This Thesis Is Accepted by the Dean of Graduate Studies

THE SITE FIDELITY OF MUMMICHOGS (FUNDULUS HETEROCLITUS) IN AN ATLANTIC CANADIAN ESTUARY: IMPLICATIONS FOR USE AS A SENTINEL SPECIES IN ENVIRONMENTAL MONITORING by Marc Anthony Skinner

Bachelor of Science, University College of Cape Breton 2001

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

Master of Science

In the Graduate Academic Unit of Biology

Supervisors: R. A. Curry, Ph.D., Canadian Rivers Institute, UNB Fredericton S.C. Courtenay, Ph.D., Canadian Rivers Institute Fisheries, UNB Fredericton and Fisheries and Oceans Canada, Moncton

Examining Board: A.W. Diamond, Ph.D., Department of Biology, UNB Fredericton, Chair D. MacLatchy, Ph.D., Canadian Rivers Institute, UNB Saint John B. Hudgins, Ph.D., Institute of Biomedical Engineering, UNB Fredericton

This thesis is accepted by the Dean of Graduate Studies

THE UNIVERSITY OF NEW BRUNSWICK

April, 2005

Small-bodied fish were proposed for use as a sentinel species in environmental monitoring because they are usually more abundant and generally less mobile than large bodied fish, therefore increasing catch rate while also increasing their ability to reflect local conditions. The primary objective of this thesis was to describe the spatial and temporal movement patterns of mummichogs (Fundulus heteroclitus) in a large river estuary and assess their usefulness as a sentinel species in environmental programs such as Environmental Effects Monitoring (EEM) for the Canadian pulp and paper industry. Experiments identified Visible Implant Elastomer (VIE) as the most suitable marking technique for mummichogs due to its high mark retention and lack of negative effects on growth and survival. Next, a two year mark- recapture study conducted along a ~10 km span of the Miramichi River estuary concluded mummichogs in this region display distinct site-fidelity with the vast majority (96.6%) of specimens marked up to 16 months previously consistently found within 200 m of point of initial release from April-November. Subsequent stable isotope analyses (SIA) of samples from sites located upstream and downstream of multiple anthropogenic inputs into the same span of estuary confirmed the results of low mummichog mobility from the mark-recapture study.

Simultaneously, these results demonstrated SIA was a useful method to determine the site-specificity of organisms along relatively small spatial scales

(< 10 km) in areas receiving multiple anthropogenic inputs. With regards to mobility, these results add to the growing body of literature supporting the

ii usefulness of mummichogs as a sentinel species in environmental monitoring programs for point source impacts in Canadian Atlantic estuaries.

iii ACKNOWLEDGMENTS

I would like to begin by thanking my supervisors, mentors and friends

Drs. Allen Curry and Simon Courtenay for their patience, counsel, support and most of all their patience for the past two and a half years. Thank you both for always letting me try things my way first and then being there afterwards to help me sort out the mess.

I would like to thank my committee members, Dr. Kelly Munkittrick and

Mr. Roy Parker for their constructive criticism, encouragement and for asking the tough questions when they were needed most. I would also like to thank Mr.

Phil Riebel and UPM-Kymmene for access to mill resources and information and for sponsoring my NSERC scholarship. Thanks also to the Atlantic and

National EEM offices of Environment Canada for funding my research.

I would have been impossible to do this alone, so I’d like to thank the many friends, fellow grad students, technicians, summer students and faculty who helped me along the way with field help, input, and advice. Thanks to Kirk

Roach for the long hours and pep talks; Remi Gionet, Richard Francis, Jon

Freedman, and Venitia Joseph for field help; Mary Murdoch and Dr. Malcolm

Stephenson of Jacques Whitford Environmental Ltd. for their input and for exposing me to the consulting world; Harry Collins and MREAC for logistical help and teaching me just about all I know about the Miramichi; Drs. Myriam

Barbeau and Eric Marchand for study design and statistical input; Tim Jardine for his sage-like isotope advice and putting a roof over my head; Chad Doherty for great discussions and advice; and Rachel Keeler for being a true friend and

iv helping me keep my sanity. Special thanks also to David Robertson. Without his help, advice and constant friendship this project would never have gotten off the ground.

I would like to finish by thanking my family for their love and support.

Carol, thank you being the big sister I never had and for giving me the gifts that exposed me to science at a young age. Sarah, thank you for your patience sifting through sediment all hours of the night, listening all about mummichogs over and over again, your understanding of my odd habits, your genuine interest in my work and your caring determination to curb my excessive procrastination.

Finally, I would like to thank my parents for teaching me the value of a hard day’s work, always supporting my decisions in life, and constantly encouraging me to learn.

v TABLE OF CONTENTS

ABSTRACT...... ii

ACKNOWLEDGMENTS ...... iv

LIST OF FIGURES ...... xi

1 GENERAL INTRODUCTION...... 1

1.1 Environmental Effects Monitoring in Atlantic Canada...... 1

1.2 Mummichog Biology and Environmental Effects Monitoring .... 2

1.3 Objectives and Outline of Thesis...... 4

1.4 Literature Cited...... 5

2 EVALUATION OF TECHNIQUES FOR THE MARKING OF

MUMMICHOGS (FUNDULUS HETEROCLITUS) WITH EMPHASIS ON

VISIBLE IMPLANT ELASTOMER (VIE) ...... 10

2.1 Abstract...... 10

2.2 Introduction ...... 11

2.3 Methods ...... 13

2.3.1 Marking and Tagging Trial (2001-2002)...... 13

2.3.2 VIE Marking Experiment (2002-2003)...... 15

2.3.3 Statistical Analyses ...... 17

2.4 Results ...... 18

2.4.1 Marking and Tagging Trial (2001-2002)...... 18

2.4.2 VIE Marking Experiment (2002-2003)...... 19

2.5 Discussion...... 20

2.6 Acknowledgements ...... 25

vi 2.7 Literature Cited...... 26

3 SITE FIDELITY OF MUMMICHOGS (FUNDULUS

HETEROCLITUS) IN AN ATLANTIC CANADIAN ESTUARY ...... 36

3.1 Abstract...... 36

3.2 Introduction ...... 37

3.3 Materials and Methods...... 40

3.3.1 Study area ...... 40

3.3.2 Fish collection and marking ...... 41

3.4 Results ...... 44

3.4.1 Marking and recapturing by site...... 44

3.4.2 Movements of marked individuals...... 44

3.5 Discussion...... 45

3.6 Acknowledgements ...... 52

3.7 Literature Cited...... 53

4 USE OF STABLE ISOTOPES TO EXAMINE THE SITE

FIDELITY OF MUMMICHOGS (FUNDULUS HETEROCLITUS) IN AN

ATLANTIC CANADIAN ESTUARY RECEIVING MULTIPLE

ANTHROPOGENIC INFLUENCES ...... 73

4.1 Abstract...... 73

4.2 Introduction ...... 74

4.3 Materials and Methods...... 77

4.3.1 Sampling sites...... 77

4.3.2 Sample collections, processing and isotopic analyses...... 78

vii 4.3.3 Data Analysis...... 80

4.4 Results ...... 82

4.5 Discussion...... 83

4.6 Acknowledgements ...... 91

4.7 Literature Cited...... 92

5 GENERAL DISCUSSION...... 111

5.1 Recommendations and Suggested Research Needs...... 116

5.2 Literature Cited...... 117

6 APPENDICES...... 120

6.1 Appendix 1 ...... 120

CURRICULUM VITAE

viii LIST OF TABLES

Table 1 – Mean lengths, mean weights, mean daily increases in length and weight, mean condition factor, sample sizes, percent survival, and percent mark retention for 2001-2002 marking and tagging trial (L = length; W = weight; SD = standard deviation; 1 = start of the experiment; 2 = end of 144 d experiment; * = statistically significant difference from controls, Tukey’s Honestly Significant Test, α = 0.05)...... 32 Table 2 - Mean lengths, mean weights, mean daily increases in length and weight, mean condition factor, sample sizes, percent survival, and percent mark retention for 2002-2003 marking and tagging trial (L = length; W = weight; SD = standard deviation; 1 = start of the experiment; 2 = end of 167 d experiment; * = statistically significant difference from tank controls, Tukey’s Honestly Significant Test, α = 0.05, ** = statistically significant difference from tank controls, Bonferroni Test, α = 0.05)...... 33 Table 4 – Numbers of mummichogs retained and survived 48 hrs after marking from May 23 – August 30, 2002. Percent survived by site and date and percent mark retention by site also included (a=deaths due to holding conditions, CH=Chatham Head, BKPM= Bleached Kraft Pulp Mill, SM=Strawberry Marsh, UFC=Upstream Flett Cove)...... 61 Table 5 - Number of mummichogs recaptured by site from May 27, 2002 – November 27, 2003 (CH=Chatham Head, BKPM= Bleached Kraft Pulp Mill, SM=Strawberry Marsh, UFC=Upstream Flett Cove, GW=Groundwood Mill, LB=Little Bog, MC=McKay Cove, GB=Green Bridge, U-Haul=U-Haul; see Figure 3). Numbers of fish marked by site are shown in Table 1. Values in bold indicate fish that were recaptured multiple times and have been included in totals once. Values with superscripts indicate numbers and recapture sites for fish recaptured >200m from release point at sites other than the four original marking sites. Column recapture totals refer to all fish marked at that site (including those recaptured at other sites)...... 62 Table 6 – Recapture summary of marked mummichogs found to have moved from May 27, 2002 – November 26, 2003 (CH=Chatham Head, BKPM=Bleached Kraft Pulp Mill, SM=Strawberry Marsh, UFC=Upstream Flett Cove, GW=Groundwood Mill, LB=Little Bog, MC=McKay Cove, GB=Green Bridge)...... 64

ix Table 7 – Total length (mean + SD) of mummichogs sampled from upper Miramichi River estuary for stable isotopic ratios of carbon and nitrogen. Isotopic ratios of anthropogenic effluents also provided. (CH = Chatham Head, BKPM = Bleached Kraft Pulp Mill, SM = Strawberry Marsh, UFC = Upstream Flett Cove, GW = Groundwood Mill, MC = McKay Cove, UH = U-Haul, LB = Long Beach, WSB = Whitesand Beach, BKPMe = BKPM effluent, BKPMrec = BKPM receiving environment, GWe = GW effluent, Sew1 = effluent from sewage facility 1, ND = not detectable)...... 103 Table 8 - Total number of mummichogs caught, number of attempts, and catch per unit effort (CPUE) from May 2002 – November 2003...... 120

x LIST OF FIGURES

Figure 1 - Map of Atlantic Canada showing pulp and paper mills which used mummichogs as a sentinel species during Cycle 2 fish surveys. (adapted from Courtenay et al., 2002 with permission). Inset shows geographic distribution of mummichogs...... 9 Figure 2a – Aerial view of experimental tank design for 2002-2003 VIE marking experiment showing plastic cages used to hold different specimen groups...... 34 Figure 2b – Experimental basket dimensions for 2002-2003 VIE marking experiment...... 35 Figure 3 – Map of the upper Miramichi River Estuary (MRE) indicating mummichog sampling sites for 2002 (CH=Chatham Head, BKPM= Bleached Kraft Pulp Mill, SM=Strawberry Marsh, UFC=Upstream Flett Cove). Triangles represent additional 2002 recapture sites. Diamonds and circles represent expanded 2003 recapture sites (sampled from April – June, 2003). Black shading indicates approximate 1% concentration area of BKPM effluent plume (Natech Environmental Services Inc., 2002)...... 66 Figure 4 – Movement directions of Fundulus heteroclitus marked at (a) Strawberry March (SM) (b) Chatham Head (CH), (c) Bleached Kraft Pulp Mill (BKPM), and (d) Upstream Flett’s Cove (UFC) with number of fish moved indicated. Additional recovery sites were: GW=Groundwood Mill, LB=Little Bog, MC=McKay Cove, GB=Green Bridge and U-Haul=U-Haul...... 68 Figure 5 – Map of Newcastle/Chatham area of the upper Miramichi River estuary with isotope sampling sites (diamonds) indicated. White circles and triangles with numbers (1-3) represent approximate locations of pulp mill and wastewater outfalls, respectively. Shaded regions approximate 1% concentration area of effluent plumes (no data available for Sew3) (Natech, 1998). Arrows indicate direction of waterflow. (CH = Chatham Head, BKPM = Bleached Kraft Pulp Mill, SM = Strawberry Marsh, UFC = Upstream Flett Cove, GW = Groundwood Mill, MC = McKay Cove, UH = U-Haul, LB = Long Beach, WSB = Whitesand Beach)...... 105 Figure 6 – Mean ( + 99% confidence intervals; sexes pooled, n shown in Table 7) δ13C (a) and δ15N (b) ratios for mummichog white muscle and bone from sites along the upper Miramichi River Estuary (MRE). Values for most effluents and receiving waters are also shown (GWe and BKPMe were not included in figure b and are found in Table 7). Site locations are shown on Figure 5. Statistical significance denoted by lowercase letters was determined using Tukey’s Honestly Significant Test (α = 0.01). (BKPMe = BKPM effluent, BKPMrec = BKPM receiving environment, GWe = GW effluent, Sew1 = effluent from sewage facility 1)...... 106

xi Figure 7a – Dendrogram showing percent similarity of δ13C and δ15N ratios among sites...... 108 Figure 7b – Cross plot of mean (+ 99% confidence intervals) δ13C vs. δ15N ratios for mummichog white muscle and bone from sites along the upper MRE. Values for effluents and receiving waters are not shown and are found in Table 7...... 109 Figure 8 – Scatter plot of δ13C vs. δ15N ratios for mummichog white muscle tissue from sites along the upper MRE. Arrows indicate outliers identified by Grubbs Test. (♦ = U-Haul, ■ = Bleached Kraft Pulp Mill, ▲= MacKay Cove, x = Groundwood Mill, ○ = Whitesand Beach, + = Long Beach, ● = Strawberry Marsh , □ = Upstream Flett Cove, ▬ = Chatham Head)...... 110

xii 1 GENERAL INTRODUCTION

1.1 Environmental Effects Monitoring in Atlantic Canada

Canada’s Environmental Effects Monitoring (EEM) program was incorporated in the 1992 modifications to the Pulp and Paper Effluent

Regulations of 1972. These changes were implemented to regulate total suspended solids, biological oxygen demand, and acute toxicity of pulp and paper mill effluents being discharged into aquatic environments while also requiring these mills to conduct a monitoring program. The goal of this EEM program was to determine the effectiveness of these new regulations for protecting aquatic resources by reducing or removing effects in receiving waters. It is designed to accomplish this goal in three year cycles by assessing effects of mill effluents on fish via a fish survey; fish habitat via a benthic macroinvertebrate survey; and the use of fisheries resources by conducting tainting evaluations and dioxin/furan assays (Courtenay et al., 2002).

The fish survey component requires mills to sample fish of both sexes from two species of fish living within the influence of their outfall and compare these to fish living at one or more appropriate reference locations (Courtenay et al., 2002). These fish are assessed to determine potential effects of mill effluents on their growth, survival and reproduction through measurements of various endpoints (Courtenay et al., 2002). For Canadian pulp and paper EEM fish surveys, fish found within effluent plumes having concentrations equal to or greater than one percent are considered exposed (Environment Canada, 1998).

- 1 - During Cycle 1 (1993-1996), mills discharging into marine and estuarine environments reported problems capturing enough specimens of appropriate sentinel species and quantifying their degree of exposure to effluents due to mobility and the complex physical and chemical characteristics of these environments. Subsequently, a Fish Survey Expert Working Group recommended the use of small-bodied sentinel species for use in Cycle 2 Fish

Surveys (1997-2000) (Munkittrick et al., 1997) as these fish are more likely to be abundant, ubiquitous, and have lower mobility than longer-lived species and therefore be more likely to accurately reflect local environmental conditions

(Gibbons et al., 1998).

1.2 Mummichog Biology and Environmental Effects Monitoring

The mummichog, Fundulus heteroclitus, is a euryhaline, littoral species of killifish (Family Cyprinodontidae) distributed along the eastern coast of North

America from Gulf of St. Lawrence region to Florida mainly in tide pools and estuaries (Abraham, 1985; Fig. 1). Adult mummichogs range in size from 35-

125 mm and may live for up to four years (Abraham, 1985). They spawn from

April to August in conjunction with spring tides of the new or full moon and are capable of spawning multiple times (Kneib, 1984). Mummichogs also sustain an endogenous circadian feeding pattern overlaid on tidal rhythms, thus feeding primarily on high tide during daylight (Weisberg and Lotrich, 1980). They are also generally the most abundant member of their genus in most areas.

Meredith and Lotrich (1979) reported abundances of 130,000-136,000 mummichogs longer than 30 mm along a 3 km long, Delaware tidal creek in July

- 2 -

and summer densities ranging from 0.35–6.04 fish/m2 for individuals longer than

40 mm have been noted in some estuarine habitats (Kelso, 1979). The name

“mummichog” reflects this, as it is a Native American word meaning “going in crowds” (Abraham, 1985).

Mummichogs are also believed to be sedentary with reports of low mobility in populations from the central part of the species range in the eastern United

States. Lotrich (1975) reported a 36 m home range for mummichogs in a

Delaware salt marsh tidal creek and Sweeney et al. (1998) found mummichogs in a New Jersey salt marsh tidal creek to move distances <650 m. As such, the mummichog was proposed for use for Cycle 2 fish surveys at mills discharging into marine and estuarine environments in Atlantic Canada.

In Cycle 2 fish surveys, mummichogs were captured in sufficient numbers and sampled for all endpoints at three mills in northern New Brunswick (Fig. 1)

(Courtenay et al., 2002). However, consultants noted specimens could only be captured immediately preceding and after high tides while sampling in the littoral zone at mills discharging into the Miramichi and Restigouche estuaries (Jacques

Whitford Environmental Limited, 2000a, b). These findings caused the consultants to hypothesize northern populations of mummichogs may not display site-fidelity similar to those noted along their central distribution because offshore movement into river channels with ebbing tides could cause advection hundreds to thousands of meters downstream. This would therefore preclude the use of mummichogs as sentinel species in large river systems (Jacques

Whitford Environmental Limited, 2000a, and b).

- 3 -

This situation led to a re-evaluation of the previous assumption of low mobility in mummichogs. On closer examination, it appeared the previous mummichog movement studies were of limited relevance to areas such as the

Miramichi River estuary (MRE) which has a very different tidal amplitude and size. Of the two studies cited above, Lotrich (1975) is the more comprehensive but the section of creek studied had a 0.3-0.6 m tidal amplitude which is not comparable to the 2 m tidal amplitude encountered in the MRE (Reinson, 1977).

Also, while the salt marsh examined by Sweeney et al. (1998) had a tidal amplitude more comparable to the MRE (>3 m), it and the study by Lotrich

(1975) were conducted in the tidal creeks of salt marshes. The MRE by contrast, is fed by two major rivers: the Northwest Miramichi and Southwest

Miramichi, with respective drainage areas of 3900 km2 and 7700 km2 (Chiasson,

1995); monthly discharges ranging from 86 m3.s-1 in August to 620 m3.s-1 in May

(Rashid and Reinson, 1979); and a 10 m deep channel that can occupy up to half its width (Lafleur et al., 1995). Therefore, the tidal creeks from the previous studies do not accurately reflect the physical conditions encountered in the MRE and as such, neither of these studies is adequate for inferring the potential distances moved by mummichogs in the MRE.

