MERCURY TN FISH AND FISH-EATING BIRDS, WITH SPECIAL REFERENCE TO THE PINCHI L A K E REGION OF BRITISH COLUMBIA, CANADA by SHARI ANN WEECH B.Sc , The University of British Columbia, 1999 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES THE F A C U L T Y OF AGRICULTURE Department of Animal Science We accept this thejjs,,as conforming tp^he required standard The University of British Columbia April, 2003 © Shari Ann Weech, 2003 In presenting t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree that the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r reference and study. I f u r t h e r agree that permission f o r extensive copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the head of my department or by h i s or her r e p r e s e n t a t i v e s . It i s understood that copying or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l gain s h a l l not be allowed without my w r i t t e n permission. Department of The U n i v e r s i t y of B r i t i s h Columbia Vancouver, Canada A B S T R A C T Previous studies have shown that fish-eating wildlife risk elevated methylmercury (meHg) exposure in environments where Hg concentrations, chemical speciation, and/or water chemistry favor Hg methylation and accumulation in fish. Prior to this study, however, it was not known whether Hg from a natural geologic source or contamination resulting from Hg-mining activities would bioaccumulate through the aquatic food chain to concentrations critical to fish-eating bird survival and reproduction. To investigate this, fish and fish-eating birds from five lakes in close proximity to the Pinchi fault in central British Columbia, an area of elevated natural geologic Hg and Hg mining contamination (Pinchi Lake), were studied. In addition to Hg analyses, selenium (Se) concentrations in muscle, age, length, and trophic position were used to possibly explain variations in Hg concentrations in rainbow trout {Oncorhynchus mykiss) and northern pikeminnow (Ptychocheilus oregonensis). Results showed fish Hg concentrations were positively related to age and length, and in northern pikeminnow specifically, Hg concentrations were positively related to trophic position as measured using stable nitrogen isotopes (5 1 5N; n = 42, P < 0.0001) and negatively related to Se concentrations (n = 46, P = 0.014). Mercury concentrations in red-necked grebe (Podiceps grisgena) eggs (n = 24), and in blood and feathers of adult bald eagles (Haliaeetus leucocephalus; n = \3) and eaglets (n = 43) were also determined. A l l grebe eggs, including those from Pinchi Lake (n = 6), were below 0.5 pg/g Hg wet weight, often cited as the lowest observed adverse effect level for Hg developmental toxicity in birds. Mercury concentrations in the blood and feathers of eaglets, including those from Pinchi Lake (n = 12, from 8 nests), were similar or lower than previous studies conducted in areas of non-point source Hg contamination. Analysis of Hg concentrations of one-inch subsections along the length of 12 adult eagle secondary feathers from 12 individual birds also showed a significant trend of decreasing Hg concentrations from the tip to the base of the feather (P < 0.001 to P = 0.022). Eagles were also monitored for reproductive success and productivity during the summers of 2000, 2001 and 2002 (n = 13 to 15 breeding pairs annually). Over the three seasons, bald eagle reproductive success (the total number of active territories found at the beginning of May that produced 8-week-old eaglets) was similar at Pinchi Lake compared with all other study lakes combined [8/13 vs. 18/28, respectively, P = 0.95]. Average productivity (the total number of chicks produced per active u territory) over the three-year study was 0.98 at Pinchi Lake (n = 12 chicks) compared to 1.17 on all other study lakes combined (n = 32 chicks, P = 0.483). These values are also comparable to reproductive success and productivity of bald eagles from reference regions outside the study area. Based on these results, increased dietary Hg exposure due to proximity to a Hg mining and natural geologic Hg source, does not appear to result in critically elevated Hg concentrations in fish-eating birds nor does it adversely affect bald eagle reproductive success. An additional study of 82 bald eagles found dead or dying in other areas of British Columbia (1987 to 1994) showed evidence of elevated Hg exposure and possible poisoning of bald eagles. Post-mortem examinations were conducted on all eagles and livers were analyzed for total Hg content, as well as meHg and Se in 17 individuals. In total, 67 eagles were classed as "low exposure", 14 eagles were classed as "moderate exposure", and one eagle was classed as "high exposure" to Hg. This latter individual was judged to have likely died of Hg poisoning, with a liver Hg content of 130.3 pg/g dry weight, of which approximately 77% was meHg. The 17 bald eagles examined for meHg and Se in their livers had a higher percentage of liver total Hg present as meHg compared to other bird species with similar total Hg concentrations in the liver. In addition, the molar ratio