Doctor of Philosophy - UNB … · Doctor of Philosophy In the Graduate Academic Unit of Biology Supervisor: Kelly Munkittrick, PhD, Dept. of Biology, UNBSJ Co-Supervisor: R. Allen

Evaluating the Suitability of Fish Species for Environmental Monitoring Programs.

by

Brendan John Galloway

BSc (Honours) UNBSJ, 1997

MSc UNBSJ, 2000

A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF

Doctor of Philosophy

In the Graduate Academic Unit of Biology

Supervisor: Kelly Munkittrick, PhD, Dept. of Biology, UNBSJ Co-Supervisor: R. Allen Curry, PhD, Dept. of Biology, UNBF Examining Board: Deborah MacLatchy, PhD, Dept. of Biology, UNBSJ Tillman Benfey, PhD, Dept. of Biology, UNBF Rob Moir, PhD, Dept. of Social Sciences, UNBSJ External Examiner: Douglas Holdway, PhD, School of Science, University of

Ontario Institute of Technology

This thesis is accepted by the Dean of Graduate Studies

THE UNIVERSITY OF NEW BRUNSWICK

March, 2005

© Brendan J. Galloway, 2005

ii

ABSTRACT

Research was conducted to assess the suitability of different fish species

as sentinels for monitoring the effects of multiple stressors on fish populations in

the upper Saint John River and to further evaluate the use of small-bodied fish

for environmental monitoring programs. The effluent discharge areas were

associated with nutrient enrichment, and slimy sculpin showed as much as a

50% increase in condition factor downstream of the effluents, as well as

increases in growth and liver size. Stable isotope data indicated slimy sculpin did

not move between sites, and while the sculpin were capable of demonstrating the

increases in nutrients downstream of the effluents, white sucker showed few

differences between sites. A follow-up study re-examined those two species, as

well as blacknose dace and yellow perch in the same area, and found that

neither species showed as dramatic a response as the sculpin. The differences

in species responses may be related to differences in life history characteristics,

or mobility, or exposure. Results clearly demonstrated the ability of slimy sculpin

to reflect local conditions, but there may be a trade-off between sensitivity and

relevance, depending on the questions being addressed. Slimy sculpin can be a

useful sentinel species for monitoring large rivers that receive multiple industrial

and municipal effluents being discharged in close proximity.

Blacknose dace showed increased variability in gonad size as spawning

time approached, reducing the power and ability to detect differences between

sites. A detailed study examined the biology of four multiple spawning, small-

iii

bodied species (i.e., blacknose dace, northern redbelly dace, golden shiner,

mummichog) to determine the influence of sampling time on data variability. This

information was useful to identify the most suitable time of the year to focus

sampling efforts for these fish species, and determine whether there are

additional challenges to using these species as sentinels.

This PhD project showed that the sentinel fish species selected for a

particular environmental monitoring program will be depend on the a priori study

objectives. The project provided sampling and data interpretation guidance for

the use of small-bodied, multiple-spawning fish species in future environmental

monitoring programs.

iv

ACKNOWLEDGMENTS

Returning to graduate school after completing my M.Sc. seemed like a

crazy thought to me 4 years ago - so crazy that I did it! It’s amazing how plans

can change when great opportunities present themselves. It’s been a long road

(but, fun!!) and I can finally extended my gratitude to all the people who have

helped me along the way.

First and foremost I would like to thank Dr. Kelly Munkittrick for providing

me the opportunity to conduct research towards my Ph.D. degree. Kelly has

been supportive in all aspects of my research and always provided enthusiasm

and optimism (and a Keith’s) when things seemed like they were heading for the

crapper. I greatly appreciate all of the opportunities and advice that Kelly has

provided over the years and I know I’m a better scientist because of it. Thanks

for singing Neil Diamond tunes with me during SETAC karaoke nights too! I

would also like to thank the rest of the Munkittrick clan (Patty, Sara, Jessica) for

allowing me to crash on the couch when I was travelling between Saint John and

Fredericton.

Thanks to Dr. R. Allen Curry for agreeing to be my co-supervisor. I am

sure it was with some uncertainty that he accepted this role, but my thesis work

has greatly benefited from his involvement. I would also like to thank Dr. Simon

Courtenay for being part of my supervisory committee and for his valuable

contributions throughout my studies.

Big thanks to Mr. Craig Wood for his assistance in securing industry

funding for my Industrial Post-Graduate NSERC Scholarship. Additional thanks

v

to Craig for his encouragement, advice and friendship through my studies – and

providing the fine single malt scotch after TAP meetings too.

Thanks to Sue Dunn and all the folks at NexFor/Fraser Papers

(Edmundston Operations) for their financial assistance, support, and their

willingness to participate in these studies.

My research was all field work and I wish to thank all those people who

unknowing volunteered for “Brendan’s Biology Boot-Camp”. We had some long

days lugging heavy equipment through some rough terrain in bad weather (there

were some nice days too!). Your enthusiasm and hard work made the things

easier, fun, and each trip a success. In no specific order I thank: Kelly

Munkittrick, Steve Currie, Kirk Roach, Michelle Gray, Sandra Brasfield, Mark

Gautreau, Genevieve Vallieres, Jenn Peddle, Nicole Duke, Chris Cronin, Tim

Rees, Megan Findlay, Coral Cargill, Lisa Peters, my Bhutanese buddy Karma

Tenzin, and Andrew “weasel” Halford.

My research was funded through a number of research grants awarded to

Dr. Kelly Munkittrick, including; Toxic Substances Research Initiative (TSRI),

NSERC Discovery Grant, Sir James Dunn Wildlife Research Centre Fund, New

Brunswick Wildlife Council Trust Fund, NWRI, Environment Canada (Atlantic

Region), New Brunswick Innovation Fund, and the Canadian Water Network.

Additional thanks for extensive in-kind support from NexFor, Noranda

Technology Centre, Fraser Papers, New Brunswick Department of Environment,

New Brunswick Department of Natural Resources, Barry Mower and the Maine

vi

Department of Environmental Protection, DFO, and the Maine Chapter of the

Nature Conservancy.

I would also like to thank everyone in my family for all the encouragement

and support they have given me throughout my studies – even though they’re still

not sure what I do! Finally, and most important, my biggest thanks goes to my

wife, Tiffanny – for all her love, support, patience and laughter throughout this

journey. Thanks for taking time to help me in the field and lab when I needed

help the most. It’s been a long journey - we can finally move on and seek out

new adventures together.

vii

TABLE OF CONTENTS

ABSTRACT ......................................................................................................... ii

ACKNOWLEDGMENTS ..................................................................................... iv

TABLE OF CONTENTS .................................................................................... vii

LIST OF TABLES ................................................................................................x

LIST OF FIGURES ............................................................................................ xii

CHAPTER 1. GENERAL INTRODUCTION.................................................. 15

1.1 Environmental Effects Monitoring Program.............................. 15

1.2 Cumulative Effects Assessment .............................................. 17

1.3 Fish Species Selection ............................................................ 20

1.4 Research Hypothesis, Thesis Objectives and Outline ............. 22

1.5 References .............................................................................. 27

CHAPTER 2. Examination of the Responses of Slimy Sculpin (Cottus

cognatus) and White Sucker (Catostomus commersoni) Collected on the Saint

John River Downstream of Pulp mill, Paper mill, and Sewage Discharges1. .. 33

2.1 Abstract ................................................................................... 33

2.2 Introduction.............................................................................. 34

2.3 Materials and Methods ............................................................ 37

2.3.1 Study area and mill characteristics ............................................ 37

2.3.2 Fish collections .......................................................................... 38

2.3.3 Statistical analyses .................................................................... 42

2.4 Results..................................................................................... 42

2.4.1 October 1999............................................................................. 42

viii

2.4.2 December 1999 ......................................................................... 44

2.4.3 Edmundston Pulp Mill - Fall 2000 Slimy sculpin ........................ 46

2.5 Discussion ............................................................................... 47

2.5.1 White sucker.............................................................................. 49

2.5.2 Ecological significance .............................................................. 51

2.6 Conclusions ............................................................................. 53

2.7 Acknowledgements.................................................................. 54

2.8 References .............................................................................. 70

CHAPTER 3. Identifying a suitable fish species for monitoring a large river

receiving effluents from a pulp and paper mill, municipal sewage wastewater

facilities, and agricultural runoff2. .................................................................... 75

3.1 Abstract ................................................................................... 75

3.2 Introduction.............................................................................. 76

3.3 Materials and Methods ............................................................ 78

3.3.1 Study area and mill characteristics ............................................ 78

3.3.2 Fish collections .......................................................................... 79

3.3.3 Data Analyses ........................................................................... 81

3.4 Results..................................................................................... 82

3.5 Discussion ............................................................................... 85

3.6 Conclusions ............................................................................. 89

3.7 Acknowledgements.................................................................. 91

3.8 References ............................................................................ 100

ix

CHAPTER 4. Influence of seasonal changes in relative liver size, condition,

relative gonad size, and variability in ovarian development of multiple spawning

freshwater fish for use in environmental monitoring programs3. ................... 103

4.1 Abstract ................................................................................. 103

4.2 Introduction............................................................................ 104

4.3 Materials and Methods .......................................................... 107

4.3.1 Data Analyses ......................................................................... 108

4.4 Results................................................................................... 109

4.4.1 Blacknose dace ....................................................................... 109

4.4.2 Golden shiner .......................................................................... 112

4.4.3 Northern Redbelly Dace .......................................................... 113

4.4.4 Mummichog............................................................................. 114

4.5 Discussion ............................................................................. 116

4.5.1 Reproductive Development ..................................................... 116

4.5.2 Liversomatic Index and Condition Factor ................................ 119

4.5.3 Data Variability, Sample Sizes, and Power ............................. 121

4.6 Conclusions ........................................................................... 125

4.7 Acknowledgements................................................................ 126

4.8 References ............................................................................ 126

CHAPTER 5. GENERAL DISCUSSION..................................................... 149

5.1 Conclusions ........................................................................... 171

5.2 References ............................................................................ 173

x

LIST OF TABLES Table 2.2. Means ± SE (n) of various parameters of adult male and female slimy

sculpin (Cottus cognatus) collected downstream of a pulp mill (DS Pulp), paper mill (DS Paper), and municipal sewage (DS Mad) and from reference sites located at Clair, St. Hilaire, and immediately upstream of the mill (US Pulp). Within a row, differences (p < 0.05) among sites are denoted by different uppercase letters. .................57

Table 2.3. Means ± SE (n) of various parameters of adult male and female white sucker (Catostomus commersoni) collected in October 1999. Within a row, differences (p < 0.05) among sites are denoted by different uppercase letters. ............................................................................60

Table 2.4. Regression estimates for adult male slimy sculpin (Cottus cognatus) condition factor and adult female slimy sculpin length-at-age..........61

Table 2.5. Comparative summary of slimy sculpin (Cottus cognatus) and white sucker (Catostomus commersoni) collected downstream of the sulphite pulp mill relative to fish collected from St. Hilaire (reference site), Saint John River, October 1999, New Brunswick, Canada (modified from Gibbons et al. 1998)a. .............................................62

Table 3.1. Means ± SE (n) of various parameters measured in adult male and female slimy sculpin (Cottus cognatus) and adult male and female blacknose dace (Rhinichthys atratulus) collected downstream of a sulphite pulp mill (D/S Pulp), upstream of the pulp mill diffuser (U/S Pulp), and downstream of municipal sewage inputs (D/S Mad), and from reference sites located downstream of the St. Francis River (Capone) and at St. Hilaire. Within a row, differences (p ≤ 0.05) among sites are denoted by different uppercase letters. (*Significant interaction within ANCOVA model; N/A = data not available). .........92

Table 3.2. Regression estimates for adult male slimy sculpin length-at-age (Cottus cognatus) and adult male blacknose dace (Rhinichthys atratulus) gonad development. ........................................................95

Table 3.3. Means ± SE (n) of various parameters measured in adult yellow perch (Perca flavescens) and white sucker (Catostomus commersoni) collected on the Saint John River, New Brunswick, Canada. Within a row, differences (p ≤ 0.05) among sites are denoted by different uppercase letters. ............................................................................96

Table 3.4. A comparison of species responses to the combined effluents at Edmundston (sewage, pulp mill effluent) and a comparison of fish responses upstream and downstream of the pulp mill discharge. ...98

Table 4.1. Microscopic characteristics for the determination of oocyte developmental stages (modified from Blazer 2002). ......................132

xi

Table 4.2. Means ± SE (n) and minimum and maximum (in parentheses) values for length, weight, condition factor (k), liversomatic index (LSI), and gonadosomatic index (GSI) in adult female and male blacknose dace (Rhinichthys atratulus), golden shiner, mummichog (Fundulus heteroclitus), and northern redbelly dace (Phoxinus eos) collected from various sites in southern New Brunswick, Canada. Within a column, differences (p < 0.05) among sampling dates are denoted by different superscript uppercase letters. (*Significant interaction within ANCOVA model)............................................................................133

Table 4.3. Log10 regression estimates for adult female and male blacknose dace (Rhinichthys atratulus) condition and gonad size and condition factor for adult female mummichog (Fundulus heteroclitus) gonad size collected from sites in southern New Brunswick in 2003 and 2004.139

Table 4.4. Relationship of log10 ovary weight to log10 adjusted body weight in female fish on various dates in 2003 and 2004..............................141

Table 4.5. Sample sizes required to detect a 25% difference in gonad size at different levels of power for female northern redbelly dace Phoxinus eos and blacknose dace Rhinichthys atratulus. .............................142

Table 4.6. Comparison of the relationship of log10 ovary weight to log10 body weight between all female blacknose dace (Rhinichthys atratulus) with 2 year old fish and fish weighing 2 – 4 g 3 year old collected from Milkish Brook on May 20, 2003......................................................143

Table 5.1. Percent differences in condition factor (k), liversomatic index (LSI), and gonadosomatic index (GSI) for slimy sculpin, white sucker, yellow perch, and blacknose dace collected at various times from 1999-2003 (“% Difference” is for municipal sewage and pulp mill effluent exposed fish relative to reference fish at St. Hilaire). (-) data not available; NS – not significantly different. ......................................177

xii

LIST OF FIGURES

Figure 2.1. Map of the study area showing the relative location of reference and exposure fish collection sites (not to scale)......................................63

Figure 2.2. Relative liver size (liversomatic index [LSI]; % of body weight) of adult male (black bars) and female (white bars) white sucker collected downstream of the Edmundston pulp mill (DS Pulp) and at reference sites located upstream of the pulp mill (US Pulp), St. Hilaire, Ogilvie Lake (males only), and First Lake during the Fall, October 1999. Refer to Figure 2.1 map for location. ...............................................64

Figure 2.3. Stable-isotope ratios of carbon and nitrogen in adult male (bold dashed lines) and female (solid lines) slimy sculpin collected from reference sites at Clair, St. Hilaire, and upstream of the Madawaska River (US Mad) and from downstream of the Edmundston pulp mill (DS Pulp) and Madawaska paper mill (DS Paper), October 1999. Values are expressed as deviations (δ) from standards. Refer to Figure 2.1 map for location. .............................................................65

Figure 2.4. Summary of "ecologically relevant" changes in condition factor in female slimy sculpin collected during the Fall 2000 fish survey of the Saint John River. Fish were collected from various sites, including: Moody Bridge (MB), Priestly Brook (PB), a site downstream of the St. Francis River (Capone), sites located upstream (US Nad) and downstream (DS Nad) of a poultry processing plant, upstream of the international bridge at Clair, downstream of Baker Brook (DS Baker), St. Hilaire, upstream of the Madawaska River (US Mad), downstream of the Madawaska River (DS Mad), upstream of the Iroquois River (US Iroq), Iroquois River (IR), upstream of the pulp mill diffuser (US Pulp), downstream of the pulp mill diffuser (DS Pulp), and the Little Forks (LF). Cross-hatched bars represent reference sculpin collected from sites on the Saint John River. Black bars represent sculpin from sites exposed to either poultry processing waste effluent (DS Nad), municipal sewage wastewater (DS Mad), and pulp mill effluent (DS Pulp). White bars represent sculpin collected from tributaries. Refer to Figure 2.1 map for location. .........................................................66

Figure 2.5. Summary of "ecologically relevant" changes in liver size female slimy sculpin collected during the Fall 2000 fish survey of the Saint John River. Solid and stippled horizontal lines represent data (i.e., mean ± 25%, mean ± 2SD, respectively) collected at St. Hilaire. Values that fall between the horizontal lines are considered “normal” for the SJR at the time of sampling. Cross-hatched bars represent reference sculpin collected from sites on the Saint John River. Black bars represent sculpin from sites exposed to either poultry processing waste effluent (DS Nad), municipal sewage wastewater (DS Mad), and pulp mill effluent (DS Pulp). White bars represent sculpin collected from tributaries. Asterisk (*) indicates possible upstream source of contamination. Refer to Figure 2.1 map for location........68

xiii

Figure 3.1. The study area near Edmundston, NB. River flow is from left to right on the map.......................................................................................99

Figure 4.1. Mean monthly ambient temperature for Milkish Brook during the study period. Upper bars depict monthly high temperatures and lower bars depict low temperatures.........................................................144

Figure 4.2. Percentage of each oocyte occurrence (except stage 1, primary oocytes) in female blacknose dace Rhinichthys atratulus caught in Milkish Brook. Dark grey bars represent stage 2 oocytes; light grey bars represent stage 3 oocytes; white bars represent stage 4 oocytes; bars with a diamond-grid represent stage 5 oocytes; bars with horizontal-grid represent stage 6 oocytes. .....................................145

Figure 4.3. Percentage of each oocyte occurrence (except stage 1, primary oocytes) in female golden shiner Notemigonus crysoleucas caught in Little Chamcook Lake. Dark grey bars represent stage 2 oocytes; light grey bars represent stage 3 oocytes; white bars represent stage 4 oocytes. ......................................................................................146

Figure 4.4. Percentage of each oocyte occurrence (except stage 1, primary oocytes) in female northern redbelly dace Phoxinus eos caught in beaver pond in Keswick River. Dark grey bars represent stage 2 oocytes; light grey bars represent stage 3 oocytes; white bars represent stage 4 oocytes; bars with a diamond-grid represent stage 5 oocytes; bars with a horizontal-grid represent stage 6 oocytes. .147

Figure 4.5. Percentage of each oocyte occurrence (except stage 1, primary oocytes) in female mummichog Fundulus heteroclitus caught in a salt marsh located adjacent to Taylor's Island. Dark grey bars represent stage 2 oocytes; light grey bars represent stage 3 oocytes; white bars represent stage 4 oocytes; bars with a diamond-grid represent stage 5 oocytes; bars with a horizontal-grid represent stage 6 oocytes. .148

Figure 5.1. Sample sizes required to detect a 25% difference in gonad size at different levels of power for all adult female blacknose dace collected on May 20 (dashed lines) and June 11 (solid lines), 2003. The solid curved line numbered “1” represents α = 0.05 and β = 0.05; the solid curved line numbered “2” represents α = 0.10 and β = 0.10; and the solid curved line numbered “3 represents α = 0.05 and β = 0.20...179

Figure 5.2. Sample sizes required to detect a 25% difference in gonad size at different levels of power for 2 year old female blacknose dace collected on May 20 (dashed lines) and June 11 (solid lines), 2003. The solid curved line numbered “1” represents α = 0.05 and β = 0.05; the solid curved line numbered “2” represents α = 0.10 and β = 0.10; and the solid curved line numbered “3 represents α = 0.05 and β = 0.20................................................................................................180

xiv

Figure 5.3. Sample sizes required to detect a 25% difference in gonad size at different levels of power for 2 – 4 g female blacknose dace collected on May 20 (dashed lines) and June 11 (solid lines), 2003. The solid curved line numbered “1” represents α = 0.05 and β = 0.05; the solid curved line numbered “2” represents α = 0.10 and β = 0.10; and the solid curved line numbered “3 represents α = 0.05 and β = 0.20...181

Figure 5.4. Sample sizes required to detect a 25% difference in gonad size at different levels of power for all female northern redbelly dace collected on May 27 (dashed lines) and June 22 (solid lines), 2003. The solid curved line numbered “1” represents α = 0.05 and β = 0.05; the solid curved line numbered “2” represents α = 0.10 and β = 0.10; and the solid curved line numbered “3 represents α = 0.05 and β = 0.20...182

Figure 5.5. Sample sizes required to detect a 25% difference in gonad size at different levels of power for 2 year old female northern redbelly dace collected on May 27 (dashed lines) and June 22 (solid lines), 2003. The solid curved line numbered “1” represents α = 0.05 and β = 0.05; the solid curved line numbered “2” represents α = 0.10 and β = 0.10; and the solid curved line numbered “3 represents α = 0.05 and β = 0.20................................................................................................183

Figure 5.6. Sample sizes required to detect a 25% difference in gonad size at different levels of power for 2 – 4 g female northern redbelly dace collected on May 27 (dashed lines) and June 22 (solid lines), 2003. The solid curved line numbered “1” represents α = 0.05 and β = 0.05; the solid curved line numbered “2” represents α = 0.10 and β = 0.10; and the solid curved line numbered “3 represents α = 0.05 and β = 0.20................................................................................................184

15

CHAPTER 1. GENERAL INTRODUCTION

1.1 Environmental Effects Monitoring Program

In Canada, the pulp and paper industry is one of the largest dischargers of

wastewater effluent and has been identified as having potential to disrupt the

growth and reproductive performance of fish [Environment Canada 1997]. In the

early 1990’s, a new regulatory package was developed that included

requirements for pulp and paper mills in Canada to conduct a cyclical

Environmental Effects Monitoring (EEM) program. The objectives of the EEM

program are to assess whether receiving environment changes are associated

with pulp and paper mill effluent exposure to fish, fish habitat, and use of the

fisheries resource. The first two cycles of EEM monitoring identified a variety of

responses in fish populations near pulp mills [Munkittrick et al. 2002]. At some

Canadian sites, exposure of fish to pulp mill effluent has been associated with

delayed sexual maturity, decreased gonad size, increased liver size, and

decreased secondary sex characteristics [Munkittrick et al. 1991; McMaster et al.

1991]. A recent review of the adult fish population survey data from the

Canadian Pulp and Paper Mill EEM program showed that the predominant

response patterns observed in fish populations across Canada were a decrease

in gonad weight and increases in liver weight, condition factor, and weight-at-age

[Environment Canada 2003]. Together, these responses are indicative of

metabolic disruption or endocrine disruption in association with a nutrient

enrichment effect [Environment Canada 2003]. Some of the responses observed

16

in Canada have also been documented in fish exposed to pulp mill effluent in

Scandinavia [Sandstrom et al. 1988; Sandstrom 1994; Tana and Lehtinen 1996;

Sandstrom et al. 2003] and the United States [Adams et al. 1992].

Since the late 1980’s, the Canadian pulp and paper industry has made

significant financial contributions to improving effluent treatment which has

reduced toxicity, nutrient enrichment, and biochemical oxygen demand

[Munkittrick et al. 2000]. Today, the effects of pulp mill effluent on fish

populations are sub-lethal and often fall within the range of natural variability that

can be associated with natural stressors (e.g., temperature) that are not directly

related to anthropogenic inputs [Munkittrick et al. 2000]. As such, separating the

effects of pulp mill effluent exposure on fish health from the effects associated

with natural stressors can be difficult. Other wastewater effluents, such as

municipal sewage, are often discharged in close proximity to pulp and paper mill

discharges. It has been well-documented that wastewater from municipal

sewage facilities contains compounds that can alter the reproductive system of

fish [Jobling and Sumpter 1993; Jobling et al. 1996; Harries et al. 1996; Allen et

al. 1999]. The acute impacts of pulp and paper mill effluents in the past were

gross to the point that the potential impacts of other wastewater discharges (e.g.,

sewage) on the health of aquatic organisms were not recognized. Today,

distinguishing between the effects of sewage and/or other anthropogenic

stressors and pulp and paper mill effluents on fish growth, reproduction, and

survival can be difficult. Attributing responsibility for the existing impacts within a

river responding to multiple stressors to a specific industry is problematic. In

17

Edmundston, New Brunswick, both municipal sewage wastewater and

agricultural inputs (i.e., animal manure runoff) are located upstream of the pulp

mill effluent diffuser. A high priority for research is associated with the need to be

able to separate the relative contributions of multiple discharges and assess the

potential cumulative effects of these discharges at the watershed scale. In order

to address this issue, the objectives of Chapters 2 and 3 were to compare the

whole-organism responses of small-bodied fish (i.e., slimy sculpin, Cottus

cognatus, and blacknose dace, Rhinichthys atratulus) and large-bodied fish (i.e.,

white sucker, Catostomus commersoni, and yellow perch, Perca flavescens)

along a downstream gradient in a section of the Saint John River receiving pulp

and paper mill effluents, municipal sewage, and agricultural runoff (i.e., animal

manure). The goal of the fish collections was to identify a suitable fish species

that could be used to assess the relative contribution of individual anthropogenic

stressors in a section of river receiving multiple stressors.

1.2 Cumulative Effects Assessment

In 1995, the Canadian Environmental Assessment Act (CEAA) required

development proponents to include a cumulative effects assessment (CEA) as

part of future environmental impacts assessments (EIAs). Changes to the CEAA

were based on the fact that long-term environmental impacts occur as a result of

the combined effects of multiple anthropogenic stressors [Hegmann et al. 1999].

Historically, environmental assessments have been conducted without clearly

identifying cumulative effects [Munkittrick et al. 2000]. Instead, traditional

approaches to environmental assessments have focused on stressor-based

18

predictive methods, which concentrate on single stressors (e.g., a single

chemical within a discharge or the discharge of a single effluent) [Munkittrick et

al. 2000]. Specifically, stressor-based CEAs have been conducted as desktop

exercises focused on documenting existing conditions and attempting to identify

potential future stressors, and identifying valued ecosystem components (VECs)

[Munkittrick et al. 2000]. Stressor-based CEAs use the assumed pathways of

impacts to make predictions about the future and propose possible mitigation

strategies [Munkittrick et al. 2000]. It is important to recognize that stressor-

based risk assessments and EIAs have successfully helped to reduce the gross

environmental impacts of the past; the main problem with these approaches is

that relatively little site-specific data are used and the predictions made during

pre-development assessments are often not validated with follow up monitoring

during operational and post-operational phases [Munkittrick et al. 2000].

Although proponents have been required to include a CEA within the scope of

the EIA since 1995, there are currently no widely accepted, scientifically-

defensible methods for analyzing and evaluating cumulative environmental

impacts [Munkittrick et al. 2000].

Munkittrick et al. [2000] proposed an effects-based cumulative effects

assessment (CEA) program to determine the cumulative impacts of multiple

aquatic stressors on fish populations. The effects-based approach involves

examining factors associated with the status of fish populations in developed and

undeveloped reaches of a large river basin. This approach uses an iterative

assessment program to define the integrated responses of fish to existing

19

conditions and to determine the factors currently limiting fish performance in the

system. The main objectives of an effects-based assessment of cumulative

effects are to document the baseline conditions of fish performance (e.g., growth,

condition, and gonad development), examine yearly variability, and examine

trends in responses over time in order to make some assessment of the existing

“accumulated environmental state” (i.e., the existing state of the system as it has

integrated all existing stressors, both natural and anthropogenic). Performance

is not limited when fish exposed to wastewater effluents exhibit no differences in

age, growth (e.g., weight-at-age), and reproduction when compared to upstream

reference fish. However, if fish showed changes in any of the performance

parameters noted above then one can conclude that limiting or enhancing factors

are present and will need to be confirmed and investigated in more detail through

follow-up studies. In contrast, stressor-based CEA’s do not consider that the

system may already be stressed by natural factors (e.g., resource competition,

predator-prey interactions) and/or other human activities (e.g., fishing pressure)

and only focus on areas where stakeholders perceive a potential stressor exists

(e.g., pulp mill, hydroelectric dam, mining operation). Under the stressor-based

approach, erroneous conclusions regarding potential impacts associated with an

identified stressor (e.g., pulp and paper mill) are likely because other factors

(e.g., natural food limitation or other anthropogenic stressors) which may limited

or enhance the system were not considered. The identification of factors that are

limiting (or enhancing) the existing performance is critical to subsequent

predictive attempts, the design of potential mitigation strategies and the design of

20

future monitoring strategies.

Overall, the main difference between the stressor-based and effects-driven

approaches to CEA’s are the kinds of data required [Munkittrick et al. 2000].

Effects-driven CEA’s are committed to baseline monitoring, adaptive

management, and post-operational monitoring that are not part of stressor-based

assessments [Munkittrick et al. 2000].

One of the key components of many environmental monitoring programs,

including the effects-driven CEA, is the selection of an appropriate sentinel fish

species for monitoring the potential biological impacts of wastewater effluents on

the receiving environment. It is critical that the selected sentinel species offer the

greatest potential for defining stressor influences within the system being studied

[Munkittrick et al. 2000]. Species are selected primarily according to abundance,

exposure, and the ability to measure growth, reproduction and age [Munkittrick et

al. 2000].

1.3 Fish Species Selection

In Canada, large-bodied fish species have commonly been used as

sentinels for monitoring the potential impacts of wastewater effluents (e.g., pulp

and paper mill effluent) on freshwater environments, and include: white sucker

spp. [McMaster et al. 1991; Hodson et al. 1992; Gagnon et al. 1994; Munkittrick

et al. 1994], lake whitefish, Coregonus clupeaformis [Munkittrick et al. 1992],

longnose sucker, Catostomus catostomus [Swanson et al. 1994], and mountain

whitefish, Prosopium williamsoni [Swanson et al. 1994]. Response to pulp mill

effluent exposure have been successfully measured in large-bodied fish, but

21

primarily in situations where fish movement between reference and exposure

sites was impeded by man-made barriers [Hodson et al. 1992; Munkittrick et al.

2000]. Using large-bodied fish for environmental monitoring in large, open rivers

has become a concern for a number of reasons, including: many large-bodied

fish species are capable of extensive movement beyond effluent exposure areas

[Swanson et al. 1994], insufficient numbers of fish captured to properly assess

potential impacts of pulp mill effluent exposure [Hodson et al. 1992], and

exploitation of commercially important species (e.g., lake whitefish) may obscure

or confound impacts associated with environmental stressors such as pulp mill

effluent [Gibbons et al. 1998a].

Recent evidence has shown that small, non-migratory fish species can be

used as alternative sentinel species in large, open water systems where mobility

may be an issue [Gibbons et al. 1998a; Munkittrick et al. 2000]. Many small-

bodied fish species, such as cottids and cyprinids, are less mobile relative to

large-bodied species, possess a smaller home range, and exhibit territorial

behaviour [Van Vliet 1964; Hill and Grossman 1987; Minns 1995]. Together,

these characteristics increase the probability that the measured physiological,

biochemical and whole-organism responses in small-bodied fish species will

reflect the local environmental conditions where they were caught [Gibbons et al.