1.3 Objectives and Outline of Thesis

The main objective of this thesis is to describe the spatial and temporal movement patterns of mummichogs in a large river estuary and assess their usefulness as a sentinel species in environmental programs such as EEM as it was hypothesized that mummichogs moving offshore daily with falling tides

- 4 -

would be displaced great distances thus reducing their actual exposure to pulp mill effluent.

The thesis is organized into five chapters: a general introduction, three main body chapters, and a final discussion chapter. Chapter two details experiments conducted to determine the most appropriate method for marking mummichogs. Chapter three describes a mark-recapture study conducted to determine the movement patterns of mummichogs in the Miramichi River estuary. Chapter four details stable-isotope analyses (SIA) conducted to evaluate SIA as a tool to determine the site-fidelity of organisms along small spatial scales using mummichogs in the Miramichi River estuary as an example.

Chapter five is a general discussion that integrates chapters 2-4.

1.4 Literature Cited

Abraham BJ. 1985. Species profiles: life histories and environmental

requirements of coastal fishes and invertebrates - Mummichog and striped

killifish. US Fish and Wildlife Service Biological Report (US Army Corps of

Engineers) 82: 1-23.

Chiasson AG. 1995. The Miramichi bay and estuary: an overview, p. 11-27. In

Chadwick M (ed.), Water, Science, and the Public: The Miramichi

Ecosystem. Canadian Special Publication of Fisheries and Aquatic Sciences

123.

Courtenay SC, Munkittrick KR, Dupuis HMC, Parker WR, and J Boyd. 2002.

- 5 -

Quantifying impacts of pulp mill effluent on fish in Canadian marine and

estuarine environments: problems and progress. Water Quality Research

Journal of Canada 37: 79-99.

Environment Canada. 1998. Pulp and paper technical guidance for aquatic

environmental effects monitoring. National EEM Office, Environment

Canada. EEM/1998/1.

Gibbons WN, Munkittrick KR, McMaster ME, and Taylor WD. 1998. Monitoring

aquatic environments receiving industrial effluents using small fish species 1.

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

kraft pulp mill. Environmental Toxicology and Chemistry 17: 2227-2237.

Jacques Whitford Environmental Limited. 2000a. Project No. JWEL 11153.

Final Report to: AV Cell Inc. on: Second Cycle Aquatic Environmental

Effects Monitoring Dissolving Grade Pulp Mill PP1112. JWEL, 711

Woodstock Road, Fredericton, NB.

Jacques Whitford Environmental Limited. 2000b Project No. JWEL 89584. Final

Report to: REPAP New Brunswick Inc. on: Second Cycle Aquatic

Environmental Effects Monitoring Study Kraft Mill (PP1112). JWEL, 711

Woodstock Road, Fredericton, NB.

- 6 -

Kelso WE. 1979. Predation on soft-shell clams, Mya arenaria, by the common

mummichog, Fundulus heteroclitus. Estuaries 2: 249-254.

Kneib RT. 1984. Patterns in the utilization of the intertidal salt marsh by larvae

and juveniles of Fundulus heteroclitus (Linnaeus) and Fundulus luciae

(Baird). Journal of Experimental Marine Biology and Ecology 83: 41-51.

Lafleur C, Pettigrew B, St-Hilaire A, Booth D, Chadwick M. 1995. Seasonal

and short term variations in the estuarine structure of the Miramichi, p. 45-73.

In: Chadwick M (ed.), Water, Science, and the Public: The Miramichi

Ecosystem. Canadian Special Publication of Fisheries and Aquatic Sciences

123.

Lotrich VA. 1975. Summer home range and movements of Fundulus

heteroclitus (Pisces: Cyprinodontidae) in a tidal creek. Ecology 56: 191-198.

Meredith WH and Lotrich VA. 1979. Production dynamics of a tidal creek

population of Fundulus heteroclitus (Linnaeus). Estuarine and Coastal

Marine Science 8: 99-118.

Munkittrick KR, Megraw SR, Colodey A, Luce S, Courtenay S, Paine M, Servos

M, Spafford M, Langlois C, Martel P, and C Levings. 1997. Fish survey

expert working group final report. Recommendations from cycle 1 review.

- 7 -

Environment Canada, EEM/1997/6.

Rashid MA, Reinson GE. 1979. Organic matter in surficial sediments of the

Miramichi Estuary, New Brunswick, Canada. Estuarine and Coastal Marine

Science 8: 23-36.

Reinson GE. 1977. Tidal-current control of submarine morphology at the mouth

of the Miramichi estuary, New Brunswick. Canadian Journal of Earth

Sciences 14: 2524-2532.

Sweeney J, Deegan L, and R Garritt. 1998. Population size and site fidelity of

Fundulus heteroclitus in a macrotidal saltmarsh creek. Biological Bulletin

195: 238-239.

Weisberg SB and Lotrich VA. 1980. Food limitation of the mummichog Fundulus

heteroclitus in a Delaware salt marsh. American Zoologist 20: 880.

- 8 -

Althoville Newfoundland Bathurst

Miramichi New Brunswick N ∗ Nova Scotia 0 150 300 kilometers

Figure 1 - Map of Atlantic Canada showing pulp and paper mills which used mummichogs as a sentinel species during Cycle 2 fish surveys. (adapted from

Courtenay et al., 2002 with permission). Inset shows geographic distribution of mummichogs.

- 9 -

2 EVALUATION OF TECHNIQUES FOR THE MARKING OF MUMMICHOGS (FUNDULUS HETEROCLITUS) WITH EMPHASIS ON VISIBLE IMPLANT ELASTOMER (VIE)

2.1 Abstract

The goal of these experiments was to determine the most appropriate method for use in a mark-recapture study of movements of mummichogs in the southern Gulf of St. Lawrence. In the first of two experiments, 315 mummichogs (Fundulus heteroclitus) were marked with fin clips, fingerling tags, acrylic paint or visible implant elastomer (VIE) and held 144 d. Percent tag retention (98.5%) and survival (100%) were highest for the VIE group and no negative effect was observed on growth in terms of length or condition factor.

This led to a more detailed experiment (167 d) to assess for potential negative effects of VIE on the growth and survival of 144 mummichogs marked in various body locations while assessing mark retention and readability. No negative effects on growth (length and condition factor) were detected: mean daily length increases ranged from 0.04 – 0.12 mm and mean daily weight increases ranged from 0.01 – 0.02 g. Mean survival of all groups was 90.8% (range: 66.7 –

100%) and was higher in marked groups than controls. All marks were retained

*Manuscript to be submitted for publication to North American Journal of Fisheries Management by Skinner MA, Courtenay SC, Parker WR, and Curry RA.

- 10 -

and visible without the aid of LED light at the end of the experiment, thus demonstrating VIE is a suitable method for the marking of mummichogs.

2.2 Introduction

The ability to mark and identify animals permits wildlife researchers and managers to label animals for handling, determine movement patterns, and gather valuable demographic information for populations of interest. External marks or tags, in particular, are beneficial for obtaining these data because they are minimally invasive to study organisms, generally portable, relatively quick and easy to administer, and more economically feasible to employ than internal marking or tagging alternatives (Dewey and Zigler, 1996).

Among the more popular external marking methods used for fisheries research are fin clipping, anchor tags, and injections/tattooing with readily visible materials. Fin clipping consists of removing a portion of a specimen’s fin and was one of the earliest methods developed for fish marking (McFarlane et al.,

1990). It is very possible, however, to confuse fish naturally missing portions of fins with those marked (Guy et al., 1996) and studies have demonstrated this technique may adversely affect growth and survival in certain species (Ricker,

1949; Saunders and Allen, 1967; Coble, 1971; Vincent-Lang, 1993; Pratt and

Fox, 2002). Similar observations have been made in studies of small fish using anchor tags (Nielsen, 1992) which are plastic or thin wire tags that are inserted into white muscle, usually below the dorsal fin (Guy et al., 1996). While injection of acrylic or latex paint has proven to be a useful method (McFarlane et al.,

- 11 -

1990), tissue damage (Lotrich and Meredith, 1974; Goforth and Foltz, 1998), mark fading (Lotrich and Meredith, 1974; Forrester, 1990; Wellington, 1992), increased mortality (Forrester, 1990), and decreased growth (Malone et al.,

1999) have all been noted, making the use of this method questionable.

Visible Implant Elastomer (VIE; Northwest Marine Technologies, Inc.,

Shaw Island, WA, USA) is a liquid elastomer that cures into a pliable solid after being injected. This elastomer will also fluoresce in the presence of deep blue

LED (light emitting diode) light adding greater tag readability in organisms with pigmented tissue. Recent studies using VIE in amphibians (Nauwelaerts et. al.,

2000), reptiles (Penney et. al., 2001), crustaceans (Ulgem et. al., 1996; Linnane and Mercer, 1998), and various taxa of fish (Godin et al., 1996; Frederick, 1997;

Willis and Babcock, 1998; Bruyndoncx et al., 2002; Roberts and Angermeier,

2004) have demonstrated high rates of mark retention, mark readability, and specimen survival with very limited side-effects. Studies with juvenile Barbus barbus (Farooqi and Morgan, 1996) and Salmo trutta (Olsen and Vollestad,

2001) have further reported no negative effects of VIE marking on growth.

The mummichog (Fundulus heteroclitus) is a euryhaline, littoral species of killifish (Family Cyprinodontidae) distributed along the eastern coast of North

America from the Gulf of St. Lawrence region to Texas mainly in tide pools and estuaries (Scott and Scott, 1998). As part of a larger project examining the site fidelity of this species, there was a need to mark large numbers of these fish.

Previous studies involving the marking of mummichogs have employed fin clipping (Butner and Brattstrom, 1960) and injections of acrylic (Lotrich and

- 12 -

Meredith, 1974; Lotrich, 1975; Meredith and Lotrich, 1979; Smith and Able,

1994; Sweeney et al., 1998) or latex paint (Murphy, 1991). Of these, only

Lotrich and Meredith (1974) reported on the effectiveness of the acrylic paint marking technique, stating certain colours were difficult to distinguish immediately after marking due in part to the dark pigmentation of mummichogs.

Furthermore, other colours were also hard to identify after 2-4 months with mark fading noted (Lotrich and Meredith, 1974).

These issues, combined with the recent development of VIE technology highlight the need for laboratory experimentation using this marking technique with mummichogs. The objectives of this study were to determine the most appropriate method (in terms of mark retention and readability) for marking large numbers of mummichogs and to assess whether this method is associated with any negative effects on growth or survival.

2.3 Methods

2.3.1 Marking and Tagging Trial (2001-2002)

This experiment was conducted at the Gulf Fisheries Centre aquarium

(Department of Fisheries and Oceans, Moncton, New Brunswick, Canada) over a 144 d period from August, 2001 until January, 2002. The 315 adult (>50 mm total length) mummichogs used in the experiment were captured by beach seine

(25 m in length; 1.5 m in height; 5 mm mesh size) at Horton’s Creek, New

Brunswick, Canada from June - July, 2001 and acclimated until the start of the experiment in a 1000 L fibreglass tank with re-circulated artificial seawater (Kent

- 13 -

sea salt in reverse osmosis treated Moncton City water; Kent Marine Inc.,

Acworth, Georgia, USA) at 10 ppt and 22 oC. Fish were fed Nutrafin fish food pellets daily (Rolf C. Hagen Inc., Montreal, Canada), supplemented with frozen bloodworms (chironomid larvae) enriched with multivitamins (Bio-pure, Hikari

Sales, Hayward, CA, USA).

Prior to marking, specimens were transferred to a 15 L bucket and anaesthetized with tricaine methanesulfonate (160 mg/L; Western Chemical,

Ferndale, WA, USA) until they exhibited a complete loss of orientation. Weight, total length, sex, and external condition of the fish (parasites, wounds, and fin erosion) were recorded before fish were randomly allocated to one of five groups and processed as follows:

1. Acrylic paint mark (n = 65) - marked with acrylic paint injected

intramuscularly on ventral surface behind anus using a 3 cc syringe with

a 25 g needle;

2. Anal fin clip (n = 58) - anal fin completely removed with scissors;

3. Visible Implant Elastomer (VIE) (n = 66) - marked with VIE injected

intramuscularly on the ventral surface behind the anus using a standard 1

cc insulin syringe and specialized hand injector;

4. Fingerling tag (plastic disc tag) (n = 63) – 8 lb. test fishing line with an

alpha numeric tag (Floy Tag Inc., Seattle, Washington, USA) passed

through the dorsal musculature anterior of the dorsal fin; and

5. Tank control (n = 64) – no mark or tag applied.

- 14 -

After marking, specimens were placed in an aerated 15 L bucket to aid in recovery from the effects of the anaesthetic, after which they were placed into separate nylon mesh baskets (52 cm x 37 cm x 37 cm; 5 mm mesh) in an experimental tank. Weight (g), total length (mm), tag readability, and overall condition were recorded monthly and baskets were checked daily for mortalities until the end of the experiment.

2.3.2 VIE Marking Experiment (2002-2003)

This experiment was conducted at the Gulf Fisheries Centre aquarium over a 167 d period from October, 2002 until April, 2003. The 144 (>40 mm TL) mummichogs used in the experiment were captured by beach seine (20 m in length; 1.5 m in height; 5 mm mesh size) and minnow traps (6.35 mm mesh size) baited with cat food at Strawberry Marsh, Miramichi, New Brunswick,

Canada during September, 2002 and acclimated to laboratory conditions until the start of the experiment.

Prior to the start of the experiment, specimens were allocated to one of four marked groups, four procedural control groups, or one tank control group.

Specimens were then transferred to a 15 L bucket and anaesthetized with tricaine methanesulfonate in groups of eight or less (160 mg/l; Western

Chemical, Ferndale, WA, USA) until they exhibited a complete loss of orientation. Sex and total length (mm) were recorded for all specimens.

Each marked group was marked bilaterally with VIE as described above in one of four body locations: both sides of the dorsal fin origin; behind both pectoral fins; lateral aspect of the caudal peduncle on both sides; or both sides

- 15 -

of the anus. Procedural control fish were processed in groups of two, as per the procedure outlined above for marked fish with the exception that when pierced with the needle, no dose of elastomer was delivered. Tank control fish were only anaesthetized. Maximum handling time per fish was less than one minute for the marking process and there was no immediate mortality of specimens. All fish were handled equally to avoid introducing a stress bias.

After handling, specimens were placed in an aerated 15 L bucket to aid in recovery from the effects of the anaesthetic. The fish were then divided into three replicate 120 L tanks (n=48 each) which were further subdivided as follows: one tank control group (n=8), one dorsal marked group (n=8), one caudal marked group (n=8), one pectoral marked group (n=8), one anal marked group (n=8), and subsequent procedural control groups (n=2) for each of the marked groups (Fig. 2a). All groups were held in plastic mesh baskets (6.35 mm mesh size) with larger baskets reserved for groups with larger sample sizes

(Fig. 2b).

Tanks shared a common supply of re-circulated water maintained at 24 oC and 25 ppt salinity. Fish were fed daily with a mixture of ground beef hearts

(Garden Province Meats, Charlottetown, PEI, Canada) and lobster meat

(donated by Dr. J.M. Hanson, Gulf Fisheries Centre, Department of Fisheries and Oceans) supplemented weekly with frozen bloodworms enriched with multivitamins (Bio-pure, Hikari Sales, Hayward, CA, USA). Weight (g), total length (mm), tag readability, and overall condition were recorded monthly and baskets were checked daily for mortalities.

- 16 -

2.3.3 Statistical Analyses

Statistical analyses using repeated measures were not employed as specimens in this experiment were not uniquely marked. Therefore, growth was assessed by statistically comparing specimens’ condition factors and lengths at the beginning and end of the experiment separately. Condition factor (k) is defined by the equation:

k = weight (g) / length (cm)3 x 100

This information was then used to describe trends among treatment groups over the course of the experiment. Lengths and weights for procedural control groups (dorsal, anal, pectoral, and caudal) were pooled for use in each analysis. Length and weight data were log-transformed prior to analysis to meet assumptions of normality (Zar, 1999). Analyses of variance (ANOVA) with

Bonferroni post-hoc analyses were used to test for differences in lengths among groups. Analyses of covariance (ANCOVA) with Tukey’s Honestly Significant post-hoc tests were used to test for differences in condition factor among groups with weight and length as dependent variable and covariate, respectively. Tests for significant differences between initial and final condition factors for each group were also conducted using ANCOVA. If significant interactions in slopes were detected during ANCOVAs, the data were graphed and interpreted for biological significance. As no biological significance was found when this case arose, a sub-sample of data from a narrow range of lengths (70-90 mm) was

- 17 -

chosen for a subsequent ANOVA (weight as dependent variable and length as covariate) to minimize the influence of length. These analyses were completed using SYSTAT® (ver. 10.2, SPSS Inc., Chicago, Il., U.S.A.) and all were judged to be significant at α = 0.05. Chi-Square tests were used to test for differences in percent mark retention and survival among groups and were completed using

MINITAB® (ver. 14.12.0, MINITAB Inc., U.S.A.).

2.4 Results

2.4.1 Marking and Tagging Trial (2001-2002)

Percent mark retention was significantly different among groups (χ2 =

154.755, df = 3, p < = 0.001) with the highest values observed for the VIE group and the lowest for the anal fin clip and fingerling tag groups (Table 1). Twenty percent of fish from the fingerling tag group showed substantial dorsal fin tears and two other specimens from this group had dorsal flesh damage. Although no significant differences were noted in survival among groups (χ2 = 0.304, df = 4, p = 0.990), VIE also had one of the highest values noted (Table 1). Length also did not vary significantly among groups at the beginning (ANOVA: F4,319 =

0.170, p = 0.954) or end of the experiment (ANOVA: F4,309 = 0.129, p = 0.972), however, the VIE group were longest at the conclusion. Condition factor, however, was different among groups at the experiment’s beginning (ANCOVA:

F4,318 = 5.887, p < 0.001) with both the VIE (n = 66) and acrylic paint groups (n =

64) having values significantly lower than control (Tukey’s Honestly Significant

- 18 -

Test, p = 0.001; Table 1). These same results were observed at the end of the experiment (ANCOVA: α = 0.05, p < 0.001, F4,308 = 29.681) for VIE (n = 66) and acrylic paint groups (n = 63) (Tukey’s Honestly Significant Test, p < 0.001; Table

1). Despite having the lowest condition factor, VIE was selected as the most appropriate marking method as this group displayed positive growth and had the highest values for mark retention and survival. A subsequent, detailed examination of possible effects of this method on the growth and survival of mummichogs was conducted.

2.4.2 VIE Marking Experiment (2002-2003)

There was no difference in mark retention among groups with 100% of marks retained and all observed without the aid of LED light. Mean daily length increases ranging from 0.04 – 0.12 mm (Table 2). At the start of the experiment, differences in length (ANOVA: F5,138 = 7.550, p < 0.001) were noted with pooled procedural control specimens having significantly greater lengths than tank controls (Bonferroni Test, n = 24, p = 0.045). Differences among groups were also noted at the end of the experiment (ANOVA: F5,129 = 3.033, p

= 0.013; Table 2) however no groups were significantly different from tank controls (Bonferroni Test, n = 24, p = 0.747). Mean daily weight increases ranged from 0.01 – 0.02 g (Table 2). Initial differences in condition factor

(ANCOVA: F5,137 = 4.908, p < 0.001) were detected among groups, with anal marked fish having significantly higher values than tank controls (Tukey’s

Honestly Significant Test, p = 0.028, n = 24; Table 2). Differences were also recorded among groups for final condition factor (ANOVA on subset of fish 70-

- 19 -

90 mm TL, n = 12-22 per treatment: F5,128 = 7.995, p < 0.001) with pooled procedural control groups having a final value significantly lower than tank controls (Bonferroni Test, n = 20, p < 0.001). Decreased condition factors were noted in four of six groups, however no significant differences were noted between initial and final condition factors for any group (Table 2).