of Hg to Se in these bald eagle livers was higher compared to other bird species. These two factors may make eagles more vulnerable to the toxic effects of meHg compared to other bird species, although further research would be required to confirm their susceptibility. A potential genetic biomarker of meHg exposure was also investigated using livers from adult ring doves (Streptopelia risoria; n = 40) and juvenile common loons (Gavia immer; n = 31). This was the first study to examine both adult and immature avian species for DNA breakage in response to dietary meHg exposure. Results indicated that chronic consumption of diets containing environmentally relevant concentrations of meHg does not result in a significantly increased incidence of liver DNA strand breakage in dosed birds compared to controls. Therefore, it does not appear that measurement of DNA breakage would be a useful biomarker of meHg exposure in wild birds at this point. i n T A B L E O F CONTENTS ABSTRACT ii TABLE OF CONTENTS iv LIST OF TABLES viii LIST OF FIGURES x LIST OF APPENDICES xv LIST OF ABBREVIATIONS xvi ACKNOWLEDGEMENTS xvii INTRODUCTION '. 1 Mercury Cycling in the Environment 2 Mercury in Fish 5 Mercury in Birds 7 Mercury in the Pinchi Lake Region of British Columbia 10 Distribution of Hg in Adult Eagle Secondary Feathers 14 Mercury in Bald Eagles throughout British Columbia 14 Genotoxicity as a Potential Biomarker of meHg Exposure in Birds 15 Thesis Rationale and Organization 16 References 17 CHAPTER 1 - MERCURY AND SELENIUM IN FISH FROM THE PINCHI L A K E REGION OF BRITISH COLUMBIA, CANADA: RELATION TO AGE, FORK LENGTH AND TROPHIC POSITION 26 Introduction 26 Materials and Methods 31 Fish Sampling 31 Sediment Sampling 32 Water Sampling and Analysis 32 Mercury Analysis 32 Selenium Analysis 34 Fish Aging 35 Stable Nitrogen (8N) and Carbon (8C) Analysis 35 Quality Assurance 37 Statistical Analyses 38 Results 39 Sediments and Water Chemistry 39 Fish 40 Discussion 44 iv Sediments and Water Chemistry 44 Fish 46 Conclusions and Implications 55 Acknowledgements 57 References 57 CHAPTER 2 - MERCURY EXPOSURE IN B A L D EAGLES (HALIAEETUS LEUCOCEPHALUS) AND RED-NECKED GREBES (PODICEPS GRISGENA) BREEDING NEAR PINCHI L A K E , BRITISH COLUMBIA, CANADA, 2000-2002 .. 87 Introduction 87 Methods 88 Study Area 88 Choice of Indicator Species 89 Red-Necked Grebe Egg Collection 90 Adult Eagle Capture and Sampling 90 Eaglet Sampling 91 Additional Adult Eagle Blood Samples 92 Mercury Analysis 92 Selenium Analysis 94 Eagle Reproductive Success 94 Quality Assurance 95 Statistical Analyses 96 Results 97 Mercury in Red-necked Grebe Eggs 97 Mercury in Bald Eagles 97 Selenium in Bald Eagles 99 Eagle Productivity and Reproductive Success 100 Discussion 101 Mercury in Red-Necked Grebe Eggs 101 Mercury in Bald Eagles 104 Selenium in Bald Eagles I l l Eagle Productivity and Reproductive Success 114 Conclusions 117 Acknowledgements 118 References 118 CHAPTER 3 - THE ECOLOGY OF FISH-EATING BIRDS IN THE PINCHI L A K E REGION OF BRITISH COLUMBIA IN RELATION TO TROPHIC TRANSFER OF MERCURY FROM SEDIMENTS TO FISH AND FISH-EATING BIRDS -DEVELOPMENT OF HYPOTHESES 144 Introduction 144 Methods 147 Mercury Analyses 147 Stable Isotope Analyses 147 Bald Eagle Feeding Habits 148 Bald Eagle, Common Loon and Red-necked Grebe Productivity 148 v Statistics 148 Results 150 Mercury Relationships between Birds, Fish and Sediments 150 Stable Isotopes 150 Bald Eagle Feeding Habits 152 Ecology of Bald Eagles, Common Loons and Red-necked Grebes in the study area 152 Discussion 155 Mercury Relationships between Birds, Fish and Sediments 155 Stable Isotopes 158 Ecology of Common Loons and Red-necked Grebes in the study area 162 Summary and Potential Future Work 164 Acknowledgements 165 References 166 CHAPTER 4 - DISTRIBUTION OF MERCURY AND SELENIUM IN ADULT B A L D EAGLE (HALIAEETUSLEUCOCEPHALUS) SECONDARY FEATHERS 187 Introduction 187 Materials and Methods 188 Adult Eagle Capture and Sampling 188 Mercury Analysis 188 Selenium Analysis 189 Quality Assurance 190 Statistical Analyses 190 Results 191 Mercury in Adult Eagle Feathers 191 Selenium in Adult Eagle Feathers 191 Mercury in Feather Vanes vs. Shaft 191 Discussion 192 Acknowledgements 198 References 198 CHAPTER 5 - MERCURY RESIDUES IN LIVERS OF B A L D EAGLES {HALIAEETUS LEUCOCEPHALUS) FOUND DEAD OR DYING IN BRITISH COLUMBIA, C A N A D A (1987-1994) 215 Introduction 215 Materials and Methods 216 Results 218 Discussion 220 Acknowledgements 226 References 226 CHAPTER 6 - ASSESSMENT OF DNA STRAND BREAKAGE AS A POTENTIAL BIOMARKER OF METHYLMERCURY EXPOSURE IN BIRDS 236 Introduction 236 Methods 238 vi Doves and MeHg Exposure 238 Common Loons and MeHg Exposure 238 DNA Isolation 239 DNA Damage Assay 239 Gel Analysis 240 Blood Mercury Analyses 241 Statistics 242 Results 242 Ring Doves 242 Common Loons 243 Discussion 244 Acknowledgements 249 References 249 GENERAL SUMMARY AND CONCLUSIONS 261 Research in the Pinchi Lake Region 262 Distribution of Mercury and Selenium in Adult Bald Eagle Secondary Feathers 264 Mercury Exposure in Bald Eagles found Dead or Dying in British