1998a]. In addition, small-bodied fish species are more abundant than large-

bodied fish species, which facilitates sampling; they are short-lived and therefore

exhibit alterations in growth and reproduction quicker than longer-lived, large-

22

bodied species; and they are not subjected to sport or commercial fishing

pressures [Gibbons et al. 1998a]. The main disadvantage of using small-bodied

species is the lack of understanding of life-history characteristics [Munkittrick et

al. 2000]. To address this issue, seasonal changes in liver size, condition factor,

gonad size, and variability in female gonad development were monitored in four

multiple spawning, small-bodied fish species (see Chapter 4).

Few studies have compared the whole-organism responses of small-bodied

and large-bodied fish exposed to anthropogenic stressors such as pulp mill

effluent. Gibbons et al. [1998b] compared the responses of trout-perch

(Percopsis omiscomaycus) and white sucker downstream of a pulp mill on the

Kapuskasing River in Northern Ontario. Results from this study showed small-

bodied fish (i.e., trout-perch) and large-bodied fish (i.e., white sucker) exposed to

pulp mill effluent exhibited different whole-organism responses, but the

underlying mechanisms (i.e., size-specific mortality and/or recruitment problems)

responsible for the responses were similar. It is evident that small-bodied and

large-bodied fish in some rivers can exhibit differences in whole-organism

responses. However, due to a lack of comparative studies, it not known whether

small-bodied and large-bodied fish in other rivers will show differences in whole-

organism responses.

1.4 Research Hypothesis, Thesis Objectives and Outline

My research hypothesis was to determine whether fish can be used to

assess the relative contribution of individual anthropogenic stressors to the

existing environmental conditions in a river exposed to multiple anthropogenic

23

stressors. More specifically, the objectives of my thesis were to compare the

whole-organism responses of small-bodied fish (i.e., slimy sculpin and blacknose

dace) and large-bodied fish (i.e., white sucker and yellow perch) along a

downstream gradient in a river exposed to pulp and paper mill effluents,

municipal sewage wastewater, and agricultural runoff (i.e., manure); identify the

fish species that is best suited to assess the relative contribution of individual

anthropogenic stressors in a river exposed to multiple anthropogenic stressors;

determine which life history characteristics most influenced the ability of a fish

species to exhibit the measured whole organism responses; and provide

sampling guidance for the use of multiple spawning, small-bodied fish species for

use in environmental monitoring programs.

The objective of the first chapter (Chapter 2) was to compare the whole-

organism responses of white sucker and slimy sculpin (Cottus cognatus)

exposed to multiple wastewater effluents in the Saint John River near

Edmundston. The null (Ho) hypotheses were:

Ho1: There are no differences in the mean age, energy use (i.e., length-at-age,

gonad size), or energy stores (i.e., liver size, condition factor) of slimy sculpin

and white sucker exposed to pulp mill effluent, paper mill effluent, municipal

sewage wastewater, and manure relative to upstream reference fish.

Ho2: There are no differences in the stable carbon (13C/12C) and nitrogen

(15N/14N) isotope signatures of slimy sculpin exposed to pulp mill effluent, paper

mill effluent, municipal sewage wastewater, and agricultural runoff (i.e., manure)

24

relative to upstream reference fish.

The Saint John River near Edmundston is a complex receiving

environment. The river receives effluent from a pulp mill, a paper mill, three

treated sewage discharges, and other untreated sewage releases enter the river

over a 10-km reach. In addition, two tributaries carry manure into the system.

Results from previous work on the Moose River system in Ontario showed large-

bodied fish and small-bodied fish exposed to pulp mill effluent could exhibit

differences in whole-organism responses [Munkittrick et al. 2000]. Because

there was a lack of comparative studies, it was unknown whether different fish

species would exhibit different whole organism responses in a river that is more

industrialized (i.e., Saint John River) relative to the Moose River system - could

fish be used to discriminate individual wastewater effluents in a relatively small

section of the Saint John River receiving multiple wastewater effluents? The

results from the research on white sucker and slimy sculpin collected from the

Saint John River near Edmundston are presented in Chapter 2, “Examination of

the responses of slimy sculpin (Cottus cognatus) and white sucker (Catostomus

commersoni) collected on the Saint John River downstream of pulp mill, paper

mill, and sewage discharges”.

After identifying differences in the whole-organism responses of white

sucker and slimy sculpin collected from the upper Saint John River near

Edmundston, I wanted to test the hypothesis that body size, rather than other

inter-specific differences, was responsible for the differences in the whole

25

organism responses of slimy sculpin and white sucker documented in Chapter 2.

Would white sucker and yellow perch show similar responses? Would the whole

organism responses of slimy sculpin and blacknose dace be similar? Would the

whole organism responses of white sucker and slimy sculpin be consistent

between years? The null (Ho) hypotheses were:

Ho1: There are no differences in the mean age, energy use (i.e., length-at-age,

gonad size), energy stores (i.e., liver size, condition factor), or stable carbon and

nitrogen isotope signatures of slimy sculpin and blacknose dace exposed to pulp

mill effluent, municipal sewage wastewater, and manure relative to upstream

reference fish.

Ho2: There are no differences in the mean age, energy use (i.e., length-at-age,

gonad size), energy stores (i.e., liver size, condition factor), or stable carbon and

nitrogen isotope signatures of white sucker and yellow perch exposed to pulp mill

effluent, municipal sewage wastewater, and manure relative to upstream

reference fish.

Ho3: There are no differences in the stable carbon (13C/12C) and nitrogen

(15N/14N) isotope signatures of small-bodied fish and large bodied fish.

Ho4: There are no inter-annual differences in the whole-organism responses of

slimy sculpin and white sucker.

In addition, it was also important to further contribute to the scientific

understanding of the suitability of small-bodied fish species for environmental

monitoring programs. Results from this work are presented in chapter 3,

“Identifying a suitable fish species for monitoring a large river receiving effluents

26

from a pulp and paper mill, municipal sewage wastewater facilities, and

agricultural runoff”.

One of the main challenges of incorporating small-bodied fish species in

environmental monitoring programs is a lack of basic biological information.

Blacknose dace (Chapter 3) showed increased variability in the reproductive

endpoints and the data suggested sampling was conducted at a time of the year

when gonad development was highly variable. There was a need to investigate

this further in order to develop better guidance on how to address reproductive

investment in multiple-spawning fish species. To address this issue, blacknose

dace (Rhinichthys atratulus), golden shiner (Notemigonus crysoleucas), northern

redbelly dace (Phoxinus eos), and the estuarine mummichog (Fundulus

heteroclitus) were collected at various times during the pre-spawning, spawning,

and post-spawning seasons. The objectives of this work were to examine the

annual reproductive cycles of blacknose dace, golden shiner, northern redbelly

dace, and the estuarine mummichog; examine seasonal variability in gonad

development by monitoring changes in the coefficient of determination values

(i.e., r2 values) for regressions of ovary weight and body weight; increase our

understanding of natural source(s) of variability in gonad development; identify

suitable techniques to minimize data variability; and reduce sample size

requirements for detecting a critical effect sizes. The null hypothesis (Ho) was:

Ho: There are no differences in the variability of female ovarian development

during the pre-spawning, spawning, and post-spawning seasons for blacknose

dace, golden shiner, northern redbelly dace, and the estuarine mummichog.

27

Results of this work are presented in chapter 4, “Influence of seasonal

changes on the suitability of multiple spawning freshwater fish species for

examining reproductive impacts of stress”.

The final chapter (Chapter 5) provides a discussion and summary of the

significance of the whole-organism response patterns of slimy sculpin, blacknose

dace, white sucker, and yellow perch exposed to multiple anthropogenic

stressors in a section of the Saint John River near Edmundston, New Brunswick.

The suitability of the slimy sculpin for monitoring sections of rivers receiving

multiple wastewater effluents discharges is discussed. Sampling guidance for

use of multiple spawning, small-bodied fish species in environmental monitoring

programs is also considered. Suggestions for future research needs and

conclusions from the research are also provided.

For this thesis, length-at-age was used to estimate growth; fecundity (i.e.,

total number of eggs per female) and relative gonad size were used to estimate

reproduction; and condition factor and relative liver size were used to estimate

energy storage.

1.5 References

Adams, M.S., Crumby, W.D., Greeley, M.S., Shugart, L.R., Saylor, C. 1992.

Responses of fish populations and communities to pulp mill effluents: a

holistic assessment. Ecotox Environ Safe 24: 347-360.

Allen, Y., Scott, A.P., Matthiessen, P., Haworth, S., Thain, J.E., and Fiest, S.

1999. Survey of estrogenic activity in United Kingdom estuarine and

28

coastal waters and its effects on gonadal development of the flounder

Platichthys flesus. Environ Toxicol Chem 18: 1791-1800.

Environment Canada. 1997. Environment Canada’s national strategy for

addressing endocrine disrupting substances in the environment.

http://www.ec.gc.ca/eds/strat_e.htm.

Environment Canada. 2003. National assessment of pulp and paper

environmental effects monitoring data: a report synopsis. National Water

Research Institute, Burlington, Ontario. NWRI Scientific Assessment

Report Series No. 2. 28p.

Gagnon, M.M., Dodson, J.J., Hodson, P.V., Van Der Kraak, G., and Carey, J.H.

1994. Seasonal effects of bleached-kraft mill effluent on reproductive

parameters of white sucker (Catostomus commersoni) populations of the

St. Maurice River, Quebec, Canada. Can J Aquat Sci 51: 337-347.

Gibbons, W.N., Munkittrick, K.R., and Taylor, W.D. 1998a. Monitoring aquatic

environments receiving industrial effluents using small fish species 1:

Response of spoonhead sculpin (Cottus ricei) downstream of a bleached-

kraft pulp mill. Environ Toxicol Chem 17: 2227-2237.

Gibbons, W.N., Munkittrick, K.R., and Taylor, W.D. 1998b. Monitoring aquatic

environments receiving industrial effluents using small fish species 2:

Comparison between responses of trout-perch (Percopsis omiscomaycus)

and white sucker (Catostomus commersoni) downstream of a pulp mill.

Environ Toxicol Chem 17: 2238-2245.

29

Harries, J.E., Sheahan, D.A., Jobling, S., Matthiessen, P., Neall, P., Routledge,

E.J., Rycroft, R., Sumpter, J.P., and Tylor, T. 1996. A survey of

estrogenic activity in United Kingdom inland waters. Environ Toxicol

Chem 15: 1993-2002.

Hill, J., and Grossman, G.D. 1987. Home range estimates for three North

American stream fishes. Copeia 1987: 376-380.

Hodson, P.V., McWhirter, M., Ralph, K., Gray, B., Thivierge, D., Carey, J.H., Van

Der Kraak, G., Whittle, D.M., and Levesque, M. 1992. Effects of

bleached kraft mill effluent on fish in the St. Maurice River, Quebec.

Environ Toxicol Chem 17: 2227-2237.

Jobling, S., and Sumpter, J.P. 1993. Detergent components in sewage effluent

are weakly oestrogenic to fish : An in vitro study using rainbow trout

(Oncorhynchus mykiss) hepatocytes. Aquat Toxicol 27 : 361-372.

Jobling, S., Sheahan, D.A., Osborne, J.A., Matthiessen, P., and Sumpter, J.P.

1996. Inhibition of testicular growth in rainbow trout (Oncorhynchus

mykiss) exposed to estrogenic alkylphenolic chemicals. Environ Toxicol

Chem 15: 194-202.

Minns, C.K. 1995. Allometry of home range size in lake and river fishes. Can J

Fish Aquat Sci 52: 1499-1508.

McMaster ME, Van Der Kraak GJ, Portt CB, Munkittrick KR, Sibley PK, Smith IR,

Dixon DG. 1991. Changes in hepatic mixed-function oxygenase (MFO)

activity, plasma steroid levels and age at maturity of a white sucker

30

(Catostomus commersoni) population exposed to bleached kraft pulp mill

effluent. Aquat Toxicol 21: 199-218.

McMaster, M.R., Munkittrick, K.R., Jardine, J.J., Robinson, R.D., and Van Der

Kraak, G. 1995. Protocol for measuring in vitro steroid production by fish

gonadal tissue. Can Tech Rep Fish Aquat Sci 1961: 78.

Munkittrick, K.R., Portt, C.B., Van Der Kraak, G., Smith, I.R., Rokosh, D.A. 1991.

Impact of bleached kraft mill effluent on population characteristics, liver

MFO activity, and serum steroid levels of a Lake Superior white sucker

(Catostomus commersoni) population. Can J Fish Aquat Sci 48:1371-

1380.

Munkittrick, K.R. 1992. A review and evaluation of study design considerations

for site-specifically assessing the health of fish populations. J Aquat

Ecosys Health 1: 283-293.

Munkittrick, K.R., Van Der Kraak, G., McMaster, M.E., Portt, C.B. 1992.

Response of hepatic mixed function oxygenase (MFO) activity and plasma

sex steroids to secondary treatment of bleached kraft pulp mill effluent and

mill shutdown. Environ Toxicol Chem 11: 1427-1439.

Munkittrick, K.R., Van Der Kraak, G., McMaster, M.E., Portt, C.B., van den

Heuvel, M.R., and Servos, M.R. 1994. Survey of receiving-water

environmental impacts associated with discharges from pulp mills. 2.

Gonad size, liver size, hepatic EROD activity and plasma sex steroid

levels in white sucker. Environ Toxicol Chem 13: 1089-1101.

31

Munkittrick, K.R., McMaster, M.E., Van Der Kraak, G., Portt, C., Gibbons, W.N.,

Farwell, A., and Gray, M. 2000. Development of methods for effects-

driven cumulative effects assessment using fish populations: Moose River

Project. SETAC Press, Pensacola, FL, USA.

Munkittrick, K.R., McGeachy, S.A., McMaster, M.E., Courtenay, S.C. 2002.

Review of cycle 2 freshwater fish studies from the pulp and paper

Environmental Effects Monitoring program. Water Quality Res J Can 37:

49-77.

Sandstrom, O., Neuman, E., and Karas, P. 1988. Effects of a bleached pulp mill

effluent on growth and gonad function in Baltic coastal fish. Water Sci

Tech 20: 107-118.

Sandstrom, O. 1994. Incomplete recovery in a coastal fish community exposed

to effluent from a modernised Swedish bleached kraft mill. Can J Fish

Aquat Sci 51: 2195-2205.

Sandstrom, O., Forlin, L., Grahn, O., Lander, L., Larsson, A., and Lindesjoo, E.

2003. Assessments of the environmental impact of Swedish pulp and

paper mill effluents at the beginning of the next century. In Stuthridge,

T.R., van den Heuval, M.R., Marvin, N.A., Slade, A.H., and Gifford, J.,

eds, Environmental Impacts of Pulp and Paper Waste Streams. SETAC,

Pensacola, FL, USA (CD) pp 499-504.

Swanson, S.W., Schryer, R., Shelast, R., Kloepper-Sams, P.J., Owens, J.W.

1994. Exposure of fish to biologically treated bleached-kraft mill effluent.

32

3. Fish habitat and population assessment. Environ Toxicol Chem 13:

1497-1507.

Tana, J., and Lehtinen, K. 1996. The aquatic environmental impact of pulping

and bleaching operations – an overview. The Finnish Environment,

Environmental Protection, Helsinki, Finland.

Van Vliet, W.H. 1964. An ecological study of Cottus cognatus (Richardson,

1836) in Northern Saskatchewan. MA Thesis, Department of Biology,

University of Saskatchewan. 155p.

Wassenar, L.I. and Culp, J.M. 1996. The use of stable isotope analyses to

identify pulp mill effluent signatures in riverine food webs. In Servos,

M.R., Munkittrick, K.R., Carey, J.H., and Van Der Kraak, G., eds,

Environmental Fate and Effects of Pulp and Paper Mill Effluents. St. Lucie

Press, Delray Beach, FL, USA. pp 413-424.

33

CHAPTER 2. Examination of the Responses of Slimy Sculpin (Cottus

cognatus) and White Sucker (Catostomus commersoni) Collected on

the Saint John River Downstream of Pulp mill, Paper mill, and

Sewage Discharges1.

1Published: Galloway, B.J., Munkittrick, K.R., Currie, S., Gray, M.A., Curry, R.A.,

and Wood, C. 2003. Examination of the responses of slimy sculpin (Cottus

cognatus) and white sucker (Catostomus commersoni) collected on the Saint

John River downstream of pulp mill, paper mill, and sewage discharges.

Environmental Toxicology and Chemistry 22: 2898-2907.

2.1 Abstract

As part of a larger survey on cumulative effects within the Saint John River

basin, a fish survey was conducted near Edmundston, New Brunswick, in the fall

of 1999 using slimy sculpin (Cottus cognatus) and white sucker (Catostomus

commersoni). The discharge environment receives effluent from a pulp mill, a

paper mill, three sewage discharges, and tributaries receiving agricultural runoff.

Sculpin collected downstream of the sewage discharges and pulp mill effluent

had greater growth, condition, and liver size than reference fish, but no significant

differences in gonad size. Stable isotope data indicated slimy sculpin did not

move across the river to an area exposed to paper mill effluent or to sites

upstream of the pulp mill effluent diffuser. Female sculpin collected downstream

of the paper mill showed no significant differences in length, body weight, age,

condition factor, liver size, and gonad size compared to fish from reference sites.

34

Female white sucker collected downstream of the pulp mill did not differ

significantly in any measured parameter compared to reference fish. Liver sizes

of white sucker from the Saint John River were outside of the range considered

to be indicative of uncontaminated riverine sites. In 2000, sculpin collected

downstream from a poultry-processing facility had larger livers and lower

condition factors, suggesting that the site is contaminated. We found no

significant differences in sculpin length, weight, condition (except for males), and

liver size in sculpin collected downstream from the pulp mill in October 2001.

The responses of slimy sculpin and white sucker differed and may be related to

differences in life history characteristics. Results from this study indicate the

slimy sculpin is a suitable fish species for monitoring rivers that receive multiple

industrial and municipal effluents.

2.2 Introduction

Over the last seven years, we have been developing an effects-driven

cumulative effects assessment (CEA) framework for the Moose River basin in

Northern Ontario [Munkittrick et al. 2000; Munkittrick and McMaster 2000]. This

evaluation is an iterative approach: in it, the population-level responses to

aquatic stressors are used to direct diagnostic studies to identify factors limiting

performance of fish in the system. Population response patterns of fish to aquatic

stressors have been developed during previous work on metal mining effluent

[Munkittrick and Dixon 1989a; Munkittrick and Dixon 1989b], pulp mill effluents

[Munkittrick et al. 1991], interactions of hydroelectric and pulp mill discharges

[Munkittrick et al. 2000] and sewage and pulp mill effluents [Frank et al. 1998]. To

35

continue the development of this CEA framework, several questions were

highlighted by the Moose River studies, including (1) would the approach be as

successful in a system with more complicated waste inputs? and (2) how do

potential contributions from non-point sources affect interpretation?

The Saint John River basin was selected as the study site for a number of

reasons. Among these are the facts that it has more development than the

Moose River basin, and a larger population base. There are hydroelectric

developments and pulp mill discharges within the system (similar to the Moose

River basin study site). There is also considerable agricultural development

along the system, plus multiple sewage discharges, and other industrial

developments (e.g., poultry-processing facility). The estuarine area near the

mouth of the Saint John River is complicated by multiple discharges, and offers

the potential for future expansion of the work into estuarine and marine areas.

Various monitoring and research programs have been implemented on the Saint

John River, but few linkages exist between them.

Canadian pulp and paper mills are required to conduct a cyclical

Environmental Effects Monitoring (EEM) program to determine if environmental

effects are evident when the mill discharges effluents that are in compliance with

effluent guidelines. The monitoring program includes requirements for

monitoring effluents, benthic invertebrate communities and fish populations; the

fish program has recently been reviewed [Munkittrick et al. 2002]. The first cycle

of EEM for pulp and paper mills reported in 1996; cycle 2 reported April 2000. As

part of the 1996 Cycle I EEM report for the Fraser Inc., Edmundston pulp mill

36

(Edmundston, New Brunswick, Canada), the adult fish survey identified a number

of responses in fish downstream of the mill discharge [BAR Environmental Inc.

1996]. Yellow perch (Perca flavescens) collected downstream of the pulp mill

diffuser had lower growth compared to fish from upstream reference sites.

Gonads in male yellow perch were also smaller and females were less fecund.

White sucker (Catostomus commersoni) below the mill had greater growth rates

and gonad size, but lower indices of liver size.

The Saint John River near Edmundston receives effluent from a pulp mill,

a paper mill, and three treated sewage discharges, and other untreated sewage

releases enter the river over a 10-km reach. In addition, two tributaries carry

agricultural runoff into the system. The objectives of this study were to develop a

database to identify the background fish performance levels of the upper Saint

John River basin, examine whether the differences in fish reported in 1996 were

still present in 1999, and examine the selection of reference sites. Specifically,

we hypothesized that large-bodied and small-bodied fish from the upper Saint

John River would show differences in performance, and that small-bodied fish

would be the most suitable sentinel species for monitoring anthropogenic impacts

in the upper Saint John River.

Fraser Papers Inc. (Edmundston pulp mill, Edmundston, New Brunswick,

Canada), Nexfor (Noranda Technology Centre, Pointe-Claire, Quebec, Canada),

and Environment Canada (National Water Research Institute, Saskatoon,

Saskatchewan, Canada) participated in a variety of studies as part of Fraser

Papers Inc. commitment to EEM studies for cycle 2. These on-site projects

37

included a 90-d flow-through fathead minnow (Pimephales promelas) bioassay

[Parrott et al. 2000], a flow-through microcosm invertebrate exposure [Culp et al.

2003], and the fish collections (present study). Additional studies are

documenting the chemicals in pulp mill effluent that can be accumulated by fish

[Hewitt et al. 2003].

Based on results from the 1999 fish survey, the objectives of the fish

survey conducted in 2000 were to locate potential sources of contamination on

the Saint John River upstream of our reference site in Clair, and to examine

annual variability at sampling sites located on the mainstem of the Saint John

River near Edmundston, N.B.

2.3 Materials and Methods

2.3.1 Study area and mill characteristics

The Saint John River originates in headwater lakes in northern Maine

(USA) and travels almost 700 km to the mouth, located at the city of Saint John,

at the Bay of Fundy. Most of the drainage area (approximately 51%) is located in

New Brunswick; while the remaining area is located in the state of Maine (36%)

and the province of Quebec (13%) [BAR Environmental Inc. 1994]. Near

Edmundston, the Saint John River forms part of the international border between

Canada and the State of Maine. The flow of the Saint John River near

Edmundston is unregulated and undergoes considerable annual fluctuations; the

dominant substrate in the river is gravel [BAR Environmental Inc. 1994]. There

are no barriers to fish movement on the mainstem of the Saint John River near

38

Edmundston. However, a dam located near the mouth of the Madawaska River,

upstream of the mill discharge, restricts the movement of fish into the tributary.

The sulphite pulp mill (Edmundston, New Brunswick, Canada) and paper

mill (Madawaska, Maine, USA) are located on the north and south sides of the

Saint John River, respectively (Figure 2.1). The pulp mill produces

approximately 550 air-dried metric tonnes (ADMT) of sulphite pulp per day, 350

ADMT of ground wood pulp per day, 120 ADMT of boxboard per day, and 100

tonnes of de-inked product per day [BAR Environmental Inc. 1994]. Most of the

pulp produced at the Edmundston mill is shipped across the Saint John River to

the paper mill, which produces uncoated groundwood paper and coated fine

paper [BAR Environmental Inc. 1994]. The pulp mill discharges approximately

75,500 m3 of secondary-treated effluent per day from an aerated lagoon facility

(retention time ~7 days) into the north side of the Saint John River, about 4 km

downstream from the mill. The paper mill discharges primary-treated effluent into

the south side of the Saint John River, about 4 km upstream of the pulp mill

effluent diffuser.

2.3.2 Fish collections

White sucker (Catostomus commersoni) and slimy sculpin (Cottus

cognatus) were consistently captured in our field collections in the upper Saint

John River. In the fall of 1999, several large back-to-back rainfall events led to

very high water levels and prevented collection of yellow perch near the effluent

outfall.

39

We attempted to capture 20 adult fish of each sex at each site. Reference

fish were collected from sites located ~35 km above the pulp mill outfall near

Clair; at St. Hilaire (~16 km upstream), and at a site immediately upstream of the

Edmundston pulp mill diffuser (Figure 2.1). Fish were also collected from sites

located downstream of the pulp mill, paper mill, and municipal sewage

discharges. White sucker were collected in areas with deep pools and moderate

flowing water between October 21 and 29, 1999, using a gasoline generator-

powered electrofishing unit (Smith-Root GPP 7.5, Smith-Root Inc., Vancouver,

WA, USA) mounted in an inflatable raft boat. White sucker were collected from

St. Hilaire, at a site ~0.5 -1.0 km upstream of the pulp mill effluent outfall and a

site downstream (~0.5 -2.0 km) of the Edmundston pulp mill diffuser. White

sucker were also collected from reference sites: Ogilvie Lake (1999) and First

Lake (2000) (Table 2.1). White sucker are benthic feeders that have been

extensively used in other monitoring programs [Munkittrick et al. 2000;

Munkittrick et al. 2002; Environment Canada 1997; Munkittrick et al. 1998].

Slimy sculpin were collected between October 1-16 and December 14-16,

1999 using a battery-powered backpack electrofisher (Smith-Root Model 12B,

Smith-Root Inc., Vancouver, WA, USA). Sculpin were collected with dip nets

where rock/rubble substrates could be found in fast-moving water (runs and

riffles) approximately 0.5-1.5 m deep. Attempts to obtain sculpin from other

downstream sites were unsuccessful due to unsuitable habitat.

Slimy sculpin are widely distributed throughout shallower portions of the

Saint John River basin [Scott and Crossman 1998]. Slimy sculpin feed primarily

40

on aquatic invertebrates; they can reach a maximum length of ~13 cm and have

a life span of 5-6 years [Van Vliet 1964]. Sculpin spawn in spring and can reach

sexual maturity at age one, but body size appears to be the determinant of

maturity [Van Vliet 1964].

Fish were rendered unconscious by concussion, followed by spinal

severance. We then measured length (i.e., total length for sculpin, and fork-

length for white sucker), body weight, liver weight, and gonad weight. White

sucker were aged by counting annuli on clean, dried opercula under a dissecting

microscope. The annuli counts were verified by two readers. Slimy sculpin were

aged by counting annuli on sagittal otoliths according to the methods of MacKay

et al. [1990] and Gibbons et al. [1998a]. Briefly, sagittal otoliths were removed

surgically from each fish and placed in propylene glycol for ≥ 24 h before being

inspected under a dissecting microscope. If annuli were difficult to count, the

sagittal otoliths were mounted on slides and ground thinner using 400 grit sand

paper until the annuli became visible.

White muscle samples were taken from male and female sculpin for stable

isotope analysis. These analyses were used to determine residency patterns of

slimy sculpin collected at each site during the fall 1999 study on the Saint John

River. Approximately 0.5 g of white muscle tissue was removed from each fish

and placed in drying oven at 65 °C for 48 h. The dried tissue samples were

individually ground into a fine powder with a mortar and pestle, and were stored

in clean glass scintillation vials before isotopic analysis. Stable-isotope ratios

(e.g., 13C/12C, 15N/14N) are expressed as delta (δ) values and are measures of a

41

parts-per-thousand (‰) difference between the sample isotope ratio and that of

an international standard (carbon – Pee Dee Belemnite; nitrogen – atmospheric

nitrogen) [Peterson and Fry 1987]. Delta is expressed in terms of the numerator

(i.e., heavier stable isotope; e.g., 13C). Samples that have more negative delta

values contain less 13C or 15N relative to samples with higher delta values and

are termed depleted for an isotope. Samples that have higher delta values are

considered enriched.

Relative abundance of slimy sculpin collected during December 14-16,

1999, was estimated using catch-per-unit-effort (CPUE). In 2000, slimy sculpin

were collected between August 22 and September 13 from the same sites

described for the 1999 fish survey; these fish were sampled for whole-body

characteristics as outlined above. Additional sites were sampled during the 2000

fish survey (Table 2.1). In 2001, slimy sculpin were collected during August 20-

29 from a site downstream of the pulp mill effluent discharge point, and from

reference sites immediately upstream of the pulp mill diffuser and St. Hilaire

(Figure 2.1). Age data for male and female slimy sculpin collected in 2000 and

2001 was not available.

It is important to note that, in a previous fish survey, exposure to pulp mill

effluent was confirmed by the presence of 12-chlorodehydroabietic and 14-

chlorodehydroabietic resin acids in the bile of yellow perch and white sucker

captured downstream of the pulp mill effluent diffuser (same site used in the

present survey) [BAR Environmental Inc. 1996]. For the present study, stable-

42

isotope ratios of carbon (13C/12C) and nitrogen (15N/14N) measured in sculpin were

used to assess site fidelity and infer exposure.

2.3.3 Statistical analyses

Analysis of variance (ANOVA) was used to test for differences in mean

length, weight, and age of fish between sites; Tukey’s post hoc comparisons

were used to examine individual site differences when more than two sites were

compared. Analysis of covariance (ANCOVA) was used to assess site

differences in length-at-age, and the relationships between weight and length

(condition factor), liver size, and gonad size. Except for length-at-age and

condition, adjusted body weight (total wt – organ wt) was used as a covariate in

the ANCOVA model. Data were log10 transformed where appropriate before

performing ANOVA and ANCOVA, and sexes were analyzed separately. All data

analyses were done using SYSTAT® (SPSS, SYSTAT, Chicago, IL, USA)

statistical software. Gonad weights and liver weights were calculated as percent

adjusted body weight for summary purposes.