Mean survival for specimens in this experiment was 90.8%, with values ranging from 66.7 – 100% (Table 2). The lowest mean survival percentages were noted for the small (n = 6) procedural control groups. However, when these smaller groups were pooled (n = 24), these minimum and mean survival percentages increased to 83.3% and 93.0%, respectively with no significant differences detected among treatments and controls (χ2 = 0.212, df = 5, p =

0.999). While differences were not observed in survival percentages between marked and control groups, it is interesting to note that marked fish had higher survival percentages (Table 2).

2.5 Discussion

Visible implant elastomer proved to be the best performing marking method of those tested for mummichogs and was subsequently shown to have repeatedly high percentages of mark retention and readability combined with high survival rates and no negative effects on growth. These findings are consistent with various studies involving VIE marking of other species of fish

(Farooqi and Morgan, 1996; Godin et al., 1996; Frederick, 1997; Willis and

- 20 -

Babcock, 1998; Olsen and Vollestad, 2001; Bruyndoncx et al., 2002; Roberts and Angermeier, 2004).

No significant differences were detected in length among groups at the beginning or end of the initial marking experiment while the VIE and acrylic paint groups had significantly lower initial and final condition factors (Table 1). This similarity in results between the start and finish of the experiment indicate none of these marking methods had negative impacts on growth.

Mummichogs marked with fingerling tags, however, had very poor mark retention (Table 1) with tissue damage observed on many specimens. Anal fin clipping was also shown to be unsatisfactory in terms of mark retention (Table

1) due to the ability of mummichogs to quickly regenerate fin tissue. Also, differences in anal fin morphology between males and females introduced the potential of confusing these marks with natural variation. Conversely, mummichogs marked by either of these methods had high survival rates (Table

1), unlike results reported in previous studies with other species (Ricker, 1949;

Coble, 1971; Saunders and Allen, 1967; Neilsen, 1992; Vincent-Lang, 1993;

Pratt and Fox, 2002). This difference may be attributable to differences in holding conditions and feeding methods between this study and those mentioned.

Results from the acrylic paint group were significantly better, with high mark retention and survival rates noted (Table 1). Specimens from this group were also free of tissue damage or mark fading, which was previously reported

(Lotrich and Meredith, 1974; Forrester, 1990; Wellington, 1992; Goforth and

- 21 -

Foltz, 1998; Malone et al., 1999). This favourable performance, however, was overshadowed by results from the VIE group which had higher tag retention and fewer mortalities (Table 1). The increased mark retention in this group was most likely due to the consistency and chemical properties of VIE compared to acrylic paint. The elastomer was more viscous than acrylic paint when injected and this may have allowed for less marking material to be lost from specimens immediately after marking. Also, VIE cured into a pliable solid with rubber-like consistency whereas acrylic paint dried hard and non-pliable. This difference in flexibility may have caused fewer VIE marks to be shed, leading to the high mark retention observed for this group.

At the beginning of the second experiment, significant differences were observed in specimen length among groups, with the pooled procedural control groups having the highest lengths (Table 2). When initial and final lengths were compared, the pooled procedural control groups again had the highest values, suggesting no influence of VIE marking on specimen length as the same general trend was observed at the experiment’s start and finish. Furthermore, the mean daily increases for both length and weight were higher for marked fish than procedural controls thus providing more evidence against VIE negatively affecting growth (Table 2). Similar results have also been reported in the literature. Farooqi and Morgan (1996) noted no significant effect of VIE marking on the length of barbel (Barbus barbus) held 57 d under laboratory conditions and Olsen and Vollestad (2001) made similar observations on juvenile brown trout (Salmo trutta) in both laboratory and field studies.

- 22 -

Variation in length of fish decreased over the course of the experiment

(Table 2) and it appears this was caused because procedural control specimens were closer to average maximum length for this species (125 mm; Abraham,

1985) at the start of the experiment. This situation then would have permitted treatment specimens to “catch up” to procedural control specimens in terms of length. Meredith and Lotrich (1979) reported mummichog growth rates of 0.115 mm/day-1 and 0.065 mm/day-1 for 60-69 mm and 70-79 mm size classes, respectively. These growth differences between age classes are comparable to those noted in the current study where fish with initial lengths from 60.8 – 67.8 mm and 73.2 – 81.7 mm had mean daily length increases of 0.11 and 0.07 mm, respectively (Table 2). Therefore, the longer fish were growing at a much slower rate throughout the experiment than shorter fish living under the same conditions and this allowed the smaller fish from other groups to attain final lengths similar to the originally larger procedural control fish.

Similar to the results for length, condition factor was significantly different among groups at the start of the experiment, with the anal marked group having the highest values (Table 2). Again, this was purely due to chance as specimens were randomly allocated to groups. At the end of the experiment, unexpected decreases were noted in condition factor in four of six groups. On closer examination, the largest decreases in condition factor were found among the procedural control groups. While it could be argued this result may have been caused by the smaller sizes of the baskets in which these groups were held, these fish were actually held in lower densities than the remaining

- 23 -

specimens and therefore it could, more reasonably, have been expected that these groups should have had greater condition factors. Regardless, these changes in condition factor were not statistically significant and when these results and the observed increases in both length and weight are accounted for, it is clear that VIE marking had no negative effect on the growth of mummichogs throughout the experiment.

Mean survival and mark retention among groups of mummichogs were high with marked groups having higher survival rates than all control groups

(Table 2). These high rates are in accordance with values reported in the literature for VIE: 100% survival and 98.2% mark retention in Cottus gobio held four weeks (Bruyndoncx et al. 2002); 99.5% survival and 100% mark retention in young-of-the-year Salmo trutta held 77 days (Olsen and Vollestad, 2001); and

Willis and Babcock (1998) who found 71% mark retention and no mortality attributable to marking in Pagrus auratus marked over 2 weeks in the wild. In a mark-recapture study related to this experiment (Skinner et al., submitted), five percent of marked mummichogs (n = 228) were held 48 hours post-marking and had 100% mark retention and 96% survival, with all mortalities related to excessive handling or adverse holding conditions rather than VIE marking.

While only examining short-term effects in the field, these results combined with the results of the longer term laboratory monitoring in this study clearly demonstrate this method of marking has no negative effect on mummichog survival. Simultaneously, VIE marked fish had perfect mark retention rates

(Table 2). All marks were so clearly readable LED light was not required during

- 24 -

the experiment to identify marks, regardless of the degree of mummichog skin pigmentation.

In the current study, higher rates of mark retention and survival in mummichogs marked with VIE compared to other popular marking alternatives along with results indicating no negative effect on growth demonstrate VIE is the most suitable method of marking mummichogs. With its relatively easy application and low cost of materials, VIE provides an excellent method to quickly mark thousands of fish in an economically feasible manner and as such should be strongly considered for use in the marking of small-bodied from a variety of taxa.

2.6 Acknowledgements

This project was funded by the National and Atlantic Region Environmental

Effects Monitoring offices of Environment Canada, Department of Fisheries and

Oceans Canada (DFO Science Subvention Grant awarded to RAC), and UPM-

Kymmene Miramichi Inc. (NSERC Industrial Post-graduate Scholarship awarded to MAS). We thank C. Roderick for assisting in the design and execution of the initial marking experiment; P. Riebel (UPM-Kymmene Miramichi

Inc.) and K. Munkittrick (UNB Saint John) for logistic support and input on study design; M. Barbeau for further study design input; and E. Marchand for statistical guidance. Additional field and in-kind support were provided by the

Miramichi River Environmental Assessment Committee (MREAC) and

Department of Fisheries and Oceans Canada.

- 25 -

2.7 Literature Cited

Abraham BJ. 1985. Species profiles: life histories and environmental

requirements of coastal fishes and invertebrates - Mummichog and striped

killifish. US Fish and Wildlife Service Biological Report (US Army Corps of

Engineers) 82: 1-23.

Bruyndoncx L, Knaepkens G, Meeus W, Bervoets L, and Eens M. 2002. The

evaluation of passive integrated transponder (PIT) tags and visible implant

elastomer (VIE) marks as new marking techniques for the bullhead. Journal

of Fish Biology 60: 260-262.

Butner A and Brattstrom BH. 1960. Local movement in Menidia and Fundulus.

Copeia 2: 139-141.

Coble DW. 1971. Effects of fin clipping and other factors on survival and growth

of smallmouth bass. Transactions of the American Fisheries Society 100:

460–473.

Dewey MR and Zigler SJ. 1996. An evaluation of fluorescent elastomer for

marking bluegills in experimental studies. The Progressive Fish Culturist 58:

219-220.

- 26 -

Farooqi MA and Morgan CE. 1996. Elastomer visible implant (EVI) tag retention

and the effect of tagging on the growth and survival of barbel, Barbus barbus

(L.). Fisheries Management and Ecology 3: 181-183.

Forrester GE. 1990. Factors influencing the juvenile demography of a coral reef

fish. Ecology 71: 1666-1681.

Frederick JL. 1997. Evaluation of fluorescent elastomer injection as a method

for marking small fish. Bulletin of Marine Science 61: 399-408.

Godin DM, Carr WH, Hagino G, Segura F, Sweeney JN, and Blankenship L.

1996. Evaluation of a fluorescent elastomer internal tag in juvenile and adult

shrimp Penaeus vannamei. Aquaculture 139: 243-248.

Goforth RR and Foltz JW. 1998. Movements of the yellowfin shiner, Notropis

lutipinnis. Ecology of Freshwater Fish 7: 49-55.

Guy CS, Blankenship HL, and Nielsen LA. 1996. Tagging and Marking, p. 353-

383. In Murphy BR and Willis DW (eds.), Fisheries Techniques. American

Fisheries Society, Bethesda, Maryland

Linnane A and Mercer JP. 1998. A comparison of methods for tagging juvenile

lobsters (Homarus gammarus L.) reared for stock enhancement. Aquaculture

- 27 -

163: 195-202.

Lotrich VA and Meredith WH. 1974. A technique and the effectiveness of

various acrylic colours for subcutaneous marking of fish. Transactions of the

American Fisheries Society 103: 140-142.

Lotrich VA. 1975. Summer home range and movements of Fundulus

heteroclitus (Pisces: Cyprinodontidae) in a tidal creek. Ecology 56: 191-198.

McFarlane GA, Wydoski RS, and Prince ED. 1990. Historical review of the

development of external tags and marks, p. 9–29. In Parker NC, Giorgi AE,

Heidinger RC, Jester DB, Jr., Prince ED, and Winans GA (eds), Fish-marking

techniques. American Fisheries Society, Symposium 7, Bethesda, Maryland.

Meredith WH and Lotrich VA. 1979. Production dynamics of a tidal creek

population of Fundulus heteroclitus (Linnaeus). Estuarine and Coastal

Marine Science 8: 99-118.

Malone J C, Graham EF, and Steele MA. 1999. Effects of subcutaneous

microtags on the growth, survival, and vulnerability to predation of small reef

fishes. Journal of Experimental Marine Biology and Ecology 237: 243-253.

Murphy S. 1991. The ecology of estuarine fishes in southern Maine high salt

- 28 -

marshes; access corridors and movement patterns. M.Sc. Thesis, University

of Massachusetts.

Nauwelaerts S, Coeck J and Aerts P. 2000. Visible implant elastomers as a

method for marking adult anurans. Herpetological Review 31: 154-155.

Penney KM, Gianopulos KD, and McCoy ED. 2001. The visible implant

elastomer technique in use for small reptiles. Herpetological Review 32: 236-

241.

Nielsen LA. 1992. Methods of marking fish and shellfish. American Fisheries

Society, Special Publication 23, Bethesda, Maryland.

Olsen EM and Vollestad LA. 2001. An evaluation of visible implant elastomer for

marking Age-0 brown trout. North American Journal of Fisheries

Management 21: 967-970.

Pratt TC and Fox MG. 2002. Effect of fin clipping on overwinter growth and

survival of age-0 walleyes. North American Journal of Fisheries Management

22: 1290–1294.

Ricker WE. 1949. Effects of removal of fins upon growth and survival of spiny-

rayed fishes. Journal of Wildlife Management 13: 29–40.

- 29 -

Roberts JH and Angermeier PL. 2004. A comparison of injectable fluorescent

marks in two genera of darters: effects on survival and retention rates. North

American Journal of Fisheries Management 24: 1017-1024

Saunders RL and Allen KR. 1967. Effects of tagging and of fin clipping on the

survival and growth of Atlantic salmon between smolt and adult stages.

Journal of the Fisheries Research Board of Canada 24: 2595–2611.

Skinner MA, Courtenay SC, Parker WR and Curry RA. Site specificity of

mummichogs (Fundulus heteroclitus) in an Atlantic Canadian Estuary.

Submitted to Water Quality Research Journal of Canada.

Smith, KJ and Able KW. 1994. Salt-marsh tide pools as winter refuges for the

mummichog Fundulus heteroclitus, in New Jersey. Estuaries 17: 226-234.

Sweeney J, Deegan L, and Garritt R. 1998. Population size and site fidelity of

Fundulus heteroclitus in a macrotidal saltmarsh creek. Biol. Bull. 195:

238-239.

Ulgem I, Noess H, Farestveit E, and Jorstad KE. 1996. Tagging of juvenile

lobsters (Homarus gammarus (L.)) with Visible Implant Fluorescent

Elastomer Tags. Aquacultural Engineering 15: 499-501.

- 30 -

Vincent-Lang D. 1993. Relative survival of unmarked and fin-clipped coho

salmon from Bear Lake, Alaska. Progressive Fish Culturist 55: 141–148.

Wellington GM. 1992. Habitat selection and juvenile persistence control the

distribution of two closely related Caribbean damselfishes. Oecologia 90:

500-508.

Willis TJ and Babcock RC. 1998. Retention and in situ detectability of visible

implant fluorescent elastomer (VIFE) tags in Pagrus auratus (Sparidae). New

Zealand Journal of Marine and Freshwater Research 32: 247-254.

- 31 -

Table 1 – Mean lengths, mean weights, mean daily increases in length and weight, mean condition factor, sample sizes,

percent survival, and percent mark retention for 2001-2002 marking and tagging trial (L = length; W = weight; SD

= standard deviation; 1 = start of the experiment; 2 = end of 144 d experiment; * = statistically significant

difference from controls, Tukey’s Honestly Significant Test, α = 0.05).

Mean Mean daily daily Mark L L Condition Condition Survival 1 2 length W (g) W (g) weight Marked Retention (mm) (mm) 1 2 factor (k) factor (k) (%) increase increase 1 2 (%) (mm) (g) Acrylic

-32- 66.9 74.2 0.05 3.42 5.01 0.01 1.13* 1.21* 64 98.4 90.5 Anal Clip 67.5 74.2 0.05 3.64 5.28 0.01 1.16 1.28 65 89.2 46.6 VIE 67.6 74.3 0.05 3.50 4.82 0.01 1.12* 1.16* 66 100 98.5 Fingerling 67.3 73.8 0.05 3.85 5.36 0.01 1.19 1.33 64 100 3.2 Control 67.4 73.9 0.04 3.75 5.47 0.01 1.22 1.31 67 94 … Mean 67.3 74.1 0.05 3.63 5.19 0.01 1.19 1.31 … 96.3 59.7 SD 0.3 0.2 0.00 0.18 0.27 0.00 0.03 0.03 … 4.67 44.0

- 32 -

Table 2 - Mean lengths, mean weights, mean daily increases in length and weight, mean condition factor, sample sizes,

percent survival, and percent mark retention for 2002-2003 marking and tagging trial (L = length; W = weight; SD

= standard deviation; 1 = start of the experiment; 2 = end of 167 d experiment; * = statistically significant

difference from tank controls, Tukey’s Honestly Significant Test, α = 0.05, ** = statistically significant difference

from tank controls, Bonferroni Test, α = 0.05).

Mean Mean daily daily Mark

-33- L L Condition Condition Marked Survival 1 2 length W (g) W (g) weight Readability (mm) (mm) 1 2 factor (k) factor (k) (n) (%) increase increase 1 2 (%) (mm) (g) Dorsal Control 75.7 87.0 0.07 4.66 6.62 0.01 1.06 0.98 6 100 … Anal Control 81.7 88.8 0.04 6.33 7.85 0.01 1.13 1.03 6 83.3 … Pectoral Control 73.2 88.8 0.09 4.49 7.8 0.02 1.12 1.07 6 66.7 … Caudal Control 77.8 88.8 0.07 5.29 7.83 0.02 1.10 1.09 6 100 … Pooled 77.1** 88.8** 0.07 5.19 7.48 0.01 1.10 1.04* 24a 83.3 …

Tank Control 67.8 84.5 0.10 3.75 7.13 0.02 1.14 1.16 24 87.5 … Dorsal Marked 67.3 83.6 0.10 3.78 6.82 0.02 1.18 1.15 24 100 100 Anal Marked 66.4 85.2 0.11 3.87 7.52 0.02 1.22* 1.19 24 95.8 100 Pectoral Marked 60.8 80.2 0.12 2.91 6.12 0.02 1.20 1.18 24 95.8 100 Caudal Marked 63.0 82.8 0.12 3.15 6.91 0.02 1.19 1.20 24 95.8 100 Mean (bold) 65.1 83.3 0.10 3.78 7.00 0.02 1.16 1.18 24 93.0b 100 SD (bold) 3.0 1.9 0.02 0.79 0.52 … 0.04 0.02 … 6.27 … asum of column bmean survival percentage is 90.8% when procedural control values not pooled.

- 33 -

Pectoral Control Dorsal Control n =2 n = 2 Caudal Control n =2 Tank Control n =8

Caudal Marked Standpipe n = 8

Dorsal Marked n = 8 Pectoral Marked n = 8 Anal Marked n = 8 Anal Control n = 2

Figure 2a – Aerial view of experimental tank design for 2002-2003 VIE marking experiment showing plastic cages used to hold different specimen groups.

- 34 -

30 cm 50 cm 50 cm

15 cm 15 cm

Figure 2b – Experimental basket dimensions for 2002-2003 VIE marking experiment.

- 35 -

3 SITE FIDELITY OF MUMMICHOGS (FUNDULUS HETEROCLITUS) IN AN ATLANTIC CANADIAN ESTUARY

3.1 Abstract

The range of movement of mummichogs (Fundulus heteroclitus) within the upper Miramichi River estuary, New Brunswick, Canada, was investigated to assess the value of using this fish as a sentinel species for monitoring effects of point source anthropogenic effluents such as pulp and paper mill effluent.

During the ice-free season (May – November) of 2002, 4123 adult mummichogs

(>30 mm TL) were captured by beach seine and minnow trap biweekly from four sites within the estuary and marked using Visible Implant Elastomer (Northwest

Marine Technologies, Inc., Shaw Island, WA, USA). Recaptures were made at the tagging sites and elsewhere during this period and again during the ice free season of April – November, 2003. A total of 639 (15.5% of those marked) mummichogs were recaptured with 617 (96.6%) found within 200 m of the point of initial release. Twenty-nine of the 617 were recaptured 2 or 3 times at sites of original marking. The remaining 22 recaptured fish moved distances ranging from 600-3600 m up and downstream of initial marking sites. Eighty-two percent of recaptures were made within 12 weeks of the start of marking with the remainder recovered up to 72 weeks later. These findings are consistent

*Manuscript submitted for publication to Water Quality Research Journal of Canada by Skinner MA, Courtenay SC, Parker WR, and Curry RA.