Columbia 265 Potential use of DNA breakage as a Biomarker of Dietary meHg Exposure in Birds 266 Future Research 267 References 268 vii LIST OF T A B L E S Page Table 1.1 Mercury concentrations in sediments from Pinchi, Tezzeron, Smart, Great Beaver and Fraser Lakes, 2001 65 Table 1.2 Water chemistry results for Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes, 2001 66 Table 1.3 Summary of Hg and Se concentrations in fish collected from Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes, 2000-2001 67 Table 1.4 Summary of Hg concentrations in fish from past studies conducted in the Pinchi Lake area 68 Table 2.1 Recovery data for certified reference materials run during Hg analysis of bald eagle blood and feather and red-necked grebe eggs 125 Table 2.2 Total mercury in red-necked grebe eggs collected from Pinchi, Tezzeron, Great Beaver and Fraser Lakes (2000 - 2001) 126 Table 2.3 Concentrations of Hg and Se in blood, and morphometric measurements of bald eagle chicks sampled on Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes (2000 - 2002) 127 Table 2.4 Total Hg in feathers of bald eagle chicks sampled on Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes (2000 - 2002) 129 Table 2.5 Concentrations of Hg and Se in blood, and morphometric measurements of adult bald eagles sampled on Pinchi, Tezzeron, Stuart and Fraser Lakes (2001 - 2002) 130 Table 2.6 Total Hg in #2 secondary feathers of adult bald eagles sampled on Pinchi, Tezzeron, Stuart and Fraser Lakes (2001 - 2002) 131 Table 2.7 Bald eagle reproductive success and productivity from Pinchi Lake compared to all other study lakes combined (2000-2002). Tezzeron, Staart, Great Beaver and Fraser Lakes are combined for illustrative purposes but were not combined for statistical analyses 132 Table 3.1, Results of stable nitrogen (8N) and carbon (5C) analyses on adult bald eagle blood from Pinchi, Tezzeron, Stuart and Fraser Lakes (2001) 171 Vlll Table 3.2 Results of stable nitrogen (5N) and carbon (SC) analyses on eaglet blood from Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes (2000) Page 172 Table 3.3 Results of stable nitrogen (5N) and carbon (5C) analyses on red-necked grebe eggs from Pinchi, Tezzeron, Great Beaver and Fraser Lakes(2001) 173 Table 3.4 Food remains found in bald eagle nests from Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes (2001 -2002) 174 Table 3.5 Fate of common loon nests found on Great Beaver, Tezzeron and Fraser Lakes in 2000 175 Table 3.6 Fate of additional common loon families found in and around the study area in 2000 176 Table 4.1 Mercury analytical results for 1 -inch pieces of adult bald eagle #2 secondary feathers from tip to base. Feathers obtained from eagles sampled on Pinchi, Tezzeron, Stuart and Fraser Lakes (2001-2002) 202 Table 4.2 Selenium analytical results for 1-inch pieces of feather from tip to base of an adult eagle #2 secondary feather from Pinchi Lake (2001) 205 Table 4.3 Mercury analytical results for 1-inch pieces of feather from tip to base separated into vanes and shaft for an adult eagle #2 secondary feather from Stuart Lake (2002) 206 Table 5.1 Total mercury (Hg), methylmercury (MeHg) and selenium (Se) in bald eagle livers from British Columbia (pg/g dry weight) 230 Table 6.1 Median liver DNA lengths and blood Hg concentrations for ring doves exposed to 0, 1, 2.5 or 5 pg/g meHg 253 Table 6.2 Median liver DNA lengths and blood Hg concentrations for common loons exposed to 0, 0.1 or 0.5 pg/g meHg 254 ix LIST OF FIGURES Page Figure A Cycling of Hg in a freshwater environment (Source: Winfrey and Rudd, 1990) 3 Figure 1.1 Study Area. The location of the mercury mine is indicated by a black square on the north side of Pinchi Lake 71 Figure 1.2 Sediment and water sampling locations on Pinchi Lake, 2001. 72 Figure 1.3 Mean concentrations of Hg and Se (± SE) in northern pikeminnow from Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes, 2000-2001 73 Figure 1.4 Mean concentrations of Hg and Se (± SE) in rainbow trout from Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes, 2000-2001 74 Figure 1.5 Relationship between fork length and Hg concentration in northern pikeminnow and rainbow trout from Pinchi Lake, 2000-2001 75 Figure 1.6 Relationship between fork length and Hg concentration in northern pikeminnow from Tezzeron, Stuart, Great Beaver and Fraser Lakes, 2000-2001 76 Figure 1.7 Relationship between age and Hg concentration in rainbow trout from Pinchi Lake compared to Tezzeron, Stuart, Great Beaver and Fraser Lakes combined, 2000-2001 77 Figure 1.8 Relationship between fork length and age for northern pikeminnow and rainbow trout from Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes, 2000-2001 78 Figure 1.9 Relationship between trophic position (5N) and fork length in northern pikeminnow and rainbow trout from Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes, 2000-2001. Hg concentrations corrected for lake differences using ANCOVA 79 Figure 1.10 Relationship between trophic position (5N) and Hg concentration in northern pikeminnow from Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes, 