2.4 Results

2.4.1 October 1999

Male and female slimy sculpin and white sucker collected downstream of

the pulp mill showed few differences in age, when compared to fish from

reference sites (Tables 2.2, 2.3). Male sculpin collected downstream of the pulp

mill were significantly longer, heavier, had larger livers and higher condition

factors and had greater length-at-age (data not shown) relative to reference fish

43

from St. Hilaire and Clair (Tables 2.2, 2.4). Female sculpin from pulp mill

exposed and reference sites did not differ in length (Table 2.2), but exposed

female sculpin were heavier, had higher condition factors, greater length-at-age

and larger livers (livers and body weight higher relative to one reference site

only) (Tables 2.2, 2.4). No significant differences were found for male and female

gonad weights for either sculpin or white sucker collected from the Saint John

River. Both male and female white sucker downstream of the pulp mill showed

no significant differences in fork length, body weight, condition (males only), liver

size or length-at-age (data not shown) compared to reference sites. Male and

female white sucker collected at St. Hilaire and upstream and downstream of the

pulp mill had larger livers than fish collected from First Lake and Ogilvie Lake

(reference sites; see Figure 2.2). Male white sucker from First Lake and Ogilvie

Lake exhibited a LSI similar to reference suckers collected in northern Ontario

during the Moose River studies (Missinaibi, 0.85 ± 0.03 [Munkittrick et al. 2000]).

We found no significant differences in male and female sculpin length,

body weight, age, and gonad size for fish collected downstream of paper mill

effluent versus fish from Clair and St. Hilaire (Table 2.2). Female sculpin also

showed no differences in condition factor or liver size. However, male sculpin

from effluent-exposed sites had greater length-at-age compared to fish from Clair

and St. Hilaire, and larger livers relative to fish from St. Hilaire.

Male sculpin exposed to sewage effluent were longer, heavier, had higher

condition factors and had larger livers than fish from reference sites (Table 2.2).

Female sculpin collected near the sewage outfall were longer, heavier, older and

44

had greater length-at-age compared to reference fish, but did not differ in liver

size (Table 2.2). Female gonad development in exposed fish was reduced

relative to fish at Clair, but not St. Hilaire (Table 2.2).

The trends in stable isotope signatures for male and female sculpin

collected at each site were similar. Male and female sculpin downstream of the

Edmundston pulp mill and Madawaska paper mill and upstream of the

Madawaska River were enriched in 13C compared to fish collected at Clair and

St. Hilaire (Figure 2.3). Male and female sculpin downstream of the Edmundston

pulp mill were less 15N-depleted than fish collected downstream of the

Madawaska paper mill and upstream reference sites (Figure 2.3).

2.4.2 December 1999

Due to small gonad sizes in October, sculpin were collected again in

December when gonad development had progressed [prespawning female

sculpin had a gonadosomatic index (GSI) of 35.9% ± 2.8; Gray unpublished data]

and liver sizes were larger. We also added another sampling site, ~ 0.4 km

upstream of the pulp mill diffuser. Similar to October, male sculpin downstream

of the pulp mill were longer, heavier, and younger than fish at St. Hilaire (Table

2.2). These fish also had greater length-at-age (data not shown), but did not

differ significantly in gonad size. Female sculpin collected from within the effluent

plume were of similar age to reference fish (p=0.054), but were longer, heavier,

had larger livers, greater gonad size and higher condition and length-at-age,

compared to fish at St. Hilaire.

45

Approximately 85 sculpin were captured per h of electrofishing at the pulp

mill effluent exposed site, and 43 sculpin were captured per h of electrofishing at

the reference site (immediately upstream of the pulp mill diffuser). But only about

27 sculpin were caught per h of electrofishing at the reference site in St. Hilaire

(data not shown).

Results from the 1999 fish survey showed that white sucker collected from

various sites on the Saint John River exhibited few differences in liver size, but

white sucker from the Saint John River sites had larger livers compared to white

sucker from lake sites in Northern New Brunswick and reference rivers in

Northern Ontario. Slimy sculpin collected downstream of the Madawaska River

and pulp mill discharge had larger livers and faster growth (as evident from

otoliths analyses) compared to fish from upstream reference sites. Stable

isotope analysis suggested that the sculpin reflect the environmental conditions

at each sampling site; thus, they seem to have a relatively small home range.

The relatively small differences among sites, for white sucker, suggest either that

(1) white sucker are mobile, and thus do not reflect local environmental

conditions, or (2) there may be additional upstream inputs of pollutants. The

larger livers (without a concomitant increase in condition factor) of sculpin

collected at Clair, relative to Saint Hilaire reference sites, suggested that

upstream areas should be evaluated for potential contamination. This was done

in the fall of 2000.

46

2.4.3 Edmundston Pulp Mill - Fall 2000 Slimy sculpin

Female sculpin from sites where pulp mill effluent was present, in dilute

form, showed no difference in length and body weight relative to fish from

reference sites. However, they did have greater condition factors (Table 2.2;

Figure 2.4). Female sculpin from downstream of the pulp mill diffuser had a LSI

that was reduced relative to fish at Clair, but not relative to fish collected

immediately upstream of the pulp mill (p=0.06) or from St. Hilaire (p=0.08) (Table

2.2; Figure 2.5).

Sculpin were collected 15 km above the Clair collection site, immediately

downstream of a poultry processing operation in Saint-Francois-de-Madawaska,

NB. These fish had larger liver sizes (Figure 2.5) compared to fish from any of

the other sites sampled; condition factor was not different from sites immediately

upstream or downstream (Figure 2.4). Condition factor was lower in fish

downstream of major tributaries at the St. Francis River (Capone), Baker Brook,

and the Iroquois River, but not the Madawaska, which receives municipal

sewage (Figure 2.5).

Health-related parameters for sculpin from the Saint John River showed

significant differences within and between years (Table 2.2). Differences in the

magnitude of the parameters measured in fish between 1999 and 2000/2001

may have been related to inconsistencies in the sampling times; seasonal

changes in liver size, and possible changes in exposure would be expected due

to seasonal changes as well.

47

2.5 Discussion

The general response of slimy sculpin downstream of the sulphite pulp mill

suggests an overall increase in energy storage and utilization [Munkittrick et al.

2000; Gibbons and Munkittrick 1994]. Male sculpin exposed to pulp mill effluent

were longer, heavier, and in better condition than males from reference sites.

Female sculpin collected downstream of the pulp mill diffuser were heavier than

reference fish collected at St. Hilaire. Increases in body size have been

documented in fish exposed to pulp mill effluent [Gibbons et al. 1998a;

Munkittrick et al. 1994; Gibbons et al. 1998b], and may result from increased

water temperatures and/or indirect food web effects related to increased nutrient

concentrations in the exposure area [Gibbons et al. 1998b]. Fish collected near

some other pulp mills have shown delayed maturity, reduced body size, reduced

gonad size, and increased liver size [McMaster et al. 1991; Munkittrick et al.

1991; Munkittrick et al. 1992]. Although reproductive investment was difficult to

estimate, more sculpin were captured per h of electrofishing (as reflected by

CPUE) downstream of the pulp mill suggesting increased productivity.

The whole-organism parameters measured in slimy sculpin collected

downstream of the Edmundston pulp mill diffuser were not due to effluent toxicity.

A fathead minnow test was conducted during the same time period as our fish

survey [Parrott et al. 2000]. Results from the fathead minnow test showed that

when fish were exposed to 10% final effluent, an environmentally relevant

concentration survival was not affected, fish exhibited increased growth (length

and weight), and liver size was unaffected [Parrott et al. 2000]. The increased

48

energy storage and utilization observed in slimy sculpin collected downstream of

the pulp mill diffuser appeared to be the result of a nutrient enrichment effect

associated with the mill effluent. Benthic invertebrates contained in an artificial

mesocosm and exposed to Edmundston pulp mill effluent had a higher level of

emergence, compared to controls and N and P concentrations were greater

downstream of the mill, suggesting increased production due to enrichment [Culp

et al. 2003]. An enrichment effect also was noted in a previous environmental

assessment of the Edmundston pulp mill, which showed an increase in benthos

abundance and benthos richness downstream of the pulp mill diffuser [BAR

Environmental Inc. 1996]. Nutrient enrichment effects on benthic invertebrates

have been observed in other Canadian studies monitoring the potential impacts

of pulp mill effluent on the receiving environment. Lowell et al. [1996], for

example, reported that pulp mill effluent exposure resulted in significant

increases in growth and molting frequency of mayflies.

Establishing the residency patterns and home ranges of slimy sculpin is

important. The stable isotope data suggest that sculpin collected in 1999 from

the SJR near Edmundston had a small home range and the measured whole-

organism responses reflected local conditions. The δ13C and δ15N values of

sculpin collected downstream of the pulp mill suggest these sites were

isotopically enriched relative to reference sites in Clair and St. Hilaire (see Figure

2.3). Pulp mill effluent has been shown to be isotopically distinct from

background levels within a river, and may be a useful pulp mill effluent tracer

[Wassenar and Culp 1996]. In fact, δ13C values of sculpin downstream of the mill

49

closely resemble the carbon signature of terrestrial plants [Peterson and Fry

1987], indicating the terrestrial source of the carbon enrichment downstream of

the mill.

In 1999, water discharge was 7351 CFS during the fish survey

(http://waterdata.usgs.gov/me/nwis/uv/?site_no=01014000) and female sculpin

collected immediately upstream of the pulp mill did not differ much from fish

exposed to pulp mill effluent. However, during the fish survey in 2000, water

discharge was 1551 CFS and female sculpin collected upstream of the pulp mill

had greater LSI and lower condition factor scores, relative to fish that were

exposed to pulp mill effluent. Low water levels during the 2000 fish survey would

have resulted in sculpin being exposed to higher concentrations of pulp mill

effluent, compared to fish collected in 1999. If effluent from the Fraser

Edmundston pulp mill had a negative impact on fish performance in the receiving

environment, we would have expected to see fish with increased liversomatic

index (LSI) and a concomitant decrease in condition. However, this was not the

case. Understanding the potential influence of municipal sewage and non-point

sources of pollution on fish growth and reproductive performance in the upper

SJR is important and warrants further investigation.

2.5.1 White sucker

Male and female white sucker demonstrated different responses to the

effluent in terms of overall body characteristics when compared to slimy sculpin.

This was surprising, because both species were collected from the same sites

50

and should have had similar exposure regimes (see Table 2.5). Liver size did

not appear to be affected in white sucker collected on the Saint John River.

However, white sucker liver size was greater, relative to white sucker collected

from two pristine New Brunswick lakes (see Figure 2.2). The LSI of white sucker

from the two New Brunswick reference lakes were similar to LSI values for white

sucker collected from reference sites in Northern Ontario, and white suckers from

the Saint John River had livers similar in size to those of white sucker collected

downstream of a sulphite pulp mill in Kapuskasing, Northern Ontario [Munkittrick

et al. 2000]. The increase in white sucker liver size might be reflective of the

“normal” situation for the Saint John River, or could be due to one or more

unidentified upstream source(s) of contamination. Alternatively, the condition is

associated with the mobility of white sucker in this river system. Additional

investigation is needed to resolve among these possibilities. Presently, it is

evident that the elevated LSI of white sucker in the SJR near Edmundston is not

reflective of a “normal” situation. A potential source of contamination was

identified upstream of Clair during the 2000 fish survey, but, it is not yet clear

how this source limits or enhances fish performance in the upper SJR. The

upper SJR near Edmundston has no natural or man-made barriers to fish

movement; therefore, it is likely that the responses of white sucker reflect

movement of the species throughout the river. If so, measured differences

between reference fish and fish captured in the exposure area may not be

attributed to a particular source of contamination.

51

2.5.2 Ecological significance

Identifying ecologically relevant changes in fish organ size, growth rates, or

energy storage for the SJR will be important for understanding the potential of

future stressors to impact local fish populations [Munkittrick et al. 2000]. The

largest difference observed for LSI occurred in female sculpin in 1999; mean LSI

values went from 1.14 ± 0.13 at St. Hilaire to 2.47 ± 0.33 downstream of the pulp

mill, a 117% increase (see Table 2.2). The largest difference in condition factor

seen occurred in female sculpin during the fall of 2000; mean condition factor

went from 1.03 ± 0.02 at St. Hilaire to 1.18 ± 0.05 downstream of the pulp mill, a

15% increase (see Table 2.2). The Canadian EEM program for pulp mill

effluents recently completed the second cycle of monitoring [Munkittrick et al.

2002]. Fish comparisons are available in this data set for 65 pulp mill locations.

The 5% of sites that have the largest increase in liver size were above a 70%

increase, and the 5% of sites that had the greatest decrease in liver size were

greater than 31% decreases [Munkittrick et al. 2002]. For condition factor, the

5% of sites that demonstrated the largest increases in condition were greater

than 17% increase, and the 5% with the greatest decreases were more than a

5% decrease. Clearly, the differences seen in some years downstream of the

city of Edmundston are large, relative to changes seen at other sites.

No guidelines have yet been developed for interpreting the ecological

relevance of changes in organ size and condition. It is widely accepted that

individual or suborganismal changes that result in population or community level

changes are ecologically relevant. However, it can be very difficult to quantify

52

changes at population and community levels when sites are not far apart and

there are no barriers to fish movement between sites. The EEM program

recommends being able to detect a difference in parameters of >25% (10% of

condition factor) [Ribey et al. 2002]. The EEM program is working towards

developing critical effect sizes to determine when changes are considered

serious.

Other suggested means for determining if changes have ecological relevance

are based on the magnitude of differences that are observed. For example,

ecological relevance may be inferred if changes are outside the range seen at

reference sites, or if differences are greater than two standard deviations from

the average for fish from reference sites [Munkittrick et al. 2000]. The long-term

average for condition factor at St. Hilaire was 0.93 ± 0.11 for female sculpin. The

range of “normal” levels, based on 2 SD would be 0.71 to 1.15, and averages

greater than this were found in some collections. From this, we can conclude

that the differences downstream of the sewage and pulp mill discharges are

large, relative to what has been recorded at other sites, and can be outside of

what would be considered “normal” for this species in New Brunswick.

Overall, the results for slimy sculpin suggest that fish collected downstream of

municipal sewage and pulp mill effluent live in a nutrient-rich environment that

promotes growth. The site differences in sculpin energy storage and utilization

observed may also have been related to differences in water temperature, but

this is not known. At this point, the interim conclusions regarding the factors that

may limit or enhance fish performance in the SJR should be viewed as

53

hypotheses and will need to be confirmed through follow-up studies. The health-

related parameters measured in slimy sculpin that were not ecologically relevant

still need to be considered when evaluating the potential risks associated with

existing and future industrial developments that may impact fish performance

[Munkittrick et al. 2000].

2.6 Conclusions

Slimy sculpin downstream of the sewage discharges and pulp mill effluent

had greater growth, condition, and liver size, but no significant differences in

gonad size relative to reference fish. The results for white sucker were not

consistent with slimy sculpin, suggesting white sucker may not be a suitable

environmental monitoring species for this portion of the SJR. The responses of

slimy sculpin and white sucker in the present study were not consistent with

results from the cycle 1 EEM. The stable isotope signatures of slimy sculpin

exposed to pulp mill and paper mill effluent were different, which confirmed

previous plume delineation studies that demonstrated the paper mill and pulp mill

effluent were not mixing within the study area. The differences observed in the

whole-organism response of sculpin and suckers may be related to fish mobility.

Slimy sculpin whole-organism responses reflect site conditions and are a suitable

sentinel species for an effects-driven cumulative effects assessment of the Saint

John River. Together, results from the fish survey (present study), fathead

minnow (Pimephales promelas) flow-through bioassay [Parrott et al. 2000], and

the microcosm flow-through invertebrate exposure [Culp et al. 2003] suggest that

54

the increased energy storage and utilization of organisms exposed to pulp mill

effluent is related to increased nutrients.

2.7 Acknowledgements

This project is receiving funding from TSRI (Project 205), NBCFWRU,

NexFor and Fraser Papers Inc., Sir James Dunn Wildlife Research Centre Fund,

New Brunswick Wildlife Council Trust Fund, NWRI, Environment Canada

(Atlantic Region), Saint John River ACAP. We also acknowledge extensive in-

kind support from NexFor, Noranda Technology Centre, Fraser Papers, NBDOE,

NBDNRE, Maine DEP, DFO, and the Maine Chapter of the Nature Conservancy.

We acknowledge Stable Isotopes in Nature Lab at UNB. Student support from

NSERC, UNB, NB Female Mentorship Program, NB Student Employment

Program. KRM receives support from a NSERC Discovery Grant, and from the

Canada Research Chairs program.

55

Table 2.1. Location and description of Saint John River sites sampled for fish in 1999-2001. All sites are located in

Canada, except for Moody Bridge and Priestly Brook, which are located in Maine, USA.

Site ID Site numbera

Year sampled Site location Approximate distance from pulp mill outfall (km)

Description

Ogilvie L 18 1999 46°57.394’ N 66°54.769’ W

110 km northeast Reference lake located in northern New Brunswick.

First L 17 1999 47°30.567’ N 68°15.072’ W

31 km north Reference lake located on the Green River.

LF 16 2000 47°35.480’ N 68°07.462’ W

21 km northeast Little Forks (LF) Branch of Green River, a tributary of the Green River.

DS Pulp 15 1999-2001 47°21.337’ N 68°16.235’ W

1.3 km downstream Site located on the SJR downstream of the Fraser Inc. Edmundston pulp mill effluent diffuser.

DS Paper 14 1999 47°21.272’ N 68°16.358’ W

0.2 km across SJR Site located on the SJR ~ 4400 m downstream of the Fraser Inc. Madawaska paper mill effluent diffuser.

US Pulp 13 1999-2001 47°21.465’ N 68°16.541’ W

0.4 km upstream Site located on the SJR immediately upstream of the pulp mill diffuser and downstream of the Iroquois River.

IR 12 2000 47°21.683’ N 68°16.558’ W

0.6 km upstream Iroquois River (IR). A tributary of the SJR.

US Iroq 11 2000 47°21.594’ N 68°16.762’ W

0.7 km upstream Site located on the SJR upstream of the Iroquois River.

DS Mad 10 1999-2000 47°21.612’ N 68°19.532’ W

4 km upstream Site located on the SJR downstream of the Madawaska River

US Mad 9 1999-2000 47°21.505’ N 68°19.578’ W

4.7 km upstream Site located on the SJR upstream of the Madawaska River.

St. Hilaire 8 1999-2001 47°17.243’ N 68°22.862’ W

16 km upstream Site near St. Hilaire located upstream of the city of Edmundston.

DS Baker 7 2000 47°18.038’ N 68°42.605’ W

32 upstream Site located downstream of Baker Brook.

Clair 6 1999-2000 47°14.866’ N 68°36.317’ W

39 km upstream Site located near the town of Clair, New Brunswick.

DS Nad 5 2000 47°14.483’ N 68°42.605’ W

44 km upstream Site located on the SJR downstream of poultry processing facility.

US Nad 4 2000 47°12.865’ N 68°48.956’ W

44 km upstream Site located upstream of poultry processing facility.

56

Site ID Site numbera

Year sampled Site location Approximate distance from pulp mill outfall (km)

Description

Capone 3 2000 47°11.196’ N 68°52.912’ W

52 km upstream Site located downstream of the St. Francis River.

PB 2 2000 46°49.533’ N 68°32.700’ W

101 km upstream Priestly Brook (PB). Reference site located on the SJR in the State of Maine.

MB 1 2000 46°37.733’ N 69°46.800’ W

123 km upstream Moody Bridge (MB). Reference site located on the SJR in the State of Maine.

aSite numbers correspond to sites shown in Fig. 1. DS=Downstream; US=Upstream.

57

Table 2.2. Means ± SE (n) of various parameters of adult male and female slimy sculpin (Cottus cognatus) collected

downstream of a pulp mill (DS Pulp), paper mill (DS Paper), and municipal sewage (DS Mad) and from reference

sites located at Clair, St. Hilaire, and immediately upstream of the mill (US Pulp). Within a row, differences (p <

0.05) among sites are denoted by different uppercase letters.

Sex, Date

Parameter Clair St. Hilaire DS Mad US Pulp DS Pulp DS Paper

Male Length (mm) 65.3 ± 1.2 (22)A 68.2 ± 1.4 (47)AB 73.7 ± 1.5 (20)BC - 77.3 ± 2.0 (11)C 69.4 ± 3.4 (17)ABC Oct

1999 Weight (g) 2.38 ± 0.13 (22)A 2.98 ± 0.19 (47)A 4.19 ± 0.24 (20)BC - 4.65 ± 0.29 (11)C 3.46 ± 0.53 (17)AB Age (y) 2.3 ± 0.2 (18)A 2.3 ± 0.2 (34)A 2.0 ± 0.0 (11)A - 2.1 ± 0.2 (11)A 2.1 ± 0.3 (17)A Ka 0.84 ± 0.01 (22)* 0.89 ± 0.01 (43)* 1.03 ± 0.02 (20)* - 1.00 ± 0.02(11)* 0.92 ± 0.02(17)* LSIb 0.96 ± 0.05 (22)AB 0.83 ± 0.07 (43)A 1.09 ± 0.08 (20)BC - 1.52 ± 0.12(11)C 1.23 ± 0.13(17)BC GSIc 1.16 ± 0.07 (22)A 1.32 ± 0.09 (42)A 1.22 ± 0.11 (20)A - 1.30 ± 0.08(11)A 1.42 ± 0.14(17)A

Length (mm) - 72.5 ± 1.0 (18)A - - 81.9 ± 2.1 (20)B - Dec 1999 Weight (g) - 3.62 ± 0.13 (18)A - - 6.04 ± 0.38 (20)B - Age (y) - 2.9 ± 0.2 (18)A - - 2.4 ± 0.2 (20)B - K - 0.95 ± 0.02 (18)A - - 1.08 ± 0.02 (20)B - LSI - 1.20 ± 0.07 (18)A - - 1.44 ± 0.06 (19)A - GSI - 1.32 ± 0.08 (18)A - - 1.41 ± 0.08 (20)A -

Length (mm) 82.6 ± 1.1(18)B 73.4 ± 3.5 (9)A - 78.3 ± 3.1 (10)AB - - Sept 2000 Weight (g) 6.04 ± 0.24(18)B 4.58 ± 0.65 (9)A - 5.38 ± 0.78 (10)AB - -

K 1.07 ± 0.02(18)A 1.10 ± 0.02 (9)A - 1.06 ± 0.03 (10)A - - LSI 2.08 ± 0.15(18)B 1.60 ± 0.15 (9)AB - 1.38 ± 0.13 (10)B - -

58

Sex, Date

Parameter Clair St. Hilaire DS Mad US Pulp DS Pulp DS Paper

GSI 0.49 ± 0.05(18)B 0.49 ± 0.06 (9)AB - 0.72 ± 0.06 (10)B - -

Length (mm) - 71.6 ± 1.9 (21)A - 73.6 ± 1.5 (19)A 73.7 ± 1.7 (25)A - Aug 2001 Weight (g) - 3.89 ± 0.38 (21)A - 4.34 ± 0.28 (19)A 4.45 ± 0.37 (25)A - K - 1.01 ± 0.01 (21)A - 1.06 ± 0.02 (19)B 1.07 ± 0.02 (25)B - LSI - 1.85 ± 0.10 (21)A - 1.61 ± 0.09 (19)B 2.10 ± 0.11 (25)A - GSI - - - - - -

Female Length (mm) 62.4 ± 1.8 (21)A 57.9 ± 1.5 (17)A 75.1 ± 2.2 (19)B - 63.4 ± 2.5 (13)A 63.3 ± 2.2 (12)A Weight (g) 2.09 ± 0.17 (21)AC 1.65 ± 0.13 (17)A 4.35 ± 0.42 (19)B - 2.60 ± 0.31 (13)C 2.28 ± 0.23 (12)AC Oct

1999 Age (y) 1.5 ± 0.2 (20)A 1.4 ± 0.2 (12)A 2.6 ± 0.2 (14)B - 1.5 ± 0.3 (13)A 1.5 ± 0.2 (12)A K 0.83 ± 0.02(21)A 0.83 ± 0.02(13)A 0.97 ± 0.02 (19)BC - 0.96 ± 0.03(13)C 0.85 ± 0.02(12)AC LSI 1.61 ± 0.08(21)AC 1.14 ± 0.13(13) B 1.40 ± 0.12 (19)AB - 2.47 ± 0.33(13)C 1.56 ± 0.14(12)ABC GSI 1.38 ± 0.10(21)A 1.10 ± 0.11(13)AB 1.20 ± 0.12 (20)B - 1.35 ± 0.23(13)AB 1.30 ± 0.12(12)A

Length (mm) - 64.6 ± 1.6 (17)A - 71.7 ± 3.0 (9)B 70.7 ± 1.8 (18)B - Dec 1999 Weight (g) - 2.43 ± 0.18 (17)A - 3.67 ± 0.41 (9)B 3.52 ± 0.26 (18)B -

Age (y) - 2.6 ± 0.2 (16)A - 2.6 ± 0.2 (9)A 2.6 ± 0.2 (18)A -

K - 0.88 ± 0.02 (17)A - 0.98 ± 0.04 (9)B 0.98 ± 0.02 (18)B -

LSI - 2.94 ± 0.09 (17)A - 3.17 ± 0.11 (9)AB 3.34 ± 0.09 (18)B -

GSI - 3.79 ± 0.20 (17)A - 5.44 ± 0.25 (9)B 4.52 ± 0.26 (18)AB -

Length (mm) 79.2 ± 2.4 (10)A 66.1 ± 2.2 (15)B - 74.1 ± 2.1 (16)A 70.4 ± 1.2 (5)AB - Sept 2000 Weight (g) 5.46 ± 0.47 (10)A 3.13 ± 0.35 (15)B - 4.47 ± 0.39 (16)A 4.10 ± 0.07 (5)AB -

K 1.08 ± 0.03 (10)AB 1.03 ± 0.02 (15)B - 1.05 ± 0.02 (16)B 1.18 ± 0.05 (5)A -

LSI 2.93 ± 0.19 (10)B 2.33 ± 0.19 (15)AB - 2.30 ± 0.12 (16)AB 1.61 ± 0.20 (4)A -

59

Sex, Date

Parameter Clair St. Hilaire DS Mad US Pulp DS Pulp DS Paper

GSI 0.69 ± 0.03 (10)A 0.79 ± 0.06 (10)AB - 1.08 ± 0.08 (13)B 0.78 ± 0.08 (5)AB -

Length (mm) - 64.8 ± 1.9 (18)A - 72.5 ± 1.9 (19)B 60.5 ± 1.5 (11)A - Aug 2001 Weight (g) - 2.75 ± 0.26 (18)A - 4.12 ± 0.34 (19)B 2.30 ± 0.18 (11)A -

K - 0.97 ± 0.02 (18)A - 1.04 ± 0.02 (19)B 1.02 ± 0.02 (11)AB -

LSI - 2.48 ± 0.17 (18)A - 2.03 ± 0.16 (19)A 2.39 ± 0.30 (11)A -

GSI - - - - - - aK = 100000*(body weight/(total length3)); bLSI = 100*(liver weight/(body weight – liver weight)); cGSI = 100*(gonad weight/(body weight – gonad weight)); *Significant interaction term within ANCOVA model; (-) Data not available.

60

Table 2.3. Means ± SE (n) of various parameters of adult male and female white

sucker (Catostomus commersoni) collected in October 1999. Within a

row, differences (p < 0.05) among sites are denoted by different

uppercase letters.

Reference sites Study site Sex Parameter St. Hilaire Upstream Pulp Mill Downstream Pulp Mill

Male Fork length (cm) 37.0 ± 0.5(20)A 39.6 ± 0.4(20)B 38.4 ± 0.5(20)AB Body weight (g) 676 ± 26(20)A 793 ± 23(20)B 736 ± 29(20)AB Age (y) 5.7 ± 0.3(18)A 5.9 ± 0.2(20)A 5.8 ± 0.2(19)A Ka 1.33 ± 0.02(20)A 1.28 ± 0.02(20)A 1.29± 0.02(20)A LSIb 1.36 ± 0.06(20)A 1.28 ± 0.03(20)A 1.38 ± 0.07(20)A GSIc 4.10 ± 0.19(20)A 4.59 ± 0.25(20)A 4.91 ± 0.18(20)A Female Fork length (cm) 42.4 ± 0.3(20)A 41.5 ± 0.6(20)A 41.9 ± 0.5(21)A Body weight (g) 1025 ± 27(20)A 913 ± 34(20)B 943 ± 33(21)AB Age (y) 7.4 ± 0.4(20)A 6.2 ± 0.3(19)A 6.2 ± 0.3(20)A K 1.33 ± 0.01(20)B 1.28 ± 0.02(20)A 1.28 ± 0.02(21)A LSI 1.83 ± 0.06(20)A 1.67 ± 0.05(20)A 1.73 ± 0.04(21)A GSI 6.90 ± 0.35(20)A 6.00 ± 0.26(20)A 6.36 ± 0.30(21)A Fecundity, no. eggs 22960 ± 1103(20)A 20724 ± 1096(20)A 22336 ± 1359(20)A aK = 100000*(body weight/(fork length3)); bLSI = 100*(liver weight/(body weight – liver weight)); cGSI = 100*(gonad weight/(body weight – gonad weight)).

61

Table 2.4. Regression estimates for adult male slimy sculpin (Cottus cognatus)

condition factor and adult female slimy sculpin length-at-age.

Date Sex Parameter Site type Site location

Slope Intercept n p r2

October 1999

M Condition Factor Reference Clair 2.81 -4.74 22 <0.0001 0.95

M Condition Factor Reference St. Hilaire 3.21 -5.43 47 <0.0001 0.96 M Condition Factor Study DS Mad 2.57 -4.18 20 <0.0001 0.88 M Condition Factor Study DS Pulp 2.46 -3.98 11 <0.0001 0.96 M Condition Factor Study DS Paper 3.11 -5.23 17 <0.0001 0.98 F Length-at-age Reference Clair 0.25 1.76 20 <0.0001 0.82 F Length-at-age Reference St. Hilaire 0.21 1.75 12 <0.0001 0.89 F Length-at-age Study DS Mad 0.45 1.70 14 <0.0001 0.89 F Length-at-age Study DS Pulp 0.28 1.74 13 <0.0001 0.81 F Length-at-age Study DS Paper 0.21 1.78 12 <0.0001 0.75

62

Table 2.5. Comparative summary of slimy sculpin (Cottus cognatus) and white

sucker (Catostomus commersoni) collected downstream of the sulphite

pulp mill relative to fish collected from St. Hilaire (reference site), Saint

John River, October 1999, New Brunswick, Canada (modified from

Gibbons et al. 1998)a.

Slimy sculpin White sucker Parameter Male Female Male Female Length + 0 0 0 Body weight + + 0 0 Condition + + 0 - Age 0 0 0 0 Length-at-age + + 0 0 GSI 0 0 0 0 LSI + + 0 0 a (0) = no change; (+) = significant increase; (-) = significant decrease.

63

Figure 2.1. Map of the study area showing the relative location of reference and

exposure fish collection sites (not to scale).