- 36 -

with results from studies of mummichog movement in smaller water bodies and other parts of the species’ range. With regards to mobility, these results add to the growing body of literature supporting the usefulness of mummichogs as a sentinel species in environmental monitoring programs for point source impacts in Canadian Atlantic estuaries.

3.2 Introduction

Amendments made to the federal Pulp and Paper Effluent Regulations in

1992 require Canadian mills discharging effluent into aquatic receiving environments to conduct an Environmental Effects Monitoring (EEM) program.

This monitoring program was designed to be conducted in three year cycles and was intended to assess the effectiveness of the amended effluent regulations in protecting fish, fish habitat, and the use of fisheries resources (Courtenay et al.,

2002). Following problems with catching sufficient fish that were reliably exposed to effluent during EEM Cycle 1 Fish Surveys (1993-1996), a Fish

Survey Expert Working Group recommended the use of small-bodied sentinel species in Cycle 2 Fish Surveys (1997-2000) (Munkittrick et al., 1997). The mummichog (Fundulus heteroclitus) was suggested as a potential sentinel species for mills discharging into Atlantic coastal and estuarine waters

(Courtenay and Couillard, 1998).

One of the main criteria for a sentinel species of point source impacts is low or reduced mobility to ensure maximum exposure to a specific receiving environment (e.g., Gibbons et al., 1998a, b). For Canadian pulp and paper

- 37 -

EEM fish surveys, fish found within effluent plumes having concentrations equal to or greater than one percent are considered exposed (Environment Canada,

1998). Such plumes in dynamic estuarine and coastal waters can be small, on the order of hundreds of meters, even for pulp mills discharging large volumes of effluent. Several facilities along the Atlantic coast of Canada selected the mummichog for monitoring based on its availability. There are reports of low mobility in populations from the central part of the species range in the eastern

United States. Lotrich (1975) reported a 36 m home range for mummichogs in a

Delaware salt marsh tidal creek and Sweeney et al. (1998) found mummichogs in a New Jersey salt marsh tidal creek to move distances <650 m.

Mummichogs were used successfully as a sentinel species for Cycle 2

Fish Surveys at three Atlantic Canadian pulp and paper mills in northern New

Brunswick between 1997 and 2000 (Courtenay et al., 2002). However, while sampling at mills discharging into the Miramichi and Restigouche estuaries, consultants noted that mummichogs could only be captured immediately preceding and after high tides (Jacques Whitford Environmental Limited, 2000a, b). These observations led the consultants to speculate that northern populations of mummichogs may not be as site-specific as those studied along their central distribution because movement offshore into river channels with falling tides could result in advection hundreds to thousands of meters downstream (Jacques Whitford Environmental Limited, 2000a, b). If northern populations were thereby mobile, either actively or passively, then their

- 38 -

exposure to point-source effluents in these large rivers and their value as biosentinels might be minimal.

Miramichi River estuary (MRE) which has a very different tidal amplitude and size. Of the two studies cited above, Lotrich (1975) is the more comprehensive but the section of creek studied had a tidal amplitude of 0.3-0.6 m which is not comparable to the 2 m tidal amplitude encountered in the MRE (Reinson, 1977).

Also, while the salt marsh examined by Sweeney et al. (1998) had a tidal amplitude more comparable to the MRE (>3 m), it and the study by Lotrich

(1975) were conducted in the tidal creeks of salt marshes. The MRE by contrast, is fed by two major rivers: the Northwest Miramichi and Southwest

Miramichi, with respective drainage areas of 3900 km2 and 7700 km2 (Chiasson,

1995); monthly discharges ranging from 86 m3.s-1 in August to 620 m3.s-1 in May

This information combined with the observations made by consultants caused the Canadian EEM National Science Working Group to classify the study of mummichog movement in Atlantic Canada as a research priority. This

- 39 -

recommendation led to the creation of a project involving industry, Environment

Canada, Fisheries and Oceans Canada, and the University of New Brunswick.

Our objective was to describe the spatial and temporal movement patterns of mummichogs in a large river estuary and assess their usefulness as a sentinel species in environmental programs such as EEM. No previous studies have attempted to address the issue of mummichog movement over such a time period and this is the first to examine this topic at the northern extent of this species’ latitudinal range.

3.3 Materials and Methods

3.3.1 Study area

This study was conducted in the upper Miramichi River estuary (MRE), located in northeastern New Brunswick, Canada (Fig. 3). The Miramichi River catchment has an approximate area of 14,000 km2 and measures 95 km north to south and 190 km east to west. River flow throughout the catchment averages 306 m3.s-1 and may vary from 25 m3.s-1 to 6000 m3.s-1. The estuary is subject to mixed semi-diurnal tides with a tidal range between 0 and 2 m

(Reinson, 1977), is exposed to a freshwater inflow of 5-10% of the volume of incoming tide water (Chiasson, 1995), and is ice covered from mid-December to early April. Salt water intrudes into the upper estuary as a salt wedge, the upper extent of which varies in position with lunar tide and freshwater discharge, being downstream of the study area in early spring and far upstream of the study area in summer.

- 40 -

3.3.2 Fish collection and marking

Mummichogs were collected for marking from four sites along the MRE near the influence of a bleached kraft pulp mill: Bleached Kraft Pulp Mill

(BKPM; N 46o58.334’, W 65o35.195’), Strawberry Marsh (SM; N 46o59.499’, W

65o34.084’), Upstream Flett Cove (UFC; N 46o57.440’, W 65o34.775’), and

Chatham Head (CH; N 46o59.680’, W 65o33.240’; Fig. 3). The 1% effluent plume for the BKPM has been shown to move 1400 m up and 1700 m downstream from the outfall with flood and ebb tides, respectively (Jacques

Whitford Environmental Limited, 2004; Fig. 3). Fish were collected bi-weekly from May 23-August 30, 2002 by beach seine (25 m in length; 1.5 m in height; 5 mm mesh size) approximately 1.5 hours before and after peak high tide.

Collections were also made using minnow traps (6.35 mm mesh size) baited with cat food set during the same periods and left until the next high tide at least. Beach seining and minnow trapping were carried out at SM when it was accessible approximately 1.5 hours before and after low tide.

Captured fish were immediately transferred to an aerated holding tank and then to a 15 L bucket for anaesthesia with tricaine methanesulfonate (160 mg/l; Western Chemical, Ferndale, WA, USA) until they exhibited a complete loss of orientation. Sex and total length (mm) were recorded and fish were marked bilaterally in one of four body locations with Visible Implant Elastomer

(Northwest Marine Technologies, Inc., Shaw Island, WA, USA) prepared according to manufacturers specifications. Marks were injected intramuscularly

- 41 -

using a standard 1cc insulin syringe and specialized hand injector (Northwest

Marine Technologies, Inc., Shaw Island, WA, USA). Generally, fish >30 mm TL were marked because specimens below this size were difficult to mark without causing mortality. Total handling time per fish was less than one minute for the marking process. After marking, specimens were placed in an aerated bucket until orientation was regained to aid in recovery from the effects of the anaesthetic and then released back into the estuary at the point of capture. For this portion of the study, 5% of all marked fish were retained for a minimum of

48 h in an aerated, temperature regulated holding tank to monitor mark retention and survival. These fish were not subsequently released into the estuary.

Each body marking location corresponded to a specific site along the

MRE: both sides of the dorsal fin origin (SM); behind both pectoral fins (CH); lateral aspect of the caudal peduncle on both sides (BKPM); or both sides of the anus (UFC). Fish were marked bilaterally to increase the number of colour combinations available and fish were marked with a different colour combination during each field excursion. This method allowed for the date that a particular fish was marked to be determined, providing an estimate of the time required for movement among sites. Preliminary experiments had shown similar high mark retention in all body locations without negative effects on growth or survival

(Skinner et al., in preparation).

Recaptures were attempted at the four tagging sites and also at an additional five sites (Fig. 3) along the MRE bi-weekly from May 27 to August 30,

2002 and then once per month from September - November, 2002. Recaptured

- 42 -

specimens were distinctively remarked by marking the body location which corresponded to the site of recapture with a colour pattern which denoted the date captured. Fish recaptured at the same site multiple times had new marks placed adjacent to previous marks while fish recaptured at non-original marking sites were marked on the body location which corresponded to the site of recapture and were given an extra mark at a 45o angle to these marks to distinguish between the original marking site and future recapture sites.

Bi-weekly recapture attempts resumed the following spring from April 17

– August 21, 2003 and then once per month from September - November, 2003.

Extra recaptures were also made May 22-23, 2003 while sampling mummichogs at BKPM and UFC for the BKPM’s Environmental Effects Monitoring (Cycle 3) conducted by Jacques Whitford Environmental Ltd. (2004).

The length of river surveyed for marked fish from April – June, 2003 was expanded to ~10 km in an attempt to recapture mummichogs that might have overwintered at different locations (Fig. 3). The rationale for this expansion was that if mummichogs had migrated to overwintering areas away from original marking locations, they might be recaptured at these areas or while migrating back to a summer home range. The portion of river surveyed was reduced to approximate the 2002 range in early July, 2003 (Fig. 3).

- 43 -

3.4 Results

3.4.1 Marking and recapturing by site

Between May 29 and August 30, 2002, 4123 marked fish were released into the upper MRE with the highest number marked at SM where fish were most vulnerable to sampling methods (Table 3). Eighty-two percent of recaptures were made within 12 weeks of the start of marking with the remainder recovered up to 72 weeks later (Table 5). Mark retention was 100% for fish held for 48 hrs after marking (Table 4). Overall survival for these fish was 89.1% (Table 4).

Six hundred and thirty-nine fish (15.5% of those marked) were recaptured in 2002 and in 2003 (Table 5). Twenty-five fish were recaptured twice and four fish were recaptured three times for a total of 29 multiple recaptures (Table 5).

Mummichogs from SM were almost twice as likely to be recaptured as mummichogs marked at other sites (20.1% compared to 9.2%, 11.6%, and

13.4%, at SM, CH, BKPM, and UFC respectively; Table 3). Recapture rates declined in 2003. The number of mummichog movements from each site appears to have been independent of the numbers of fish found (density) at these sites as BKPM had the greatest number of fish moved (Fig. 4c) but also had the second lowest total number of fish captured by site (Table 3).

3.4.2 Movements of marked individuals

Mummichogs in the upper MRE displayed distinct site fidelity; 96.6% of marked fish remained within 200 m of original marking sites (Table 5). From

- 44 -

May-November, 2002 and April-November, 2003, 22 marked mummichogs

(3.4%) moved distances greater than 200 m (Table 6). While a higher proportion of fish moved were recaptured >200m from original marking sites in

2003 than 2002 (6.5% vs 2.8%), the low proportion in both years suggests a high degree of interannual site fidelity (Tables 5 and 6). Of the 22 fish found to move, those from BKPM had the highest proportion of mobile individuals (41%) and most movement occurred along the northern shoreline between this location and Strawberry Marsh (Fig. 4b and c; Table 6).

3.5 Discussion

Intensive marking and recapturing of mummichogs during two successive ice-free seasons (May-November) supported previous findings that mummichogs have generally high site fidelity, even in larger estuarine systems such as the Miramichi River. Sweeney et al. (1998) noted the movements of marked mummichogs in Massachusetts tidal creeks were not greater than 650 m from point of release while Murphy (1991) found ~85% of marked mummichogs remained in the same panne within a Maine salt marsh. Through repeated field experiments, Lotrich (1975) consistently found that the majority of mummichogs (> 60 mm TL) marked and released were found within a 36 m home range in Delaware tidal creeks, with the greatest distance moved being

375 m by just three fish. Teo and Able (2003) concluded mummichogs in a restored New Jersey salt marsh have a maximum areal home range of ~15 ha. during high tide which corresponded to a linear homerange of ~500 m. The high rate of recapture (96.6%) within 200 m of the point of their release in the

- 45 -

current study corroborates previous studies and indicates that no substantial movement of mummichogs occurred within the upper MRE.

Fish retained immediately after marking in the field had 100% mark retention and 89.1% survival (Table 4). This is in accordance with Skinner et al.

(in preparation) who observed mummichogs marked and held under laboratory conditions for 167 d had 100% mark retention and mean survival of 90.8%.

During the current study, mortalities occurred when aeration or chilling failed on days with high temperatures or when accidentally killed (i.e., crushed during handling not related to marking). These rare events aside, all fish survived and we have no evidence of mortality from the marking procedure.

During this study, 22 of the 639 recaptures occurred at locations 600-

3600 m away from the original marking sites (mean = 2023 m; Table 6). No trends related to the size or sex of these fish were noted (Table 6). It is plausible these movements may be due to intraspecific competition for food or reproductive resources. No previous studies, however, have examined intraspecific competition among fundulid species and the findings of this study and others mentioned suggest this topic deserves further study. Gerking (1958) describes possible reasons for the extensive movements of certain fish. He states “stray fish” should be expected in any movement study as a consequence of biological variation within a given population. In all populations, it is this variant mobile behaviour that leads individuals to disperse to new locations or re-colonize areas previously vacated due to natural disturbance or biotic interactions (Gerking, 1958). As such, the movement of these 22 mummichogs

- 46 -

may be explained as being due to variant behaviour and not some other specific mechanism.

The results of previous studies suggest this high incidence of site-fidelity may be due to the availability of suitable habitat for reproduction, feeding and predator avoidance. During the spawning season, female mummichogs lay their eggs among intertidal vegetation at levels only reached by high spring tides

(Taylor and DiMichelle, 1983). Mummichogs in the upper MRE were found in abundance only where this vegetation existed before, during, and after the spawning period (pers. obs.), thus highlighting their dependence on this habitat for feeding and predator avoidance (Vince et al., 1976; Heckert et al., 1999) as well as reproduction. Teo and Able (2003) noted that the 15 ha. home range observed constricted as the water level receded from the marsh surface.

Mummichogs in that study restricted their movements to a shallow section of sub-tidal creek less than 200 m long and the authors concluded that this limited movement was due to the presence of predators such as striped bass (Morone saxatilis) and Atlantic croaker (Micropogonias undulatus) downstream. These predators actively feed on mummichogs from this marsh when available (Tupper and Able, 2000; Nemerson, 2001) yet, for the most part, are unable to enter this shallow creek section. Similar observations were made at Strawberry Marsh during the current study. During low tide mummichogs were found to aggregate in the shallow creeks and pools in groups consisting of up to tens of thousands of individuals (pers. obs.). While potential predators such as striped bass and

Atlantic tomcod (Microgadus tomcod) are present in this region of the MRE for

- 47 -

extended periods (Hanson and Courtenay, 1995), none were ever observed or captured in the marsh (pers. obs.) perhaps because they were unable to access

Strawberry Marsh. These results and those of previous studies suggest that mummichogs remain in areas of suitable habitat and these areas are defined, at least in part, by the presence of appropriate vegetation for feeding, reproduction, and predator avoidance.

With regard to movement, Gibson (1988) describes intertidal fishes as being either residents or visitors, with mummichogs being classified as the latter as many studies have noted adult mummichogs retreat from the intertidal zone as the tide ebbs and take refuge in deep intertidal pools and subtidal creeks

(Butner and Brattstrom, 1960; Weisberg et al., 1981; Kneib, 1984, 1987; Kneib and Wagner, 1994). These studies, however, were conducted in salt marshes rather than the upper river portion of the estuary as was the case in the current study. As such, some similarities and differences were noted. Mummichogs at

Strawberry Marsh were consistently found in shallow tide pools during low tide as was expected due to the similarity of this location to the habitats sampled in the previous studies. Conversely, mummichogs at all remaining locations were only captured in significant numbers just before, during, and immediately after high tide, as also noted by consultants during Cycle 2 Fish Surveys at sites

BKPM and UFC of the current study (Jacques Whitford Environmental Ltd,

2000b). These differences are most likely due to the difference in topography between marshes and the main estuary. Specifically, depressions allowing for the formation of tide pools were only present at Strawberry Marsh and no

- 48 -

intertidal creeks exist at the remaining sites sampled. As such, the only option for mummichogs retreating from the intertidal zone during ebb tide is to move into the main body of the estuary. Thus, previous observations of the low tide behaviour and position of mummichogs are of limited applicability in this system.

These findings, however, do not support the hypothesis that mummichogs are being advected great distances during the tidal cycle (Jacques

Whitford Environmental Ltd, 2000b). First, if the advection hypothesis were correct, it would have been expected that marked mummichogs would not be consistently recaptured at original marking sites as their location in the estuary at high tide would fluctuate due to the variance in tide height. One would expect those fish, or a large proportion of them, to be displaced some distance downstream. This was not the case during this study as marked mummichogs were consistently found within 200 m of their original point of release (Table 5).

Second, had these fish been advected downstream with the falling tide, large proportions of fish marked at upstream sites (e.g. BKPM and/or UFC) would have been expected to be captured by seining or minnow traps at downstream sites (SM and/or CH) during low tide. No such proportions of mobile fish were captured. For instance, Strawberry Marsh was sampled exclusively at low tide; therefore, the proportion of mobile fish recaptured at this site would have been expected to be much higher than noted (Table 5).

Lastly, the absence of mummichogs from most sites in the MRE during low tide periods and their consistent reappearance at the same sites during rising tides also discredits the concept of significant tidal displacement.

- 49 -

Following the advection hypothesis, one would be forced to explain the disappearance and subsequent reappearance as being due to mummichogs returning to the marking sites after being displaced by tidal currents. However, it would not seem energetically efficient for small-bodied fish to migrate great distances (i.e. 600-3600 m) to these sites twice daily as they would surely be exposed to other suitable habitat along the way. Instead, this observed reappearance suggests mummichogs in the MRE may be maintaining position offshore during low tides while avoiding the fast flow of the river channel.

Many studies on various species of fish have demonstrated distinct home ranges and the role of homing mechanisms to maintain these ranges

(summarized by Gerking, 1958; Gibson 1986). For mummichogs, this may be accomplished by retreating from shore yet avoiding the river channel; moving to the river channel but seeking refuge from the current along the substrate in a manner similar to eels (Davidson, 1949); or some combination of the two behaviours. However, attempts to test the prediction that mummichogs in the

MRE were moving offshore with the falling tide while avoiding the river channel were fruitless as mummichogs could not be captured in any abundance during low tide. The lone exception was Strawberry Marsh, which provided no insight into this question as it was isolated from the higher water velocity of the main river channel. Minnow traps set at depths of one to three meters up to ~50 m offshore during low tides at the remaining sites failed to capture mummichogs

(unpublished data). During these low tide trapping periods, repeated attempts were made to capture mummichogs along the entire study area (~10 km) by

- 50 -

seine. However, mummichogs moving as little as 5 m offshore during low tides would be residing in areas too deep for beach seining at most sites along the

MRE. Morrison et al. (2002), using beach seines and outrigger trawls in an estuarine mudflat, noted a number of fish species concentrated in areas adjacent to deep channels. Thus, the inability to capture mummichogs by seine in shallow areas and minnow traps in deeper water at low tide suggests they may have moved to deeper areas but are not susceptible to minnow traps, perhaps because they were higher in the water column. Increased fishing effort with alternative gear types will be needed to fully address the question of mummichog location during low tide.