2000-2001. Hg concentrations corrected for lake differences using ANCOVA 80 x Page Figure 1.11 Relationship between Hg and Se concentrations in northern pikeminnow from Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes, 2000-2001. Hg concentrations corrected for lake differences using ANCOVA 81 Figure 2.1 Study Area. The location of the mercury mine is indicated by a black square on the north side of Pinchi Lake 133 Figure 2.2 Mercury concentrations [± standard error (SE)] in red-necked grebe eggs from Pinchi, Tezzeron, Great Beaver and Fraser Lakes (2000-2001). Eggs from Pinchi Lake have significantly more Hg than eggs from Fraser Lake only (P = 0.001) 134 Figure 2.3 Mean Hg and Se concentrations (± SE) in blood of eaglets from Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes (2000-2002) 135 Figure 2.4 Mean Hg and Se concentrations (± SE) in blood of adult bald eagles from Pinchi, Tezzeron, Stuart and Fraser Lakes (2001-2002) 136 Figure 2.5 Relationship between eaglet blood and feather Hg concentrations. Results are for individual eaglets sampled from Pinchi, Tezzeron, Smart, Great Beaver and Fraser Lakes (2000-2002) 137 Figure 2.6 Relationship between Hg concentrations in blood of adult bald eagles and chicks from the same nest (Pinchi, Tezzeron, Stuart and Fraser Lakes combined). Sibling blood Hg levels averaged for the comparison 138 Figure 2.7 Relationship between Hg and Se concentrations in blood of eaglets from Pinchi, Tezzeron, Smart, Great Beaver and Fraser Lakes. Hg concentrations corrected for lake differences using ANCOVA 139 Figure 2.8 Active Bald Eagle nesting locations on Pinchi, Tezzeron, Smart, Great Beaver and Fraser Lakes, 2000 140 Figure 2.9 Active Bald Eagle nesting locations on Pinchi, Tezzeron, Smart, Great Beaver and Fraser Lakes, 2001 141 Figure 2.10 Active Bald Eagle nesting locations on Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes, 2002 142 Figure 2.11 Relationship between mean blood Hg levels in eaglets and eagle productivity for Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes, 2000 - 2002 (r = -0.254, P = 0.403) 143 xi Page Figure 3.1 Relationships between logio(mean lake sediment Hg) and average Hg concentrations in red-necked grebe eggs, and adult and chick bald eagle blood from Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes(2000-2002) 177 Figure 3.2 Relationships between mean Hg concentrations in northern pikeminnow and adult and chick bald eagle blood from Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes (2000-2002) 178 Figure 3.3 Relationship between mean Hg concentrations in northern pikeminnow <25cm in fork length and red-necked grebe eggs from Pinchi, Tezzeron, Great Beaver and Fraser Lakes (2000-2001) 179 Figure 3.4 Relationships among mean Hg concentrations in red-necked grebe eggs, and adult and chick bald eagle blood from Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes (2000-2002) 180 Figure 3.5 Relationship between 8N and Hg concentrations in red-necked grebe eggs from Pinchi, Tezzeron, Great Beaver and Fraser Lakes. Hg concentrations corrected for differences among lakes using ANCOVA 181 Figure 3.6 Mean 8C + SD in bald eagle adults and chicks, red-necked grebes, northern pikeminnow, rainbow trout, mountain whitefish and kokanee salmon from the study area. Bars that share the same '*' symbols are not significantly different 182 Figure 3.7 Mean 5N + SD in bald eagle adults and chicks, red-necked grebes, northern pikeminnow, rainbow trout, mountain whitefish and kokanee salmon from the study area. Bars that share the same '*' symbols are not significantly different 183 Figure 3.8 Mean 8C + SD in bald eagle adults and chicks, red-necked grebes, northern pikeminnow, rainbow trout, mountain whitefish and kokanee salmon from Pinchi Lake. Bars that share the same '*' symbols are not significantly different 184 Figure 3.9 Mean 8N + SD in bald eagle adults and chicks, red-necked grebes, northern pikeminnow, rainbow trout, mountain whitefish and kokanee salmon from Pinchi Lake. Bars that share the same '*' symbols are not significantly different 185 xn Page Fi gure 3.10 Relationship between Hg and 8N for individual mountain whitefish, kokanee salmon, rainbow trout, northern pikeminnow, red-necked grebes and bald eagles from Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes 186 Figure 4.1 Relationship between Hg concentration and distance along the length of 4 adult bald eagle #2 secondary feathers (adults #1 - #4) 207 Figure 4.2 Relationship between Hg concentration and distance along the length of 4 adult bald eagle #2 secondary feathers (adults #5, #9 - #11) 208 Figure 4.3 Relationship between Hg concentration and distance along the length of 5 adult bald eagle #2 secondary feathers (adults #6 - #8, #12, recapture) 209 Figure 4.4 Selenium concentration along the length of an adult bald eagle #2 secondary feather 210 Figure 4.5 Relationship between Hg and Se concentrations along the length of an adult bald eagle #2 secondary feather 211 Figure 4.6 Mercury concentrations in whole secondary sections compared to concentrations in the vanes and shaft from each