Site 1, 123 kmSite 2, 101 km

St. Francis River

Conners

St. Francois-de-MadawaskaClair

St. Hilaire

Madawaska River Iroquois River

Baker Brook

Green River

Saint John Rive

r

Quebec, Canada

Provincial Border

New Brunswick, Canada

Maine, USA

Pulp mill

Paper mill

Pulp mill effluent lagoon system and outfall

Paper mill effluent lagoon system and outfall

Municipal sewage lagoonand outfall

34

56

78

910

1112 13

1514

1617

Site 18,180 km

N

Site 1, 123 kmSite 2, 101 km

St. Francis River

Conners

St. Francois-de-MadawaskaClair

St. Hilaire

Madawaska River Iroquois River

Baker Brook

Green River

Saint John Rive

r

Quebec, Canada

Provincial Border

New Brunswick, Canada

Maine, USA

Pulp mill

Paper mill

Pulp mill effluent lagoon system and outfall

Paper mill effluent lagoon system and outfall

Municipal sewage lagoonand outfall

34

56

78

910

1112 13

1514

1617

Site 18,180 km

N

64

Figure 2.2. Relative liver size (liversomatic index [LSI]; % of body weight) of

adult male (black bars) and female (white bars) white sucker collected

downstream of the Edmundston pulp mill (DS Pulp) and at reference

sites located upstream of the pulp mill (US Pulp), St. Hilaire, Ogilvie

Lake (males only), and First Lake during the Fall, October 1999. Refer

to Figure 2.1 map for location.

0.0

0.4

0.8

1.2

1.6

2.0

First L Ogilvie L St. Hilaire US Pulp DS Pulp

LSI (

%)

65

Figure 2.3. Stable-isotope ratios of carbon and nitrogen in adult male (bold

dashed lines) and female (solid lines) slimy sculpin collected from

reference sites at Clair, St. Hilaire, and upstream of the Madawaska

River (US Mad) and from downstream of the Edmundston pulp mill (DS

Pulp) and Madawaska paper mill (DS Paper), October 1999. Values

are expressed as deviations (δ) from standards. Refer to Figure 2.1

map for location.

7.5

8

8.5

9

9.5

10

10.5

11

11.5

12

-28 -27 -27 -26 -26 -25 -25 -24 -24 -23 -23 -22 -22 -21 -21 -2013C 0/00

15N

0/0

0

DS Pulp

DS Paper

US Mad

Clair St. Hilaire

66

Figure 2.4. Summary of "ecologically relevant" changes in condition factor in

female slimy sculpin collected during the Fall 2000 fish survey of the

Saint John River. Fish were collected from various sites, including:

Moody Bridge (MB), Priestly Brook (PB), a site downstream of the St.

Francis River (Capone), sites located upstream (US Nad) and

downstream (DS Nad) of a poultry processing plant, upstream of the

international bridge at Clair, downstream of Baker Brook (DS Baker), St.

Hilaire, upstream of the Madawaska River (US Mad), downstream of

the Madawaska River (DS Mad), upstream of the Iroquois River (US

Iroq), Iroquois River (IR), upstream of the pulp mill diffuser (US Pulp),

downstream of the pulp mill diffuser (DS Pulp), and the Little Forks (LF).

Cross-hatched bars represent reference sculpin collected from sites on

the Saint John River. Black bars represent sculpin from sites exposed

to either poultry processing waste effluent (DS Nad), municipal sewage

wastewater (DS Mad), and pulp mill effluent (DS Pulp). White bars

represent sculpin collected from tributaries. Refer to Figure 2.1 map for

location.

67

0.60

0.70

0.80

0.90

1.00

1.10

1.20

1.30M

B

PB

Cap

one

US

Nad

DS

Nad

US

Cla

ir

DS

Bak

er

St.H

ilaire

US

Mad

DS

Mad

US

Iroq

US

Pul

p

DS

Pul

p IR LF

Con

ditio

n Fa

ctor

(K)

Mainstem Saint John River Tributaries

68

Figure 2.5. Summary of "ecologically relevant" changes in liver size female slimy

sculpin collected during the Fall 2000 fish survey of the Saint John

River. Solid and stippled horizontal lines represent data (i.e., mean ±

25%, mean ± 2SD, respectively) collected at St. Hilaire. Values that fall

between the horizontal lines are considered “normal” for the SJR at the

time of sampling. Cross-hatched bars represent reference sculpin

collected from sites on the Saint John River. Black bars represent

sculpin from sites exposed to either poultry processing waste effluent

(DS Nad), municipal sewage wastewater (DS Mad), and pulp mill

effluent (DS Pulp). White bars represent sculpin collected from

tributaries. Asterisk (*) indicates possible upstream source of

contamination. Refer to Figure 2.1 map for location.

69

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

4.00

4.50

MB

PB

Cap

one

US

Nad

DS

Nad

US

Cla

ir

DS

Bak

er

St.H

ilaire

US

Mad

DS

Mad

US

Iroq

US

Pul

p

DS

Pul

p IR LF

LSI (

%)

Mainstem Saint John River Tributaries

*

70

2.8 References

BAR Environmental Inc. 1994. Environmental effects monitoring pre-design

historical information for Fraser Inc., Edmundston sulphite pulp mill.

Guelph, ON, Canada.

BAR Environmental Inc. 1996. Environmental effects monitoring for Fraser Inc.

Edmundston sulphite pulp mill final report. Guelph, ON, Canada.

Culp, J.M., Cash, K.J., Glozier, N.E., and Brua, R.B. 2003. Effects of pulp mill

effluent on benthic assemblages in along the Saint John River, Canada.

Environ Toxicol Chem 22: 2916-2925.

Environment Canada. 1997. Fish Survey Expert Working Group:

Recommendations from Cycle 1 review. Ottawa ON: Environment Canada.

EEM/1997/6. 262p.

Frank, M., McMaster, M.E., Munkittrick, K.R., Savoie, M.C., and Wood, C.

Effects of sulphite and bleached kraft pulp and paper mill effluents on

yellow perch and Johnnie darters. 25th Aquatic Toxicity Workshop,

Quebec, QC, Canada, October 18-21, 1998, pp 53.

Gibbons, W.N., and Munkittrick, K.R. 1994. A sentinel monitoring framework for

identifying fish population responses to industrial discharges. J Aquat

Ecosyst Health 3: 227-237.

Gibbons, W.N., Munkittrick, K.R., and Taylor, W.D. 1998a. Monitoring aquatic

environments receiving industrial effluents using small fish species 1:

Response of spoonhead sculpin (Cottus ricei) downstream of a bleached-

kraft pulp mill. Environ Toxicol Chem 17: 2227-2237.

71

Gibbons, W.N., Munkittrick, K.R., McMaster, M.E., and Taylor, W.D. 1998b.

Monitoring aquatic environments receiving industrial effluents using small

fish species 2: Comparison between responses of trout-perch (Percopsis

omiscomaycus) and white sucker (Catostomus commersoni) downstream

of a pulp mill. Environ Toxicol Chem 17: 2238-2245.

Hewitt, L.M., Pryce Hobby, A.C., Parrott, J.L., Marlatt, V., Wood, C., Oakes, K.,

and Van Der Kraak, G. 2003. Accumulation of ligands for aryl

hydrocarbon and sex steroid receptors in fish exposed to treated effluent

from a bleached sulphite/groundwood pulp mill. Environ Toxicol Chem 22:

2890-2897.

Lowell, R.B., Culp, J.M., Wrona, F.J., and Bothwell, M.L. 1996. Effects of pulp

mill effluents on benthic freshwater invertebrates: food availability and

stimulation of increased growth and development. In Servos MR,

Munkittrick KR, Carey JH, Van Der Kraak GJ, (eds.), Environmental fate

and effects of pulp and paper mill effluents. St. Lucie Press, Delray

Beach, FL, USA, pp 525-532.

MacKay, W.C., Ash, G.R., and Norris, H.J. (eds.). 1990. Fish ageing methods

for Alberta. RL&L Environmental Services, Edmonton, AB, Canada.

McMaster, M.E., Van Der Kraak, G., Portt, C.B., Munkittrick, K.R., Sibley, P.K.,

Smith, I.R., and Dixon, D.G. 1991. Changes in hepatic mixed-function

oxygenase (MFO) activity, plasma steroid levels and age at maturity of a

white sucker (Catostomus commersoni) population exposed to bleached

kraft pulp mill effluent. Aquat Toxicol 21: 199-218.

72

Munkittrick, K.R., and Dixon, D.G. 1989a. An holistic approach to ecosystem

health assessment using fish population characteristics. Hydrobiologia

188/189: 122-135.

Munkittrick, K.R., and Dixon, D.G. 1989b. Use of white sucker populations to

assess the health of aquatic ecosystems exposed to low-level contaminant

stress. Can J Fish Aquat Sci 46: 1455-1462.

Munkittrick, K.R., Portt, C.B., Van Der Kraak, G.J., Smith, I.R., and Rokosh, D.A.

1991. Impact of bleached kraft mill effluent on population characteristics,

liver MFO activity, and serum steroid levels of a Lake Superior white

sucker (Catostomus commersoni) population. Can J Fish Aquat Sci

48:1371-1380.

Munkittrick, K.R., McMaster, M.E., Portt, C.B., Van Der Kraak, G., Smith, I.R.,

and Dixon, D.G. 1992. Changes in maturity, plasma sex steroid levels,

hepatic mixed function oxygenase activity, and the presence of external

lesions in lake whitefish (Coregonus clupeaformis) exposed to bleached

kraft mill effluent. Can J Fish Aquat Sci 49: 1560-1569.

Munkittrick, K.R., Van Der Kraak, G., McMaster, M.E., Portt, C.B., van den

Heuvel, M.R., and Servos, M.R. 1994. Survey of receiving-water

environmental impacts associated with discharges from pulp mills. 2.

Gonad size, liver size, hepatic EROD activity, and plasma sex steroid

levels in white sucker. Environ Toxicol Chem 13: 1089-1101.

Munkittrick, K.R., McMaster, M.E., McCarthy, L.H., Servos, M.R., and Van Der

Kraak, G. 1998. An overview of recent studies on the potential of pulp

73

mill effluents to impact reproductive function in fish. J Toxicol Environ

Health Part B 1:101-125.

Munkittrick, K.R., and McMaster, M.E. 2000. Effects-driven assessment of

multiple stressors using fish populations. In: Ferenc, S.A., and Foran,

J.A., (eds.), Multiple stressors in ecological risk and impact assessment:

Approach to risk estimation. SETAC Press, Pensacola, FL, USA, pp 27-

65.

Munkittrick, K.R., McMaster, M.E., Van Der Kraak, G., Portt, C., Gibbons, W.N.,

Farwell, A., and Gray, M. 2000. Development of methods for effects-

driven cumulative effects assessment using fish populations: Moose River

Project. SETAC Press, Pensacola, FL, USA.

Munkittrick, K.R., McGeachy, S.A., McMaster, M.E., Courtenay, S.C. 2002.

Overview of cycle 2 freshwater fish studies from the pulp and paper

Environmental Effects Monitoring program. Water Quality Res J Can 37:

49-77.

Parrott, J.L., Wood, C.S., Boutot, P., Blunt, B.R., Baker, M.A., and Dunn, S.

Fathead minnow long-term growth/reproductive tests to assess final

effluent from a bleached sulphite mill. 4th International conference on

environmental impacts of the pulp and paper industry, Helsinki, Finland,

June 12-15, 2000, pp 207-212.

Peterson, B.J., and Fry, B. 1987. Stable isotopes in ecosystem studies. Ann

Rev Ecol Syst 18: 293-320.

74

Ribey, S.C., Munkittrick, K.R., McMaster, M.E., Courtenay, S., Langlois, C.,

Munger, S., Rosaasen, A., and Whitley, G. 2002. Development of a

monitoring design for examining effects in wild fish associated with

discharges from metal mines. Water Quality Res. J. Can. 37: 229-249.

Scott, W.B., and Crossman, E.J. 1998. Freshwater Fishes of Canada. Galt

House, Oakville, ON, Canada.

Van Vliet, W.H. 1964. An ecological study of Cottus cognatus (Richardson,

1836) in Northern Saskatchewan. MA Thesis. University of

Saskatchewan, Saskatoon, Saskatchewan, Canada.

Wassenar, L.I., and Culp, J.M. 1996. The use of stable isotope analyses to

identify pulp mill effluent signatures in riverine food webs. In Servos,

M.R., Munkittrick, K.R., Carey, J.H., Van Der Kraak, G., (eds),

Environmental Fate and Effects of Pulp and Paper Mill Effluents. St. Lucie

Press, Delray Beach, FL, USA, pp 413-424.

75

CHAPTER 3. Identifying a suitable fish species for monitoring a large

river receiving effluents from a pulp and paper mill, municipal

sewage wastewater facilities, and agricultural runoff2.

2Published: Galloway, B.J., Munkittrick, K.R., Curry, R.A., Wood, C., and Dunn,

S. 2004. Identifying a suitable fish species for monitoring a large river receiving

effluents from pulp and paper mill, municipal sewage wastewater facilities, and

agricultural runoff. In Borton, D.L., Hall, T.J., Fisher, R.P., and Thomas, J.F.,

(eds) Fifth International Conference on Fate and Effects of Pulp and Paper Mill

Effluents, June 1-4, Seattle, WA: 169-181.

3.1 Abstract

Identifying the impacts of individual aquatic stressors on fish health in rivers

exposed to multiple stressors is difficult and has become the focus of recent

Canadian studies on environmental effects. There has been a movement over

the past 5 years towards using small-bodied fish species for monitoring, because

of their presumed lower mobility. There have not been many direct comparisons

between the results of monitoring studies using small and large-bodied species.

The objective of this project was to determine whether differences in species

responses exist, and whether they can be explained by differences in life history

characteristics and mobility. This paper will compare the responses of 4 fish

species upstream and downstream of a city in a river receiving wastes from a

pulp mill, paper mill, multiple sewage discharges and agricultural runoff. The

study outlines data gaps associated with ongoing studies.

76

3.2 Introduction

Identifying the potential impacts of individual aquatic stressors on fish

health in rivers exposed to multiple stressors is a difficult task. Over the past few

years there has been increased interest in using small-bodied fish (e.g., Cottids

and Cyprinids) for monitoring the potential impacts of point source [Frank et al.

1998; Gibbons et al. 1998a; Gibbons et al. 1998b; Munkittrick et al. 2000;

Munkittrick et al. 2002] and non-point sources of contaminants [Gray et al. 2002].

In Canada, a recent review of cycle 2 fish studies (1997-2000) for the pulp and

paper Environmental Effects Monitoring (EEM) program noted an increase in the

use of small-bodied fish from cycle 1 (1993-1996) [Munkittrick et al. 2002]. Many

small-bodied fish species are well suited for monitoring complex receiving

environments since they are presumed to be less mobile and more abundant

than large-bodied fish species (e.g., Catostomids). The main disadvantage of

using small-bodied fish species for EEM programs is the lack of basic life-history

information [Munkittrick et al. 2002]. A recent literature survey of the Aquatic

Sciences and Fisheries Abstracts (Cambridge Scientific Abstracts, Bethesda,

MD.), yielded a paucity of detailed reproductive information for some of the small-

bodied fish used in the cycle 2 pulp and paper EEM program, including:

blacknose dace (Rhinichthys atratulus), bluntnose minnow (Pimephales notatus),

logperch (Percina caproides), and creek chub (Semotilus corporalis). The lack of

detailed life history information for small-bodied fish can hinder the design and

interpretation of environmental monitoring programs.

77

In 1999, a collaborative research agreement was initiated between Fraser

Papers Inc. (Edmundston pulp mill, Edmundston, New Brunswick, Canada),

Nexfor (Noranda Technology Centre, Pointe-Claire, Quebec, Canada), and

Environment Canada (National Water Research Institute, Saskatoon,

Saskatchewan, Canada) to participate in a variety of studies as part of Fraser

Papers Inc. commitment to Environmental Effects Monitoring (EEM) studies for

cycle 2. The studies including an on-site 90-d flow-through fathead minnow

(Pimephales promelas) bioassay [Parrott et al. 2003], a flow-through microcosm

invertebrate exposure [Culp et al. 2003], an assessment of bioavailable pulp mill

effluent chemicals accumulated by fish during short term exposures [Hewitt et al.

2003], and wild fish collections [Galloway et al. 2003].

The objectives of the present study were to clarify local impacts on wild

fish, expand field collections to include other species to examine potential

species differences in responses, continue to de-couple other confounding

factors at this study area using less mobile small-bodied fish, and contribute to

furthering the scientific understanding of the suitability of small-bodied fish

species for environmental monitoring programs. We compared the responses of

blacknose dace and slimy sculpin (small-bodied fish species) with yellow perch

and white sucker (large-bodied fish species) to determine whether the responses

of small-bodied fish and large-bodied fish would be different.

78

3.3 Materials and Methods

3.3.1 Study area and mill characteristics

The Saint John River originates at Fifth St. John Pond in northern Maine

(USA) and travels ~700 km to the mouth, located at the city of Saint John, at the

Bay of Fundy. Approximately 51% of the drainage area is located in New

Brunswick; while the remaining area is located in the state of Maine (36%) and

the province of Quebec (13%) [11]. Near Edmundston (approximately 450 km

from the river mouth), the Saint John River forms part of the international border

between Canada and the State of Maine and extends for ~120 km. The flow of

the Saint John River near Edmundston is unregulated and undergoes

considerable annual fluctuations; the dominant substrate in the river is gravel

[11]. There are no barriers to fish movement on the mainstem of the Saint John

River near Edmundston. However, a dam located near the mouth of the

Madawaska River, upstream of the mill discharge, restricts the movement of fish

in the tributary.

The bleached-sulphite pulp mill (Edmundston, New Brunswick, Canada)

and paper mill (Madawaska, Maine, USA) are located on the north and south

sides of the Saint John River, respectively (Figure 3.1). The pulp mill produces

daily approximately 550 air-dried metric tonnes (ADMT) of sulphite pulp, 350

ADMT of ground wood pulp, 120 ADMT of boxboard, and 100 tonnes of de-inked

product [BAR Environmental Inc. 1994]. At the time of our survey, the wood

furnish consisted of ~70% fir and ~30% spruce and the bleaching sequence was

(Dc-Ep-DH-): Dc = chlorination stage with 75% chlorine dioxide substitution; Ep =

79

extraction stage with sodium hydroxide reinforced with hydrogen peroxide; DH =

bleaching stage with chlorine dioxide and sodium hypochlorite; (-) = washing

stage. The pulp mill discharges more than 73,000 m3 of secondary-treated

effluent per day from an aerated lagoon facility (retention time ~7 days) into the

north side of the Saint John River, about 4 km downstream from the mill. The

paper mill discharges secondary-treated effluent into the south side of the Saint

John River, about 4 km upstream of the pulp mill effluent diffuser. A plume

delineation study has shown the paper mill effluent and pulp mill effluent do not

mix in our fish collection area [11].

3.3.2 Fish collections

Adult fish of each sex were targeted during each site collection.

Prespawning slimy sculpin (Cottus cognatus) were collected between March 4 to

26, 2002 and prespawning blacknose dace (Rhinichthys atratulus) were collected

between June 11 to 14, 2002 using a backpack electrofisher (Smith-Root Model

12B, Smith-Root Inc., Vancouver, WA, USA). Fish were collected from a site

located below a bleached-sulphite pulp mill (D/S Pulp; lat 47°21.337’N, long

68°16.235’W) and a site located downstream of municipal sewage inputs (D/S

Mad; 47°21.612’W, long 68°19.532’W). Reference fish were collected from a site

immediately upstream of the pulp mill diffuser (U/S Pulp; lat 47°21.465’N,

68°16.541’W), a site ~ 20 km upstream of the mill diffuser (St. Hilaire; lat

47°17.243’N, long 68°22.862’W), and a site located ~ 52 km upstream of the

town of Edmundston, below the St. Francis River (“Capone”; lat 47°11.196’N,

80

long 68°52.912’W). Ice cover in March prohibited the collection of slimy sculpin

from the reference site located at St. Hilaire. Male blacknose dace were not

captured from the site below the St. Francis River or immediately upstream of the

pulp mill effluent diffuser. Adult yellow perch (Perca flavescens) were collected

by angling between September 20-24, 2002 and white sucker (Catostomus

commersoni) were collected using 3” monofilament gillnets between October 18-

23, 2002 from a site located downstream of the pulp mill and from a reference

site at St. Hilaire.

Fish were rendered unconscious by concussion, followed by spinal

severance, and length, body weight, liver weight, and gonad weights were

recorded. Aging material was collected for determination of survival (i.e., mean

age) and growth (i.e., length-at-age). The age of white sucker and yellow perch

was determined by counting annuli on clean, dried opercula under a dissecting

microscope. Slimy sculpin and blacknose dace were aged by counting annuli on

sagittal otoliths. The fecundity of female slimy sculpin was estimated in the field

by counting the total number of eggs per fish. For female white sucker and

yellow perch, approximately 1 g of ovarian tissue was collected in the field and

placed in a 10 % buffered formalin solution. In the lab, the ovarian tissue was

blotted dry, reweighed, and the total number of eggs was counted. These results

were used to estimate the total number of eggs per fish (total fecundity).

Fecundity estimates for female blacknose dace were not available. White

muscle samples were obtained for stable isotope analyses and were used to

estimate residency patterns of fish collected at various sites in our survey.

81

3.3.3 Data Analyses

Analysis of variance (ANOVA) was used to test for differences in mean

length, weight, and age of fish between sites; Bonferroni post hoc comparisons

were used to examine individual site differences when more than two sites were

compared. Analysis of covariance (ANCOVA) was used to assess site

differences in length-at-age, and the relationships between weight and length

(condition factor), liver size, and gonad size. Except for length-at-age, condition

and stable isotope ratio analyses, adjusted body weight (total wt – organ wt) was

used as a covariate in the ANCOVA model. All data were log10 transformed

before performing ANOVA and ANCOVA, and sexes were analyzed separately.

All data analyses were done using SYSTAT® (SPSS, SYSTAT, Chicago, IL,

USA) statistical software. For ANCOVA, slopes were consider different when p <

0.01. Gonad weights and liver weights were calculated as percent-adjusted body

weight for summary purposes.

When gonad size of female dace was analyzed, it became evident that

females could be divided into two groups: smaller fish had a GSI less than 8%,

and exhibited large variations in gonad development while larger fish with a GSI

≥ 8% exhibited much less variability. For this study, comparisons for this species

were restricted to larger fish and female dace with a GSI less than 8% were

excluded from the analyses.

White muscle samples were obtained from female fish for stable 13C/12C

and 15N/14N isotope analyses and were used to determine site fidelity of fish

collected at each site. A small piece of white muscle was removed from each

82

fish and placed in drying oven at 65°C for at least 48 h. Dried tissue samples

were individually ground into a fine powder and stored in clean glass scintillation

vials before isotopic analyses. Stable-isotope ratios (e.g., 13C/12C, 15N/14N) are

expressed as delta (δ) values and are measures of a parts-per-thousand (‰)

difference between the sample isotope ratio and that of an international standard

(carbon – Pee Dee Belemnite; nitrogen – atmospheric nitrogen) [12]. Delta is

expressed in terms of the numerator (i.e., heavier stable isotope; e.g., 13C).

Samples that have more negative delta values contain less 13C or 15N relative to

samples with higher delta values and are termed depleted for an isotope.

Samples that have higher delta values are considered enriched.

3.4 Results

Male and female slimy sculpin collected downstream of the sewage and

pulp mill effluent were significantly longer and heavier than fish collected from a

reference site located downstream of the St. Francis River (i.e., Capone) (Table

3.1). In contrast, male and female yellow perch and white sucker exhibited no

site differences in length and body weight (Table 3.3). Female blacknose dace

collected downstream of the pulp mill effluent diffuser and municipal sewage

wastewater were significantly shorter relative to reference fish at Capone, but not

St. Hilaire (Table 3.1). Female dace collected downstream of the sewage inputs

from the Madawaska River (i.e., D/S Mad) showed reduced body weight relative

to fish collected at upstream reference sites (Table 3.1). Male dace exhibited no

significant site differences in length (p=0.19) or weight (p=0.31) (Table 3.1).

83

Female sculpin collected downstream of sewage inputs from the

Madawaska River and pulp mill effluent diffuser were significantly older relative to

fish at Capone (Table 3.1). Male sculpin exposed to pulp mill effluent were the

same age as reference fish from Capone, but sewage-exposed males were older

(Table 3.1). For male and female sculpin, the slopes of the regression lines for

length-at-age at the site located downstream of the Madawaska River, upstream

of the pulp mill, and downstream of the pulp mill were higher than the reference

site at Capone (Table 3.2). Male and female white sucker exhibited no

significant site differences in age (Table 3.3) and length-at-age (data not shown).

Male yellow perch collected downstream of the pulp mill were older relative to

fish at St. Hilaire, but this was not the case for females (Table 3.3). Female dace

collected downstream of the pulp mill were the same age as reference fish at

Capone and St. Hilaire, but males collected downstream of the pulp mill were

younger than fish at St. Hilaire (Table 3.1). Perch and dace exhibited no

significant site differences in length-at-age (data not shown).

Liver size in male sculpin from the pulp mill site and sewage site was

significantly larger compared to fish from Capone, but this was not the case for

fish captured immediately upstream of the pulp mill diffuser (Table 3.1). There

were no significant site differences in female sculpin liver size (Table 3.1). Liver

size in male and female dace exposed to pulp mill effluent was significantly

greater compared to reference fish at Capone (females only) and St. Hilaire

(Table 3.1). Male yellow perch downstream of the pulp mill showed a significant

increase in liver size relative to upstream reference fish, but not females (Table

84

3.3). In contrast, male and female white sucker exposed to pulp mill effluent

exhibited significant decreases in liver size compared to upstream reference fish

(Table 3.3).

The condition factor of male and female sculpin downstream of the pulp

mill, upstream of the pulp mill, and downstream of the sewage was significantly

higher relative to reference fish at Capone. In addition, sculpin collected

downstream of the sewage discharges had significantly higher condition factors

relative to fish exposed to pulp mill effluent (Table 3.1). Male and female dace

exhibited no significant site differences in condition factor (Table 3.2). Male

yellow perch captured downstream of the pulp mill had significantly higher

condition factors compared to reference fish, but this was not the case for female

perch or male and female white sucker (Table 3.3).

There were no significant site differences in gonad size for male and

female slimy sculpin, yellow perch, and white sucker (Table 3.1). Likewise, there

were no significant site differences in fecundity of female sculpin, yellow perch,

and white sucker (Tables 3.1, 3.3). Female dace exposed to pulp mill effluent

and sewage wastewater exhibited no significant differences in ovary size

compared to reference fish at Capone and St. Hilaire (Table 3.1). For male dace,

the slope of the regression lines for gonad development were higher at the sites

located downstream of the Madawaska River and St. Hilaire when compared to

fish exposed to pulp mill effluent (Table 3.2).

Female sculpin captured downstream of the pulp mill effluent diffuser were

significantly more enriched in 13C than reference fish at Capone and fish

85

immediately downstream of sewage, but not fish captured immediately upstream

of the pulp mill diffuser (Table 3.1). Female sculpin collected immediately

downstream of the sewage discharges from the Madawaska River and upstream

of the pulp mill diffuser were significantly more enriched in 15N relative to

reference fish at Capone and fish exposed to pulp mill effluent (Table 3.1).

Female dace collected downstream of the pulp mill were significantly enriched in

13C relative to reference fish at St. Hilaire, but showed no site differences in 15N

(Table 3.1). Interestingly, female yellow perch showed no significant site

differences in either 13C or 15N (Table 3.3). Female white sucker exposed to pulp

mill effluent were enriched in 13C and depleted in 15N relative to reference fish at

St. Hilaire (Table 3.3).

3.5 Discussion

The general response patterns of slimy sculpin are consistent with our

previous work and suggest an overall increase in energy storage and utilization

via a nutrient enrichment effect [Munkittrick et al. 2000; Gibbons and Munkittrick

1994]. Male and female sculpin exposed to pulp mill effluent were longer,

heavier, and were in better condition and showed increased length-at-age

relative to reference fish. Similar to our results from 1999 [Galloway et al. 2003],

male and female sculpin exposed to pulp mill effluent exhibited no significant site

differences in gonad size relative to reference fish. Slimy sculpin invest most

their energy into gonad development during the winter when the river is ice-

covered, water temperatures and flow are low, and pulp mill effluent

86

concentrations are elevated. The absence of site differences in March indicates

that pulp mill effluent exposure was not adversely affecting sculpin gonad

development.

A series of fathead minnow life cycle studies were conducted on site at the

time of the fish collections [Parrott et al. 2003], and fathead minnow exposed to

10% effluent in a transportable laboratory showed increased growth (length and

weight), increased condition factor, and increased liver size. The wild fish

responses were not attributable to the effluent. Although sculpin showed

increased body size, liver size and condition relative to the upstream reference

sites, there were no differences relative to a site immediately upstream of the

diffuser, and condition was lower than fish collected downstream of the sewage

discharges in Edmundston.

Sculpin living downstream of the pulp mill had a unique 13C signature

relative to reference fish at Capone and fish downstream of municipal sewage

(i.e., D/S Mad) indicating a carbon-enriched environment. It is important to note

that female sculpin 13C and 15N signatures from the present study were similar to

our 1999 results (data not shown) [Galloway et al. 2003], providing further

evidence that sculpin movements are limited and the integrated response

patterns are site-specific. Sculpin exposed to municipal sewage wastewater (i.e.,

D/S Mad) had a distinct 15N signature relative to fish at Capone and downstream

of the pulp mill effluent diffuser, but not fish captured upstream of the pulp mill. If

sculpin captured downstream of the pulp mill were moving to upstream sites then

15N signatures should have been similar, but this was not the case. Furthermore,

87

our stable isotope results support other studies that have shown sites receiving

pulp mill effluent and municipal sewage wastewater can have distinct stable

isotope signatures relative to upstream sites and can be used to trace the

geographical extent of specific wastewater effluent streams [Wayland and

Hobson 2001; Wassenar and Culp 1996]. We also know that sculpin on the north

shore of the river do not mix with the sculpin on the south shore [Galloway et al.

2003].

White sucker and blacknose dace also showed different stable isotope

signatures between St. Hilaire and Edmundston, showing that these fish were not

mixing between the reference site and the site downstream of the pulp mill.

Similar to sculpin, blacknose dace showed larger livers (both sexes) downstream

of the pulp mill, relative to the reference site, but these differences were also

apparent upstream at the sewage sites and can not be attributed to the pulp mill.