While these results demonstrate the need for further work on the location of mummichogs during low tide, the most important matter of this study is the duration of exposure of mummichogs to pulp mill effluent. In the MRE, the 1% effluent plume for the BKPM has been shown to move 1400 m upstream and

1700 m downstream from the outfall with flood and ebb tides, respectively

(Jacques Whitford Environmental Limited, 2004; Fig. 3). Whether the fish remain near the mill outfall or are advected with the tides, it is reasonable to assume mummichogs in the MRE would be constantly exposed to >1% pulp mill effluent throughout the tidal cycle. As previously mentioned, 96.6% of mummichogs recaptured during this study were found within 200 m of the point of original release. Therefore, if mummichogs in the pulp mill receiving environment (e.g., at the BKPM outfall) are being advected with tides and homing back to that site, they would be continuously exposed as they and the

- 51 -

effluent would be moving together with the tide. In the alternative and more probable scenario, mummichogs are moving just a few meters offshore into deeper water during falling tides and remain within the effluent plume as they are not leaving the outfall site. Thus, mummichogs in the MRE are consistently found in very high abundance in areas considered small relative to the size of the 1% BKPM effluent plume, and therefore, with regard to mobility, are a suitable sentinel species for use in environmental monitoring in this area.

The results of this study, like those of Lotrich (1975), Sweeney et al.

(1998), and Teo and Able (2003) clearly demonstrate that mummichogs display distinct site fidelity. Excluding fluctuations noted in the distances moved by individual fish in this study and others, mummichogs occurring in greatly varying habitats throughout their range of distribution all exhibit similar site-specific behaviour, whereby very large proportions of the populations examined reside in discrete areas. As such, it would appear populations of F. heteroclitus sampled at any site should accurately reflect local environmental conditions and, therefore, be useful in environmental monitoring programs.

3.6 Acknowledgements

This project was funded by the National and Atlantic Region Environmental

Effects Monitoring offices of Environment Canada, Department of Fisheries and

Oceans Canada (DFO Science Subvention Grant awarded to RAC), and UPM-

Kymmene Miramichi Inc. (NSERC Industrial Post-graduate Scholarship awarded to MAS). We thank P. Riebel (UPM-Kymmene Miramichi Inc.) and

- 52 -

Kelly Munkittrick (UNB Saint John) for logistic support and input on study design. We also thank M. Stephenson and M. Murdoch (Jacques Whitford

Environmental Ltd.) for their valued input. Additional field and in-kind support were provided by the Miramichi River Environmental Assessment Committee

(MREAC) and Department of Fisheries and Oceans Canada.

3.7 Literature Cited

Butner A and Brattstrom BH. 1960. Local movement in Menidia and Fundulus.

Copeia 2: 139-141.

Chiasson AG. 1995. The Miramichi bay and estuary: an overview, p. 11-27. In

Chadwick M (ed.), Water, Science, and the Public: The Miramichi

Ecosystem. Canadian Special Publication of Fisheries and Aquatic Sciences

123.

Courtenay SC and Couillard C. 1998. The mummichog as a sentinel species

for pulp and paper mill EEM surveys in the Atlantic coastal environment,

p. 50-55. In: Courtenay SC, Parker WR and Rawn GP (ed.), Proceedings of

a workshop to assess alternatives to the fish survey component of the

Environmental Effects Monitoring Program for Canadian pulp and paper

mills. Canadian Technical Report of Fisheries and Aquatic Sciences 2233.

Courtenay SC, Munkittrick KR, Dupuis HMC, Parker WR, and Boyd J. 2002.

- 53 -

Quantifying impacts of pulp mill effluent on fish in Canadian marine and

estuarine environments: problems and progress. Water Quality Research

Journal of Canada 37: 79-99.

Davidson DM. 1949. Salmon and eel movement in constant circular current.

Journal of the Fisheries Research Board of Canada 7: 432-448.

Environment Canada. 1998. Pulp and paper technical guidance for aquatic

environmental effects monitoring. National EEM Office, Environment

Canada. EEM/1998/1.

Gerking SD. 1958. The restricted movement of fish populations. Biological

Reviews of the Cambridge Philosophical Society 34: 221-242.

Gibbons WN, Munkittrick KR, McMaster ME, and Taylor WD. 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. Environmental Toxicology and Chemistry 17: 2227-2237.

Gibbons WN, Munkittrick KR, McMaster ME, and Taylor WD. 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.

- 54 -

Environmental Toxicology and Chemistry 17: 2238-2245.

Gibson RN. 1986. Intertidal teleosts: life in a fluctuating environment, p. 388-

408. In: Pitcher TJ (ed.), The behaviour of teleost fishes. Croom Helm Ltd.:

Beckenham.

Gibson RN. 1988. Patterns of movement in intertidal fishes, p. 55-63. In:

Chelazzi G Vannini M (ed.), Behavioural adaptation to intertidal life. Plenum

Press: New York.

Hanson JM and Courtenay SC. 1995. Seasonal abundance and distribution of

fishes in the Miramichi Estuary. Canadian Special Publication of Fisheries

Aquatic Sciences 123: 141-160.

Heckert M, Fuselier L, and Horwitz RJ. 1999. Habitat use by Fundulus

heteroclitus and F. diaphanus and effects of species co-occurrence. Journal

of the Pennsylvania Academy of Science 73: 22-26.

Jacques Whitford Environmental Limited. 2000a. Project No. JWEL 11153.

Final Report to: AV Cell Inc. on: Second Cycle Aquatic Environmental

Effects Monitoring Dissolving Grade Pulp Mill PP1112. JWEL, 711

Woodstock Road, Fredericton, NB.

- 55 -

Jacques Whitford Environmental Limited. 2000b Project No. JWEL 89584. Final

Report to: REPAP New Brunswick Inc. on: Second Cycle Aquatic

Environmental Effects Monitoring Study Kraft Mill (PP1112). JWEL, 711

Woodstock Road, Fredericton, NB.

Jacques Whitford Environmental Limited. 2004. Cycle 3 EEM report for the

UPM-Miramichi Inc. Kraft/Paper mill (PP1142), Miramichi, NB. Project No.

NBF14126 report to UPM-Miramichi Inc. JWEL, 711 Woodstock Road,

Fredericton, NB, E3B 5N8.

Kneib RT. 1984. Patterns in the utilization of the intertidal salt marsh by larvae

and juveniles of Fundulus heteroclitus (Linnaeus) and Fundulus luciae

(Baird). Journal of Experimental Marine Biology and Ecology 83: 41-51.

Kneib RT. 1987. Predation risk and use of intertidal habitats by young fishes

and shrimp. Ecology 68: 379-386.

Kneib R and Wagner SL. 1994. Nekton use of vegetated marsh habitats at

different stages of tidal inundation. Marine Ecology Progress Series 106:

227-238.

Lafleur C, Pettigrew B, St-Hilaire A, Booth D, and Chadwick M. 1995. Seasonal

and short term variations in the estuarine structure of the Miramichi, p. 45-73.

- 56 -

In Chadwick M (ed.), Water, Science, and the Public: The Miramichi

Ecosystem. Canadian Special Publication of Fisheries and Aquatic Sciences

123.

Lotrich VA. 1975. Summer home range and movements of Fundulus

heteroclitus (Pisces: Cyprinodontidae) in a tidal creek. Ecology 56: 191-198.

Morrison MA, Francis MP, Hartill BW, and Parkinson DM. 2002. Diurnal and

tidal variation in the abundance of the fish fauna of a temperate tidal mudflat.

Estuarine, Coastal and Shelf Science 54: 793-807.

Munkittrick KR, Megraw SR, Colodey A, Luce S, Courtenay S, Paine M, Servos

M, Spafford M, Langlois C, Martel P, and Levings C. 1997. Fish survey

expert working group final report. Recommendations from cycle 1 review.

Environment Canada, EEM/1997/6.

Murphy S. 1991. The ecology of estuarine fishes in southern Maine high salt

marshes; access corridors and movement patterns. M.Sc. Thesis, University

of Massachusetts.

Natech Environmental Services Inc. 2002. Tracer and modeling study for UPM-

Miramichi Kraft Mill.

Nemerson DM. 2001. Trophic dynamics and habitat ecology of the dominant

- 57 -

fish of Delaware Bay (USA) marsh creeks. Ph.D. Dissertation, Rutgers

University, New Brunswick, New Jersey.

Rashid MA and Reinson GE. 1979. Organic matter in surficial sediments of the

Miramichi Estuary, New Brunswick, Canada. Estuarine and Coastal Marine

Science 8: 23-36.

Reinson GE. 1977. Tidal-current control of submarine morphology at the mouth

of the Miramichi estuary, New Brunswick. Canadian Journal of Earth

Sciences 14: 2524-2532.

Skinner MA, Courtenay SC, Parker WR, and Curry RA. 2004. Evaluation of

techniques for the marking of mummichogs (Fundulus heteroclitus) with

emphasis on Visible Implant Elastomer (VIE). In preparation.

Sweeney J, Deegan L, and Garritt R. 1998. Population size and site fidelity of

Fundulus heteroclitus in a macrotidal saltmarsh creek. Biological Bulletin

195: 238-239.

Taylor MH, Leach GJ, DiMichelle L, Levitan WM, and Jacob WF. 1979. Lunar

spawning cycle in the mummichog Fundulus heteroclitus (Pisces:

Cyprinodontidae). Copeia 2: 291-297.

- 58 -

Taylor MH and DiMichelle L. 1983. Spawning site utilization in a Delaware

population of Fundulus heteroclitus (Pisces: Cyprinodontidae). Copeia 3:

719-725.

Teo SLH and Able KW. 2003. Habitat use and movement of the mummichog

(Fundulus heteroclitus) in a restored marsh. Estuaries 26: 720-730.

Tupper M and Able KW. 2000. Movements and food habits of striped bass

(Morone saxatilis) in Delaware Bay (USA) salt marshes: Comparison of a

restored and a reference marsh. Marine Biology 137: 1049-1058.

Vince S, Valiela I, Backus N, and Teal JM. 1976. Predation by the salt marsh

killifish Fundulus heteroclitus (L.) in relation to prey size and habitat

structure: consequences for prey distribution and abundance. Journal of

Experimental Marine Biology and Ecology 23: 255-266.

Weisberg SB, Whalen R, and Lotrich VA. 1981. Tidal and diurnal influence on

food composition of a salt marsh killifish Fundulus heteroclitus. Marine

Biology 61: 243-246.

- 59 -

Table 3 - Number of mummichogs marked by site from May 23 – August 30, 2002. Mean length of fish marked by site

presented with length range of marked fish in brackets (CH=Chatham Head, BKPM= Bleached Kraft Pulp Mill,

SM=Strawberry Marsh, UFC=Upstream Flett Cove).

Date Site Total CH BKPM SM UFC -60- Marked Released Marked Released Marked Released Marked Released Marked Released May 23-29 12 11 47 47 132 122 126 126 317 306 June 10-14 72 71 152 144 558 523 358 339 1140 1077 June 24-28 184 175 83 79 324 308 152 144 743 706 July 8-12 164 156 68 64 312 297 87 82 631 599 July 22-26 20 19 46 44 236 224 104 99 406 386 Aug 12-16 55 53 161 153 201 181 28 27 445 414 Aug 26-30 75 71 221 210 204 194 169 160 669 635 Total 582 556 778 741 1967 1849 1024 977 4351 4123 Mean length (mm) 66.6 (34-121) 68.7 (38-115) 55.6 (23-120) 64.6 (39-118) 61.9 (23-121)

- 60 -

Table 4 – Numbers of mummichogs retained and survived 48 hrs after marking from May 23 – August 30, 2002. Percent

survived by site and date and percent mark retention by site also included (a=deaths due to holding conditions,

CH=Chatham Head, BKPM= Bleached Kraft Pulp Mill, SM=Strawberry Marsh, UFC=Upstream Flett Cove).

Date Site Date total CH BKPM SM UFC Retained Survived Retained Survived Retained Survived Retained Survived Retained Survived Percent -61- May 23-29 1 1 0 0 10 9 0 0 11 10 90.9 June 10-14 1 1 8 8 35 24a 19 17 63 50 79.4 June 24-28 9 9 4 4 16 16 8 8 37 37 100 July 8-12 8 8 4 4 15 10 5 5 32 27 84.4 July 22-26 1 1 2 2 12 10a 5 5 20 18 90 Aug 12-16 2 1a 8 7a 20 3a 1 0a 31 11 35.5 Aug 26-30 4 4 11 11 10 10 9 9 34 34 100 Site total 26 25 37 36 118 82 47 44 Percent survived 96.1 97.3 69.5 93.6 Percent mark retention 100 100 100 100

- 61 -

Table 5 - Number of mummichogs recaptured by site from May 27, 2002 –

November 27, 2003 (CH=Chatham Head, BKPM= Bleached Kraft Pulp Mill,

SM=Strawberry Marsh, UFC=Upstream Flett Cove, GW=Groundwood Mill,

LB=Little Bog, MC=McKay Cove, GB=Green Bridge, U-Haul=U-Haul; see Figure

3). Numbers of fish marked by site are shown in Table 1. Values in bold indicate fish that were recaptured multiple times and have been included in totals once. Values with superscripts indicate numbers and recapture sites for fish recaptured >200m from release point at sites other than the four original marking sites. Column recapture totals refer to all fish marked at that site

(including those recaptured at other sites).

- 62 -

Site Date Total a (CH) b (BKPM) c (SM) d (UFC) May 27-29 0 0 0 1 d June 10-14 0 4 b 14 c 12d June 24-28 9a+1aGB 9b+1c 125c+1 cGW+1a 59d+1dGW+1c July 8-12 17a 18b+1d 35c 22d July 22-26 3a+1aGB 1b+1a+1bUH 30c 20d Aug 12-16 3a 14b+2bUH+1c 16c 1dGW Aug 26-30 2a 11b 79c+1c 6d Sep-30 0 0 5c+1b 0

2002 Oct-29 0 0 1c 0 Nov-26 0 0 0 0 Fish recaptured 38 61 309 123 531 from each site Percent recaptured 6.8 8.2 16.7 12.6 12.9 from each site Fish moved 4 4 4 3 15 >200 m May 12-16 0 0 16c+6c+1b 0 May 22-23 0 4b+1b+1c 0 3d May 26-30 0 1 bMC 4c+2c 1d June 16-20 3a 12b+1b+1bLB 5c+1c+1b 1d June 30-July4 9a 1b 12c+2c 2d July 14-18 1a 2b+1c+2b 6c+3c+1b 1d July 28-Aug1 0 3b 13c+4c 0 Aug 19-21 1a 1b+1b 2c+1c 1d Oct-02 0 0 1c+2c 0 2003 Oct 29-30 0 0 1c 0 Nov 26-27 0 0 0 0 Fish recaptured 12 26 62 8 108 from each site Percent recaptured 2.2 3.5 3.4 0.82 2.6 from each site Fish moved 0 5 2 0 7 >200 m Overall recapture 50 87 371 131 639 total Fish moved >200 m 4 9 6 3 22 Total % Recaptured 9.2 11.6 20.1 13.4 15.5

- 63 -

Table 6 – Recapture summary of marked mummichogs found to have moved from May 27, 2002 – November 26, 2003 (CH=Chatham Head,

BKPM=Bleached Kraft Pulp Mill, SM=Strawberry Marsh, UFC=Upstream Flett

Cove, GW=Groundwood Mill, LB=Little Bog, MC=McKay Cove, GB=Green

Bridge).

- 64 -

Time from mark Marking Recap Marking Distance Length Site of Recap Sex to recapture Date Date Site moved (m) (mm) (days) 14-June 24-June CH SM 1400 64 unknown 10 12-June 25-June SM BKPM 2250 60 unknown 13 28-May 26-June SM GW 2300 95 unknown 29 23-May 26-June UFC GW 2000 56 unknown 34 14-June 26-June CH GB 600 71 unknown 12 13-June 27-June SM UFC 3600 57 unknown 14 23-May 10-July UFC BKPM 1900 88 M 48 unknown 23-July CH GB 600 n/a unknown n/a

-65- 12-June 24-July BKPM U-haul 1000 83 F 42 14-June 25-July CH BKPM 3500 81 M 41 13-June 13-Aug SM BKPM 2250 77 F 61 09-July 14-Aug BKPM U-haul 1000 88 F 35 25-July 14-Aug BKPM U-haul 1000 72 F 19 23-July 14-Aug UFC GW 2000 n/a M 21 29-Aug 30-Sep BKPM SM 2250 49 M 32 Mean 1843.3 72.4 29 15-Aug-02 16-May-03 BKPM SM 2250 78 M 275 28-May-02 22-May-03 SM BKPM 2250 86 M 358 12-June-02 30-May-03 BKPM MC 2500 62 M 352 29-Aug-02 18-June-03 BKPM SM 2250 96 M 293 2003 15-Aug-02 19-June-03 BKPM LB 3100 88 F 302 12-June-02 15-July-03 SM BKPM 2250 101 F 371 09-July-02 15-July-03 BKPM SM 2250 80 M 344 Mean 2407 84.4 328 Overall Mean 2023 76.6 129

- 65 -

Figure 3 – Map of the upper Miramichi River Estuary (MRE) indicating mummichog sampling sites for 2002 (CH=Chatham Head, BKPM= Bleached

Kraft Pulp Mill, SM=Strawberry Marsh, UFC=Upstream Flett Cove). Triangles represent additional 2002 recapture sites. Diamonds and circles represent expanded 2003 recapture sites (sampled from April – June, 2003). Black shading indicates approximate 1% concentration area of BKPM effluent plume

(Natech Environmental Services Inc., 2002).

- 66 -

SM CH

BKPM

UFC N

2 km

- 67 -

Figure 4 – Movement directions of Fundulus heteroclitus marked at (a)

Strawberry March (SM) (b) Chatham Head (CH), (c) Bleached Kraft Pulp Mill

(BKPM), and (d) Upstream Flett’s Cove (UFC) with number of fish moved indicated. Additional recovery sites were: GW=Groundwood Mill, LB=Little Bog,

MC=McKay Cove, GB=Green Bridge and U-Haul=U-Haul.

- 68 -

(a)

4 CH SM UHaul 1 BKPM GW MC

1 N UFC

LB 2 km

- 69 -

(b)

2 GB 1 CH SM UHaul

BKPM 1 GW MC

UFC N

2 km

- 70 -

(c)

CH SM UHaul 4 1 BKPM GW MC 3 UFC N 1

2 km

- 71 -

(d)

CH SM UHaul

BKPM GW MC 2 1 N UFC

LB 2 km

- 72 -

4 USE OF STABLE ISOTOPES TO EXAMINE THE SITE FIDELITY OF MUMMICHOGS (FUNDULUS HETEROCLITUS) IN AN ATLANTIC CANADIAN ESTUARY RECEIVING MULTIPLE ANTHROPOGENIC INFLUENCES

4.1 Abstract

The goal of this study was to evaluate the usage of SIA as a method to determine the site-specificity of organisms on a relatively small spatial scale (~

10 km). Samples were collected from a section of the upper Miramichi River estuary (MRE), New Brunswick influenced by multiple anthropogenic impacts: two pulp mills and three municipal wastewater facilities. White muscle and bone from mummichogs sampled at 9 sites along the upper MRE (n = 198) had overall mean ratios of -21.03 + 1.45 ‰ (SD) δ13C and 11.37 + 1.02 ‰ (SD) δ15N.

Mean δ13C and δ15N ratios were significantly different among sites but not between sexes within sites. Mean δ13C increased progressively in a downstream direction while two distinct δ15N groups representing the northern and southern shores were apparent. These differences appear to be related to the influence of anthropogenic inputs to the system, specifically polycyclic

*Manuscript to be submitted for publication to Marine Ecology Progress Series by Skinner MA, Courtenay SC, Dube M, Parker WR, and Curry RA.