section of an adult bald eagle #2 secondary feather 212 Figure 4.7 Total Hg concentration in an adult eagle #2 secondary feather separated into Hg contribution from the vanes and shaft of each section 213 Figure 4.8 Percentage of total Hg and weight contributed by the shaft compared to the vanes of each feather section from an adult bald eagle #2 secondary feather 214 Figure 5.1 Locations where 82 bald eagles were collected around southern British Columbia (1987-1994) 231 Figure 5.2 Relationship between total Hg and Se in 17 bald eagle livers 232 Figure 5.3 Total Hg in livers of bald eagles with poor (BC = 0-2) versus good (BC = 3-5) body condition; error bars represent standard deviation 233 Figure 5.4 Total mercury levels in relation to age for 81 bald eagles 234 Figure 5.5 Cause of death for 82 bald eagles examined in this study 235 xiii Page Figure 6.1 Characteristic ring dove liver gel photographed under U V light. Lanes (numbered starting from left) 1 and 14 are standards, lanes 2-4 are controls, lanes 5-7, 8-10, and 11-13 are from the 1, 2.5 and 5 pg/g meHg dose groups, respectively 255 Figure 6.2 Average median liver DNA lengths (± SD) for male and female ring doves dosed with 0, 1, 2.5 or 5 pg/g meHg 256 Figure 6.3 Mean blood Hg concentrations (± SD) for male and female ring doves dosed with 1, 2.5 or 5 pg/g meHg 257 Figure 6.4 Characteristic common loon liver gel photographed under U V light. Lanes (numbered starting from left) 1 and 14 are standards, and lanes 2 through 13 are samples 1463, 1465, 1467, 1455, 1449, 1464, 1440, 1468, 1462, 1436, 1470 and 1446, respectively 258 Figure 6.5 Average median liver DNA lengths (± SD) for common loons dosed with 0, 0.1 or 0.5 pg/g meHg 259 Figure 6.6 Mean blood Hg concentrations (± SD) for male and female common loons dosed with 0, 0.1 or 0.5 pg/g meHg 260 xtv LIST OF APPENDICES Appendix 1.1 Raw data for individual fish collected from Pinchi, Tezzeron, Great Beaver, Stuart and Fraser Lakes, 2000-2001 82 Appendix A The Neurotoxicology of Mercury - Oxidative Stress and Excitotoxicity 271 xv LIST OF ABBREVIATIONS ANCOVA Analysis of Covariance A N O V A Analysis of Variance BC Body Condition CFIRMS Continuous Flow Isotope Ratio Mass Spectrometry CVAAS Cold Vapor Atomic Absorption Spectrophotometry CWS Canadian Wildlife Service DOC Dissolved Organic Carbon dw Dry Weight 5C/5 I 3C Ratio of Stable Carbon Isotopes 8N/8 1 5N Ratio of Stable Nitrogen Isotopes GFAAS Graphite Furnace Atomic Absorption Spectrophotometry GSH Glutathione Hg Total Mercury Hg° Elemental Inorganic Mercury H g 2 + Ionic Inorganic Mercury HgS Mercury Sulphide/Cinnabar IRMS Isotope Ratio Mass Spectrometer LOAEL Lowest Observed Adverse Effect Level MITE-NSERC Metals in the Environment - Natural Sciences and Engineering Research Council of Canada MeHg Methylmercury NIH National Institute of Health NOAEL No Observed Adverse Effect Level NRCC National Research Council of Canada NWRC National Wildlife Research Centre SD Standard Deviation SE Standard Error Se Selenium UMESC Upper Midwest Environmental Sciences Center ww Wet Weight x v i A C K N O W L E D G E M E N T S I would like to thank my supervisory committee, John Elliott, Kim Cheng, Tony Scheuhammer and Chris Kennedy for providing guidance, support and assistance with this project. Special personal thanks go to Tony Scheuhammer who has given so much of his valuable time helping me through all phases of this project, both as a friend and mentor. I could not have done it so well without his guidance and support! I would also like to thank Walter Kuit and Bruce Donald of TeckCominco Ltd. for providing the funding necessary to complete the fieldwork in the Pinchi Lake region, and for having faith in my ability as a researcher to complete this work. The Canadian Wildlife Service provided additional funding, which assisted in the completion of this project. Many people provided much appreciated assistance both in the field and in the laboratory. Acknowledgements specific to each part of the research are given at the end of each chapter. In particular, I would like to thank my friend Lori Smith who took time out of her own Ph.D. research to assist me in the field in both 2001 and 2002. Lori was also a great friend throughout the completion of this thesis, helping me recover my sanity when I thought it was gone. I am also thankful for Darcy Haycock and Gregg Howald, who both took time out of their own busy work schedules to assist me with bald eagle sampling. Andy Roessler kindly loaned me his trailer for my zodiac when my other boat died a terrible death on Stuart Lake. Sandi Lee was also a big help in several aspects of this project. Tony Scheuhammer supplied laboratory funding, space, equipment and supplies for mercury and selenium analyses, and partial funding for fieldwork through a Metals in the Environment (MITE)-NSERC Research Networks grant. Ewa Neugebauer and Delia Bond provided valuable assistance in completing metals analyses. Angela Lorenzen, Sean Kennedy and Stephanie Jones provided laboratory space, equipment, supplies and expertise for the completion of the genetics