We can not expect that the larger-bodied species do not mix between the

south and north shores of the river (approximately 50 m at this location). White

sucker showed no differences between the upstream reference site and the pulp

mill site, except for smaller livers downstream of the pulp mill. This species did

not respond to the combined effects of the sewage and mill discharges. Yellow

perch males showed increased liver size and condition factor, but no other

significant differences were present in males or females.

Using changes in mean age, energy expenditure (e.g., Length-at-age,

gonad weight, fecundity), and energy storage (e.g., condition factor, liver size)

between exposed and reference sites it is possible to identify the potential

88

mechanisms that resulted in the measured response patterns of each fish

species [Munkittrick et al. 2000; Gibbons et al. 1994]. For example, sculpin

collected downstream of the pulp mill effluent diffuser showed no differences in

age (males only), but showed increased length-at-age, condition, body size, and

liver size (males only) when compared to reference fish suggesting an increase

in food and/or habitat [Munkittrick et al. 2000; Gibbons et al. 1994]. The sculpin,

dace and perch all responded to better food conditions downstream of the city of

Edmundston (Table 3.4), although the changes can be attributed more to the

sewage inputs than to the input of the pulp mill effluent. Culp et al. [2003]

demonstrated that invertebrate and algal production was increased by the pulp

mill effluent relative to upstream river situations, but the fish apparently did not

detect the increase associated with the mill. Furthermore, Parrott et al. [2003]

demonstrated that the effluent alone was sufficient to further increase growth and

liver size in fathead minnow life cycle studies, although this again was not

translated into a field effect. Liver size and condition did not increase in dace or

sculpin above that seen immediately upstream of the outfall, and stable isotope

levels demonstrated that the sculpin were not moving between sites.

The decrease in liver size of white sucker downstream of the pulp mill

outfall is puzzling; both male and female liver sizes are more than 20% smaller at

the downstream site. The white sucker are probably the species in this study

that best integrates the stressors within the reach of the area, and would mix

between sides of the rivers with some upstream and downstream movement.

Studies conducted further downstream on the Saint John River have

89

demonstrated that the white sucker is not very mobile outside of the spawning

season with a home range outside of the spawning season of 1 to <2 km

[Doherty et al. 2003]. Furthermore, the stable isotopic ratios reported here show

that the downstream and upstream fish were not mixing. White sucker were

remarkably similar between the two sites outside of the reduction in liver sizes.

The liver sizes downstream are identical to those reported previously for white

sucker during the fall at this site [Galloway et al. 2003], and the difference reflects

a change at the reference site. This will be further investigated.

3.6 Conclusions

• Slimy sculpin downstream of the sewage discharges and pulp mill effluent

had greater growth, condition, and liver size, but no significant site

differences in gonad size relative to reference fish. The results of the

present survey along with previous work at these sites [see Parrott et al.

2003; Culp et al. 2003; Hewitt et al. 2003] support a nutrient enrichment

effect in fish exposed to the combined effluents from the sewage

discharges and the pulp mill. Stable isotope data showed slimy sculpin

have a very limited home range, the whole-organism responses are site-

specific, and as such, slimy sculpin are suitable sentinel species for an

effects-driven cumulative effects assessment of the Saint John River.

• White sucker and yellow perch showed few site differences in any of the

measured whole-organism parameters and the responses were not

consistent with slimy sculpin or previous work done at these sites,

suggesting white sucker and yellow perch may not be as sensitive as the

90

small-bodied species for picking up changes within a limited reach of the

river.

• White muscle stable 13C and 15N isotope signatures in white sucker, yellow

perch, and blacknose dace indicated movement is limited. However,

stable 13C and 15N isotope signatures in blacknose dace and yellow perch

did not show as much isotopic separation as white sucker and slimy

sculpin suggesting their prey items may not be incorporating pulp mill

effluent derived-carbon into their diet. Future work will need to be

conducted to understand the influence of pulp mill effluent and municipal

sewage wastewater on the local food web.

• There is substantial lack of basic biological information for many small-

bodied fish (e.g., Cyprinids), which can hinder scientifically sound data

interpretation and recommendations. Many cyprinids are multiple

spawners; Environment Canada’s Environmental Effects Monitoring (EEM)

regulations suggest these fish should be sampled prior to the initial

spawning event when all fish are presumed to be at the same stage of

reproductive development. For the present study, pre-spawning female

blacknose dace exhibited a great deal of variability in gonad development

relative to slimy sculpin. Currently, we are working to provide additional

guidance for sampling and interpreting data for multiple spawning small-

bodied used for environmental monitoring programs.

91

3.7 Acknowledgements

This project received funding from the Toxic Substances Research Initiative

(Project 205), NexFor and Fraser Papers, and NBCFWRU. We also

acknowledge extensive in-kind support from NexFor, Noranda Technology

Centre, and Fraser Papers. The invaluable help of K. Roach, A. Halford, K.

Tenzin, and J. McPhee in the field is appreciated. Graduate student support

from NSERC and UNB. KRM receives support from a NSERC Discovery Grant,

the Canadian Water Network National Centres of Excellence and from the

Canada Research Chairs program.

92

Table 3.1. Means ± SE (n) of various parameters measured in adult male and female slimy sculpin (Cottus cognatus) and

adult male and female blacknose dace (Rhinichthys atratulus) collected downstream of a sulphite pulp mill (D/S

Pulp), upstream of the pulp mill diffuser (U/S Pulp), and downstream of municipal sewage inputs (D/S Mad), and

from reference sites located downstream of the St. Francis River (Capone) and at St. Hilaire. Within a row,

differences (p ≤ 0.05) among sites are denoted by different uppercase letters. (*Significant interaction within

ANCOVA model; N/A = data not available).

Sites

Species Sex Parameter Capone St. Hilaire D/S Mad U/S Pulp D/S Pulp

Sculpin M Fork length (mm) 65.2 ± 1.0(22)A N/A 92.7 ± 1.4(29)B 85.8 ± 2.1(26)BC 82.9 ± 3.0(24)C

Body weight (g) 2.85 ± 0.17(22)A N/A 10.6 ± 0.5(29)B 7.71 ± 0.50(26)C 7.37 ± 0.77(24)C

Age 2.8 ± 0.1(22)A N/A 3.5 ± 0.1 (29)B 3.3 ± 0.1(26)AB 2.9 ± 0.2 (24)A

Ka 1.01 ± 0.01(22)A N/A 1.30 ± 0.02(29)C 1.17 ± 0.02(26)B 1.16 ± 0.03(24)B

LSIb 1.84 ± 0.10(22)A N/A 3.06 ± 0.15(29)BC 2.28 ± 0.11(26)AC 2.68 ± 0.11(23)C

GSIc 1.46 ± 0.05(22)AB N/A 1.56 ± 0.05(29)A 1.71 ± 0.06(26)B 1.51 ± 0.07(24)AB

F Fork length (mm) 56.3 ± 0.09(33)A N/A 83.9 ± 1.7 (30)B 77.6 ± 1.5(29)B 80.0 ± 2.5 (22)B

93

Sites

Species Sex Parameter Capone St. Hilaire D/S Mad U/S Pulp D/S Pulp

Sculpin F Body weight (g) 1.82 ± 0.10(33)A N/A 7.79 ± 0.45 (30)B 5.42 ± 0.32(29)C 6.27 ± 0.67(22)C

Age 2.4 ± 0.1(33)A N/A 3.6 ± 0.1 (30)B 2.9 ± 0.1(28)C 3.2 ± 0.2 (22)BC

K 0.99 ± 0.02(33)A N/A 1.31 ± 0.04 (30)B 1.13 ± 0.02(29)C 1.13 ± 0.02(22)C

LSI 4.25 ± 0.11(33)A N/A 4.62 ± 0.13 (30)A 4.39 ± 0.09(29)A 4.30 ± 0.16 (21)A

GSI 14.9 ± 0.5(33)A N/A 15.7 ± 0.8 (30)A 16.9 ± 0.4(29)A 16.0 ± 1.3(22)A

Fecundity, no. eggs 60.9 ± 3.5(33)A N/A 266 ± 16 (30)A 185 ± 12(29)A 208 ± 21 (22)A

δ13C (‰) -28.7 ± 0.3 (8)A N/A -28.3 ± 0.6 (8)AB -26.4 ± 0.4 (8)BC -25.0 ± 0.9 (8)C

δ15N (‰) 8.9 ± 0.1 (7)A N/A 11.5 ± 0.1 (7)B 11.0 ± 0.1 (7)B 8.4 ± 0.2 (7)C

Dace M Fork length (mm) N/A 57.9 ± 1.0(33)A 57.1 ± 0.9(20)A N/A 55.3 ± 1.0(18)A

Body weight (g) N/A 2.00 ± 0.11(33)A 1.92 ± 0.09(20)A N/A 1.76 ± 0.13(18)A

Age N/A 2.4 ± 0.1(33)B 2.3 ± 0.1(20)AB N/A 2.1 ± 0.01(17)A

K N/A 1.00 ± 0.01(33)A 1.02 ± 0.01(20)A N/A 1.02 ± 0.02(18)A

LSI N/A 1.19 ± 0.08(33)A 1.67 ± 0.11(19)B N/A 2.27 ± 0.18(17)B

GSI N/A 0.83 ± 0.04(30)* 0.66 ± 0.06(17)* N/A 1.02 ± 0.10(18)*

F Fork length (mm) 64.0 ± 0.8(20)A 63.5 ± 1.3(24)AB 58.5 ± 1.5(14)B 59.7 ± 1.3(12)AB 59.2 ± 1.0(22)B

94

Sites

Species Sex Parameter Capone St. Hilaire D/S Mad U/S Pulp D/S Pulp

Dace F Body weight (g) 2.90 ± 0.11(20)A 2.91 ± 0.19(24)A 2.22 ± 0.19(14)B 2.36 ± 0.18(12)AB 2.33 ± 0.13(22)AB

Age 3.0 ± 0.1(20)A 2.8 ± 0.1(23)AB 2.4 ± 0.1(14)B 2.8 ± 0.1(12)AB 2.6 ± 0.1(22)AB

K 1.10 ± 0.01(20)A 1.10 ± 0.02(24)A 1.08 ± 0.01(14)A 1.09 ± 0.03(12)A 1.10 ± 0.01(22)A

LSI 2.46 ± 0.19(20)A 2.68 ± 0.15(24)AC 3.20 ± 0.27(14)C 3.36 ± 0.22(12)BC 3.36 ± 0.14(22)BC

GSI 14.0 ± 0.8(20)A 18.6 ± 1.2(24)B 14.1 ± 1.2(14)AB 15.8 ± 1.5(12)AB 14.9 ± 0.9(22)AB

δ13C (‰) N/A -25.1 ± 0.1 (8)A N/A N/A -24.6 ± 0.2 (8)B

δ15N (‰) N/A 10.4 ± 0.1 (8)A N/A N/A 10.1 ± 0.1 (8)A

95

Table 3.2. Regression estimates for adult male slimy sculpin length-at-age

(Cottus cognatus) and adult male blacknose dace (Rhinichthys

atratulus) gonad development.

Species Sex Parameter Site Locations

Site Types Slope Intercept n p r2

Sculpin M Length-at-age

Capone Reference 0.16 1.74 22 0.017 0.25

Length-at-age

D/S Mad Study 0.38 1.76 29 <0.0001 0.44

Length-at-age

U/S Pulp Reference 0.51 1.67 26 <0.0001 0.81

Length-at-age

D/S Pulp Study 0.62 1.63 24 <0.0001 0.82

F Length-at-age

Capone Reference 0.30 1.64 33 <0.0001 0.70

Length-at-age

D/S Mad Study 0.45 1.68 30 <0.0001 0.80

Length-at-age

U/S Pulp Reference 0.34 1.73 28 <0.0001 0.80

Length-at-age

D/S Pulp Study 0.50 1.65 22 <0.0001 0.81

Dace M Gonad weight

St. Hilaire Reference 0.98 -2.09 33 <0.0001 0.54

Gonad weight

D/S Mad Study 2.04 -2.50 20 <0.0001 0.56

Gonad weight

D/S Pulp Study 0.38 -1.88 18 0.28

96

Table 3.3. Means ± SE (n) of various parameters measured in adult yellow perch

(Perca flavescens) and white sucker (Catostomus commersoni)

collected on the Saint John River, New Brunswick, Canada. Within a

row, differences (p ≤ 0.05) among sites are denoted by different

uppercase letters.

Reference site Study site Species Sex Parameter St. Hilaire Downstream Pulp mill Yellow perch

M Fork length (cm) 19.1± 1.2(14)A 19.4 ± 0.4(19)A

Body weight (g) 115 ± 22(14)A 114 ± 8(19)A Age 2.8 ± 0.4(14)A 3.6 ± 0.3(19)B Ka 1.42 ± 0.02(14)A 1.52 ± 0.03(19)B LSIb 0.86 ± 0.05(14)A 1.01 ± 0.04(19)B GSIc 5.52 ± 0.60(14)A 7.12 ± 0.69(19)A White sucker

M Fork length (cm) 39.6 ± 0.5(13)A 39.3 ± 0.4(19)A

Body weight (g) 839 ± 33 (13)A 798 ± 24(19)A Age 6.6 ± 0.5(13)A 6.1 ± 0.4(19)A K 1.35 ± 0.02(13)A 1.31 ± 0.02(19)A LSI 1.97 ± 0.10 (13)A 1.50 ± 0.13 (19)B GSI 5.24 ± 0.21 (13)A 4.98 ± 0.15(19)A Yellow perch

F Fork length (cm) 25.2 ± 1.6 (10)A 22.5 ± 0.5 (42)A

Body weight (g) 243 ± 45 (10)A 169 ± 11 (42)A Age 2.8 ± 0.1 (38)A 2.9 ± 0.4 (10)A K 1.34 ± 0.03 (10)A 1.41 ± 0.02 (42)A LSI 1.20 ± 0.07 (10)A 1.26 ± 0.05 (42)A GSI 2.56 ± 0.17 (10)A 2.41 ± 0.11 (42)A Fecundity, no. of

eggs 37556 ± 5854 (9)A 30446 ± 2460 (31)A

δ13C (‰) -23.6 ± 0.3(8)A -23.0 ± 0.2 (7)A δ15N (‰) 11.8 ± 0.3(7)A 10.3 ± 0.3(7)A White sucker

F Fork length (cm) 42.7 ± 0.5(22)A 41.6 ± 0.6 (26)A

Body weight (g) 1015 ± 30(22)A 933 ± 35(26)A Age 8.0 ± 0.4(17)A 7.2 ± 0.4(22)A K 1.30 ± 0.02(22)A 1.29 ± 0.02(26)A LSI 2.30 ± 0.06(22)A 1.81 ± 0.06(26)B GSI 6.95 ± 0.33(22)A 6.08 ± 0.32(26)A Fecundity, no. of

eggs 18667 ± 943(22)A 16803 ± 1294 (21)A

97

Reference site Study site Species Sex Parameter St. Hilaire Downstream Pulp mill White sucker

F δ13C (‰) -24.8 ± 0.8(8)A -22.6 ± 0.3 (8)B

δ15N (‰) 10.0 ± 0.4(8)A 8.4 ± 0.3(8)B

98

Table 3.4. A comparison of species responses to the combined effluents at

Edmundston (sewage, pulp mill effluent) and a comparison of fish

responses upstream and downstream of the pulp mill discharge.

Species Sex Reference* versus downstream pulp mill

Upstream pulp mill versus downstream

Age Body Size

Gonad size

Liver size

Condition Age Body Size

Liver size

Gonad size

Condition

Sculpin* Male 0 ↑ 0 ↑ ↑ 0 0 0 0 0

Female ↑ ↑ 0 0 ↑ 0 0 0 0 0

Dace Male ↓ 0 0 ↑ 0

Female 0 0 ↑ 0 0 0 0 0

Perch Male ↑ 0 0 ↑ ↑

Female 0 0 0 0 0

Sucker Male 0 0 0 ↓ 0

Female 0 0 0 ↓ 0

* Reference site is Capone.

99

Figure 3.1. The study area near Edmundston, NB. River flow is from left to right

on the map.

Maine, USA

Quebec, Canada

D/S Mad

New Brunswick, Canada

St. Francis River

Pulp mill

Paper mill

Saint Jo

hn River

Provincial Border

Madawaska River

Iroquois River

Pulp mill effluent lagoon and diffuser

Paper mill effluent lagoon and diffuser

Municipal sewage facilities

Capone

St. Hilaire

D/S Pulp

U/S Pulp

Maine, USA

Quebec, Canada

D/S Mad

New Brunswick, Canada

St. Francis River

Pulp mill

Paper mill

Saint Jo

hn River

Provincial Border

Madawaska River

Iroquois River

Pulp mill effluent lagoon and diffuser

Paper mill effluent lagoon and diffuser

Municipal sewage facilities

Capone

St. Hilaire

D/S Pulp

U/S Pulp

100

3.8 References

BAR Environmental Inc. 1994. Environmental effects monitoring pre-design

historical information for Fraser Inc., Edmundston sulphite pulp mill.

Guelph, ON, Canada.

Culp,J.M., Cash, K.J., Glozier, N.E., and Brua, R.B. 2003. Effects of pulp mill

effluent on benthic assemblages in along the Saint John River, Canada.

Environ Toxicol Chem 22: 2916-2925.

Doherty, C.A., Curry, R.A., and Munkittrick, K.R. 2003. Tracking adult white

sucker movements near point source discharges in the Saint John River,

New Brunswick, Canada. Water Quality Res J Can: Submitted.

Frank, M., McMaster, M.E., Munkittrick, K.R., Savoie, M.C., and Wood, C.

Effects of sulphite and bleached kraft pulp and paper mill effluents on

yellow perch and Johnnie darters. 25th Aquatic Toxicity Workshop,

Quebec, QC, Canada, October 18-21, 1998, pp 53.

Galloway, B.J., Munkittrick, K.R., Currie, S., Gray, M.A., Curry, R.A., and Wood,

C. 2003. Examination of the responses of slimy sculpin (Cottus cognatus)

and white sucker (Catostomus commersoni) collected on the Saint John

River downstream of pulp mill, paper mill, and sewage discharges.

Environ Toxicol Chem 22: 2898-2907.

Gibbons, W.N., and Munkittrick, K.R. 1994. A sentinel monitoring framework for

identifying fish population responses to industrial discharges. J Aquat

Ecosyst Health 3: 227-237.

101

Gibbons, W.N., and Munkittrick, K.R., and Taylor, W.D. 1998a. Monitoring

aquatic environments receiving industrial effluents using small fish species

1: Response of spoonhead sculpin (Cottus ricei) downstream of a

bleached-kraft pulp mill. Environ Toxicol Chem 17: 2227-2237.

Gibbons, W.N., Munkittrick, K.R., McMaster, M.E., and Taylor, W.D. 1998b.

Monitoring aquatic environments receiving industrial effluents using small

fish species 2: Comparison between responses of trout-perch (Percopsis

omiscomaycus) and white sucker (Catostomus commersoni) downstream

of a pulp mill. Environ Toxicol Chem 17: 2238-2245.

Gray, M.A., Curry, R.A., and Munkittrick, K.R. 2002. Non-lethal sampling

methods for assessing environmental impacts using a small-bodied

sentinel fish species. Water Quality Res J Can 37: 195-211.

Hewitt, L.M., Pryce Hobby, A.C., Parrott, J.L., Marlatt, V., Wood, C., Oakes, K.,

Van Der Kraak, G. 2003. Accumulation of ligands for aryl hydrocarbon

and sex steroid receptors in fish exposed to treated effluent from a

bleached sulphite/groundwood pulp mill. Environ Toxicol Chem 22: 2890-

2897.

Munkittrick, K.R., McMaster, M.E., Van Der Kraak, G., Portt, C., Gibbons, W.N.,

Farwell, A, and Gray, M. 2000. Development of methods for effects-

driven cumulative effects assessment using fish populations: Moose River

Project. SETAC Press, Pensacola, FL, USA.

Munkittrick, K.R., McGeachy, S.A., McMaster, M.E., Courtenay, S.C. 2002.

Overview of cycle 2 freshwater fish studies from the pulp and paper

102

Environmental Effects Monitoring program. Water Quality Res J Can 37:

49-77.

Parrott, J.L., Wood, C.S., Boutot, P., Blunt, B.R., Baker, M.A., and Dunn, S.

2003. Fathead minnow long-term growth/reproductive tests to assess

final effluent from a bleached sulphite mill. Environ Toxicol Chem 22:

2908-2915.

Peterson, B.J., and Fry, B. 1987. Stable isotopes in ecosystem studies. Ann

Rev Ecol Syst 18: 293-320.

Scott, W.B., and Crossman, E.J. 1998. Freshwater Fishes of Canada. Galt

House, Oakville, ON, Canada.

Wayland, M., and Hobson, K.A. 2001. Stable carbon, nitrogen, and sulphur

isotope ratios in riparian food webs on rivers receiving sewage and pulp-

mill effluents. Can. J. Zool. 79: 5-15.

Wassenar, L.I., and Culp, J.M. 1996. The use of stable isotope analyses to

identify pulp mill effluent signatures in riverine food webs. In Servos,

M.R., Munkittrick, K.R., Carey, J.H., Van Der Kraak, G., (eds),

Environmental Fate and Effects of Pulp and Paper Mill Effluents. St. Lucie

Press, Delray Beach, FL, USA, pp 413-424.

103

CHAPTER 4. Influence of seasonal changes in relative liver size,

condition, relative gonad size, and variability in ovarian development

of multiple spawning freshwater fish for use in environmental

monitoring programs3.

3Submitted for Publication: Galloway, B.J., and Munkittrick, K.R. 2005. Influence

of seasonal changes in relative liver size, condition, relative gonad size, and

variability in ovarian development of multiple spawning freshwater fish for use in

environmental monitoring programs. Journal of Fish Biology.

4.1 Abstract

The use of forage fish for environmental monitoring programs has

increased over the past few years because they are relatively easy to capture,

many species exhibit site fidelity, and they are usually relatively abundant. The

main disadvantage of using forage fish for environmental monitoring programs is

the lack of basic life-history information, which can hinder study design and data

interpretation. The objectives of this study were to collect basic biological

information to assist in understanding the biology of selected forage fish for use

in environmental monitoring programs. Specifically, we examined seasonal

changes in condition factor, liversomatic index (LSI), gonadosomatic index (GSI),

and ovarian histology in four common cyprinids found in Atlantic Canada:

blacknose dace Rhinichthys atratulus, northern redbelly dace Phoxinus eos,

golden shiner Notemigonus crysoleucas and the mummichog Fundulus

heteroclitus. All four species are batch-spawners that have an extended

104

spawning season (spring-summer). GSI and LSI profiles of female blacknose

dace, mummichog, and golden shiner were similar, decreasing gradually during

the spawning season. GSI and LSI of female northern redbelly dace peaked

twice suggesting oocyte recruitment was continuous throughout the spawning

season. In addition, regressions of ovary weight to adjusted body weight were

more variable in northern redbelly dace relative to the other fish species. For

female blacknose dace, variability in the relationship between ovary weight and

adjusted body weight was lowest during the early prespawning period (r2=0.93)

and highest just before spawning (r2=0.43), and this variability was minimized by

selecting either 2-year-old fish or fish that weighed 2 – 4 g. Minimizing sources

of natural variability in metrics required for environmental monitoring studies is

particularly important since it will increase the probability of detecting impacts

associated with contaminant exposure.

4.2 Introduction

In Canada, the use of large-bodied fish species in many Environmental

Effects Monitoring programs has decreased [Munkittrick et al. 2002] because of

concerns over mobility issues (i.e., fish moving out of the area receiving

wastewater effluents). The use of small-bodied fish for environmental monitoring

programs has increased over the past few years because they are relatively easy

to capture, many species exhibit site fidelity, and they are usually relatively

abundant. Small-bodied fish have also been used in environmental monitoring

programs in the United States [Yeardley, Jr 2000; Fraker et al. 2002; Noggle et

al. 2004] and New Zealand [Richardson 1997]. The main disadvantage of using

105

small-bodied fish for environmental monitoring programs is the lack of basic life-

history information, which can hinder study design and data interpretation.

There are a number of challenges to using small-bodied species for

environmental assessments, including populations can sometimes show

localized differences in performance because of their reliance on small habitat

patches. Additionally, there is often a paucity of background biological

information on which to base study designs. Furthermore, several small-bodied

species available for use in monitoring programs have multiple spawning periods

during the summer. It becomes difficult to ensure synchronous sampling

between exposed and reference populations that may have relatively small

differences in environmental conditions [LeBlanc et al. 1997]. In an attempt to

avoid the complications of multiple spawning periods and differences in timing

between groups of fish, Environment Canada has provided guidance for

monitoring programs that suggests that the sampling of multiple spawning fish be

conducted in the prespawning period prior to the onset of the first spawn of the

year [Environment Canada, 1997, 2001]. At this time, it was assumed that

development would be synchronized between individuals, and if changes were

not evident prior to the first spawn, the consequences of small changes later in

the season would be less important.

Recently, blacknose dace Rhinichthys atratulus (Hermann) were used in a

monitoring program examining the consequences of multiple wastewater

106

discharges in the upper Saint John River, New Brunswick, Canada [Galloway et

al. 2004]. During these studies it was apparent that the strength of the

relationship between gonad size and body size in this multiple spawning fish

species decreased as the sampling season was approached. This decreased

statistical power and complicated the interpretation of data. The present study

examined four species of multi-spawning, small-bodied cyprinids found in Atlantic

Canada to examine the influence of season and reproductive activity on the

potential power of statistical comparisons and the ability to distinguish site

differences.

Seasonal changes in condition factor, liversomatic index (LSI),

gonadosomatic index (GSI), and ovarian histology in blacknose dace R.

atratulus, northern redbelly dace Phoxinus eos (Cope), golden shiner

Notemigonus crysoleucas (Mitchell), and estuarine mummichog Fundulus

heteroclitus L. were examined. The mummichog is a common minnow found

along the Atlantic coast and been used in environmental monitoring programs

[Leblanc et al. 1997; Couillard and Nellis 1999; Dube and MacLatchy 2000] and

its life history has been well studied [Taylor and DiMichele 1980; Wallace and

Selman 1981; Shimizu 1997]. However, there is a paucity of detailed biological

information (especially for reproduction) for blacknose dace, northern redbelly

dace, and golden shiner in Canada and specifically for the Maritime Provinces.

This is particularly surprising for blacknose dace since it is considered to be one

of the most common minnow species in the Maritime Provinces and may serve

107

as an important food source for economically important trout species [Scott and

Crossman 1998].

Seasonal variability in the ovary weight to body weight relationship (i.e.,

coefficient of determination [r2] values for linear regressions equations) of female

fish was also monitored to: (1) identify the time of the year when the relationship

between ovary weight and adjusted body weight (i.e., total body minus gonad

weight) was most variable (i.e., low r2 values), (2) increase our understanding of

natural source(s) of variability in gonad development, (3) identify suitable

techniques to minimize data variability and (4) reduce sample size requirements

for detecting a critical effect size when designing cost-effective studies in the

future.

4.3 Materials and Methods

Fish were collected from various sites in southern New Brunswick

between May 2003 and April 2004. Blacknose dace were collected using a

backpack electrofisher [Smith-Root Model 12B, Smith-Root Inc., Vancouver, WA,

USA] from Milkish Brook (N 45.37977°, W 66.13306°). Northern redbelly dace

were collected from a beaver pond in the Keswick River watershed (N 46.09523°,

W 66.97490°; females only); golden shiner were collected from Little Chamcook

Lake (N 45.15228°, W 67.107320°); baited minnow traps were used to collect

mummichogs from a salt marsh located adjacent to Taylor's Island in Saint John

(N 45.22953°, W 66.13082°). Captured fish were transferred to the laboratory in

108

coolers filled with ambient water, rendered unconscious by concussion, followed

by spinal severance, and length (± 1 mm), body weight (± 0.01 g), liver (± 0.001

g) weight, and gonad weights (0.001 g) were recorded. Fish were aged by

counting annuli on sagittal otoliths. The annuli counts were verified by two

readers.

Ovaries were fixed in 10% buffered formalin and transferred to 70%

ethanol for storage. Paraplast-embedded tissues were sectioned (4 µm) with a

rotary microtome and stained with Ehrlich's haematoxylin and counterstained

with eosin Y. Ovarian sections were examined with a light microscope to

determine the percent of various oocyte stages present at each sampling time.

One hundred oocytes sectioned through the nucleus were counted to determine

the relative proportion of different developmental stages. The ovarian stage

terminology follows Blazer [2002] (see Table 4.1). Ovarian section images were

obtained from an Olympus™ BX40 microscope with Sony™ Exwave HSD digital

video camera attachment.

4.3.1 Data Analyses

Analysis of variance (ANOVA) was used to test for differences in mean

length, weight, and age of fish between collection dates; Bonferroni post hoc

comparisons were used to compare means. Analysis of covariance (ANCOVA)

was used to assess differences in the relationships between weight and length

109

(condition factor), and body weight and liver size and gonad size. Except for

condition, adjusted body weight (total wt – organ wt) was used as a covariate in

the ANCOVA model. All data were log10 transformed before performing ANOVA

and ANCOVA, and sexes were analyzed separately. All data analyses were

done using SYSTAT® (SPSS, SYSTAT, Chicago, IL, USA) statistical software.

ANCOVA is robust even when the slopes are not equal, so slopes were

considered different when p < 0.01 (Hamilton et al. 1993). Gonad weights and

liver weights were calculated as percent-adjusted body weight for summary

purposes.

4.4 Results

4.4.1 Blacknose dace

Female blacknose dace exhibited few significant differences in fork length

(68.6 ± 0.54 mm, n=141), weight (3.56 ± 0.10 g, n=141) or age (mean 2.6 years,

n=90), between sampling periods (Table 4.2). Male blacknose dace showed no

significant differences in fork length (65.1 ± 0.6 mm, n=100) or body weight (3.02

± 0.09 g, n=100), (Table 4.2). Condition factors (slope of body weight versus

length) were highest for both sexes in May relative to all other sampling times,

except males in August (Table 4.3).

Initial spawning for blacknose dace occurred in early June when mean

water temperatures reached 15.0°C (Figure 4.1). In early May, females showed

a strong relationship between ovarian size and adjusted body weight (body

weight-gonad weight), with an average GSI of 9%, (Table 4.2) and r2=0.97 (see

110

Table 4.4). Ovarian size continued to increase, peaking on May 20 and

remaining high in some fish (>20%) until mid-June. Coefficients of determination

(i.e., r2) values declined to 0.80 in late May and 0.43 in early June (Table 4.4).