- 73 -

aromatic hydrocarbons and wastewater treatment influences, thus demonstrating the utility of SIA as a method to determine the site-specificity of organisms on a relatively small spatial scale. The scarcity of statistical outliers

(2%) during examination of isotopic ratios among sites supports the results of a previous mark-recapture study that showed very few mummichogs (3.4%) in the upper MRE move more than 200m.

4.2 Introduction

Previous studies have demonstrated the utility of stable isotope analysis

(SIA) to establish movement and migration patterns in a variety of organisms across great distances (Fry, 1983; Hobson and Wassenaar, 1997; Hobson,

1999). SIA is a helpful tool for such studies as it exploits differences in the isotopic composition of target organisms. These differences are due to the preferential assimilation of one isotope of a particular element over another during chemical fixation by primary producers in an organism’s food-web

(Peterson and Fry, 1987). For example, ratios of 13C vs. 12C are useful to researchers because they are habitat specific and conserved during trophic transfer (Peterson and Fry, 1987). Therefore, it can be determined whether a particular organism is a resident of a site or has recently moved into the area by examining the degree of difference between the carbon isotope ratios of the organisms’ tissues and their surroundings (DeNiro and Epstein, 1978; Michener and Schell, 1994). Additionally, ratios of 15N vs. 14N are useful for examining trophic relationships because metabolic fractionations in consumers

- 74 -

discriminate against the lighter isotope as this element is passed on to higher trophic levels (DeNiro and Epstein, 1981), resulting in higher concentrations of

15N in predators than their prey. The extra information provided by examining multiple isotopes simultaneously allows for further discrimination of differences among samples (Peterson et al., 1985).

While SIA has been useful for addressing questions related to large-scale movements of animals, there have been few studies which have used SIA to examine small-scale animal movements (Hobson, 1999). Hughes and Scherr

(1983) demonstrated intraspecific variation in the δ13C composition of estuarine consumers sampled ca. 16 km apart was related to differences in plant carbon sources between sites. Despite this finding, difficulty still remains in conducting similar studies in areas with less heterogeneous isotopic signatures (i.e. small sections of rivers or upper estuaries) as it is highly probable that food sources may show little isotopic variation among sites separated by short distances (<10 km). This homogeneity, however, may be eliminated in systems receiving anthropogenic inputs (Macko and Ostrom, 1994). Several studies have shown that sewage and wastewater can influence nitrogen isotopic signatures of biota

(Rau et al.,1981; Spies et al., 1989; Hansson et al., 1997; McClelland et al.,

1997; Griffin and Valiela, 2001; Wayland and Hobson, 2001; deBruyn and

Rasmussen, 2002; deBruyn et al., 2003; Fry et al., 2003, Gaston and Suthers,

2004; Savage and Elmgren, 2004; Steffy and Kilham, 2004) and similar responses have also been noted for carbon and nitrogen signatures of organisms residing in areas receiving pulp mill effluent (Wassenaar and Culp,

- 75 -

1996; Wayland and Hobson, 2001). Examples specific to mummichogs

(Fundulus heteroclitus) have also been recorded. Griffin and Valiela (2001) found mummichogs from rivers separated by 3-5 km and subjected to differing levels of nitrogen loading had significantly different δ15N signatures. Hughes et al. (2000) artificially enriched nitrogen levels in a New England estuary and demonstrated mummichogs separated by as little as 1400 m had distinct δ15N signatures. Therefore, potential exists to allow the examination of small-scale movements of organisms residing in areas exposed to such anthropogenic inputs as they may cause significant isotopic variation on a scale of only a few hundred meters (Hobson, 1999).

A recent mark-recapture study (Skinner et al., submitted) demonstrated mummichogs in the upper Miramichi River estuary (MRE), New Brunswick,

Canada displayed distinct site fidelity with 96.6% (n=639) of all recaptures occurring within 200 m of point of release over a 19 month period. This finding is in accordance with those of previous studies that reported limited movements by this species (Lotrich, 1975; Murphy, 1991; Sweeney et al., 1998; Teo and

Able, 2003). The goal of the current study was to evaluate the use of SIA as a method to determine the site-specificity of organisms on a relatively small spatial scale (~10 km). This study examined sites located upstream and downstream of inputs from two pulp mills and three municipal sewage facilities in the MRE. The working hypothesis was that because mummichogs in this region have been shown to display distinct site-fidelity, fish sampled near areas

- 76 -

receiving different anthropogenic inputs should show distinct carbon and nitrogen isotopic signatures.

4.3 Materials and Methods

4.3.1 Sampling sites

As part of a larger project examining the potential use of mummichogs as an aquatic monitoring species in estuarine and coastal areas (2002-2003), fish samples were collected for SIA from the upper MRE. The MRE is fed by two major rivers, the Northwest Miramichi and Southwest Miramichi, with respective drainage areas of 3900 km2 and 7700 km2 (Chiasson, 1995); is subject to mixed semi-diurnal tides with a tidal range between 0 and 2 m (Reinson, 1977); and is exposed to a freshwater inflow of 5-10% of the volume of incoming tide water

(Chiasson, 1995). Salt water intrudes into the upper estuary as a salt wedge, the upper extent of which varies in position with lunar tide and freshwater discharge, being downstream of the study area in early spring and far upstream of the study area in summer.

Sampling sites were chosen by proximity to known sources of anthropogenic inputs: a bleached kraft pulp mill (BKPM), a groundwood pulp mill (GW), and three municipal sewage facilities (Sew1-3; Fig. 5). Sewage 1 utilizes an aerated lagoon and bacterial system, Sewage 2 discharges untreated waste (Mr. Cecil Bowes, Department of Public Works, City of Miramichi, pers. com.) and Sewage 3 utilizes a settling lagoon with UV light treatment (Mr. John

Simonson, Eel Ground First Nation, pers. com.).

- 77 -

4.3.2 Sample collections, processing and isotopic analyses

In January 2001, two litre samples of final effluent were collected from

BKPM (n = 2), GW (n = 1), and Sew1 (n = 1) and water samples from the BKPM receiving environment (n = 1) were also collected. All were frozen immediately and shipped to the National Hydrology Research Institute (NHRI; Environment

Canada, Saskatoon, SK) for processing and analyses. Samples were freeze dried and ground to a fine powder with a mortar and pestle and inorganic carbonates were removed by re-dissolving freeze dried samples in 50 ml of 1% vol:vol HCl. After all inorganic carbonates had been removed, samples were re-frozen and lyophilized. Using a GV Instruments Isoprime™ and Eurovector

EA™ (GV Instruments Ltd., Manchester, UK), stable carbon and nitrogen isotope ratios were determined by conventional elemental analyzer flash combustion to CO2 and N2 and measurement by continuous-flow isotope ratio mass spectrometry. Isotopic ratios are expressed in the delta notation (δ) normalized to the ratio of the sample to Pee Dee Belemnite for 13C (Craig, 1957) and atmospheric nitrogen for 15N (Mariotti, 1983) in parts per thousand. These ratios are calculated using the following equation:

3 δX = [(Rsample/Rstandard) -1] x 10

where X is the isotope of interest (e.g. 13C) and R is the ratio of this isotope relative to its lighter isotope (e.g. 13C/12C) (Peterson and Fry, 1987).

- 78 -

Repeatability of internal homogenized working standards was better than ± 0.13 and ± 0.15 ‰ (SD) for δ13C and δ15N, respectively.

From July 29-31 and August 20-22, 2003, a minimum of 10 female and

10 male mummichogs (60-120 mm TL) were collected by beach seine (25 m in length; 1.5 m in height; 5 mm mesh size) and minnow traps (baited with cat food; 6.35 mm mesh size) from each of 9 sites along a reach of ca. 10 km (Fig.

5). Mummichogs chosen for analysis were euthanized with a quick blow to the head and frozen at -20oC.

White muscle and bone were dissected from the right flank of mummichogs, oven dried in glass scintillation vials at 50oC for 48 h and ground to a fine powder using a mortar and pestle. Sub-samples of 200 μg were analyzed by the Stable Isotopes in Nature Laboratory (SINLAB; University of

New Brunswick, Fredericton, NB), with a continuous-flow isotope-ratio mass spectrometer (Finnigan Mat Delta Plus, Thermofinnigan, Bremen, Germany) equipped with a ThermoQuest elemental analyzer (Carlo Erba NC2500, Italy).

Isotopic standards CH6, CH7, N1, and N2 were used for correction of carbon and nitrogen values. Runs of an elemental standard, Acetanilide (n = 34), resulted in a mean δ13C value of -33.57 + 0.06 ‰ (standard deviation, SD) and a mean δ15N value of -3.12 + 0.12 ‰ (SD). Precision of the mass spectrometer over time was evaluated using repeated analyses of select samples (n = 28) which resulted in a mean SD of 0.069 for δ13C (range = 0.003 – 0.360) and a mean SD of 0.098 for δ15N (range = 8.21 x 10-5 – 0.481).

- 79 -

4.3.3 Data Analysis

As mostly single samples of effluent and BKPM receiving water were collected, these data were not incorporated in statistical analyses. Statistical analyses of δ13C and δ15N data were first conducted separately to examine for differences in each isotope among sites. While comparing multivariate data using separate univariate tests rather than a single multivariate method (i.e.

MANOVA or MANCOVA) increases the possibility of committing a type I error

(Sokal and Rohlf,1995), this potential effect was minimized by judging the results of statistical analyses for each element individually to be significant at α

= 0.01 (Zar, 1999). In addition, mean δ13C and δ15N data were analyzed simultaneously with a Bray-Curtis similarity matrix followed by a group-average link cluster analysis to distinguish similarity in isotopic ratios among sites (Clarke and Warwick, 1994) using PRIMER 5 © (ver. 5.2.2, Primer-E Ltd., Plymouth,

U.K.). Isotopic data were also examined for outliers within sites using a two- tailed Grubbs Test in an effort to identify mobile fish (α = 0.05; Sokal and Rohlf,

1995).

Examination of δ13C and δ15N data revealed slight non-normality, minor heteroscedasticity, and an influence of fish length. As such, analyses of covariance (ANCOVA) were initially chosen for statistical analysis with site as factor and total fish length as covariate for each as ANCOVA is robust against slight deviations from the assumptions of normality and homogeneity of variance

(Zar, 1999). In analyses for each element, however, the length*isotopic ratio relationships were determined to be significantly different resulting in significant

- 80 -

interactions. After graphing and determining these interactions were not biologically interpretable, a sub-sample of lengths (70-90 mm; minimum 9 per site) was selected to minimize the effect of length and subsequently analyzed using analyses of variance (ANOVA) to test for significant differences between sexes for each element. Separate ANOVAs were further conducted to test for significant differences among sites (sexes pooled) for each element.

Comparison of mean δ13C values (dependent variable) vs. relative sample site position expressed as river distance (independent variable) was performed using simple linear regression (α = 0.05).

Grubbs Test (Sokal and Rohlf, 1995) was selected for outlier determination using the following equation:

(Y1 – Ymean)/s

where Y1 is the suspected outlier, Ymean is the sample mean, and s is the sample standard deviation. Samples with critical values greater than 2.644 (n =

24, p>0.05) were judged as being statistical outliers (Sokal and Rohlf, 1995).

ANOVA, ANCOVA, and simple linear regression were completed using

SYSTAT® (ver. 10.2, SPSS Inc., Chicago, Il., U.S.A.) and Grubbs test was completed by hand.

- 81 -

4.4 Results

The δ13C values obtained for water sampled from the receiving environment of BKPM, BKPM effluent and GW effluent grouped separately from all other samples (Table 7 and Fig. 6a). This same trend was noted for 15N values, however, with BKPM receiving environment water having no detectable nitrogen in the sample (Table 7). The ratios obtained from the municipal wastewater facility near Strawberry Marsh (Sew 1) were enriched in 13C and 15N relative to other liquid samples (Table 7 and Fig. 6a).

Mummichog muscle and bone (n = 198) had overall mean ratios of -21.03

+ 1.45 ‰ (SD) δ13C and 11.37 + 1.02 ‰ (SD) δ15N with distinct isotopic signatures found among some sites (Table 7; Fig. 6a and b). Mean δ13C ratios were significantly different among sites (ANOVA: F8,141= 27.77, p < 0.001; Fig.

6a) but not between sexes within sites (Tukey’s Honestly Significant Test, p =

1.00, n = 18; Table 7). Values of δ13C also increased progressively in a downstream direction. This relationship was statistically significant (r2 = 0.639, p = 0.017, n = 8; Fig. 6a) when Strawberry Marsh (SM) was omitted from analysis as an outlier (studentized residual = -4.814).

No significant relationship was noted between mean δ15N ratios of mummichogs and location of sample site along the river (r2 = 32.2%, p = 0.111, n = 9; Fig. 6b); however, differences in δ15N ratios among sites were significant

(ANOVA: F8,141 = 36.88, p < 0.001; Fig. 6b) with two mainly distinct groups apparent. The first consisted of sites found along the northern shore of the

MRE (MC, BKPM, UH, and SM), while the second group mainly consisted of

- 82 -

sites from the southern shore (UFC, GW, CH, and LB) with WSB being the lone exception (Fig. 6b). Mean δ15N ratios were not significantly different between sexes within sites (Tukey’s Honestly Significant Test, p = 0.753, n = 18; Table

7).

With δ13C and δ15N ratios of mummichogs examined together, all sites were 94% similar (Fig. 7a); however, distinct groups emerged showing a north vs. south shore pattern (Fig. 5, 7a and b). This pattern was similar to that found for δ15N values (Fig. 6b) except that SM and WSB separated independently of either group due to their differing δ13C signatures.

When each site was examined independently, statistically significant outliers were noted (Grubbs Test, p<0.05; Fig. 8). Four fish captured at CH,

MC, SM, and GW with Grubbs values of 3.29, 3.25, 3.03, and 2.88, respectively, represented ~2% of the 198 individual mummichogs sampled. One fish sampled from each of CH and MC were more isotopically similar to fish from

SM; one GW fish was more isotopically similar to fish from UFC; and one SM fish was distinct from all others.

4.5 Discussion

The use of SIA as a tool to discern movements and migrations in organisms has generally been restricted to large-scale examinations over tens to thousands of kilometres (Fry, 1981, 1983; Hobson and Wassenaar, 1997) or small-scale analyses in systems with natural isotopic variation among food- webs (Hesslein et al., 1991; Hansson et al., 1997). Due to the influence of multiple anthropogenic influences, this study has shown mummichog white

- 83 -

muscle and bone sampled from sites along a 10 km stretch of the upper

Miramichi River have distinct isotopic signatures. This finding supports the results of a mark-recapture study that also found mummichogs in this region display discrete site-fidelity (Skinner et al., submitted).

Values of 13C for both BKPM and GW effluents and water sampled from the BKPM receiving environment had isotopic signatures very close to those previously reported for terrestrial-derived material (Table 7, Fig. 6a) (France,

1995). Values of 15N for BKPM and GW effluent samples were also comparable to values previously reported for algae sampled from holding lagoons for pulp mill effluent (Wayland and Hobson, 2001) (Table 7). Similarly, both 13C (Spies et al., 1989) and 15N values (Heaton, 1986) for sewage fell within reported ranges (Table 7). Finally, as predicted, an obvious difference was noted for 13C and 15N when the isotopic signatures of both pulp mill effluents were compared to that of the sewage effluent (Table 7).

Mean δ15N values for mummichogs sampled (Table 7) are in accordance with the majority of those previously reported (Kneib et al., 1980; Hughes and

Scherr, 1983; Peterson et al., 1986; Deegan and Garritt, 1997; McClelland et al., 1997; Hughes et al., 2000; Griffin and Valiela, 2001; Pastershank, 2001).

The most noticeable trend in the 15N data for the areas sampled is the clear separation between the northern and southern shores, with the exception of

WSB (Fig. 6b). While no distinct cause has been discerned for this differing result at WSB, it may be due to an unidentified anthropogenic influence as it is just downstream from the confluence of a freshwater stream (Fig. 5). The

- 84 -

contrasting δ15N signatures between the remaining upriver sites are most likely related to anthropogenic influences in this area, perhaps due to differences in the quality of sewage effluents discharged. Numerous studies have utilized

δ15N to trace the influence of sewage discharges (Rau et al.,1981; Spies et al.,

1989; McClelland et al., 1997; Wayland and Hobson, 2001; deBruyn and

Rasmussen, 2002; deBruyn et al., 2003; Fry et al., 2003; Gaston and Suthers,

2004; Savage and Elmgren, 2004; Steffy and Kilham, 2004) with greatly varying

15N values reported depending on the level of sewage treatment employed and various physicochemical processes (such as volatilization of ammonia) related to the physical dimensions of their respective holding lagoons (Heaton, 1986).

In the current study, untreated sewage is discharged from Sew2 while sewage from Sew1 and Sew3 are treated as previously described (Fig. 5). All three effluents are tidally dispersed up and downstream twice daily with their plumes remaining close to shore, thereby remaining relatively localized (Fig. 5).

Therefore, it appears these different nitrogenous inputs are creating distinct isotopic signatures in the fish sampled at exposed sites, with south shore sites having the greater 15N values.

Previous studies (Fry and Sherr, 1984; Finlay, 2001) encompassing a wide range of organisms have demonstrated predictable differences in δ13C values along the river continuum. A progressive downstream enrichment of values has been linked to differences in the food web base (Fry and Sherr,

1984), as the carbon assimilated by riverine and upper estuarine organisms is derived mostly from sources with characteristically light isotopic signatures such

- 85 -

13 as particulate organic carbon (POC) (-27 to -25 ‰ δ C), C3-dominated terrestrial plant material (ca. -26 ‰ δ13C), and river-estuarine phytoplankton (-30 to -24 ‰ δ13C), while lower estuarine and marine organisms have heavier δ13C

13 signatures derived mainly from marine phytoplankton (-24 to -18 ‰ δ C) and C4 marsh grasses (-14 to -12 ‰ δ13C). For the current study, δ13C results fell within the expected range of values based on previous studies of the isotopic composition of mummichogs (Kneib et al., 1980; Hughes and Scherr, 1983;

Peterson et al., 1986; Deegan and Garritt, 1997; McClelland et al., 1997;

Hughes et al., 2000; Griffin and Valiela, 2001; Pastershank, 2001). Ratios for the current study were most similar to those of Deegan and Garritt (1997), as samples for both studies were collected from the upper portions of estuaries.

While the current study was conducted in an area of upper estuary only ~10 km long, downstream trends in δ13C values were also noticed, with significant enrichment occurring in fish tissue collected from sites progressively downstream (Fig. 6a). It is important to note this trend is only significant when the mean δ13C value for Strawberry Marsh samples is omitted from analysis as an outlier.

The unexpectedly low δ13C values at Strawberry Marsh may be due to historic contamination of the location, which was the site of a creosote wood preservation operation for ca. eight decades (Mining Watch Canada and Sierra

Club of Canada, 2001). High levels of the polycyclic hydrocarbons (PAHs) napthalane, phenanthrene, anthracene, fluoranthene, and pyrene were reported from groundwater, soil, and sediment samples taken from Strawberry Marsh

- 86 -

from 1987-1989 (Environment Canada, 1993). Recent research has shown

δ13C values are conserved across many PAHs of creosote origin. At multiple sampling stations within two different exposure sites, Hammer et al. (1998) found the δ13C values for 12 of 16 PAH components of creosote-contaminated groundwater varied as little as 1 ‰. The five PAHs present at Strawberry Marsh

(Environment Canada, 1993) were identified at both sampling sites in the study by Hammer et al. (1998) and had mean δ13C values ranging from -23.62 to -

24.45 ‰. As the δ13C values for mummichogs from Strawberry Marsh overlap the values from these other creosote-contaminated sites and are depleted by ~

3 ‰ relative to all other sites sampled (Table 7; Fig. 6a), it would appear the

Strawberry Marsh results are due to the influence of creosote-based PAHs.