work. Diana Little and Shayne MacLellan of the Pacific Biological Station kindly took the time to teach me how to age fish and provided me with equipment and laboratory space to complete the work. Finally, I would like to thank my family and friends for always having faith in me and keeping me in good spirits throughout this project. And special thanks go to my parents who expressed constant interest in my work despite not knowing what I was talking about most of the time. I will always be grateful for the time I got to spend fish sampling with my dad in Fort St. James in 2000 and 2001. Their support, (financial and otherwise!) made it possible to complete this work in a timely manner. xvii INTRODUCTION Mercury (Hg) is a non-essential heavy metal that is ubiquitous in the environment. Mercury can exist in a variety of forms, both inorganic and organic, with organic varieties generally accepted as being the most toxic to humans and wildlife. Methylmercury (meHg) in particular is believed to be the most toxic form of Hg to wildlife because it is able to bioaccumulate and biomagnify through the food chain (Atwell et al, 1998; Watras et al, 1998; Wolfe et al, 1998), it is efficiently absorbed from the diet (Thompson, 1996; Wiener and Spry, 1996), and results in reproductive impairment, neurotoxicity, and ultimately death in wildlife exposed to toxic concentrations (Thompson, 1996). For example, Barr (1986) found evidence of reduced reproductive success of common loons in an area of elevated Hg contamination, which was associated with Hg concentrations of 2 - 3 pg/g wet weight (ww) in eggs and adult loon brain tissue. Koeman et al. (1969) provided evidence of Hg poisoning of wild kestrels (Falco tinnunculus; kidney Hg = 73 - 125 pg/g ww), buzzards (Buteo buteo; kidney Hg = 85 - 93 pg/g ww) and long-eared owls (Asio otus; kidney Hg = 68 ug/g ww), which also exhibited signs of Hg toxicity prior to death. A review of the potential causes of neurotoxicity due to Hg poisoning (i.e. oxidative stress and excitotoxicity) is included as appendix A. Environmental Hg can be both naturally and anthropogenically-derived; although it is generally believed that Hg released from anthropogenic sources has equaled or exceeded Hg released from natural sources since the onset of the industrial period (Fitzgerald et al, 1998; Boening, 2000). Some of the main sources of Hg to the environment include volcanic activity, geologic faults, gold mining using Hg amalgamation, coal combustion, waste incineration, use of mercurial pesticides and chlor-alkali production (Johnels et al, 1979; Barr, 1986; Plouffe, 1995; Thompson, 1996; Dorea et al, 1998). Flooding, especially due to hydro-electric impoundment creation, has also been shown to mobilize Hg and increase methylation (i.e. transformation of H g 2 + to CHsHg"1") leading to elevated concentrations of Hg in fish and wildlife (Jackson, 1988; DesGranges et al, 1998). 1 Currently, Hg methylation and trophic transfer within aquatic environments is of primary concern for ecotoxicologists. There are many factors affecting the cycling of Hg in aquatic environments, and the availability of meHg. Several of these factors are discussed below. Mercury Cycling in the Environment Mercury in the environment is present in many chemical forms. The majority of atmospheric Hg is typically elemental Hg (Hg°). Photocatalytic reactions can result in the oxidation of atmospheric Hg° to H g 2 + (Iverfeldt and Lindqvist, 1986). Inorganic H g 2 + binds to particulates in the air and can then be removed from the atmosphere by precipitation (Winfrey and Rudd, 1990). Mercury may also be returned to the atmosphere from lakes as Hg°, through abiotic reduction of H g 2 + or transformation by Hg-resistant bacteria (Winfrey and Rudd, 1990). Atmospheric deposition is believed to be the major source of environmental Hg to remote lakes isolated from point source Hg pollution (Asmund and Nielsen, 2000; Riget et al., 2000). Most meHg present in lakes is formed as a result of within-lake methylation. Small concentrations of meHg have been reported in precipitation (Bloom and Watras, 1989), but are believed to be a minor contribution to the total meHg in lakes (Winfrey and Rudd, 1990). Several chemical, physical and biological processes affect the net production of meHg within lakes, including contamination by point-source Hg emissions (Fimreite et al., 1971), environmental acidification (McMurtry et al., 1989; Scheuhammer and Graham, 1999), and flooding, especially due to hydro-electric impoundment creation, has been shown to mobilize Hg and increase methylation (Jackson, 1988; DesGranges et al., 1998). Another factor that may affect the degree of Hg methylation is the chemical form of Hg present in the system. Cinnabar (HgS), a form of Hg considered to have low bioavailability, can be formed in waters in the presence of sulphide (S2~; Bjornberg et al, 1988). Under both aerobic and anaerobic conditions in soil, HgS was much more resistant to oxidation compared to other metal sulphides including CuS, MnS, ZnS and FeS, and the degree of oxidation was directly