Female blacknose dace ovaries in early May contained mostly oocytes in the

cortical alveoli and early vitellogenic stages (Figure 4.2). The appearance of mid-

vitellogenic oocytes increased in late May, and showed a slight reduction by June

11, however, some females still contained mid-vitellogenic and mature oocytes

(Figure 4.2).

The highest GSIs for female blacknose dace were recorded on May 20

(25%) and June 11 (21.2%); the highest male GSIs were recorded on May 4

(1.67%) and May 20 (1.53%) (Table 4.2). By July 28, mean female and male

GSIs were significantly reduced relative to pre-spawning values; ovaries

contained mostly previtellogenic oocytes (Figure 4.2), and the r2 value for the

ovary weight to adjusted body weight relationship in female blacknose dace was

high at 0.76 (Table 4.4).

It has been well documented and generally understood that temperature

plays a critical role in many aspects of the physiology and biochemistry of fishes

and increasing water temperatures and photoperiod are often associated with

increased gonad development and the onset of spawning activity in fishes

inhabiting temperate waters [see Pankhurst and Porter 2003]. In Milkish Brook,

mean monthly water temperatures increased from 10.7°C in May to 15.0°C in

111

June (Figure 4.1), which corresponded to the largest increase in female GSI,

increased presence of vitellogenic oocytes, and increased condition (see Table

4.2). Liver weights did not decrease during this same time period, indicating

energy investment in gonad development from May to June was probably from

active feeding. Generally, increased water temperatures are associated with

increased feeding activity in fish inhabiting temperate waters. Mean water

temperature increased ~2°C from June to reach 17.1°C in July (Figure 4.1); it

was evident that blacknose dace spawning occurred at these temperatures since

male and female GSIs, LSIs, and condition all decreased during this time (Table

4.2). Mean water temperatures remained constant from July to August at ~ 17°C

(Figure 4.1) and female GSIs showed a significant increase during this time

(Table 4.2).

Liver sizes of female and male blacknose dace varied seasonally and

decreased during spawning. Female blacknose dace had mean LSI values of ~

3% in May and were significantly reduced by June 11 with the appearance of

stage 5 oocytes (exogenous vitellogenesis) in the gonads (Table 4.2, Figure 4.2).

By July 28, female LSI values were significantly reduced relative to pre-spawning

values and remained low by August, which corresponded to a decline in the

presence of vitellogenic eggs and appearance of atretic eggs (Table 4.2, Figure

4.2). Male LSIs peaked in early May, remained virtually unchanged by June 11,

and were significantly reduced relative to pre-spawning values by July and

August (Table 4.2).

112

4.4.2 Golden shiner

Female golden shiners were a similar size during the collections (p>0.05);

mean fork length (76.5 ± 0.9 mm, n=140), body weight (4.77 ± 0.21g, n=140),

and age (3.1 ± 0.1 years, n=101). Male shiners were identical in size, with an

average fork length of 76.5 ± 1.1 (n=80) and body weight 4.77 ± 0.26 (n=80).

Female and male golden shiner GSIs peaked in late June and remained elevated

by early July (Table 4.2). Ovaries exhibited an increase in the presence of

mature oocytes on June 22 (Figure 4.3), which corresponded to the lowest

recorded r2 value for the ovary weight to adjusted body weight relationship (Table

4.4). By early August, mean GSIs for female and male golden shiners were

significantly reduced relative to fish collected in late July, however, female fish

with GSIs >10% were still present (Table 4.2). By late August, mean GSIs of

male and female golden shiners were significantly reduced relative to pre-

spawning fish and ovaries contained primary oocytes only (Table 4.2, Figure

4.3). The gonad weight to adjusted body weight relationship of females collected

on August 27 was strong with an r2 value of 0.91 (Table 4.4).

There were no significant differences in female and male condition factor

between sampling times, except males on August 29 (Table 4.2). Except for

April 24, female LSI values were similar between sampling times (Table 4.2).

Male golden shiner showed no significant changes in LSI between sampling

times (Table 4.2).

113

4.4.3 Northern Redbelly Dace

Female northern redbelly dace exhibited few differences in mean fork

length (54.6 ± 0.3 mm, n=354), body weight (1.56 ± 0.03 g, n=354) and age (2 ±

0.06 years, n=154) during the study period (Table 4.2). Gonad development in

female northern redbelly dace showed the most variability among the species

sampled. Mean GSIs in prespawning female northern redbelly dace were low in

April (<5%), increased significantly by May 27 and peaked on June 12 at almost

10% (Table 4.2). Histological sections of female P. eos ovaries confirmed that

the increase in mean GSI from April to June corresponded to an increase in the

percentage of late vitellogenic (stage 4) oocytes present within the ovary (Figure

4.4). During this same time period, r2 values for the ovary weight to adjusted

body weight relationship in female redbelly dace were consistently low (< 0.4,

Table 4.4). By late June, mean GSIs were significantly reduced relative to

earlier samples and the presence of post-ovulatory follicles (POFs) confirmed

spawning had occurred. However, it is important to note that some individual fish

still had GSIs >20% and ovaries contained late stage vitellogenic oocytes and

fully mature oocytes (Table 4.2, Figure 4.4).

The lowest r2 value for the ovary weight to adjusted body weight

relationship in female redbelly dace was recorded in late June (Table 4.4). The

highest female northern redbelly dace GSI values were measured on May 27

(21.5%), June 12 (21.5%), and June 22 (23%) (Table 4.2). After a decline over 4

weeks, mean GSI of females collected in July and August were ~ 2%, were

significantly reduced relative to prespawning fish in April, and remained low to

114

October (Table 4.2). The ovaries of fish collected in August contained

previtellogenic oocytes only (Figure 4.4). Coefficient of determination values for

the ovary weight to adjusted body weight relationship in female redbelly dace

were all >0.5 from July to October (Table 4.4).

The liversomatic index of female northern redbelly dace was variable with

few statistically significant differences throughout the season. LSI peaked at two

periods: in April, afterwards the LSI decreased 28% by June 12, which

corresponded to a 59% increase in GSI; and in late June, corresponding to the

maximum GSI observed (i.e., 23%) (Table 4.2). By July 27, LSI of female

northern redbelly dace was significantly reduced relative to pre-spawning fish

and remained low to October (Table 4.2). Condition of prespawning female

northern redbelly dace was low in April, increased in June and remained high

until late July (Table 4.2). After 4 weeks, females had returned to pre-spawning

condition and remained unchanged by October (Table 4.2).

4.4.4 Mummichog

Female mummichog collected in April and June were shorter and lighter

relative to fish collected on other dates (Table 4.2); males exhibited no significant

differences in fork length (65.7 ± 0.9 mm, n=86) and body weight (3.40 ± 0.15 g,

n=86) during our collections. The mean fork length, body weight, and age of all

female mummichog was 65.3 ± 0.7 mm (n=239), 3.39 ± 0.11 g (n=239), and 2.4

115

± 0.1 years (n=109), respectively. Female mummichogs collected in June had a

mean age of 2.0 ± 0.1 (n=42) and were significantly younger relative to females

collected in May, July, and August with a mean age of 2.5 ± 0.1 (n=69).

Mean GSIs of female mummichog were low (< 5%) in April and early May

(Table 4.2). On May 8, mummichog ovaries consisted of mostly previtellogenic

oocytes and few early vitellogenic oocytes (Table 4.2, Figure 4.5). Variability in

the ovary weight to adjusted body weight relationship for female mummichog was

low in April and May when r2 were 0.93 and 0.85, respectively (Table 4.4).

Female GSIs peaked in late June with the appearance of fully mature oocytes

(Table 4.2, Figure 4.5), which corresponded to increased variability in the ovary

weight to adjusted body weight relationship (Table 4.4). Mean GSIs of female

mummichogs were low in July (Table 4.2), but gonad development was highly

variable relative to other sampling times (Table 4.4). By late August, female

GSIs ranged from ~1-3% (similar to October), ovaries contained mostly primary

oocytes, and variability in the gonad weight to adjusted body weight relationship

was low and approaching pre-spawning conditions (Tables 4.2, 4.4; Figure 4.5).

Mean GSIs for male mummichogs peaked by late June, were significantly

reduced by late July, and remained low to October (Table 4.2).

ANCOVA results showed the slopes for ovary weight versus adjusted

body weight were significantly different (F=4.675; d.f. = 5, 227; P<0.0005, see

Table 4.3). For female mummichog, slopes of the regression for condition were

116

significantly different (ANCOVA; F=5.502; d.f.=2,227;p<0.0005). Thus there

were seasonal differences in the rate at which body weight increased with length

(see Table 4.3). The fattest male mummichogs were collected on July 28 and

October 12 and corresponded to elevated LSIs (Table 4.2).

4.5 Discussion

The objectives of our research on the four multi-spawning, small-bodied

fish species from southern New Brunswick were to document seasonal changes

in gonad development, liver size, and condition (metrics required for many

environmental monitoring programs) during the pre-spawning, spawning, and

post-spawning seasons; attempt to identify and minimize sources of natural

variability in gonad development; and provide guidance to facilitate the use of

multi-spawning, small-bodied fish species for regional, national, and international

environmental monitoring programs.

4.5.1 Reproductive Development

The results of the present study indicate blacknose dace, northern

redbelly dace, and golden shiner, like some other cyprinids, have an extended

spawning season (spring-summer), with multiple spawning events [Tartar 1969;

Delahunty and DeVlaming 1980; Heins and Rabito 1986; DeHaven et al. 1992;

117

Rinchard and Kestemont 1996; Roberts and Grossman 2001]. Mummichog

followed a similar pattern, as has been previously reported [Shimizu 1997].

Releasing small batches of eggs several times during the reproductive season

increases the probability that some offspring will survive despite fluctuating

environmental conditions, predation pressure, and/or resource competition (e.g.,

suitable spawning habitat/food/shelter). Our results also demonstrated that

oocyte development in the four fish species was asynchronous; there were

considerable differences in the patterns of ovarian maturation between northern

redbelly dace and the other fish species; and rates of oocyte development differ

between individuals of a species. Together, histology and regressions of ovary

weight to adjusted body weight were critical for understanding seasonal

variability in ovarian development both within and between species. Except for

northern redbelly dace, all female fish sampled showed the same seasonal

pattern of variability in gonad development (see Table 4.4). For example,

variability in female blacknose dace gonad development was low (r2=0.93) during

the early pre-spawning season when the ovaries contained few vitellogenic

oocytes, increased (r2=0.43) just before the onset of spawning activity when the

ovary contained two batches of maturing oocytes, and gradually the variability

decreased as the post-spawning season approached and the ovaries contained

only primary oocytes.

Histology showed the increases in variability in gonad development

resulted from individual differences in oocyte maturation as fish approached the

118

first spawning event. The exact mechanisms controlling the rate of growth of

maturing oocytes in fish is not completely understood [Tyler and Sumpter 1996].

Studies with rainbow trout [Oncorhynchus mykiss; Tyler et al. 1990], and cod

[Gadus morhua; Kjesbu et al. 1996, Tomkiewicz et al. 2003] have shown there is

considerable variability in individual oocyte growth in maturing ovaries. Thus, it is

likely that the increased variability seen in female blacknose dace, mummichog,

and golden shiner ovaries during the pre-spawning period resulted from

differences in the growth rates of individual oocytes in the early stages of

vitellogenesis that would be released later in the spawning season.

Gonad development in female northern redbelly dace was highly variable

throughout the study relative to the other three species and was highly variable

among individuals. Histological sections of ovarian tissue were invaluable to our

understanding of variability in female northern redbelly dace gonad development.

Specifically, post-ovulatory follicles (POFs) were observed in the ovaries of some

fish that also contained maturing oocytes suggesting at least one batch of

oocytes was recently released. Some female northern redbelly dace had ovaries

containing remnant ripe eggs, atretic eggs, and maturing oocytes indicating the

fish had recently spawned and was preparing to release another batch of eggs.

The histological results for female northern redbelly dace suggest individuals

release small batches of eggs daily throughout an extended spawning season.

The high variability in female northern redbelly dace gonad development is due

to the fact that ovaries contain oocytes in various maturity stages during the

119

entire reproductive season and that oocytes are continuously recruited into

vitellogenesis.

4.5.2 Liversomatic Index and Condition Factor

LSI and condition factor data for the fish species collected in the present

study provided important information about the patterns of energy use and

storage during the pre-spawning period, spawning period, and post-spawning

period. The seasonal changes in LSI were particularly pronounced in female fish

and there were differences between species. Liver energy stores in female

blacknose dace were not fully utilized during the early stages of gonadal

development. Liver weights of female blacknose dace were highest during the

pre-spawning period in May when the largest increase in gonad weight and

condition occurred. If the liver was the main energy source for oocyte

development in female blacknose dace then we should have seen a decline in

LSI from May 4 to May 20, but this was not the case. In fact, female LSI values

remained virtually unchanged in May suggesting most of the energy for early

gonad development was derived from feeding. Female blacknose dace LSIs

declined from May 20 to June 11, which corresponded to a slight decrease in

GSI and presence of mature (stage 5) oocytes within the ovary. Female LSI and

GSI values showed the largest decline from June 11 to July 28. Taken together,

seasonal changes in LSI of female blacknose dace suggests energy invested

into early gonad growth (i.e., first batch of eggs released) is probably derived

120

from active feeding, while most of the liver energy stores are utilized for

maturation of subsequent batches of eggs to be released later in the spawning

season. Future studies will be needed to improve our understanding of this

issue. The pattern of energy use for gonad development in male blacknose dace

was similar, although not as pronounced. Changes in female blacknose dace

condition factor mirrored the changes in GSI.

As in bleak [Alburnus alburnus; Rinchard and Kestemont 2003], LSIs of

northern redbelly dace did not decrease continuously during the spawning

season, but showed two peaks: in April, just before the spawning season and

again in late June. The changes in northern redbelly dace LSI occur because

batches of oocytes are going through vitellogenesis during the entire spawning

season, which was also the case for the bleak [Rinchard and Kestemont 2003].

LSI and condition remained relatively stable during the spawning season in male

and female golden shiner collected from Little Chamcook Lake suggesting liver

energy stores were not seriously depleted and fish were possibly feeding. Again,

this is difficult to confirm without stomach content analysis. Mummichog liver

energy stores declined as gonad weights increased, but it is important to note

that fish condition continuously increased during the pre-spawning period and

remained relatively high during the spawning season indicating fish were feeding

to maintain liver energy stores.

121

It is important to note that the seasonal changes in female blacknose dace

LSI differed from the other multiple spawning fish used in this study suggesting

the pattern of oocyte recruitment and maturation is different between blacknose

dace and the other species.

4.5.3 Data Variability, Sample Sizes, and Power

It is our goal to facilitate the use of multi-spawning, small-bodied fish in

environmental monitoring programs. In order to accomplish this goal, we

identified sources of data variability and selected techniques that could be used

to minimize data variability, which would allow us to determine the sample sizes

required in order to detect a 25% difference (i.e., critical effect size) in gonad size

between a reference and exposure site at a specified power level [see

Munkittrick et al. 2002; Lowell et al. 2003]. In Canada, for example, Federal

government regulations require pulp and paper mills and metal mining operations

to design and implement monitoring programs capable of detecting a 25%

difference in gonad size between an exposed site (i.e., site receiving effluent)

and reference site [Environment Canada 1997].

The number of samples required to detect a difference between

populations depends on the variability, the size of difference you want to detect,

the power required (1-β), and the statistical certainty (α). The batch-spawning

fish investigated in this paper demonstrated changes in variability during the

122

sample season that would dramatically affect the sample sizes required to detect

a standardized difference between populations. For blacknose dace collected in

early May (r2=0.97), the sample size required to detect a difference is much

smaller than during late June when the r2 is < 0.45. Based on all female

blacknose dace data collected on May 20 and June 11, the sample sizes

required to detect a 25% difference in gonad size are shown in Table 4.5. On

May 20 when the standard deviation was about 24% of the mean (i.e., CV=24%);

a sample size of 16 to 24 fish (depending on site) would be required to detect a

25% difference in gonad size when α and β were 0.05 and 0.20, and when both

α and β equalled 0.10. On June 11, the standard deviation for all female

blacknose dace data increased to about 41% of the mean (i.e., CV = 41%), as

such, future studies during this time period would require sample sizes of 47 to

71 (Table 4.5).

It is possible to reduce the variability by restricting the samples to include

fish of a similar age or body size. Sample sizes required to detect a 25%

difference in gonad size for all female blacknose dace collected on May 20 and

June 11 were compared to 2 year old fish and fish weighing 2 – 4 g at three

different levels of power (i.e., α=0.05 and β=0.05; α=0.05 and β=0.20; α=0.10

and β=0.10). The r2 for the regression of ovary weight to adjusted body weight

for all female blacknose dace collected on May 20 was 0.80 and when 2 - 4 g fish

or 2 year old fish were selected the r2 value increased to 0.88 and 0.94,

respectively (Table 4.6). When only 2 – 4 g female blacknose dace were

123

considered, the standard deviation was reduced to 20% of the mean on May 20

and 17 fish per site were required to detect a 25% difference in gonad size when

α=0.05 and β=0.05, but only 11 fish per site were required when α=0.05 and

β=0.20 or when α and β both equalled 0.10 (Table 4.5). When 2 year old female

blacknose dace were selected only, the CV was reduced to about 15% of the

mean on May 20 and less than 10 fish per site were required to detect a 25%

difference in gonad size. Selecting 2 year old fish or fish that ranged in size from

2 – 4 g would reduce some of the variability associated with gonad development

and increase our ability to detect potential reproductive impacts associated with

contaminant exposure if they exist and would also allow us to collect all of the

required reproductive end-points for the EEM program.

However, these techniques were not successful at reducing variability

during all seasons or for all species. For example, the variability in gonad

development in 2 year old female blacknose dace increased sample size

requirements by about 13.5 - 14.9 times in June (i.e., spawning season) relative

to May (i.e., pre-spawning season). Overall, it is evident that choosing a certain

age and/or size class of fish during certain times of the reproductive season will

reduce data variability and sample sizes required to detect critical effect sizes

with a specified level of power, but these techniques will not work for all species

(e.g., northern redbelly dace).

124

The timing of sampling for small-bodied fish (both single and multi-

spawners) is critical to the success of a study and can only be accomplished

when sufficient basic biological information is available. However, the existing

guidance for sampling prior to the first spawn is not always the best timing. As

the proximity of the first spawn approached, the correlation reduced in strength in

most species. This is probably associated with a reduced synchronization of the

second spawning event between individuals. Furthermore, female northern

redbelly dace showed considerable more variability in gonad development during

the entire study relative to the other fish. Using data for female northern redbelly

dace collected on May 27 (highest spring r2 value) and June 22 (lowest r2 value),

we calculated the sample sizes required to detect a 25% difference in gonad size

at the same three power levels noted above. On May 27, the CV for all female

northern redbelly dace was 57% and sample sizes between 81-134 fish per site

would be required to detect of 25% difference in gonad size (Table 4.5). On

June 22, when the CV was 75%, the sample sizes required per site were more

than double that required on May 27 (Table 4.5). Sampling only 2 year old or 2 –

4 g female northern redbelly dace on May 27 and June 22 did not substantially

reduce data variability or sample size requirements as the CV remained around

50% during each sampling time for both groups (Table 4.5).

In some species, selecting a specific age or size class of fish may be

useful to reduce the data variability associated with gonadal development during

the pre-spawning season. Reducing the variability will reduce the sample size

125

required to detect critical effect sizes with a specified level of power. Conversely,

it also apparent that the techniques used to reduce sample variability may not be

as useful during certain periods of the reproductive season and may not be

suitable for all fish species.

4.6 Conclusions

Our results demonstrate blacknose dace, northern redbelly dace, golden

shiner, and mummichog have extended spawning seasons (spring-summer),

which has been reported for some of these species in other locations. However,

there are differences in the onset of spawning, which is probably related to

geographical differences in environmental conditions. The suite of indices used

in this study provided information about the seasonal changes in energy use and

storage, particularly as it relates to gonad growth. Female blacknose dace

appear not to rely on liver energy stores when the maximum increases in gonad

size occur and may rely on energy derived from active feeding. Future studies

will be needed to clarify this issue. Understanding the basic biology of multiple

spawning, small-bodied fish potentially used in environmental monitoring

programs is important for study design and data interpretation. Overall, we know

the strength of the gonad size – adjusted body weight relationship varies during

the season for each species and that some fish species (i.e., northern redbelly

dace) show more seasonal variability in this parameter than others (i.e.,

blacknose dace). Selecting a specific age class or size class of fish will reduce

126

some of the data variability associated with gonadal development and will also

reduce sample size requirements for future studies in some fish species during

certain times of the year, but these techniques may not be suitable for all

species. Multi-spawning, small-bodied fish can be successfully used in

environmental monitoring programs, but additional basic biological research will

be required in order to continue to facilitate their use.

4.7 Acknowledgements

This project received funding from the Canadian Water Network, Toxic

Substance Research Initiative (Project 205), New Brunswick Innovation Fund,

NexFor and Fraser Papers. The invaluable help of K. Tenzin, J. Peddle, G.

Vallieres, and T. Galloway in the field is greatly appreciated. Graduate student

support for BJG from NSERC Industrial-Post Graduate Scholarship Program and

UNB. KRM receives support from a NSERC Discovery Grant, the Canadian

Water Network Networks of Centres of Excellence and from the Canada

Research Chairs Program.

4.8 References

Blazer, V.S. 2002. Histopathological assessment of gonadal tissue in wild fishes.

Fish Physio Biochem 26: 85-101.

127

Couillard, C.M., and Nellis, P. 1999. Organochlorine contaminants in

mummichog (Fundulus heteroclitus) living downstream of a bleached kraft

pulp mill in the Miramichi Estuary, New Brunswick, Canada. Environ

Toxicol Chem 18: 2545-2556.

DeHaven, J.E., Stouder, D.J., Ratajczak, R., Welch, T.J., and Grossman, G.D.

1992. Reproductive timing in three southern Appalachian stream fishes.

Ecol Fresh Fish 1: 104 - 111.

Delahunty, G., and DeVlaming, V.L. 1980. Seasonal relationships of ovary

weight, liver weight, and fat stores with body weight in the goldfish,

Carassius auratus (L.). J Fish Biol 16: 5 -13.

Dube, M.G., and MacLatchy, D.L. 2000. Endocrine responses of Fundulus

heteroclitus to effluent from a bleached-kraft pulp mill before and after

installation of reverse osmosis treatment of a waste stream. Environ

Toxicol Chem 19: 2788-2796.

Environment Canada. 1997. Fish Survey Expert Working Group:

Recommendations from Cycle 1 review. Ottawa ON: Environment Canada.

EEM/1997/6. 262p.

Environment Canada. 2001. Guidance for determining follow-up actions when

effects have been identified in environmental effects monitoring (EEM).

Environment Canada. Ottawa, ON.

Fraker, M.E., Snodgrass, J.W., and Morgan, F. 2002. Differences in growth and

maturation of blacknose dace (Rhinichthys atratulus) across an urban-rural

gradient. Copeia 4: 1122-1127.

128

Hamilton, H., Paine, M.D., Gibson, W., and Conley, D. 1993. 1992 Operational

monitoring survey of the Lesser Slave River. Volume VI. Prepared for the

Slave Lake Pulp Corporation by EVS Environmental Consultants Ltd., North

Vancouver, British Columbia.

Galloway, B.J., Munkittrick, K.R., Curry, R.A., Wood, C., and Dunn, S. 2004.

Identifying a suitable fish species for monitoring multiple effluents in the

upper Saint John River, Canada. In Borton, D.L., Hall, T.J., Fisher, R.P., &

Thomas, J.F., (eds), Pulp & Paper Mill Effluent Environmental Fate &

Effects, June 1-4, Seattle, WA, pp. 169 -181.

Heins, D.C., and Rabito Jr, F.G. 1986. Spawning performance in North

American minnows: direct evidence of the occurrence of multiple clutches

in the genus Notropis. J Fish Biol 28: 343 - 357.

Kjesbu, O.S., Kryvi, H., and Norberg, B. 1996. Oocyte size and structure in

relation to blood plasma steroid hormones in individually monitored,

spawning Atlantic cod. J Fish Biol 49: 1197 - 1215.

Leblanc, J., Couillard, C.M., and Brethes, J.C.F. 1997. Modifications of the

reproductive period in mummichog (Fundulus heteroclitus) living

downstream from a bleached kraft pulp mill in the Miramichi Estuary, New

Brunswick, Canada. Can J Fish Aquat Sci 54: 2564-2573.

Lowell, R.B., Ribey, S.C., Ellis, I.K., Porter, E.L., Culp, J.M., Grapentine, L.C.,

McMaster, M.E., Munkittrick, K.R., and Scroggins, R.P. 2003. National

assessment of the pulp and paper environmental effects monitoring data.

NWRI Publ. 03-521. National Water Research Institute, Burlington, 124 p.

129

Munkittrick, K.R., McGeachy, S.A., McMaster, M.E., and Courtenay, S.C. 2002.

Overview of freshwater fish studies from the pulp and paper environmental

effects monitoring program. Water Qual Res J Can 37: 49-77.

Noggle, J.J., Smith, J.T., Ruessler, D.S., Quinn, B.P., Holm, S.E., Sepulveda,

M.S., Gross, T.S. 2004. Paper mill process modifications reduce biological

effects on largemouth bass and eastern gambusia. In Borton, D.L., Hall,

T.J., Fisher, R.P., and Thomas, J.F. (eds), Pulp & Paper Mill Effluent

Environmental Fate & Effects, June 1-4, Seattle, WA, pp. 14 -24.

Pankhurst, N.W., and Porter, M.J.R. 2003. Cold and dark or warm and light:

variations on the theme of environmental control of reproduction. Fish

Physio Biochem 28: 385 - 389.

Richardson, J. 1997. Use of a fish health profile for monitoring pulp and paper mill

effluents. In Stuthridge, T., van den Heuval, M., Marvin, N., Slade, A., and

Clifford, J. (eds), Proceedings of 3rd International Conference on Fate and

Effects of Pulp and Paper Effluents, November 10-13, Rotorua, NZ, pp.

474-479.

Rinchard, J., and Kestemont, P. 1996. Comparative study of reproductive

biology in single- and multiple spawner cyprinid fish. I. Morphological and

histological features. J Fish Biol 49: 883 - 894.

Rinchard, J., and Kestemont, P. 2003. Liver changes related to oocyte growth in

roach, a single spawner fish, and in bleak and white bream, two multiple

spawner fish. Inter Rev Hydrobiol 88: 68 - 76.

130

Roberts, J.H., and Grossman, G.D. 2001. Reproductive characteristics of female

longnose dace in Coweeta Creek drainage, North Caroline, USA. Ecol

Fresh Fish 10: 184 - 190.

Scott, W.B., and Crossman, E.J. 1998. Freshwater fishes of Canada. Galt

House Publications Ltd. Oakville, ON, Canada.

Shimizu, A. 1997. Reproductive cycles in a reared strain of mummichog, a daily

spawner. J Fish Biol 51: 724-737.

Tartar, D.C. 1969. Some aspects of reproduction of western blacknose dace,

Rhinichthys atratulus meleagris Agassiz, in Doe Run, Meade County,

Kentucky. Trans Amer Fish Soc 98: 454 - 459.

Taylor, M.H., and DiMichele, L. 1980. Ovarian changes during the lunar

spawning cycle of Fundulus heteroclitus. Copeia 1: 118-125.

Tomkiewicz, J., Tybjerg, L., and Jespersen, A. 2003. Micro-and macroscopic

characteristics to stage gonadal maturation of female Baltic cod. J Fish

Biol 62: 253 - 275.

Tyler, C.R., and Sumpter, J.P. 1996. Oocyte growth and development in

teleosts. Rev Fish Biol Fisheries 6: 287 - 318.

Tyler, C.R., Sumpter, J.P., and Witthames, P.R. 1990. The dynamics of oocyte

growth during vitellogenesis in rainbow trout, Salmo gairdneri. Biol

Reprod 43: 202 - 209.

Wallace, R.A., and Selman, K. 1981. The reproductive activity of Fundulus

heteroclitus females from Woods Hole, Massachusetts, as compared to

more southern locations. Copeia 1: 212-215.

131

Yeardley Jr, R.B. 2000. Use of small forage fish for regional streams wildlife risk

assessment: relative bioaccumulation of contaminants. Environ Monitor

Assess 65: 559-585.

132

Table 4.1. Microscopic characteristics for the determination of oocyte developmental stages (modified from Blazer 2002).

Ovarian stage Ovarian stage code

Description of the most advanced oocytes present within ovary.

Previtellogenic (Primary Growth)

1 Includes chromatin nucleolar oocytes (single nucleolus) and perinucleolar oocytes (multiple nucleoli).

Previtellogenic (Cortical alveoli)

2 Cortical alveoli oocyte with cortical alveoli beginning to appear in the periphery of the cytoplasm.

Early vitellogenic 3 Cortical alveoli present throughout the cytoplasm but yolk globules become apparent.

Mid Vitellogenic 4 Cortical alveoli are densely packed into the cell periphery with most of the cytoplasm containing yolk globules.

Mature 5 Yolk globules fuse into a homogenous mass.

Atretic 6 Breakdown of the nucleus, vitelline envelope, and increase in the size and number of the follicular cells.

Post-spawning 7 Post-ovulatory follicles and previtellogenic oocytes.

133

Table 4.2. Means ± SE (n) and minimum and maximum (in parentheses) values for length, weight, condition factor (k),

liversomatic index (LSI), and gonadosomatic index (GSI) in adult female and male blacknose dace (Rhinichthys

atratulus), golden shiner, mummichog (Fundulus heteroclitus), and northern redbelly dace (Phoxinus eos)

collected from various sites in southern New Brunswick, Canada. Within a column, differences (p < 0.05) among

sampling dates are denoted by different superscript uppercase letters. (*Significant interaction within ANCOVA

model).

