217 333 Gymnodinium 4 11 5 6 3 1 3

264

Chockyotte Algae Sampling Period (continued) 13 10 11 16 18 19 20 21 22 23 9 MAT MAT MAT MAT MAT MAT MAT MAT MAT MAT MAT Group Genus 7.12.15 11.12.1 8.9.15 9.9.15 2.15.16 4.2.16 5.10.16 6.13.16 7.1.16 8.12.16 9.5.16 5 Streptophyta (Desmids) Chara 54 90 170 129 33 127 236 222 276 165 Closterium 8 26 12 9 37 Genicularia 333 391 33 673 22 54 78 55 18 Cosmarium Pleurotaenium Staurastrum 45 12 18 99 12 542 721 555 455 765 1,232 Peudstaurastrum Chlorophyta (Green algae) Coelastrum 44 76 34 133 33 198 544 2,343 124 745 1,200 Elakatothrix 733 173 7 12 78 522 123 785 89 55 1,187 Eudorina 2 12 46 Gloeocystis 1,200 2,090 2,880 140 234 1,199 150 Helicodictyon 13,040 9,010 31 67 98 222 9 32 22 Oedogonium Oocystis 333 46 55 9 6 Pediastrum 12 78 88 Schroederia Scenedesmus 720 1,550 12,100 13 129 345 8,800 2,309 4,500 1,812 540 Selenastrum Sphaerocystis 22 40 61 18 38

265

Chockyotte Algae Sampling Period (continued) 13 10 11 16 18 19 20 21 22 23 9 MAT MAT MAT MAT MAT MAT MAT MAT MAT MAT MAT Group Genus 7.12.15 11.12.1 8.9.15 9.9.15 2.15.16 4.2.16 5.10.16 6.13.16 7.1.16 8.12.16 9.5.16 5 Bacillariophyta (Diatoms) Achnanthes 174 44 63 5 4 231 85 12 33 236 445 Amphipleura Aulacoseira Cymbella 33 7 45 72 6 Diatoma 121 66 8 25 31 185 Navicula Stauroneis 180 31 9 Synedra 1,267 3,376 872 345 222 34 128 344 78 55 233 Cyanophyta (Blue-green algae) Anabaena Aphanocapsa 34 66 87 35 9 34 15 156 3,121 1,394 812 Microcystis Nodularia Oscillatoria 189 22 11 2 2 12 11 Pseudanabaena 2,344 991 1,500 18 15 19 222 240 1,800 2,213 1,127 Planktolyngbya 12,000 332 765 67 34 121 421 129 19

266

Chockyotte Algae Sampling Period (continued) 13 10 11 16 18 19 20 21 22 23 9 MAT MAT MAT MAT MAT MAT MAT MAT MAT MAT MAT Group Genus 7.12.15 11.12.1 8.9.15 9.9.15 2.15.16 4.2.16 5.10.16 6.13.16 7.1.16 8.12.16 9.5.16 5 Euglenophyta (Euglenoids) Euglena Phacus Trachelmonas 23 11 9 17 87 581 672 233 87 44 99 Raphidophyta Gonyostomum 9 45 8 55 9 177 187 93 18 33 9 Cryptophyta Cryptomonas 67 44 1,231 557 705 555 453 31 7 65 121 Synurophyceae Synura 428 24 17 62 194 770 1,074 844 Mallamonas 29 7 5 14 410 59 633 67 7 1 3 Chrysophyceae Chrysococcus Dinobryon 21 7 4 Dinophyta Ceratium 122 286 488 123 651 212 193 43 12 8 4 Gymnodinium 33 54 9 9 18 12

267

Appendix B.

List of algal taxa in the Hickory Meadows irrigation pond and average density (cells/mL) from the three sampling points at each sampling event.

Hickory Meadows Algae Sampling Period 1 day 2 M pre 1 M pre 2 WAT 1 MAT 2 MAT 4 MAT 6 MAT 7 MAT 8 MAT pre Group Genus 7.15.14 8.26.14 10.13.14 11.2.14 12.8.14 2.12.15 4.1.15 5.5.15 6.11.15 9.30.14

Streptophyta (Desmids) Closterium 342 510 132 9 70 32 18 Genicularia 17 32 47 426 30 6 8 32 230 123 Cosmarium 4 17 35 7 1 Pleurotaenium Staurastrum 7 7 33 125 99 Chlorophyta (Green algae) Ankistrodesmus Elakatothrix Eudorina 898 47 17 5 Coelastrum 30 Dictyosphaerium Oedogonium 30 18 4,200 156 Oocystis 105 293 225 8 34 Pediastrum 11 81 85 30 Scenedesmus 152 87 78 6 45 12 2 27 350 Tetraedron 13 24 66

268

Hickory Meadows Algae Sampling Period 1 day 2 M pre 1 M pre 2 WAT 1 MAT 2 MAT 4 MAT 6 MAT 7 MAT 8 MAT pre Group Genus 7.15.14 8.26.14 10.13.14 11.2.14 12.8.14 2.12.15 4.1.15 5.5.15 6.11.15 9.30.14