Based on previous studies involving SIA of pulp mill effluents and biota from their receiving environments, it was expected that mummichogs sampled from PME-exposed sites would have significantly depleted δ13C values

(Wassenaar and Culp, 1996). This was not the case as δ13C values in the MRE were fairly ambiguous, with BKPM and GW being isotopically similar to seven of nine sites and six of nine sites, respectively (Fig. 6a). Wayland and Hobson

(2001) concluded carbon isotopes were ineffective tracers of pulp mill-derived carbon in biota under certain circumstances, as also appears to be the case for this study. The δ13C values of mummichogs from this region of the upper MRE combined with the findings of Rashid and Reinson (1979) appear to suggest this lack of significant differences may be due to the influence of organic sediment rather than effluent directly. Rashid and Reinson (1979), via δ13C analysis of

- 87 -

surficial sediments, determined the carbon influence of the two Miramichi pulp mills (BKPM and GW) could be traced ~30 km downstream to the river mouth.

Terrestrial organic matter comprised 100% (-27.5 ‰ δ13C) of the sediments at the furthest upstream station sampled (Rashid and Reinson, 1979), which is 5 km downstream of site WSB in the current study. This percentage was found to decrease in relation to distance from the mills to a minimum of 51% at the mouth of the river. These organic sediments are mobilized by tidal mixing in the estuary, which causes them to be re-suspended in the bottom layers of the water column with salt wedge intrusion on the flood tide (Rashid and Reinson,

1979). Following spring, the freshwater discharge decreases causing the salt wedge in the upper MRE to intrude ca. >5 km upstream of BKPM on the

Northwest Miramichi River and most likely as far upstream on the Southwest

Miramichi River (Lafleur et al., 1995). Thus, sediments composed of ~100% terrestrial-derived carbon from the mill effluents are continually being re- suspended, distributed over a wide area, and potentially made available to various potential prey organisms for the omnivorous mummichog (Abraham,

1985), most likely causing the homogenous δ13C values found among sites. As neither surficial nor suspended sediment samples were collected during this study, further work would be required to test this hypothesis. This would involve sampling surficial/suspended sediments and biota, including mummichogs and potential prey, along a downstream gradient, beginning upstream from the furthest point of salt wedge intrusion. If pulp mill-derived terrestrial organic sediments are being transported with the intruding salt wedge and causing

- 88 -

homogenous δ13C values in biota, SIA of samples taken above its influence should be significantly different than those below. This type of analysis may also help determine why fish sampled from WSB are isotopically different from all other sites and why UFC δ13C values for mummichogs are more similar to

BKPM than GW, which is the opposite of what would be expected following

Rashid and Reinson (1979).

While δ13C values alone did not adequately reveal site-specific differences in mummichog tissues, they were useful in distinguishing fish from sites along a small spatial scale when used in conjunction with δ15N data (Fig.

7a and b). Some level of separation was expected based on known effects of anthropogenic influences on isotopic ratios (Macko and Ostrom, 1994;

Wassenaar and Culp, 1996; Wayland and Hobson, 2001). Sites SM and WSB separated independently at opposite ends of the δ13C and δ15N spectra while the remaining sites formed two distinct groups representing the northern and southern shores of the study area (Fig. 5, 7a and b). This clear separation in isotopic ratios is solely due to differences in the δ15N composition of the fish tissues described earlier (Fig. 6b). Thus, anthropogenic influences allowed the use of SIA to examine the extent of site-fidelity in mummichogs over a ~10 km scale, which would have been impossible to determine in a similar-sized system lacking such influences.

When δ13C and δ15N data for individual fish were examined among sites, four statistically significant outliers were detected (Fig. 8). As their isotopic signatures are not reflective of the signatures of the majority of fish captured at

- 89 -

the same site, this suggests these individual mummichogs are relatively more recent arrivals. Isotopic turnover of carbon and nitrogen occurs through the assimilation of these isotopes as an organism feeds (Peterson and Fry, 1987).

While no known studies have quantified specific isotopic turnover rates for mummichogs, simple growth dilution models can provide some insight. For example, it would take 300 d for an adult mummichog (weight = 6.00 g) with a mean daily growth rate of 0.02 g/d-1 (Skinner et al., in preparation) to have a

50% shift in its isotopic signature from one source to another. Relative to the four year maximum life expectancy of a mummichog (Abraham, 1985), this example highlights the prolonged time scale required for the isotopic signature of a mummichog to completely reflect that of a new site. It further demonstrates the majority of mummichogs sampled from each area have most likely resided in these areas since spawning while the individuals detected as being outliers are immigrants. These results corroborate those of a recent mark-recapture study (Skinner et al., submitted). The movement patterns of these few fish were similar in both studies, with fish marked at SM recaptured at CH, and fish from

UFC recaptured at GW during the mark-recapture study. Also, the four outlying mummichogs from the current study amount to 2% of those sampled compared to 3.4% of recaptured mummichogs classified as mobile during the mark- recapture study. Thus, the minor proportion of mobile individuals and their respective movement patterns determined by SIA in the current study support the results of the mark-recapture study (Skinner et al., submitted) and further illustrate the utility of SIA as a measure of movement by aquatic species.

- 90 -

This study has established the site-fidelity of fish along small spatial scales may be determined with some resolution using dual stable isotope analyses in areas receiving multiple anthropogenic influences. As was demonstrated, care must be taken in the interpretation of results from such SIA as the chemical composition of the surrounding environment may mask the true extent of an organism’s movement. It is therefore recommended that investigations of movements at fine scales using stable isotopes include analyses of a wide variety of materials (i.e. samples of potentially influential anthropogenic compounds, sediments/soils, food sources) while also utilizing as many elements as feasible (13C, 15N, 37Cl, 34S) to minimize this risk (Dube et al.,

2004). Also, based on the results of this study and others (Takai and

Sakomoto, 1999), stable isotopes coupled with conventional techniques for determination of movements (i.e. telemetry, mark-recapture, etc.) appear to provide a useful tool for inferring the site-fidelity of organisms.

4.6 Acknowledgements

This project was funded by the National and Atlantic Region Environmental

Effects Monitoring offices of Environment Canada, Department of Fisheries and

Oceans Canada (DFO Science Subvention Grant awarded to RAC), and UPM-

Kymmene Miramichi Inc. (NSERC Industrial Post-graduate Scholarship awarded to MAS). We thank P. Riebel (UPM-Kymmene Miramichi Inc.) and

Kelly Munkittrick (UNB Saint John) for logistic support and input on study design and T. Jardine for technical guidance and manuscript review. Additional field

- 91 -

and in-kind support were provided by the Miramichi River Environmental

Assessment Committee (MREAC) and Department of Fisheries and Oceans

Canada.

4.7 Literature Cited

Abraham BJ. 1985. Species profiles: life histories and environmental

requirements of coastal fishes and invertebrates - Mummichog and striped

killifish. US Fish and Wildlife Service Biological Report (US Army Corps of

Engineers) 82: 1-23.

Chiasson AG. 1995. The Miramichi bay and estuary: an overview, p. 11-27. In

Chadwick M (ed.), Water, Science, and the Public: The Miramichi

Ecosystem. Canadian Special Publication of Fisheries and Aquatic Sciences

123. viii + 300 p.

Clarke KR and Warwick RM. 1994. Change in Marine Communities: An

Approach to Statistical Analysis and Interpretation. Plymouth Marine

Laboratory: Plymouth, UK. 144p.

Craig H. 1957. Isotopic standards for carbon and oxygen and correction factors

for mass-spectrometric analysis of carbon dioxide. Geochimica et

Cosmochimica Acta 12: 133-149.

- 92 -

Currin C, Newell SA, and Paerl H. 1995. The role of standing dead Spartina

alterniflora and benthic microalgae in salt marsh food webs:

considerations based on multiple stable isotope analysis. Marine Ecology

Progress Series 121: 99-116.

deBruyne AMH and Rasmussen JB. 2002. Quantifying assimilation of sewage-

derived organic matter by riverine benthos. Ecological Applications 12:

511-520.

deBruyne AMH, Marcogliese DJ, and Rasmussen JB. 2003. The role of sewage

in a large river food web. Canadian Journal of Fisheries and Aquatic

Sciences 60: 1332-1344.

Deegan LA and Garritt RH. 1997. Evidence for spatial variability in estuarine

food webs. Marine Ecology Progress Series 147: 31-47.

DeNiro MJ and Epstein S. 1978. Influence of diet on the distribution of carbon

isotopes in animals. Geochimica et Cosmochimica Acta 42: 495-506.

DeNiro MJ and Epstein S. 1981. Influence of diet on the distribution of nitrogen

isotopes in animals. Geochimica et Cosmochimica Acta 45: 341-351.

Dube MG, Benoy GA, Wassenaar LI, Blenkinsopp S, Ferone JM and Brua RB.

- 93 -

2004. Application of multi-stable isotope (13C, 15N, 34S, 37Cl) assays to

assess exposure of fish (Longnose Sucker Catostomus catostomus) to

complex effluents in the Athabasca River, Alberta, Canada. Submitted to

Water Quality Research Journal of Canada.

Environment Canada. 1993. Canadian Environmental Protection Act: Cresote-

impregnated waste materials. Priority Substances List Assessment

Report.

Finlay JC. 2001. Stable-carbon-isotope ratios of river biota: implications for

energy flow in lotic food webs. Ecology 82: 1052-1064.

France R. 1995. Critical examination of stable isotope analysis as a means for

tracing carbon pathways in stream ecosystems. Canadian Journal of

Fisheries and Aquatic Sciences 52: 651-656.

Fry B. 1981. Natural stable carbon isotope tag traces Texas shrimp migrations.

Fishery Bulletin 79: 337-345.

Fry B. 1983. Fish and shrimp migrations in the northern Gulf of Mexico analyzed

using stable C, N, and S isotope ratios. Fishery Bulletin 81: 789-801.

Fry B and Sherr EB. 1984. δ13C measurements as indicators of carbon flow in

- 94 -

marine and freshwater ecosystems. Contributions in Marine Science 27:

13-47.

Fry B, Gace A, and McClelland JW. 2003. Chemical indicators of anthropogenic

nitrogen loading in four Pacific estuaries. Pacific Science 57: 77-101.

Gaston TA and Suthers I. 2004. Spatial variation in δ13C and δ15N of liver,

muscle and bone in a rocky reef planktivorous fish: the relative

contribution of sewage. Journal of Experimental Marine Biology and

Ecology 304: 17-33.

Griffin MPA and Valiela I. 2001. δ15N isotope studies of life history and trophic

position of Fundulus heteroclitus and Menidia menidia. Marine Ecology

Progress Series 214: 299-305.

Hammer B, Kelley C, Coffin R, Cifuentes LA, and Mueller J. 1998. δ13C values

of polycyclic aromatic hydrocarbons collected from two creosote-

contaminated sites. Chemical Geology 152: 43-58.

Hansson S, Hobbie JE, Elmgren R, Larsson U, Fry B, and Johansson S. 1997.

The stable nitrogen isotope ratio as a marker of food-web interactions and

fish migration. Ecology 78: 2249-2257.

- 95 -

Heaton THE. 1986. Isotopic studies of nitrogen pollution in the hydrosphere and

atmosphere: a review. Chemical Geology 59: 87-102.

Hesslein RH, Capel MJ, Fox DE, and Hallard KA. 1991. Stable isotopes of

sulfur, carbon, and nitrogen as indicators of trophic level and fish

migration in the Lower Mackenzie River Basin, Canada. Canadian

Journal of Fisheries and Aquatic Sciences 48: 2258-2265.

Hobson KA and Wassenaar LI. 1997. Linking breeding and wintering grounds of

neotropical migrant songbirds using stable hydrogen isotopic analysis of

feathers. Oecologia 109: 142-148.

Hobson KA. 1999. Tracing origins and migration of wildlife using stable

isotopes: a review. Oecologia 120: 314-326.

Hughes EH and Sherr EB. 1983. Subtidal food webs in a Georgia estuary:

δ13C analysis. Journal of Experimental Marine Biology and Ecology 67:

227-242.

Hughes JE, Deegan LA, Peterson BJ, Holmes RM, and Fry B. 2000. Nitrogen

flow through the food web in the oligohaline zone of a New England

estuary. Ecology 81: 433-452.

- 96 -

Keppel G and Wickens TD. 2004. Design and Analysis: A Researcher's

Handbook. Pearson and Prentice Hall: Upper Saddle River, New Jersey.

xii + 611 p.

Kneib RT, Stiven AE and Haines EB. 1980. Stable carbon isotope ratios in

Fundulus heteroclitus muscle tissue and gut contents from a North

Carolina Spartina marsh. Journal of Experimental Marine Biology and

Ecology 46: 89-98.

Lafleur C, Pettigrew B, St-Hilaire A, Booth D, and Chadwick M. 1995. Seasonal

and short term variations in the estuarine structure of the Miramichi, p. 45-

73. In Chadwick M (ed.), Water, Science, and the Public: The Miramichi

Ecosystem. Canadian Special Publication of Fisheries and Aquatic Sciences

123. viii + 300 p.

Lotrich VA. 1975. Summer home range and movements of Fundulus

heteroclitus (Pisces: Cyprinodontidae) in a tidal creek. Ecology 56: 191-

198.

Macko SA and Ostrom PH. 1994. Pollution studies using stable isotopes. In

Stable isotopes in ecology and environmental studies. Edited by K. Lathja

and R.H. Michener. Blackwell Scientific Publishers, London pp. 45-62.

- 97 -

Mariotti A. 1983. Atmospheric nitrogen is a reliable standard for natural 15N

abundance measurements. Nature 303: 685-687.

McClelland JW, Valiela I, and Michener RH. 1997. Nitrogen-stable isotope

signatures in estuarine food webs: a record of increasing urbanization in

coastal watersheds. Limnology and Oceanography 42: 930-937.

Michener RH and Schell DM. 1994. Stable isotope ratios as tracers in marine

aquatic food webs. In Stable isotopes in ecology and environmental

studies. Edited by K. Lathja and R.H. Michener. Blackwell Scientific

Publishers, London pp. 138-157.

Mining Watch Canada and Sierra Club of Canada. 2001. TOXICanada: 13 good

reasons to establish a clean Canada fund. Report by Sierra Club of

Canada. Suite 508, City Center Building, 880 Wellington St., Ottawa,

Ontario, Canada. K1R 6K7

Murphy S. 1991. The ecology of estuarine fishes in southern Maine high salt

marshes; access corridors and movement patterns. M.Sc. Thesis, University

of Massachusetts.

Natech Environmental Services Inc. 2002. Tracer and modeling study for UPM-

Miramichi Kraft Mill.

- 98 -

Pastershank G. 2001. Unifying ecosystem concepts and mercury

biomagnification in an estuarine environment using stable isotopes. PhD

Thesis, University of Ottawa.

Peterson BJ, Howarth RW, and Garritt RH. 1985. Multiple stable isotopes used

to trace the flow of organic matter in estuarine food webs. Science 227:

1361-1363.

Peterson BJ, Howarth RW and Garritt RH. 1986. Sulfur and carbon isotopes as

tracers of salt-marsh organic matter flow. Ecology 67: 865-874.

Peterson BJ and Fry B. 1987. Stable isotopes in ecosystem studies. Annual

Review of Ecology and Systematics 18: 293-320.

Rashid MA and Reinson GE. 1979. Organic matter in surficial sediments of the

Miramichi Estuary, New Brunswick, Canada. Estuarine and Coastal

Marine Science 8: 23-36.

Rau GH, Sweeney RE, Kaplan IR, Mearns AJ, and Young DR. 1981.

Differences in animal 13C, 15N and D abundance between a polluted and an

unpolluted coastal site: likely indicators of sewage uptake by a marine food

web. Estuarine, Coastal and Shelf Science 13: 701-707.

- 99 -

Reinson GE. 1977. Tidal-current control of submarine morphology at the mouth

of the Miramichi estuary, New Brunswick. Canadian Journal of Earth

Sciences 14: 2524-2532.

Savage C and Elmgren R. 2004. Macroalgal (Fucus vesiculosus) δ15N values

trace decrease in sewage influence. Ecological Applications 14: 517-526.

Skinner MA, Courtenay SC, Parker WR and Curry RA. a. Site specificity of

mummichogs (Fundulus heteroclitus) in an Atlantic Canadian Estuary.

Submitted to Water Quality Research Journal of Canada.

Skinner MA, Courtenay SC, Parker WR, and Curry RA. 2004. Evaluation of

techniques for the marking of mummichogs (Fundulus heteroclitus) with

emphasis on Visible Implant Elastomer (VIE). In preparation.

Sokal RR and Rohlf FJ. 1995. Biometry: The Principles and Practice of

Statistics in Biological Research. W.H. Freeman and Co., New York.

Spies RB, Kruger H, Ireland R, and Rice DW, Jr. 1989. Stable isotope ratios

and contaminant concentrations in a sewage distorted food web. Marine

Ecology Progress Series 54: 157-170.

- 100 -

Steffy LY and Kilham SS. 2004. Elevated δ15N in stream biota in areas with

septic tank systems in an urban watershed. Ecological Applications 14:

637-641.

Sweeney J, Deegan L, and Garritt R. 1998. Population size and site fidelity of

Fundulus heteroclitus in a macrotidal saltmarsh creek. Biological Bulletin

195: 238-239.

Takai N and Sakamoto W. 1999. Identification of local populations of Lake Biwa

catfish Silurus biwaensis in Japan on the basis of δ13C and δ15N

analyses. Canadian Journal of Zoology 77: 258-266.

Teo SLH and Able KW. 2003. Habitat use and movement of the mummichog

(Fundulus heteroclitus) in a restored marsh. Estuaries 26: 720-730.

Wassenaar LI and Culp JM. 1996. The use of stable isotopic analyses to

identify pulp mill effluent signatures in riverine food webs, pp. 413-423. In

Servos MR, Munkittrick KR, Carey JH, and Van Der Kraak GJ (eds.),

Environmental fate and effects of pulp and paper mill effluents. St. Lucie

Press, Delray Beach, Fla.

Wayland M and Hobson KA. 2001. Stable carbon, nitrogen, and sulfur isotope

ratios in riparian food webs on rivers receiving sewage and pulp mill

- 101 -

effluents. Canadian Journal of Zoology 79: 5-15.

Zar JH. 1999. Biostatistical Analysis, 4th Edition. Prentice Hall: Upper Saddle

River, New Jersey.

- 102 -

Table 7 – Total length (mean + SD) of mummichogs sampled from upper

Miramichi River estuary for stable isotopic ratios of carbon and nitrogen.

Isotopic ratios of anthropogenic effluents also provided. (CH = Chatham Head,

BKPM = Bleached Kraft Pulp Mill, SM = Strawberry Marsh, UFC = Upstream

Flett Cove, GW = Groundwood Mill, MC = McKay Cove, UH = U-Haul, LB =

Long Beach, WSB = Whitesand Beach, BKPMe = BKPM effluent, BKPMrec =

BKPM receiving environment, GWe = GW effluent, Sew1 = effluent from sewage facility 1, ND = not detectable).