Species Sex Date Length (mm) Body Weight (g) Ka LSIb GSIc

Blacknose

Dace

F May 4, 2003 69.9±1.9(16)AB

(62.0-93.0)

3.84±0.42(16)AB

(2.37-8.96)

1.07± 0.03(16)*

(0.88 - 1.29)

3.17± 0.21(16)C

(1.34 - 4.39)

9.07±0.47(16)A

(6.28 - 13.3)

F May 20, 2003 70.6±1.1(24)A

(63.0-78.0)

4.13±0.23(24)A

(2.51-6.18)

1.14± 0.02(24)*

(1.00 - 1.35)

3.33 ± 0.24(24)C

(0.42 - 5.11)

16.5±0.8(24)B

(10.3 - 25.0)

F June 11, 2003 70.9±1.0(35)A

(61.0-83.0)

4.05±0.18(35)A

(2.19-6.46)

1.11± 0.02(35)*

(0.93 - 1.35)

2.09± 0.12(35)A

(0.73 - 3.94)

12.9±0.9 (35)A

(2.92 - 21.2)

F July 28, 2003 65.3±1.1(29)B

(58.0-83.0)

2.95±0.15(29)B

(2.05-6.00)

1.04 ± 0.01(29)*

(0.89 -1.19)

0.92± 0.05(29)B

(0.47 - 1.56)

2.33±0.14(29)C

(0.31 - 3.65)

F Aug 19, 2003 67.0±1.0(37)AB

(59.0-81.0)

3.07±0.15(37)B

(2.02-5.28)

1.00 ± 0.01(37)*

(0.89 - 1.07)

0.96± 0.05(37)B

(0.25 - 1.64)

3.41±0.11(37)D

(1.11-4.56)

134

Species Sex Date Length (mm) Body Weight (g) Ka LSIb GSIc

Blacknose

Dace

M May 4, 2003 65.0±1.4(24)A

(51.0-74.0)

3.05±0.22(24)A

(1.11-4.94)

1.05±0.02(24)*

(0.84-1.27)

2.16±0.18(24)A

(0.76-3.83)

1.00±0.05(24)A

(0.66-1.67)

M May 20, 2003 68.0±0.97(21)A

(60.0-77.0)

3.27±0.15(21)A

(2.24-4.88)

1.03±0.01(21)*

(0.93-1.13)

1.76±0.10(21)A

(1.05-2.59)

1.00±0.04(21)A

(0.66-1.53)

M June 11, 2003 62.1±1.8(15)A

(51.0-73.0)

3.06±0.26(15)A

(1.61-4.81)

1.24±0.03(15)*

(1.08-1.45)

1.69±0.21(15)A

(0.68-3.23)

0.58±0.08(15)B

(0.06-1.17)

M July 28, 2003 64.7±1.0(22)A

(57.0-72.0)

2.88±0.12(22)A

(1.81-3.87)

1.05±0.02(22)*

(0.93-1.23)

0.94±0.05(22)B

(0.57-1.44)

0.30±0.03(22)C

(0.08-0.73)

M Aug 19, 2003 64.7±1.9(18)A

(50.0-77.0)

2.85±0.25(18)A

(1.20-4.97)

1.00±0.01(18)*

(0.91-1.09)

0.84±0.08(18)B

(0.38-1.48)

0.30±0.05(18)C

(0.03-0.71)

Golden

Shiner

F June 22, 2003 78.6±1.9(20)AB

(67.0-94.0)

5.02±0.45(20)AB

(2.57-9.22)

0.99± 0.02(20)A

(0.85 - 1.15)

1.48±0.10(20)AB

(0.84-2.36)

7.12±0.84(20)B

(1.11-13.4)

F July 9, 2003 74.0±1.4(41)B

(60.0-106)

4.35±0.32(41)AB

(2.15-13.9)

1.02± 0.01(41)A

(0.88-1.17)

1.35±0.06(41)A

(0.57-2.43)

5.15±0.51(41)B

(1.27-17.2)

F Aug 7, 2003 82.1±2.6(26)A

(59.0-107)

6.14±0.69(26)A

(1.80-15.7)

1.00± 0.02(26)A

(0.88-1.17)

1.62±0.09(26)AB

(0.85-2.53)

2.58± 0.42(26)AC

(1.05-10.7)

135

Species Sex Date Length (mm) Body Weight (g) Ka LSIb GSIc

Golden

Shiner

F Aug 29, 2003 77.0±1.7(19)AB

(64.0-95.0)

4.68±0.43(19)AB

(2.51-10.8)

0.98± 0.02(19)A

(0.82-1.26)

1.38±0.06(19)AB

(0.94-1.96)

1.58± 0.07(19)A

(1.01-2.23)

F April 24, 2004 73.6±1.7(34)B

(55.0-96.0)

4.15±0.35(34)B

(1.55-9.41)

0.97± 0.02(34)A

(0.73-1.17)

1.71±0.08(34)B

(0.66-2.95)

2.75±0.22(34)C

(1.06-6.09)

Golden

Shiner

M June 22, 2003 77.5±1.9(20)AB

(65.0-98.0)

4.95±0.40(20)A

(2.71-9.10)

1.02± 0.02(20)A

(0.91-1.22)

1.14± 0.05(20)A

(0.82-1.57)

2.37± 0.20(20)*

(1.09-4.75)

M July 9, 2003 79.1±2.4(21)A

(62.0-102)

5.57±0.56(21)A

(2.21-12.0)

1.05± 0.02(21)A

(0.77-1.23)

1.11± 0.05(21)A

(0.60-1.45)

1.81± 0.14(21)*

(0.74-2.66)

M Aug 7, 2003 79.3±2.5(16)A

(66.0-104)

5.37±0.70(16)A

(2.39-13.6)

1.00± 0.02(16)AC

(0.83-1.21)

1.16± 0.07(16)A

(0.54-1.79)

1.12± 0.17(16)*

(0.33-2.33)

M Aug 29, 2003 71.3±1.5(23)B

(62.0-94.0)

3.47±0.28(23)B

(2.09-8.70)

0.92± 0.01 (23)C

(0.81-1.05)

0.98± 0.05(23)A

(0.47-1.44)

0.45± 0.03(23)*

(0.14-0.87)

Northern

redbelly Dace

F May 27, 2003 56.8 ± 00.8(60)B

(43.0-72.0)

1.76±0.08(60)B

(0.71-3.98)

0.93 ± 0.01(60)C

(0.76 - 1.19)

2.08±0.08(60)AC

(0.39 - 3.30)

9.70 ± 0.71(60)A

(0.44 - 21.5)

F June 12, 2003 52.8±0.9(45)A 1.49±0.08(45)ABC 0.97±0.02(45)AD 1.77±0.09(45)A 9.94 ± 0.85(45)A

136

Species Sex Date Length (mm) Body Weight (g) Ka LSIb GSIc

(44.0-65.0) (0.58-2.87) (0.68 - 1.30) (0.77 - 4.19) (0.39 - 21.5)

F June 22, 2003 55.5±0.8(72)AB

(44.0-75.0)

1.72±0.08(72)AB

(0.75-4.13)

0.97±0.01(72)ACD

(0.83 - 1.18)

2.50 0.09(72)BC

(0.78 - 4.49)

6.98±0.62(72)CD

(0.44 - 23.0)

Northern

redbelly Dace

F July 27, 2003 54.5±1.2(31)AB

(45.0-66.0)

1.59±0.11(31)ABC

(0.88-2.79)

0.93±0.02(31)DC

(0.80 - 1.13)

1.71±0.08(31)A

(0.53 - 2.70)

2.07±0.22(31)B

(0.74 - 6.25)

F Aug 27, 2003 53.3±0.8(50)AB

(45.0-70.0)

1.36±0.07(50)C

(0.85-3.20)

0.87 ± 0.01(50)B

(0.77 - 0.99)

2.04±0.09(50)AC

(0.61 - 3.79)

2.09 ± 0.11(50)B

(0.67 - 4.09)

F Oct 12, 2003 53.4±0.8(38)AB

(42.0-63.0)

1.37±0.06(38)AC

(0.62-2.61)

0.87 ± 0.01(38)BC

(0.78 - 1.04)

1.94±0.09(38)ABC

(0.95 - 3.11)

3.20±0.22(38)BCD

(0.68 - 5.71)

F April 24, 2004 54.6±0.8(50)AB

(45.0-66.0)

1.50±0.07(50)ABC

(0.81-2.89)

0.89 ± 0.01 (50)B

(0.79 - 1.07)

2.27±0.10 (50)C

(0.87-4.12)

4.03±0.28(50)D

(1.00-9.59)

Mummichog F May 8, 2003 72.9±1.9(21)A

(57.0-82.0)

3.89±0.30(21)A

(1.66-5.62)

0.95±0.02(21)*

(0.82-1.09)

3.61± 0.22(21)A

(1.62-5.28)

4.45±0.16(21)*

(2.89-5.26)

F June 20, 2003 56.7±1.4(43)B

(42.0-83.0)

2.37±0.21(43)B

(0.94-7.03)

1.21± 0.02(43)*

(0.96-1.67)

3.28± 0.17(43)A

(1.54-6.73)

23.7± 1.0(43)*

(8.81-37.6)

F July 28, 2003 69.6±0.83(45)A 4.21±0.16(45)A 1.22± 0.01(45)* 4.03± 0.13(45)AB 1.85± 0.16(45)*

137

Species Sex Date Length (mm) Body Weight (g) Ka LSIb GSIc

(59.0-88.0) (2.47-8.08) (1.09-1.45) (2.34-5.34) (1.15-7.08)

F Aug 28, 2003 71.3±1.3(40)A

(45.0-90.0)

4.39±0.24(40)A

(1.01-8.72)

1.16± 0.01(40)*

(0.98-1.32)

3.35± 0.12(40)A

(1.88-4.72)

1.52± 0.05(40)*

(1.05-2.47)

F Oct 12, 2003 66.2±1.5(41)A

(54.0-89.0)

3.69±0.30(41)A

(1.73-9.51)

1.17± 0.01(41)*

(0.95-1.35)

5.32 ± 0.15(41)C

(2.54-7.42)

1.84± 0.05(41)*

(1.20-2.93)

F April 24, 2004 60.0±1.3(49)B

(48.0-85.0)

2.27±0.20(49)B

(0.93-7.46)

0.95± 0.01(49)*

(0.80-1.22)

4.07± 0.14(49)B

(2.02-6.12)

3.46 ± 0.10(49)*

(1.93-5.76)

Mummichog M May 8, 2003 65.5±2.8(12)A

(46.0-83.0)

2.88±0.40(12)A

(0.75-6.05)

0.94± 0.03(12)B

(0.77-1.06)

2.88± 0.28(12)A

(1.57-4.95)

1.53± 0.23(12)B

(0.36-3.17)

M June 20, 2003 73.1±2.6(8)A

(68.0-90.0)

4.09±0.49(8)A

(3.07-7.09)

1.02± 0.02(8)B

(0.97-1.18)

1.51± 0.19(8)B

(1.15-2.84)

2.33± 0.15(8)B

(1.46-2.95)

M July 28, 2003 63.1±2.7(20)A

(39.0-79.0)

3.44±0.43(20)A

(0.66-6.45)

1.22± 0.02(20)A

(1.08-1.34)

3.59± 0.23(20)A

(2.08-5.52)

0.52± 0.07(20)A

(0.10-1.13)

M Aug 28, 2003 64.5±1.3(22)A

(52.0-75.0)

3.10±0.20(22)A

(1.37-5.02)

1.12± 0.01(22)C

(0.97-1.22)

3.47± 0.14(22)A

(2.28-4.85)

0.39± 0.03(22)A

(0.15-0.87)

M Oct 12, 2003 66.5±1.3(24)A 3.67±0.24(24)A 1.20 ±0.02(24)A 5.54± 0.22(24)C 0.63± 0.10(24)A

138

Species Sex Date Length (mm) Body Weight (g) Ka LSIb GSIc

(55.0-78.0) (1.75-5.91) (1.05-1.35) (3.46-7.83) (0.13-2.01)

aK = 100000*(body weight/(total length3)); bLSI = 100*(liver weight/(body weight – liver weight)); cGSI = 100*(gonad weight/(body weight – gonad weight)).

139

Table 4.3. Log10 regression estimates for adult female and male blacknose dace

(Rhinichthys atratulus) condition and gonad size and condition factor for

adult female mummichog (Fundulus heteroclitus) gonad size collected

from sites in southern New Brunswick in 2003 and 2004.

Species Sex Parameter Collection Date Slope Intercept n p r2 Blacknose

Dace F Condition

Factor May 4, 2003 3.45 -5.80 16 <0.0005 0.94

Condition Factor

May 20, 2003 3.69 -6.22 24 <0.0005 0.96

Condition Factor

June 11, 2003 3.04 -5.02 35 <0.0005 0.90

Condition Factor

July 28, 2003 2.79 -4.60 29 <0.0005 0.93

Condition Factor

Aug 19, 2003 3.02 -5.03 37 <0.0005 0.97

M Condition

Factor May 4, 2003 3.71 -6.27 24 <0.0005 0.98

Condition Factor

May 20, 2003 3.14 -5.25 21 <0.0005 0.92

Condition Factor

June 11, 2003 3.04 -4.99 15 <0.0005 0.95

Condition Factor

July 28, 2003 2.56 -4.18 22 <0.0005 0.92

Condition Factor

Aug 19, 2003 3.21 -5.39 18 <0.0005 0.99

Golden shiner M Gonad Size June 22, 2003 0.72 -1.46 20 0.013 0.30

Gonad Size July 9, 2003 0.69 -1.56 21 0.002 0.41 Gonad Size Aug 7, 2003 1.80 -2.58 16 <0.0005 0.70 Gonad Size Aug 29, 2003 0.93 -2.34 23 0.004 0.34

Mummichog F Gonad Size May 8, 2003 1.04 -1.38 21 <0.0005 0.86 Gonad Size June 20, 2003 0.73 -0.58 43 <0.0005 0.60 Gonad Size July 28, 2003 1.33 -1.97 45 <0.0005 0.47 Gonad Size August 28,

2003 1.16 -1.92 40 <0.0005 0.89

Gonad Size October 12, 2003

1.18 -1.83 41 <0.0005 0.94

Gonad Size April 24, 2004 1.07 -1.49 49 <0.0005 0.87 F Condition

Factor May 8, 2003 3.35 -5.68 21 <0.0005 0.96

Condition Factor

June 20, 2003 2.89 -4.73 43 <0.0005 0.94

Condition Factor

July 28, 2003 3.05 -5.01 45 <0.0005 0.95

Condition Factor

August 28, 2003

3.25 -5.39 40 <0.0005 0.98

140

Species Sex Parameter Collection Date Slope Intercept n p r2 Mummichog F Condition

Factor October 12, 2003

3.36 -5.59 41 <0.0005 0.98

Condition Factor

April 24, 2004 3.39 -5.72 49 <0.0005 0.98

141

Table 4.4. Relationship of log10 ovary weight to log10 adjusted body weight in

female fish on various dates in 2003 and 2004.

Species Sex Sample Date Regression Equation p r2 n

Blacknose Dace

F May 4, 2003 Y= 0.15x – 0.20 <0.0005 0.97 16

May 20, 2003 Y= 0.27x - 0.35 <0.0005 0.80 24 June 11, 2003 Y= 0.15x - 0.08 <0.0005 0.43 35 July 28, 2003 Y= 0.04x - 0.05 <0.0005 0.76 29 August 19, 2003 Y= 0.05x - 0.04 <0.0005 0.87 37 Golden Shiner

F June 22, 2003 Y=0.104x – 0.134 <0.0005 0.58 20

July 9, 2003 Y=0.142x – 0.346 <0.0005 0.63 41 Aug 7, 2003 Y=0.041x – 0.087 <0.0005 0.61 34 Aug 29, 2003 Y=0.024x – 0.037 <0.0005 0.91 20 Northern Redbelly Dace

F May 27, 2003 Y= 0.146x – 0.073 <0.0005 0.38 60

June 12, 2003 Y=0.089x + 0.013 <0.0005 0.28 45 June 22, 2003 Y=0.062x + 0.011 <0.0005 0.19 72 July 27, 2003 Y=0.044x - 0.031 <0.0005 0.56 31 Aug 27, 2003 Y=0.027x - 0.008 <0.0005 0.64 50 Oct 12, 2003 Y=0.056x - 0.031 <0.0005 0.55 38 April 24, 2003 Y = 0.039x+0.002 <0.0005 0.27 50 Mummichog F May 8, 2003 Y=0.048x – 0.011 <0.0005 0.85 21 June 20, 2003 Y=0.158x + 0.126 <0.0005 0.58 43 July 28, 2003 Y=0.045x – 0.107 <0.0005 0.39 45 Aug 28, 2003 Y=0.016x – 0.003 <0.0005 0.79 40 Oct 12, 2003 Y=0.022x-0.012 <0.0005 0.94 41 April 24, 2004 Y = 0.038x-0.007 <0.0005 0.93 49

142

Table 4.5. Sample sizes required to detect a 25% difference in gonad size at different levels of power for female northern redbelly dace Phoxinus eos and blacknose dace Rhinichthys atratulus.

Sample sizes at specified power levels

Species Collection Date

Group Coefficient of Variation

(CV,%)

α=0.05, β=0.05

α=0.05, β=0.20

α=0.10, β=0.10

Northern

Redbelly Dace May 27 All fish 57 134 81 89

2 year old 56 126 76 83 2 – 4 g 44 79 48 52 June 22 All fish 75 237 143 156 2 year old 44 157 95 104 2 – 4 g 57 134 81 88

Blacknose Dace May 20 All fish 24 24 15 16 2 year old 15 9 6 6 2 – 4 g 20 17 11 11 June 11 All fish 41 71 43 47 2 year old 56 134 81 89 2 – 4 g 52 104 63 69

143

Table 4.6. Comparison of the relationship of log10 ovary weight to log10 body

weight between all female blacknose dace (Rhinichthys atratulus) with 2

year old fish and fish weighing 2 – 4 g 3 year old collected from Milkish

Brook on May 20, 2003.

Group Regression Equation R2 n

All fish Y= 0.27x - 0.35 0.80 24

2 year old Y= 0.18x – 0.20 0.94 8

2 – 4 g Y= 0.23x - 0.34 0.88 12

144

Figure 4.1. Mean monthly ambient temperature for Milkish Brook during the

study period. Upper bars depict monthly high temperatures and lower

bars depict low temperatures.

0

5

10

15

20

25

May June July Aug Sept OctMonth

Wat

er T

empe

ratu

re (°

C)

0

5

10

15

20

25

May June July Aug Sept OctMonth

Wat

er T

empe

ratu

re (°

C)

145

Figure 4.2. Percentage of each oocyte occurrence (except stage 1, primary

oocytes) in female blacknose dace Rhinichthys atratulus caught in

Milkish Brook. Dark grey bars represent stage 2 oocytes; light grey

bars represent stage 3 oocytes; white bars represent stage 4 oocytes;

bars with a diamond-grid represent stage 5 oocytes; bars with

horizontal-grid represent stage 6 oocytes.

0

5

10

15

20

25

30

35

40

2 3 4 2 3 4 2 3 4 5 2 6 2 6Egg stage

Per

cent

(%)

May 4 Aug 19July 28June 11May 20

146

Figure 4.3. Percentage of each oocyte occurrence (except stage 1, primary

oocytes) in female golden shiner Notemigonus crysoleucas caught in

Little Chamcook Lake. Dark grey bars represent stage 2 oocytes; light

grey bars represent stage 3 oocytes; white bars represent stage 4

oocytes.

0

5

10

15

20

25

30

2 3 4 2 3 4 2Egg stage

Per

cent

(%)

June 9 June 22 Aug 29

147

Figure 4.4. Percentage of each oocyte occurrence (except stage 1, primary

oocytes) in female northern redbelly dace Phoxinus eos caught in

beaver pond in Keswick River. Dark grey bars represent stage 2

oocytes; light grey bars represent stage 3 oocytes; white bars represent

stage 4 oocytes; bars with a diamond-grid represent stage 5 oocytes;

bars with a horizontal-grid represent stage 6 oocytes.

0

10

20

30

40

50

60

2 3 4 2 3 4 2 3 4 5 6 2 6Egg stage

Per

cent

(%)

May 27 June 12 June 22 Aug 27

148

Figure 4.5. Percentage of each oocyte occurrence (except stage 1, primary

oocytes) in female mummichog Fundulus heteroclitus caught in a salt

marsh located adjacent to Taylor's Island. Dark grey bars represent

stage 2 oocytes; light grey bars represent stage 3 oocytes; white bars

represent stage 4 oocytes; bars with a diamond-grid represent stage 5

oocytes; bars with a horizontal-grid represent stage 6 oocytes.

0

5

10

15

20

25

30

35

40

45

2 3 4 2 3 4 5 2 6 2 6Egg stage

Per

cent

(%)

May 8 June 20 Aug 28July 28

149

CHAPTER 5. GENERAL DISCUSSION

The previous chapters have presented the results of my research designed

to assess the suitability of various fish species for environmental monitoring

programs. Overall, my research hypothesis was to determine whether fish can

be used to assess the relative contribution of individual anthropogenic stressors

to the existing environmental conditions in a river exposed to multiple

anthropogenic stressors.. More specifically, the objectives of my thesis were to

compare the whole-organism responses of small-bodied fish (i.e., slimy sculpin

and blacknose dace) and large-bodied fish (i.e., white sucker and yellow perch)

along a downstream gradient in a river exposed to pulp and paper mill effluents,

municipal sewage wastewater, and agricultural runoff (i.e., manure); identify the

fish species that is best suited to assess the relative contribution of individual

anthropogenic stressors in a river exposed to multiple anthropogenic stressors;

determine which life history characteristics most influenced the ability of a fish

species to exhibit the measured whole organism responses; and provide

sampling guidance for the use of multiple spawning, small-bodied fish species for

use in environmental monitoring programs.

The first part of my thesis focused on comparing the whole-organism

responses of white sucker and slimy sculpin collected downstream of pulp and

paper mill effluents, municipal sewage wastewater, and agricultural inputs on the

Saint John River near the city of Edmundston in northwest New Brunswick.

Results from the study showed that white sucker and slimy sculpin exhibited

differences in whole-organism responses, despite being collected from the same

150

sites. Slimy sculpin collected downstream of the municipal sewage discharges

and pulp mill effluent had greater growth, condition, and liver size and exhibited

no significant differences in gonad size relative to sculpin collected from

upstream reference sites.

In some monitoring situations, there is a lot of concern from stakeholders

that the measured whole-organism responses of fish do not reflect site-specific

environmental conditions. Since the receiving environment near Edmundston is

complex, establishing the residency patterns and home ranges of slimy sculpin

was an important issue to address. The stable carbon (δ13C) and nitrogen (δ15N)

isotope data showed that slimy sculpin collected from the Saint John River near

Edmundston had small home ranges (< 500 m), and the measured whole

organism responses reflected local conditions. In addition, stable carbon (δ13C)

and nitrogen (δ15N) isotope values of sculpin collected downstream of the pulp

mill, paper mill, and municipal sewage showed these sites were isotopically

enriched relative to upstream reference sites located at Clair and St. Hilaire. The

pulp mill and paper mill are located on opposite sides of the Saint John River, in

Canada (New Brunswick) and the United States (Maine), respectively. Sculpin

collected downstream of the paper mill showed no significant differences in

length, body weight, age, condition factor, liver size, and gonad size compared to

fish from upstream reference sites. Interestingly, the sculpin collected

downstream of the paper mill had unique stable isotope signatures relative to

sculpin collected immediately across the river and downstream of the pulp mill

showing sculpin were not moving across the river either. Gray [2003], also

151

showed slimy sculpin exhibit a high degree of spatial and temporal residency

within watercourses of New Brunswick using PIT tags and stable isotopes.

It was also surprising that the relative liver sizes of white sucker collected

from sites located downstream of the pulp mill effluent, upstream of the pulp mill

effluent, and the reference site located at St. Hilaire on the Saint John River did

not appear to be affected. However, with further analyses, white sucker liver size

was greater relative to white sucker collected from First Lake and Ogilvie Lake

(two pristine New Brunswick lakes near Edmundston). The LSI of white sucker

from the two New Brunswick reference lakes were similar to LSI values for white

sucker collected from reference sites in northern Ontario, and white suckers from

the Saint John River had livers similar in size to those of white sucker collected

downstream of a sulfite pulp mill in Kapuskasing, northern Ontario [Munkittrick et

al. 2000]. The increase in white sucker liver size may be reflective of the normal

situation for the Saint John River or it could be due to one or more unidentified

upstream source(s) of contamination (i.e., poultry processing facility). On the

other hand, the white sucker liver sizes could have been associated with the

mobility of the species in this river system. It was apparent that follow-up work

was required to help determine among these possibilities and whether the

responses persisted.

One of the most challenging, controversial, and criticized aspects of

environmental monitoring programs is any attempt to determine the ecological

152

relevance of the measured changes in a particular set of parameters. The initial

fish collections for my thesis were part of a larger effects-driven cumulative

effects assessment (CEA) of the Saint John River basin. The overall goal of the

effects-driven CEA is to help identify both natural factors (e.g., water

temperature, resource competition, etc.) and anthropogenic factors (e.g., pulp

mills, municipal sewage, hydro dams, etc.) that may enhance or limit fish

survival, growth, and reproduction and determine whether the current situation is

sustainable [Munkittrick et al. 2000]. As such, white sucker and slimy sculpin

performance was used as baseline data in an initial attempt to identify

ecologically relevant changes (i.e., magnitude of changes) in fish energy storage

and utilization for the upper Saint John River and would be important for

understanding the potential of future stressors to impact local fish populations. In

addition, the patterns of whole-organism responses could help identify the

potential cause(s) of the measured responses [Gibbons and Munkittrick 1994;

Munkittrick et al. 2000]. Since baseline fish performance data for the upper Saint

John River near Edmundston were not available, fish performance data from the

national pulp and paper EEM program were used to help put into context the

relative magnitude of the differences in fish performance downstream of the

municipal sewage (i.e., DS Mad) and pulp mill (i.e., DS Pulp) relative to fish

collected at reference sites located upstream of Edmundston. Intra- and inter-

annual differences in the magnitude of fish performance responses are

presented in Table 5.1. The magnitude of responses and changes in the

patterns of the measured responses outlined in Table 5.1 provided invaluable

153

information regarding the initial assessment of fish performance in the upper

Saint John River near the city of Edmundston and will be discussed in more

detail in the following paragraphs.

Except for male GSI, sculpin collected downstream of municipal sewage

and pulp mill effluent discharge points in October 1999 exhibited increases in

condition, LSI, and GSI (Table 5.1). During this same time, male sculpin

collected downstream of the sewage and pulp mill effluent exhibited a small

decrease in GSI, but female sculpin exhibited an increase in GSI and this was

more evident for fish collected downstream of the pulp mill effluent (Table 5.1).

Some of the measured responses in sculpin downstream of the municipal

sewage and pulp mill effluent were very large relative to the average national

responses seen in the pulp and paper EEM program [Munkittrick et al. 2002;

Lowell et al. 2004]. The largest difference observed for LSI in female sculpin

collected downstream of the pulp mill occurred in 1999; mean LSI values went

from 1.14 ± 0.13 at St. Hilaire to 2.47 ± 0.33 downstream of the pulp mill, a 117%

increase, and is close to the largest measured responses reviewed in Cycle 2

and 3 of the EEM program [Munkittrick et al. 2002; Lowell et al. 2004]. The

largest measured difference in condition factor of female sculpin downstream of

the pulp mill occurred during the fall of 2000; mean condition factor went from

1.03 ± 0.02 at St. Hilaire to 1.18 ± 0.05 downstream of the pulp mill, a 15%

increase, also a large increase within the context of the EEM program. One of

the most interesting observations was in sculpin collected downstream of

154

municipal sewage wastewater. Except for female GSI, male and female sculpin

collected downstream of the sewage showed very large increases in condition

and liver size relative to reference fish and fish exposed to pulp mill effluent. In

fact, the increases in condition factor for sculpin collected downstream of the

sewage were greater than any of the measured responses outlined in the

national review of the Cycle 2 and Cycle 3 pulp and paper EEM program

[Munkittrick et al. 2002; Lowell et al. 2004].

Consistency in the patterns of the measured whole-organism responses

between species and between sexes is also important to consider. Male and

female sculpin collected downstream of the sewage and pulp mill effluent in

March 2002 (a few weeks before spawning) showed concurrent increases in

condition, LSI, and GSI which were also noted in female sculpin collected

downstream of the pulp mill in December 1999 (Table 5.1). Increases in

condition and LSI with concomitant increase in both male and female GSI have

also been noted in spoonhead sculpin and longnose sucker collected

downstream of a mill in Hinton, Alberta; yellow perch in Cabano, Quebec; and

white sucker in Jonquieres, Quebec and reflects an increase in food availability

downstream of the discharges, consistent with nutrient enrichment [Gibbons et

al. 1998; Munkittrick et al. 2002]. Inconsistent responses in fish parameters

reflecting energy use and storage (i.e., condition, LSI, GSI) can indicate

metabolic disruption. For example, the pattern of responses for male sculpin

collected downstream of sewage and pulp mill effluent in October 1999 has also

155

been noted at another Canadian pulp mill [Munkittrick et al. 1994]. This response

pattern has also been referred to as a “Jackfish Bay” type of response or

metabolic disruption as fish seem to access enough food but show decreases in

gonad size [Gibbons and Munkittrick 1994; Munkittrick et al. 2000]. It is the most

common response pattern observed in the fish population surveys for the

national pulp and paper EEM program [Environment Canada 2003]. Collecting

sculpin in late summer-early fall could lead some investigators to erroneously

conclude that metabolic disruption may be an issue in fish exposed to sewage

and pulp mill effluent in Edmundston.

Slimy sculpin show a different pattern of reproductive development than

many other Canadian spring-spawning species. Gonadal recrudescence does

not initiate in females until very late fall, and continues throughout the winter

period [Gray 2003]. When sculpin were collected closer to the spawning season

(December 1999) and just a few weeks prior to the spawning season (March

2002) the responses of both male and female sculpin were the same (i.e.,

increased condition, LSI, and GSI) and suggested that fish were living in nutrient-

rich environments that promoted growth. These responses patterns also

suggested that sculpin were not investing energy into reproduction during the late

summer – early fall and collecting fish during these times would provide no

suitable reproductive information.

For future studies with slimy sculpin in the upper Saint John River near the

city of Edmundston, sampling should be done in late fall – spring when gonadal

156

development has sufficiently progressed such that all reproductive parameters

can be measured. It would also be more likely to detect potential reproductive

impacts during a time of the year when the fish is investing increased energy into

reproduction. However, the availability of a spring, prespawning period varies

from year to year, dependent on the pace of snow melt and the timing of the start

of the spring freshet.

Together, data from my initial fish survey and the fathead minnow (P.

promelas) flow-through bioassay [Parrot et al. 2003] and microcosm flow-through

invertebrate exposure [Culp et al. 2003] suggested that the improved energy

storage and utilization of aquatic biota exposed to pulp mill effluent near the city

of Edmundston was related to increased nutrients.