Bacillariophyta (Diatoms) Achnanthes 2 9 31 46 6 345 231 143 Amphipleura 7 4 8 21 Aulacoseira Cymbella 3 6 45 16 22 Diatoma 1 29 11 4 6 Gomphonema 2 Gyrosigma 1 7 Navicula 23 156 Stauroneis 1 1 1 12 21 33 121 Synedra 15 58 Cyanophyta (Blue-green algae) Anabaena 2,690 726 5,337 1,313 90 50 7 Aphanocapsa 400 40 180 433 132 523 430 Microcystis 29,113 5,158 16,377 aeruginosa Microcystis 24,600 18,988 9,800 308 ichthyoblabe Nodularia 6 Oscillatoria 2,162 2,482 1,067 16 11 6 18 6 Pseudanabaena 15 35 444 120

269

Hickory Meadows Algae Sampling Period 1 day 2 M pre 1 M pre 2 WAT 1 MAT 2 MAT 4 MAT 6 MAT 7 MAT 8 MAT pre Group Genus 7.15.14 8.26.14 10.13.14 11.2.14 12.8.14 2.12.15 4.1.15 5.5.15 6.11.15 9.30.14 Euglenophyta (Euglenoids) Euglena 310 76 196 26 62 142 180 45 11 Phacus 1 1 Trachelmonas 15 37 175 347 126 101 45 145 18 35 Raphidophyta Gonyostomum 19 36 460 753 175 8 3 180 67 0 Cryptophyta Cryptomonas 15 23 18 58 152 64 53 354 222 18 Synurophyceae Synura 1,200 920 123 Mallamonas 19 38 70 47 56 18 0 Chrysophyceae Chrysococcus 18 21 Dinobryon 33 202 24 Dinophyta (Dinoflagellates) Ceratium 1 5 2 3 Gymnodinium 17 18 3 3 4 Peridinium

270

Hickory Meadows Algae Sampling Period (continued) 13 10 11 16 18 19 20 21 22 23 9 MAT MAT MAT MAT MAT MAT MAT MAT MAT MAT MAT Group Genus 7.12.15 11.12.1 8.9.15 9.9.15 2.15.16 4.2.16 5.10.16 6.13.16 7.1.16 8.12.16 9.5.16 5 Streptophyta (Desmids) Closterium 23 114 26 12 9 37 Genicularia 897 1,200 7 44 111 236 54 78 55 18 Cosmarium Pleurotaenium 2 3 Staurastrum 325 1,311 19 100 119 429 765 231 111 682 549 Chlorophyta (Green algae) Ankistrodesmus 15 10 22 30 Elakatothrix 11 34 45 Eudorina 345 222 198 Coelastrum Dictyosphaerium 10 20 32 Oedogonium 65 220 1,200 39 18 Oocystis 333 46 55 9 34 465 10 55 89 Pediastrum 12 78 88 Scenedesmus 720 1,243 9,800 23 44 436 8,810 989 4,500 1,234 432 Tetraedron

271

Hickory Meadows Algae Sampling Period (continued) 13 10 11 16 18 19 20 21 22 23 9 MAT MAT MAT MAT MAT MAT MAT MAT MAT MAT MAT Group Genus 7.12.15 11.12.1 8.9.15 9.9.15 2.15.16 4.2.16 5.10.16 6.13.16 7.1.16 8.12.16 9.5.16 5 Bacillariophyta (Diatoms) Achnanthes 16 7 2 560 345 231 85 12 33 4 18 Amphipleura Aulacoseira 4 Cymbella 8 Diatoma 5 172 216 121 66 8 2 81 Gomphonema 8 4 Gyrosigma 15 Navicula 8 Stauroneis 45 4 44 46 180 36 9 Synedra 12 2 1 354 34 123 344 78 55 113 Cyanophyta (Blue-green algae) Anabaena Aphanocapsa 780 777 1,200 35 8 34 15 156 3,121 1,410 812 Microcystis

aeruginosa Microcystis

ichthyoblabe Nodularia Oscillatoria 184 22 Pseudanabaena 4,300 1,196 400 18 6 19 222 240 1,800 2,203 1,123

272

Hickory Meadows Algae Sampling Period (continued) 13 10 11 16 18 19 20 21 22 23 9 MAT MAT MAT MAT MAT MAT MAT MAT MAT MAT MAT Group Genus 7.12.15 11.12.1 8.9.15 9.9.15 2.15.16 4.2.16 5.10.16 6.13.16 7.1.16 8.12.16 9.5.16 5 Euglenophyta (Euglenoids) Euglena 123 9 13 222 454 Phacus 312 128 23 89 Trachelmonas 11 6 275 164 144 18 54 76 Raphidophyta Gonyostomum 34 143 235 177 187 93 9 8 Cryptophyta Cryptomonas 3 7 55 34 8 555 453 31 7 65 121 Synurophyceae Synura 23 88 122 892 777 8 4 33 Mallamonas 15 19 111 213 22 11 Chrysophyceae Chrysococcus 8 30 Dinobryon 46 91 5 8 Dinophyta Ceratium 71 18 98 27 1 8 19 8 Gymnodinium 7 34 20 103 12 3 1 3 5 Peridinium 2 8 5

273