- 103 -

Group Sex n TL (mm) δ13C (‰) δ15N (‰) MC F 11 76.09+ 9.51 -21.33 + 0.62 10.64 + 0.53 M 11 76.00+ 6.15 -21.81 + 1.01 10.93 + 0.19 M and F 22 76.05 + 7.82 -21.57 + 0.85 10.78 + 0.42 BKPM F 12 85.92+ 10.87 -20.77 + 0.90 10.44 + 0.85 M 10 87.70+ 7.04 -20.95 + 0.55 9.87 + 0.73 M and F 22 86.73 + 9.16 -20.85 + 0.75 10.18 + 0.83 UH F 10 70.90+ 3.48 -20.64 + 0.77 10.37 + 0.70 M 10 72.00+ 7.44 -20.62 + 0.77 10.80 + 0.93 M and F 20 71.45 + 5.68 -20.63 + 0.75 10.57 + 0.82 UFC F 11 83.45+ 9.42 -21.66 + 0.70 11.94 + 0.41 M 11 74.18+ 5.10 -21.98 + 0.57 11.92 + 0.24 M and F 22 78.82 + 8.78 -21.82 + 0.65 11.93 + 0.33 GW F 12 79.33+ 7.50 -20.72 + 0.95 12.05 + 0.34 M 12 74.00+ 5.06 -20.66 + 0.48 12.05 + 0.35 M and F 24 76.67 + 6.82 -20.69 + 0.73 12.05 + 0.34 CH F 11 78.91+ 12.65 -20.65 + 0.92 11.71 + 0.58 M 11 73.64+ 5.14 -21.15 + 1.14 11.73 + 0.46 M and F 22 76.27 + 9.80 -20.90 + 1.05 11.72 + 0.51 LB F 11 82.00+ 7.72 -20.37 + 0.86 12.02 + 0.59 M 11 80.73+ 5.80 -20.51 + 0.48 12.31 + 0.51 M and F 22 81.36 + 6.69 -20.44 + 0.68 12.17 + 0.55 SM F 11 81.55+ 9.44 -23.29 + 0.60 10.15 + 0.20 M 11 81.36+ 4.61 -23.54 + 0.68 10.46 + 0.28 M and F 22 81.45 + 7.25 -23.42 + 0.64 10.30 + 0.29 WSB F 11 87.55+ 13.98 -19.19 + 1.92 12.42 + 0.84 M 11 85.09+ 7.33 -18.61 + 1.18 12.48 + 1.09 M and F 22 86.32 + 10.96 -18.90 + 1.58 12.45 + 0.95 BKPMe … 2 … -24.5 -2.4 GWe … 1 … -24.1 0.3 Sew1 … 1 … -13.0 10.7 BKPMrec … 1 … -23.2 ND

- 104 -

WSB

LB 2

CH SM 1 UH BKPM GW MC

3 UFC N

2 km

Figure 5 – Map of Newcastle/Chatham area of the upper Miramichi River estuary with isotope sampling sites (diamonds) indicated. White circles and triangles with numbers (1-3) represent approximate locations of pulp mill and wastewater outfalls, respectively. Shaded regions approximate 1% concentration area of effluent plumes (no data available for Sew3) (Natech,

1998). Arrows indicate direction of waterflow. (CH = Chatham Head, BKPM =

Bleached Kraft Pulp Mill, SM = Strawberry Marsh, UFC = Upstream Flett Cove,

GW = Groundwood Mill, MC = McKay Cove, UH = U-Haul, LB = Long Beach,

WSB = Whitesand Beach).

- 105 -

Figure 6 – Mean ( + 99% confidence intervals; sexes pooled, n shown in Table

7) δ13C (a) and δ15N (b) ratios for mummichog white muscle and bone from sites along the upper Miramichi River Estuary (MRE). Values for most effluents and receiving waters are also shown (GWe and BKPMe were not included in figure b and are found in Table 7). Site locations are shown on Figure 5.

Statistical significance denoted by lowercase letters was determined using

Tukey’s Honestly Significant Test (α = 0.01). (BKPMe = BKPM effluent,

BKPMrec = BKPM receiving environment, GWe = GW effluent, Sew1 = effluent from sewage facility 1).

- 106 -

(a)

-13 Sew1

-15

) -17

oo f / o

C ( -19 e 13 abc acef ace abc δ WSB -21 ab b UH GW LB BKPM CH MC UFC d -23 BKPMrec GWe SM BKPMe -25 -0.5 1.5 3.5 5.5 7.5 9.5 Upstream Downstream River Distance

(b) 14

b 13 b b b b ) 12 oo

/ G WSB o U W LB F C ad N ( a C H 15

δ 11 c Sew1 acd MC

S 10 U M H B KPM 9 -0.5 1.5 3.5 5.5 7.5 9.5 Upstream Downstream River Distance (km)

- 107 -

-108-

Figure 7a – Dendrogram showing percent similarity of δ13C and δ15N ratios among sites

- 108 -

13 )

12oo / GW

o LB WSB UFC

N ( CH

15 11 δ MC -109- 10 SM UH BKPM

9 -24 -23 -22 -21 -20 -19 -18 -17 13 o δ C ( /oo)

Figure 7b – Cross plot of mean (+ 99% confidence intervals) δ13C vs. δ15N ratios for mummichog white muscle and bone

from sites along the upper MRE. Values for effluents and receiving waters are not shown and are found in Table 7.

- 109 -

14.00

12.00 15N δ

10.00 -110-

8.00 -26.00 -24.00 -22.00 -20.00 -18.00 -16.00 δ13C

Figure 8 – Scatter plot of δ13C vs. δ15N ratios for mummichog white muscle tissue from sites along the upper MRE.

Arrows indicate outliers identified by Grubbs Test. (♦ = U-Haul, ■ = Bleached Kraft Pulp Mill, ▲= MacKay Cove, x =

Groundwood Mill, ○ = Whitesand Beach, + = Long Beach, ● = Strawberry Marsh , □ = Upstream Flett Cove, ▬ = Chatham

Head).

- 110 -

5 GENERAL DISCUSSION

Small-bodied fish were proposed for use as sentinel species in environmental monitoring because these species are usually more abundant, ubiquitous and generally less mobile than large bodied fish, therefore increasing probability of obtaining sufficient fish and fish that represent the environmental conditions of the capture locale (Munkittrick et al., 1997; Gibbons et al., 1998).

After problems encountered using large-bodied fish in Cycle 1 EEM fish surveys for pulp and paper mills in Atlantic Canada, this small-bodied sentinel species approach was subsequently applied to Cycle 2 surveys, using mummichogs.

While occurring in relatively high densities, it was speculated these fish were potentially being advected large distances by tidal action based on observations made by consultants sampling these fish in estuaries of large river systems.

This situation led to a re-evaluation of the previous assumption of low mobility in mummichogs. The primary objective of this thesis was to describe the spatial and temporal movement patterns of mummichogs in a large river estuary and assess their usefulness as a sentinel species in environmental programs such as EEM (Chapter 3). Secondary objectives were to determine the most appropriate method of marking mummichogs (Chapter 2) and evaluate stable- isotope analysis as a tool to determine the site-fidelity of organisms along small spatial scales (Chapter 4).

Prior to this thesis, no known experiments had been conducted assessing the suitability of multiple external marking methods in mummichogs.

The first experiment clearly demonstrated VIE was the most suitable of four

- 111 -

common external marking techniques, specifically in terms of mark retention and survival (Chapter 2). Further experimentation supported the hypothesis this method had no negative effects on the growth or survival of mummichogs marked in various body locations (Chapter 2). Thus, VIE proved to be a suitable method for marking mummichogs in various body locations, therefore allowing large numbers of fish to be marked in a manner that would allow discrimination of specimens from multiple sites.

All previous knowledge about the movement of mummichogs came from studies of tidal creeks or salt marshes within the central distribution of their range (Lotrich, 1975; Murphy, 1991, Sweeney et al., 1998; Teo and Able, 2003).

It was possible the assumption of low mobility in mummichogs was incorrect for larger aquatic systems with very different tidal amplitudes such as the Miramichi

River estuary. It was also possible that mummichogs at the northern limit of their geographic range might behave differently from southern conspecifics.

Lotrich (1975) hypothesized variation in mummichog site-specificity could exist along their distribution and further speculated this could most likely be due to the genetic composition of local populations or a combination of local conditions and fish density. Latitudinal variation has been reported in a number of attributes of mummichogs including allozyme expression (Brown and Chapman,

1991) mitochondrial DNA (Smith 1989), gene frequency (Ropson et al., 1990), timing of spawning (Kneib, 1986), and egg sizes and characteristics (Morin et al., 1983; Marteinsdottir and Able, 1988).

- 112 -

Building on the results of Chapter 2, an intensive, two year mark- recapture study was conducted along a ~10 km span of estuary. It concluded mummichogs in this region display distinct site-fidelity with the vast majority

(96.6%) of specimens remaining within very discrete areas (<200 m) from April-

November (Chapter 3). Also, the small proportion of fish found to have moved outside of these areas travelled distances of only 600-3600 m. These results are consistent with previous findings that mummichogs have generally high site fidelity (Lotrich, 1975; Murphy, 1991, Sweeney et al., 1998; Teo and Able,

2003), and extend this observation to larger estuarine systems such as the

Miramichi River. At the same time, this also confirmed it was highly probable mummichogs sampled at any site would accurately reflect local environmental conditions and, therefore, be useful in environmental monitoring programs for point source impacts.

The use of stable isotope analysis to establish movement and migration patterns across great distances in a variety of organisms has been well documented in scientific literature (Fry 1983; Hobson and Wassenaar 1997;

Hobson 1999). However, few studies have used this method to examine small- scale (>10 km) animal movements due to the lack of isotopic variation in potential food sources of target organisms along such spatial scales (Hobson

1999). This heterogeneity, however, may be increased in systems receiving anthropogenic, point source inputs. Therefore potential exists to allow the examination of movements of organisms along spatial scales of only a few hundred meters. Based on the site fidelity demonstrated in Chapter 3,

- 113 -

mummichogs in the Miramichi River estuary were chosen as a model organism to test the above hypothesis (Chapter 4). Carbon and nitrogen stable isotope analyses of samples taken from sites located upstream and downstream of inputs from two pulp mills and three municipal sewage facilities in the MRE along a 10 km scale subsequently supported the hypothesis by demonstrating mummichogs exposed to historical creosote contamination and municipal wastewaters of different treatment levels could be distinguished on the basis of their isotopic signatures. Furthermore, statistical outliers (2% of all fish) were detected during examination of individual isotopic ratios. As their isotopic signatures were not reflective of the signatures of the majority of fish captured at the same sites, this suggested these individual mummichogs were more recent arrivals at these sites. As three of four of these outliers shared isotopic signatures with fish sampled from other sites, these results were consistent with movement patterns previously observed for mobile fish in the mark-recapture study (Chapter 3). In combination, the results of these stable isotope analyses further confirmed the hypothesis of low mobility in mummichogs and also demonstrated the site-fidelity of fish along small spatial scales may be determined with some resolution in areas receiving multiple anthropogenic influences using dual stable isotope analyses.

While these results describe the site-fidelity of mummichogs from April-

November, some questions regarding their overwintering behaviour are left unanswered. During both sampling seasons of the mark-recapture study, catch per unit effort (CPUE) decreased dramatically from September to November

- 114 -

until ice formation in December (Appendix 1, Table 8). The two most likely reasons for these decreases are movement of mummichogs to more favourable areas during winter, or mummichogs burrowing into the substrate to overwinter.

Interestingly, both behaviours have been recorded in the literature. In small tidal creeks and salt marshes, mummichogs have been reported to move upstream

(Halpin, 1997; Lotrich, 1975), downstream (Chidester, 1920; Butner and

Brattstrom, 1960; Smith and Able, 1994) or to burrow into the substrate

(Chidester, 1920; Hardy, 1978; Smith and Able, 1994). Therefore, it is obvious the overwintering habits of this species may vary greatly, even within a given location. Attempts were made during this study to examine the winter movements of mummichogs. The length of estuary surveyed for marked fish during the second sampling season was expanded both up and downstream, increasing the total length surveyed to ~10 km from ~5km during the first season. The rationale for expanding the sampling area was that if mummichogs had migrated to overwintering areas away from original marking locations, then immediately after ice break-up in early spring they should be recaptured at these potential overwintering areas or recaptured while migrating back to a summer home range. Despite intensive seining and trapping effort, no mummichogs, marked or unmarked, were initially recaptured during period in early May 2003. Shortly after however, CPUE at all sites climbed rapidly at all sites with marked fish being recaptured at their respective marking sites. The most probable explanation for this abrupt reappearance of the appropriately marked mummichogs would be these fish had remained in these areas and

- 115 -

most likely burrowed into the substrate to overwinter and had re-emerged over a relatively short span of time. If this hypothesis is correct, mummichogs would have maintained the high site-fidelity observed during the ice-free season and could therefore represent the environmental conditions of these areas.

The results of this thesis have satisfied the stated objectives by describing the spatial and temporal patterns of mummichog movement.

Consistent with the findings of Lotrich (1975), Sweeney et al. (1998), and Teo and Able (2003), this thesis has clearly demonstrated that mummichogs display distinct site fidelity. Thus, mummichogs occurring in greatly varying habitats throughout their range of distribution all exhibit similar site-specific behaviour, whereby very large proportions of the populations examined reside in discrete areas. As such, it would appear populations of this species sampled at any site along its distribution should accurately reflect local environmental conditions and, therefore, be useful in environmental monitoring programs.

Although much has been learned with regards to this species’ ecology, more study is required as outlined below.

5.1 Recommendations and Suggested Research Needs

• While it is highly probable that mummichogs are residing in deeper water

during low tide, increased fishing effort with alternative gear types is

needed to fully address this hypothesis.

• Further study is required to determine the overwintering behaviour of

mummichogs in Atlantic Canada and also establish the environmental

- 116 -

conditions which cue this behaviour. This may be possible by comparing

spring and fall isotope signatures in mummichog liver and/or muscle to

food sources, sediments, water samples and any anthropogenic

compounds to determine where they may be residing during winter.

• When using stable isotopic analyses to examine movements of

organisms, researchers should incorporate a wide variety of materials

(i.e. samples of potentially influential anthropogenic compounds,

sediments/soils, food sources) while also utilizing as many elements as

feasible (13C, 15N, 37Cl, 34S) to increase the resolution of this technique.

• Stable isotopic analyses coupled with conventional techniques for

determination of movements (i.e. telemetry, mark-recapture, etc.)

appears to provide a useful tool for inferring the site-fidelity of organisms.

5.2 Literature Cited

Brown BL and Chapman RW. 1991. Gene flow and mitochondrial DNA variation

in the killifish, Fundulus heteroclitus. Evolution 45: 1147-1161.

Butner A and Brattstrom BH. 1960. Local movement in Menidia and Fundulus.

Copeia 2: 139-141.

Gibbons WN, Munkittrick KR, McMaster ME, and Taylor WD. 1998a. Monitoring

aquatic environments receiving industrial effluents using small fish species 1.

- 117 -

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

kraft pulp mill. Environmental Toxicology and Chemistry 17: 2227-2237.

Halpin PM. 1997. Habitat use patterns of the mummichog Fundulus

heteroclitus, in New England. I. Intramarsh Variation. Estuaries 20: 618-625.

Hardy JD. 1978. Development of fishes of the mid-Atlantic Bight: an atlas of

egg, larval, and juvenile stages. Vol. 2: Anguillidae though Syngnathidae.

U.S. Fish and Wildlife Service Biological Survey Program. FWS/OBS-78/12.

458 p.

Fry B. 1983. Fish and shrimp migrations in the northern Gulf of Mexico analyzed

using stable C, N, and S isotope ratios. Fishery Bulletin 81: 789-801.

Chidester F E. 1920. The behaviour of Fundulus heteroclitus on the salt

marshes of New Jersey. The American Naturalist 635: 551-557.

Hobson KA and LI Wassenaar. 1997. Linking breeding and wintering grounds of

neotropical migrant songbirds using stable hydrogen isotopic analysis of

feathers. Oecologia 109: 142-148.

Hobson KA. 1999. Tracing origins and migration of wildlife using stable

isotopes: a review. Oecologia 120: 314-326.

- 118 -

Lotrich VA. 1975. Summer home range and movements of Fundulus

heteroclitus (Pisces: Cyprinodontidae) in a tidal creek. Ecology 56: 191-198.

Marteinsdottir G and Able KW. 1988. Geographic variation in egg size among

populations of the mummichog, Fundulus heteroclitus (Pisces: Fundulidae).

Copeia 1988 2: 471-478.

Munkittrick KR, Megraw SR, Colodey A, Luce S, Courtenay S, Paine M, Servos

M, Spafford M, Langlois C, Martel P, and Levings C. 1997. Fish survey

expert working group final report. Recommendations from cycle 1 review.

Environment Canada, EEM/1997/6.

Murphy S. 1991. The ecology of estuarine fishes in southern Maine high salt

marshes; access corridors and movement patterns. M.Sc. Thesis, University

of Massachusetts.

Sweeney J, Deegan L, and Garritt R. 1998. Population size and site fidelity of

Fundulus heteroclitus in a macrotidal saltmarsh creek. Biological Bulletin

195: 238-239.

Teo SLH and Able KW. 2003. Habitat use and movement of the mummichog

(Fundulus heteroclitus) in a restored marsh. Estuaries 26: 720-730.

- 119 -

6 APPENDICES

6.1 Appendix 1

Table 8 - Total number of mummichogs caught, number of attempts, and catch

per unit effort (CPUE) from May 2002 – November 2003.

Site Date Number Number of CPUE Captured seines/traps Strawberry May-02 132 2 66.0 Marsh Jun-02 2701 28 96.5 Jul-02 1351 2 675.5 Aug-02 4140 2 2070.0 Sep-02 590 1 590.0 Oct-02 222 1 222.0 Nov-02 8 1 8.0

1-14 May-03 0 6 0 15-31 May-03 912 4 228 Jun-03 875 10 87.5 Jul-03 1780 10 178.0 Aug-03 830 5 166.0 Sep-03 0 0 0.0 Oct-03 810 11 73.6 Nov-03 3 7 0.4

Upstream Flett May-02 126 15 8.4 Cove Jun-02 926 25 37.0 Jul-02 191 11 17.4 Aug-02 660 39 16.9 Sep-02 27 3 9.0 Oct-02 15 2 7.5

1-14 May-03 0 6 0 15-31 May-03 29 8 3.6 Jun-03 117 7 16.7 Jul-03 126 21 6.0 Aug-03 174 12 14.5 Sep-03 0 0 0.0 Oct-03 50 7 7.1 Nov-03 1 7 0.1

- 120 -

Table 8 – continued.

Bleached Kraft May-02 47 5 9.4 Pulp Mill Jun-02 235 22 10.7 Jul-02 114 14 8.1 Aug-02 382 6 63.7 Sep-02 14 4 3.5 Oct-02 7 2 3.5

1-14 May-03 0 6 0 Jun-03 629 7 89.9 Jul-03 410 21 19.5 Aug-03 28 7 4.0 Sep-03 4 7 0.6 Oct-03 1 7 0.1 Nov-03 0 7 0.0

Chatham Head May-02 12 9 1.3 Jun-02 392 37 10.6 Jul-02 199 20 10.0 Aug-02 127 22 5.8 Sep-02 0 2 0.0 Oct-02 3 2 1.5

1-14 May-03 0 5 0 15-31 May-03 12 9 1.3 Jun-03 153 7 21.9 Jul-03 640 21 30.5 Aug-03 134 7 19.1 Sep-03 24 7 3.4 Oct-03 14 7 2.0 Nov-03 0 7 0.0

- 121 -