related to the solubility of the sulphides 2 (Engler and Patrick, 1975). In addition, whereas CuS, MnS, ZnS and FeS were all at least partially oxidized in anaerobic soils adjacent to the roots of a rice plant with subsequent uptake of the oxidized sulphate, HgS did not produce any sulphate suggesting HgS was not oxidized at all (Engler and Patrick, 1975). Given the very low solubility product of HgS (approximately K s p = 10~36 7 3 ; Paquette and Helz, 1995), the rate-limiting step in the production of meHg from HgS is the oxidation of HgS to H g 2 + and S2", which is very slow (Jernelov, 1972). Production of meHg was estimated to be approximately 1000 times slower for HgS compared to H g 2 + (Jernelov, 1972). In the presence of H 2S, no CH3Hg+ was produced from HgC^ in sediments, indicating H g 2 + was precipitated as HgS and no longer available for methylation (Jackson, 1987). Similarly, other studies have reported that increasing sediment-sulphide is often correlated with decreasing meHg concentration or production (Benoit et al. 1999; Winfrey and Rudd, 1990). Methylation of Hg can proceed at low sulphide concentrations partly because HgS is a neutral complex with low volatility that has a high formation constant and relatively high diffusion rate (Dryssen and Wedborg, 1991), allowing low concentrations of HgS to enter bacterial cells. Benoit et al. (1999) have suggested that, as environmental sulphide concentrations increase, the speciation of Hg shifts from predominantly neutral HgS 0 to charged HgHS2 _ and HgS22~ which cannot diffuse across microbial cell membranes and are thus unavailable for methylation. An example of the biogeochemical cycling of Hg in freshwater aquatic environments is shown in Figure A. Hg" Hg(ll) AIR WATER SEDIMENT FISH t . Hg" —Hg(ll) ^ — J J K C H Hg • Hg(ll)- . ^ C H 3 H g ORGANIC AND INORGANIC COMPLEXES HgS C H 3 H g C H 3 r C H 3 H f l C H 3 : CHgHgCHg Figure A: Cycling of Hg in a freshwater environment (Source: Winfrey and Rudd, 1990) 3 Transformation of Hg to meHg is believed to be primarily biological under anaerobic conditions, although abiotic transformation may also play a role, especially in low pH waters (Lee et al., 1985). The net amount of meHg present in sediments and water is a result of microbially mediated methylation and demethylation processes. H g 2 + can be methylated to meHg, which may then be further methylated to volatile dimethylmercury (CH 3HgCH 3) and released to the atmosphere, or again demethylated by meHg-resistant bacteria to elemental Hg, with Hg 2 + as an intermediate (Winfrey and Rudd, 1990). The processes involved are complex and not well understood, and many environmental factors have been shown to affect net meHg production in natural waters. The total amount of meHg produced by microbes in surficial sediments and the water column is inversely related to lake pH (Winfrey and Rudd, 1990). Xun et al. (1987) showed that net methylation rates are lower in high pH lakes compared to lakes with low pH. In lower pH lakes, it has also been suggested that acidity causes dissolved organic carbon (DOC), which typically binds Hg in the water column, to release H g 2 + to the water column, making it available for methylation (Winfrey and Rudd, 1990). Conversely, in higher pH lakes, increased DOC has been shown to decrease within-lake Hg-methylation (Miskimmin et al, 1992). Lake temperature has also been positively correlated to Hg-methylation rates and inversely related to Hg-demethylation rates, indicating increased net Hg-methylation with increasing water temperature (Bodaly et al., 1993). In lakes that experience seasonal changes in temperature, Hg-methylation rates peak in late summer when water temperatures are highest (Winfrey and Rudd, 1990). A water temperature of 35°C has been suggested as the optimal temperature for net meHg production (Winfrey and Rudd, 1990). Increasing the total amount of inorganic Hg present within a lake has also been shown to increase the rate of meHg production in the water column and sediments (Rudd et al., 1983; Xun et al, 1987). Rates of Hg methylation in lake water were shown to more than 4 double with each doubling of Hg concentration (Xun et al, 1987). However, while meHg concentrations may increase with added Hg 2 + ; the chemical speciation of Hg, water pH, DOC, and temperature, and general abiotic and biotic methylation and demethylation processes will ultimately determine the amount of added Hg 2 + that becomes methylated and available for uptake into the food chain. Mercury in Fish Several studies have examined Hg concentrations in fish primarily because this is the dominant route of exposure for both humans and piscivorous wildlife to meHg. Typically, almost all (-99%) of the Hg present in fish muscle is in the methylated form (Grieb et al, 1990; Scheuhammer et al, 1998). The primary route of Hg uptake by fish is through the diet (Hall et al, 1997). While the intestinal wall offish acts as an efficient barrier to Hg chloride, its permeability to meHg allows concentrations of approximately 50%> of the total dose to accumulate in muscle tissue