In contrast, male and female white sucker consistently exhibited

decreases (except male LSI and GSI in October 1999) in the measured whole-

organism responses and the magnitude of the differences varied between years

(Table 5.1). Decreases in all three parameters have also been measured in

silver redhorse sucker in Kingsey Falls, Quebec; shorthead redhorse sucker in

Trenton, Ontario; yellow perch in Nackawic, New Brunswick; and rock bass in

Shawinigan, Quebec [Munkittrick et al. 2002]. A decrease in condition, liver size,

and gonad size indicates decreased food availability, but it is not possible to

determine from the initial data which cases have reduced food at the exposure

site, versus those with increased food at the reference site [Munkittrick et al.

157

2002]. Future studies would be required to determine whether these responses

persisted; if the response pattern persisted then an alternative reference site

would be selected and the benthic invertebrate community would be examined in

detail [Munkittrick et al. 2002].

Overall, the results from my initial study showed slimy sculpin collected

downstream of municipal sewage and pulp mill effluent were living in a nutrient-

rich environment that facilitated growth and indicate the slimy sculpin is a suitable

fish species for monitoring complex receiving environments. The responses of

slimy sculpin and white sucker differed, and were perhaps in relation to

differences in life history characteristics. In my initial attempt to identify

ecologically significant changes in fish performance for the upper Saint John

River, I showed that the differences downstream of the sewage and pulp mill

discharges were quite large relative to what has been observed other sites in

Canada receiving pulp mill effluent and can be outside of what would be

considered normal for slimy sculpin in New Brunswick.

Although white sucker and slimy sculpin showed different whole-organism

responses, I wanted to determine whether other large-bodied and small-bodied

fish would show similar responses. Would white sucker and yellow perch show

similar responses? Would the responses of slimy sculpin and blacknose dace be

similar? Would the white sucker and slimy sculpin results be consistent between

years? The second objective of my thesis was to deal with these issues (see

158

Chapter 3). In this study, I attempted to clarify local impacts on wild fish, expand

fish collections to include other species to examine potential species differences

in responses, continue to de-couple other confounding factors at this study area

using less mobile small-bodied fish, and contribute to furthering the scientific

understanding of the suitability of small-bodied fish species for environmental

monitoring programs. I compared the responses of blacknose dace and slimy

sculpin (small-bodied fish species) with yellow perch and white sucker (large-

bodied fish species) to determine whether the responses of small-bodied fish and

large-bodied fish would be different.

The general response patterns of slimy sculpin were consistent with my

previous study and provided further evidence of an overall increase in energy

storage and utilization via a nutrient enrichment effect. Male and female sculpin

exposed to pulp mill effluent were longer, heavier, and were in better condition

and showed increased length-at-age relative to reference fish. Similar to my

previous study, male and female sculpin exposed to pulp mill effluent exhibited

no significant site differences in gonad size relative to reference fish. Slimy

sculpin invest most of their energy into gonad development during the winter

when the river is ice-covered, water temperatures and flow are low, and pulp mill

effluent concentrations are elevated. The absence of site differences in March

indicated that pulp mill and sewage effluent exposure did not adversely affect

sculpin gonad development. Similar to sculpin, blacknose dace showed larger

livers (both sexes) downstream of the pulp mill, relative to the reference site, but

159

these differences were also apparent upstream at the sewage sites and could not

be attributed to the pulp mill effluent. Overall, the magnitude of the responses of

male and female blacknose dace collected downstream of the sewage and pulp

mill effluent varied but some were large (e.g., 91 % increase in male LSI

downstream of the pulp mill) (Table 5.1).

Sculpin living downstream of the pulp mill had unique 13C signatures

relative to fish at the reference site located downstream of the St. Francois River

(i.e., “Capone”) and fish downstream of municipal sewage (i.e., D/S Mad)

indicating a carbon-enriched environment. The stable isotope results from this

study were similar to results from my previous study and provided conclusive

evidence that sculpin movements in the Saint John River near the city of

Edmundston were limited and the integrated response patterns were site-

specific. Another interesting finding of this study was that sculpin exposed to

municipal sewage wastewater had a distinct 15N signature relative to fish

collected at a reference site located downstream of the St. Francois River (i.e.,

“Capone”) and downstream of the pulp mill effluent diffuser, but not fish captured

upstream of the pulp mill. If sculpin captured downstream of the pulp mill were

moving to upstream sites then 15N signatures should have been similar, but this

was not the case. In fact, the stable 15N signatures of sculpin collected

immediately downstream of the sewage and at a site located immediately

upstream of the pulp mill effluent diffuser showed that these fish were feeding at

a higher trophic level relative to pulp mill effluent exposed fish and reference fish.

160

The results suggest that nutrient enrichment from the municipal sewage is so

great that sculpin are able to attain a body size that would allow them to feed at a

higher trophic level relative to other sculpin populations in the vicinity.

Furthermore, the stable isotope results support other studies [Wassenar and

Culp 1996; Wayland and Hobson 2001] that have shown sites receiving pulp

mill effluent and municipal sewage wastewater can have distinct stable isotope

signatures relative to upstream sites and can be used to trace the geographical

extent of specific wastewater effluent streams.

White sucker and yellow perch showed few site differences in any of the

measured whole-organism parameters and the responses were not consistent

with slimy sculpin or previous work done at these sites, and suggested white

sucker and yellow perch were not as sensitive as sculpin for picking up changes

within a limited reach of the river. The decreases in white sucker liver size

downstream of the pulp mill outfall were puzzling; male and female LSIs were

reduced at the pulp mill site by 24% and 27%, respectively. I could not exclude

the possibility that the larger-bodied species did not mix between the south and

north shores of the Saint John River (about 50 m at the section near

Edmundston) or other areas immediately upstream or further downstream of the

mill diffuser. The white muscle stable 13C and 15N isotope ratios reported show

that white sucker collected downstream of the pulp mill effluent diffuser were not

mixing with fish collected at the upstream reference site located at St. Hilaire.

Doherty [2003], using radio telemetry, showed white sucker movements in the

161

middle section of the Saint John River were limited outside the spawning season

and tagged fish had a winter home range of < 2.5 km. If white sucker in the

upper portion of the river near Edmundston also moved ≤ 2.5 km then it is likely

that the fish were moving between the north and south sides of the river with the

possibility of some limited upstream and downstream movements. Thus, the

measured sucker responses were indicative of the environmental conditions

within in a much larger section of the river (i.e., potentially a 2.5 km radius from

the capture site) and not those immediately downstream of the pulp mill effluent

diffuser and within about 50 m of the shore, which the sculpin reflected.

Yellow perch males showed increased liver size and condition factor, but

no other significant differences were present in males or females. Except for

female GSI, male and female yellow perch whole-organism responses indicated

a nutrient rich-environment downstream of the pulp mill (Table 5.1); these fish

were collected in a deep pool in a backwater eddy downstream of the diffuser.

Stable 13C and 15N isotope signatures in blacknose dace and yellow perch did

not show as much isotopic separation as white sucker and slimy sculpin

suggesting their prey items may not be incorporating pulp mill effluent derived-

carbon into their diet. Future work will be needed to better understand the

influence of pulp mill effluent and municipal sewage wastewater on the local food

web.

162

Taken together, the results from chapters 2 and 3 suggest white sucker

and yellow perch may not be the most suitable species for monitoring multiple

wastewater effluents discharged in close proximity, such as the situation near the

city of Edmundston. However, as a result of their life history characteristics (e.g.,

trophic position, mobility) white sucker and perch are probably the fish species

that best integrate all of the stressors (both natural and anthropogenic) in this

section of the Saint John River. Ultimately, the best choice of fish species to be

use in a monitoring program will depend on the question(s) that need to be

addressed and the guidance that can be provided. If, for example, a researcher

wants to examine the impacts of multiple effluents discharged in close proximity

within a limited river reach on fish performance where mobility is an issue, then

small-bodied fish (e.g., cottids) are the best candidate species. These species

exhibit high site-fidelity and are able to show changes in whole-organism

responses that reflect localized conditions, within a limited reach of the river.

However, if the geographic scale of a study is expanded and the objective is to

assess the integrated impacts of stressors on fish performance within a

watershed then the use of large-bodied fish species (e.g., Catostomids) would be

more appropriate.

The most challenging aspect of this study was incorporating an additional

small-bodied fish species into the field program. Many small-bodied fish species

are particularly well suited for monitoring complex aquatic receiving environments

since they are presumed to be less mobile, easily captured, and more abundant

163

than large-bodied fish species. However, the main disadvantage of using small-

bodied fish species for monitoring programs (e.g., pulp and paper and metal

mining EEM programs) is the lack of basic life-history information. In general,

there is a paucity of detailed reproductive information for many small-bodied fish

that can be used in the pulp and paper and metal mining EEM programs. This is

troublesome; many metal mines in Canada are located in remote locations and

discharge into small headwater streams where many of these fish species live

and will be the only sentinel species available. It will be difficult to design studies

and interpret data, and field costs will be higher using small-bodied fish when

sufficient basic biological information is absent. For this study, I consulted Scott

and Crossman (1998) and other scientific literature via searchable databases

(e.g., Aquatic Sciences and Fisheries Abstracts, Science Citation Index) to try

and determine the most suitable time to collected pre-spawning blacknose dace

in the upper Saint John River near Edmundston. However, none of these

information resources provided any detailed reproductive information for the

region and generically suggested that spawning occurred from May to June when

water temperatures reached about 21°C. Pre-spawning female blacknose dace

were collected from river sites near Edmundston during the first week of June

2002 and showed increased variability in gonad development relative to that

seen in sculpin. Smaller female blacknose dace showed increased variability in

gonad development relative to the larger females. The variability was not due to

differences in sampling time since all fish were collected during the same period.

It was difficult to determine the exact reason why female dace showed this

164

variability in gonad development during the pre-spawning period since there was

no relevant biological information available.

I knew that dace were capable of exhibiting site differences in some of the

measured whole-organism parameters, but I wanted to know why female dace

exhibited increased variability in gonad development. As a result of this

research, a number of interesting follow-up questions were generated, including:

Do blacknose dace spawn once or multiple times during the reproductive

season? Do other small-bodied, multiple spawning fish exhibit the same

variability in gonad development during the pre-spawning period? Is there a

better time of the year to sample these fish in order to reduce variability? These

questions were addressed in chapter 4.

The objectives of chapter 4 were to collect basic biological information for

multiple spawning, small-bodied fish species and provide sampling guidance for

their use in future environmental monitoring programs. When using multiple

spawning fish in monitoring programs, Environment Canada recommends that

sampling should be conducted in the prespawning period prior to the onset of the

first spawn of the year so that potential complications associated with differences

in synchronicity of spawning between groups of fish can be avoided

[Environment Canada, 1997]. At this time, it was assumed that development

would be synchronized between individuals, and if changes were not evident

prior to the first spawn, the consequences of small changes later in the season

165

would be less important. To address the issues associated with using multiple-

spawning, small-bodied fish in monitoring programs I examined seasonal

changes in condition factor, liversomatic index (LSI), gonadosomatic index (GSI),

and ovarian histology in blacknose dace, northern redbelly dace, golden shiner,

and estuarine mummichog. I was also interested in assessing seasonal changes

in the ovary weight to body weight relationship (i.e., coefficient of determination

[r2] values for linear regressions equations) of female fish so that I could have a

better understanding of seasonal variability in gonad development and whether

different size classes of female fish showed similar seasonal variability. More

specifically, I attempted to identify the time of the year when the relationship

between ovary weight and body weight was most variable (i.e., low r2 values);

increase our understanding of natural source(s) of variability in gonad

development; identify suitable techniques to minimize data variability; and reduce

sample size requirements for detecting a critical effect size when designing cost-

effective studies in the future.

Data from this study provided a number of interesting findings. Overall, I

found that all the fish species I collected have extended spawning seasons

(spring-summer) and the suite of indices measured provided sufficiently detailed

information about the seasonal change in energy use and storage, particularly as

it relates to gonad growth. Oocyte development in the four fish species was

asynchronous and there were considerable differences in the patterns of ovarian

maturation between northern redbelly dace and the other fish species collected.

166

Oocyte development also differed between individuals of a species. For

blacknose dace, the first spawning occurred in early June when mean water

temperatures reached about 15.0°C (min.-max: 10.3 to 21.6°C). Interestingly,

female blacknose dace appeared not to rely solely on liver energy stores when

the maximum increases in gonad size occurred (i.e., 45% increase from May 4 to

May 20) and may rely on energy derived from active feeding. Future studies will

be needed to clarify this issue.

It was evident that the strength of the gonad size – body weight

relationship varied during the season for each species and that some fish

species showed much more seasonal variability in this parameter relative to

others and suggested that the timing of collections is very important when using

these fish species in monitoring programs. The most striking example of

seasonal variability in ovary development was that of northern redbelly dace.

Gonad development in this species was highly variable throughout the

reproductive season relative to the other species. Histology was an important

tool that helped me better understand the source of this variability. Ovaries of

northern redbelly dace contained oocytes in all developmental stages; as such,

there appeared to be no synchronicity in the timing of spawning events among

individuals, which would account for the high degree of variability observed

throughout the season. During the peak of the northern redbelly spawning

season (i.e., May – June), female GSI values ranged from < 1% to > 20%. It was

evident that seasonal variability in ovary development was higher in some

167

species (i.e., northern redbelly dace) relative to other species (i.e., blacknose

dace) and that timing of sample collections would be very important to take into

consideration.

Due to poor weather conditions and/or poor river conditions, it is

sometimes difficult to collect fish during the most optimal time of the year to

assess potential reproductive impacts associated with a particular anthropogenic

stressor. One of the key objectives of this study was to have a better

understanding of some of the basic biology of the collected fish species, attempt

to facilitate their use in environmental monitoring programs, and reduce sample

size requirements for future studies. In order to do this, I identified sources of

data variability and selected basic techniques that I hoped could be employed to

reduce some of the data variability and determine the sample sizes required in

order to detect a specified critical effect size (i.e., 25%) in gonad size between a

reference and exposure site at a specified power level. I chose a critical effect

size of 25% because the Canadian Federal government regulations require pulp

and paper mills and metal mining operations to design and implement monitoring

programs capable of detecting a 25% difference in gonad size between an

exposed site and reference site [Environment Canada 1997, 2002]. By selecting

a specific age class or size class of fish I was able to reduce some of the data

variability associated with seasonal gonadal development, which helped to

substantially reduce sample size requirements for future studies with some fish

species (i.e., blacknose dace) during certain times of the year, but it was

168

apparent that these techniques were not as suitable for other species (i.e.,

northern redbelly dace). The sample size required to detect a 25% difference in

gonad size in female blacknose dace was much smaller during the pre-spawning

period in May when compared to late June when the r2 was < 0.45. More

specifically, on May 20 the coefficient of variation (CV) was about 24% and a

sample size of 16 to 24 fish was required to detect a 25% difference in gonad

size when α and β were 0.05 and 0.20, and when both α and β equalled 0.10

(Figure 5.1). On June 11, however, the CV for all female blacknose dace data

increased to about 41%, as such, future studies during this time period would

require sample sizes of 47 to 71 (depending on the power level) (Figure 5.1). I

was able to reduce the variability (and sample sizes) by selecting only fish of a

similar age or body size, but this only worked for certain times of the year and not

all species. When 2 year old female blacknose dace were selected only, the CV

was reduced to about 15% on May 20 and less than 10 fish per site were

required to detect a 25% difference in gonad size (Figure 5.2). In contrast,

selecting 2 year female blacknose dace on June 11 actually increased the

variability (CV = 56%) and greater sample sizes were required (Figure 5.2).

Selecting only 2-4 g female blacknose dace on May 20 reduced some of the

variability (CV = 20%), but not as much as selecting only 2 year old fish;

nonetheless, sample sizes were reduced (Figure 5.3). Again, this was not the

case on June 11 when selecting 2-4 g female blacknose dace increased

variability (CV = 52%) and sample sizes requirements (Figure 5.3).

169

Since female northern redbelly dace showed the most seasonal variability

in gonad development I wanted to determine whether these techniques could

substantially reduce the sample size requirements for this species. On May 27,

the CV for all female northern redbelly dace was 57% and sample sizes between

81-134 fish per site were required to detect a 25% difference in gonad size

(Figure 5.4). On June 22, variability in gonad development was higher and even

larger sample sizes were required (Figure 5.4). Unfortunately, using only 2 year

old or 2 – 4 g female northern redbelly dace on May 27 or June 22 did not

substantially reduce data variability. The CV for 2 year old female northern

redbelly dace captured on May 27 was 56% and sample size requirements were

83-126 fish per site (Figure 5.5). The CV for 2-4 g fish was reduced to 44%, but

52-79 individuals per site were still required (Figure 5.6). Regardless of the

technique, large sample sizes were still required on June 22 for 2 year old

(Figure 5.5) and 2 – 4 g (Figure 5.6) female northern redbelly dace. Taken

together, results for female northern redbelly dace suggest that this species is

not a suitable sentinel fish species for environmental monitoring programs, at

least in southern New Brunswick. Correlation coefficients of under 0.4 mean that

sample sizes are very high. Increased seasonal variability in gonad development

would make it too difficult to detect potential reproductive impacts associated with

contaminant exposure. In addition, it has been well documented that northern

redbelly dace readily hybridize with other species, which would also confound

data interpretation.

170

Addressing the female blacknose dace variability issue noted in chapter 3

was important. The variability in gonad development in small female dace noted

in Chapter 3 is likely due to the timing of sampling (variability in female gonad

development increased as the spawning time approached and we sampled

blacknose dace between June 11 to 14; the first week of the spawning season)

and the presence of sexually immature fish and/or first time spawners, which can

hinder data analysis. Young slimy sculpin [Brasfield 2004] and trout-perch

[Gibbons et al. 1998] spawning for the first time also show increased variability in

gonad development leading these authors to conclude that their inclusion would

obscure interpretation of results.

Most importantly, multi-spawning, small-bodied fish can be successfully

used in environmental monitoring programs, but additional basic biological

research may be required in order to continue to facilitate their use. The

Canadian EEM program uses the variability in gonadal size to determine sample

size requirements for sampling. It is important to have high correlations between

gonad weight and body weight to reduce the potential impacts of sampling on the

population. Variability is increased in multiple spawning fish species because:

a) Younger fish of some species only spawn once, versus multiple

times in older fish, inflating variability. The variation can be

reduced in some species by selecting restricted size ranges or

ages of fish to increase power.

b) The first spawn of the season in most species is synchronized,

171

probably by spawning temperature. However, the timing of the

second spawn is not, and variability increases as spawning time

approaches because of the size of the asynchronicity of the

second clutch of eggs. Variability can be reduced by sampling

very early in gonadal development.

c) Some multiple spawning species may have too much variability

(r2< 0.4) throughout the spawning season. This may be due to

asynchronicity of the first spawn, or the potential for some

individuals to not spawn every year. Species with low

correlations should be avoided for reproductive investigations

until we develop a better understanding of the biological basis for

the variability.

5.1 Conclusions

The differences in life-history characteristics of fish species are important

when choosing a sentinel species since these characteristics will influence the

sensitivity and/or tolerance of the species to contaminant exposure. For the

upper Saint John River, large-bodied species (e.g., white sucker) are best suited

for assessing the overall health at a watershed or reach scale since their

responses reflect a relatively wider geographical scale. Small-bodied fish

species (e.g., sculpin) are best suited to assess the relative contribution of

individual effluents in sections of rivers receiving multiple effluents discharged in

172

close proximity. The biggest challenge of using small-bodied fish in

environmental monitoring programs is the paucity of basic biological information

for both single and multiple spawning species. It is also evident from this thesis

that basic biological information can significantly reduce the numbers of fish

required for the EEM program while still collecting all the relevant biological

information.

Future research will need to focus on assessing the impact of sewage and

pulp mill effluent on local food webs. Some of the responses of sculpin exposed

to municipal sewage near Edmundston were larger than any of the responses

associated with pulp mill effluents across Canada and will need to be

investigated in more detail. Detailed studies on the life history of large bodied

fish in the upper Saint John River near Edmundston should be initiated and the

number of collection sites should be expanded in order to better understand and

continue to identify natural and anthropogenic stressors which were causing the

measured responses (or lack thereof in the case of white suckers) observed in

this thesis. Continuing to collect detailed life history information for small-bodied

fish will be particularly important when designing fish studies for the Federal

metal mining EEM program. In Canada, most metal mines are located in remote

areas and discharge into small, headwater streams that have limited fish

populations and fish diversity. Large-bodied fish are not available in many of

these locations and lethal sampling of large numbers of fish may posed a serious

threat to the very local fish community that the EEM is designed to protect.

173

Continued research on developing non lethal techniques and community surveys

and data interpretation guidance for the use of small-bodied fish species would

be particularly beneficial for metal mining studies. Finally, research partnerships

between government, academia, and industry need to continue and expand so

that we can continue to better understand the impacts of complex industrial and

municipal wastewater effluents and agricultural activities on the health of aquatic

biota. These partnerships will not only allow us to continue to identify previously

unknown areas (and contaminants) of concern, but will also facilitate the

development and implementation of socially acceptable and economically viable

mitigation strategies.

5.2 References

Brasfield SM. 2004. Examining population-level responses in small-bodied and

short-lived fishes: what Cottus has taught us. Poster presentation at 31st

Annual Aquatic Toxicity Workshop, October 24-27, Charlottetown, PEI.

Culp,J.M., Cash, K.J., Glozier, N.E., and Brua, R.B. 2003. Effects of pulp mill

effluent on benthic assemblages in along the Saint John River, Canada.

Environ Toxicol Chem 22: 2916-2925.

Doherty, C.A., R.A. Curry and K.R. Munkittrick. 2003. Tracking adult white

sucker movements near point source discharges in the Saint John River,

New Brunswick, Canada. In Borton, D.L., Hall, T.J., Fisher, R.P., &

Thomas, J.F., (eds), Pulp & Paper Mill Effluent Environmental Fate &

Effects, June 1-4, Seattle, WA, pp. 123 - 132.

Environment Canada. 1997. Fish Survey Expert Working Group:

174

Recommendations from Cycle 1 review. Ottawa ON: Environment Canada.

EEM/1997/6. 262p.

Environment Canada. 2002. Metal mining guidance document for aquatic

environmental effects monitoring. EEM/2002. June 2002.

Environment Canada. 2003. National assessment of pulp and paper

environmental effects monitoring data: a report synopsis. National Water

Research Institute, Burlington, ON. NWRI Scientific Assessment Report

Series No. 2.

Gibbons WN and Munkittrick KR. 1994. A sentinel monitoring framework for

identifying fish population responses to industrial discharges. J. Aquat.

Ecosyst. Health 3: 227-237.

Gibbons WN, Munkittrick KR, McMaster ME, Taylor WD. 1998. Monitoring

aquatic environments receiving industrial effluents using small fish species

2: Comparison between responses of trout-perch (Percopsis

omiscomaycus) and white sucker (Catostomus commersoni) downstream

of a pulp mill. Environ Toxicol Chem 17: 2238-2245.

Gray MA. 2003. Assessing non-point source pollution in agricultural regions of

the upper St. John River basin using the slimy sculpin (Cottus cognatus).

PhD thesis, University of New Brunswick. Fredericton, NB, Canada.

Lowell R, Ring B, Pastershank G, Walker S, and Trudel L. 2004. National

assessment of data from the pulp and paper EEM program: cycle 3 and

comparisons to earlier cycles. Pulp and paper EEM detailed information

session, November 25-26, Edmonton, AB, Canada.

175

Munkittrick KR, Van Der Kraak GJ, McMaster ME, Portt CB, van den Heuvel MR,

Servos MR. 1994. Survey of receiving water environmental impacts

associated with discharges from pulp mills. 2. Gonad size, liver size,

hepatic EROD activity, and plasma sex steroid levels in white sucker.

Environ Toxicol Chem 13:1089–1101.

Munkittrick, K.R., McMaster, M.E., Van Der Kraak, G., Portt, C., Gibbons, W.N.,

Farwell, A., and Gray, M. 2000. Development of methods for effects-

driven cumulative effects assessment using fish populations: Moose River

Project. SETAC Press, Pensacola, FL, USA.

Munkittrick KR, McGeachy SA, McMaster ME, Courtenay SC. 2002. Overview

of cycle 2 freshwater fish studies from the pulp and paper Environmental

Effects Monitoring program. Water Quality Res J Can 37: 49-77.

Parrott, J.L., Wood, C.S., Boutot, P., Blunt, B.R., Baker, M.A., and Dunn, S.

2003. Fathead minnow long-term growth/reproductive tests to assess

final effluent from a bleached sulphite mill. Environ Toxicol Chem 22:

2908-2915.

Scott WB, Crossman EJ. 1998. Freshwater Fishes of Canada. Galt House,

Oakville, ON, Canada.

Wassenar LI, Culp JM. 1996. The use of stable isotope analyses to identify pulp

mill effluent signatures in riverine food webs. In Servos MR, Munkittrick

KR, Carey JH, Van Der Kraak GJ, (eds), Environmental Fate and Effects

of Pulp and Paper Mill Effluents. St. Lucie Press, Delray Beach, FL, USA,

pp 413-424.

176

Wayland M, Hobson KA. 2001. Stable carbon, nitrogen, and sulphur isotope

ratios in riparian food webs on rivers receiving sewage and pulp-mill

effluents. Can. J. Zool. 79: 5-15.

177

Table 5.1. Percent differences in condition factor (k), liversomatic index (LSI),

and gonadosomatic index (GSI) for slimy sculpin, white sucker, yellow

perch, and blacknose dace collected at various times from 1999-2003

(“% Difference” is for municipal sewage and pulp mill effluent exposed

fish relative to reference fish at St. Hilaire). (-) data not available; NS –

not significantly different.

Sex Species Date Parameter Municipal Sewage

Pulp mill Effluent

M Slimy Sculpin

October 1999

K 16 12

LSI 31 83 GSI -8 (NS) -2 (NS) December

1999 K - 14

LSI - 20 (NS) GSI - 7 (NS) August

2001 K - 6

LSI - 14 (NS) GSI - - March

2002* K 29 15

LSI 66 46 GSI 7 (NS) 3 (NS) F Slimy

Sculpin October 1999

K 17 16

LSI 23 (NS) 117 (NS) GSI 9 (NS) 23 (NS) December

1999 K - 11

LSI - 14 GSI - 19 (NS) September

2000 K - 13

LSI - -31 (NS) GSI - -1.3 (NS) August

2001 K - 5 (NS)

LSI - -4 (NS) GSI - -

178

March 2002*

K 32 14

LSI 9 (NS) 1 (NS) GSI 5 (NS) 7 (NS) M White

sucker October 1999

K - -3 (NS)

LSI - 2 (NS) GSI - 20 (NS) October

2002 K - -3 (NS)

LSI - -24 GSI - -5 (NS) F White

sucker October 1999

K -4

LSI -5 (NS) GSI - -8 (NS) October

2002 K - -0.8 (NS)

LSI - -21 GSI - -13 (NS) M Yellow

Perch September 2002

K 7

LSI 17 GSI 29 (NS) F Yellow

Perch September 2002

K 5 (NS)

LSI 5 (NS) GSI -6 (NS) M Blacknose

Dace June 2002 K 2 (NS) 2 (NS)

LSI 40 91 GSI -26 23 F Blacknose

Dace June 2002 K -2 (NS) 0 (NS)

LSI 19 (NS) 25 GSI -24 (NS) -20 (NS)

179

0

50

100

150

200

250

300

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85

CV (%)

Num

ber o

f Fis

h1

23

n=24

n=15n=16

n=71

n=43n=47

Figure 5.1. Sample sizes required to detect a 25% difference in gonad size at

different levels of power for all adult female blacknose dace collected on

May 20 (dashed lines) and June 11 (solid lines), 2003. The solid

curved line numbered “1” represents α = 0.05 and β = 0.05; the solid

curved line numbered “2” represents α = 0.10 and β = 0.10; and the

solid curved line numbered “3 represents α = 0.05 and β = 0.20.

180

0

50

100

150

200

250

300

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85

CV%

Num

ber o

f Fis

h

1

23

n=9n=6

n=134

n=89

n=81

Figure 5.2. Sample sizes required to detect a 25% difference in gonad size at

different levels of power for 2 year old female blacknose dace collected

on May 20 (dashed lines) and June 11 (solid lines), 2003. The solid

curved line numbered “1” represents α = 0.05 and β = 0.05; the solid

curved line numbered “2” represents α = 0.10 and β = 0.10; and the

solid curved line numbered “3 represents α = 0.05 and β = 0.20.

181

0

50

100

150

200

250

300

0 10 20 30 40 50 60 70 80 90

CV (%)

Num

ber o

f Fis

h

n=17

n=11

n=104

n=63

n=69

1

2

3

Figure 5.3. Sample sizes required to detect a 25% difference in gonad size at

different levels of power for 2 – 4 g female blacknose dace collected on

May 20 (dashed lines) and June 11 (solid lines), 2003. The solid

curved line numbered “1” represents α = 0.05 and β = 0.05; the solid

curved line numbered “2” represents α = 0.10 and β = 0.10; and the

solid curved line numbered “3 represents α = 0.05 and β = 0.20.

182

0

50

100

150

200

250

300

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85

CV (%)

Num

ber o

f Fis

h

1

23

n=134

n=81

n=89

n=237

n=156

n=143

Figure 5.4. Sample sizes required to detect a 25% difference in gonad size at

different levels of power for all female northern redbelly dace collected

on May 27 (dashed lines) and June 22 (solid lines), 2003. The solid

curved line numbered “1” represents α = 0.05 and β = 0.05; the solid

curved line numbered “2” represents α = 0.10 and β = 0.10; and the

solid curved line numbered “3 represents α = 0.05 and β = 0.20.

183

0

50

100

150

200

250

300

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85

CV (%)

Num

ber o

f Fis

h

n=126

n=83

n=76

n=157

n=104n=95

1

2

3

Figure 5.5. Sample sizes required to detect a 25% difference in gonad size at

different levels of power for 2 year old female northern redbelly dace

collected on May 27 (dashed lines) and June 22 (solid lines), 2003.

The solid curved line numbered “1” represents α = 0.05 and β = 0.05;

the solid curved line numbered “2” represents α = 0.10 and β = 0.10;

and the solid curved line numbered “3 represents α = 0.05 and β = 0.20.

184

0

50

100

150

200

250

300

0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80

CV (%)

Num

ber o

f Fis

h

n=79

n=52

n=48

n=81n=88

n=134

1

23

Figure 5.6. Sample sizes required to detect a 25% difference in gonad size at

different levels of power for 2 – 4 g female northern redbelly dace

collected on May 27 (dashed lines) and June 22 (solid lines), 2003.

The solid curved line numbered “1” represents α = 0.05 and β = 0.05;

the solid curved line numbered “2” represents α = 0.10 and β = 0.10;

and the solid curved line numbered “3 represents α = 0.05 and β = 0.20.