over time (Boening, 2000). Some additional meHg is also obtained through uptake from water passing over the gills, although this is minor in comparison to dietary uptake (Phillips and Buhler, 1978; Wiener and Spry, 1996). Mercury itself is not methylated to any significant degree within fish tissues (Pennacchioni et al, 1976; Huckabee et al, 1978), however some degree of methylation has been shown to occur in the gut (Rudd et al, 1980). Following consumption and assimilation of meHg, the majority of meHg is redistributed from internal tissues such as the blood, liver and brain to the skeletal muscle (Giblin and Massaro, 1973; Wiener and Spry, 1996). Redistribution of meHg from internal organs to the muscle tissue may offer some degree of protection from the toxic effects of meHg because this meHg is then not available to attack more sensitive internal organs such as the brain. Muscle tissue is believed to be less sensitive to the adverse effects of meHg (Wiener and Spry, 1996). Elimination of meHg via feces, kidney and possibly gills is very slow relative to uptake with half-retention times ranging from 200 days to 2 years (Giblin and Massaro, 1973; Wiener and Spry, 1996), resulting in a net bioaccumulation of meHg in fish over time (Watras et al, 1998). 5 Concentrations of Hg in fish muscle are often related to the size and/or age of the fish. This relationship has been shown to occur in many species of fish (Wiener et al, 1990; Lathrop et al, 1991; Scheuhammer and Graham, 1999; Riget et al, 2000; EVS Environment Consultants, 2001). It should be noted that Stafford and Haines (2001) have found that biodilution of Hg can occur in fish, and that biodilution can confound Hg -vs-size relationships as a result. Biodilution can occur when growth rates exceed contaminate bioaccumulation rates, resulting in steady or possibly decreasing contaminant concentrations in fish. This may be of particular importance in fish species with a fast growth rate that occupy a lower position in the food chain. Trophic status has been shown to significantly affect the degree to which fish accumulate meHg. Fish occupying a higher trophic position, both within a given species and among different species; have been shown to have higher concentrations of Hg in muscle tissue (Bowles et al., 2001; Power et al., 2002). Several environmental factors have also been shown to affect meHg accumulation in fish. For example, fish from low pH/low alkalinity waters typically have higher concentrations of Hg in muscle tissue compared to those of the same species from higher pH/alkalinity waters (Suns and Hitchin, 1990; Wiener et al, 1990; Mason et al, 2000). Higher levels of dissolved organic carbon (DOC) have also been linked to increased fish-Hg concentrations in some species (McMurtry et al, 1989); however Grieb et al. (1990) found a negative correlation between DOC and Hg in fish from seepage lakes. Epilimnetic temperature also showed a significant positive relationship with Hg concentrations in northern pike (Esox lucius), walleye (Stizostedion vitreum), cisco (Coregonus clupeaformis) and yellow perch (Perca flavescens) from lakes in northwestern Ontario (Bodaly et al, 1993). Increased selenium (Se) availability has also been linked to decreased Hg concentrations in fish. Treatment of a Swedish lake with sodium selenite resulted in significantly decreased Hg concentrations in fish (Paulsson and Lundbergh, 1989). After a period of only one year, Hg concentrations in perch decreased 65-77%, depending on the size of the fish (Paulsson and Lundbergh, 1989). Similar decreases were also found in perch and 6 northern pike from eleven different lakes in Sweden that were treated with Se (Paulsson and Lundbergh, 1991). Chen and Belzile (2001) found a significant inverse relationship between concentrations of Hg and Se in perch and walleye in lakes near the metal smelters in Sudbury, Canada. It was suggested that Se had an antagonistic effect on the assimilation of Hg in fish (Chen and Belzile, 2001). Increasing Se concentrations in prey fish were also shown to reduce the assimilation of Hg (as 2 0 3Hg(NC>3)2) by northern pike (Turner and Swick, 1983). Typical symptoms of meHg poisoning of fish include lack of coordination, inability to feed, sluggishness and emaciation (Wiener and Spry, 1996). Most symptoms suggest that the mode of meHg toxicity in fish is primarily neurotoxicity. Possible critical tissue concentrations for meHg poisoning offish are 7 pg/g wet weight (ww) in the brain, with concentrations of 3 pg/g ww or greater likely indicating significant toxic effects, and 6 to 20 pg/g ww meHg in muscle for wild fish species, which is similar to laboratory-based findings of 5 to 8 pg/g ww in walleye and 10 to 20 pg/g ww in salmonids (Wiener and Spry, 1996). Other potential, subtler effects of lower-level meHg exposure in fish have not yet been well studied. Mercury in Birds The primary route of exposure of birds to meHg is through the diet. Similar to fish, intestinal absorption of inorganic Hg is very low compared to the almost complete absorption of meHg (Scheuhammer, 1987). Assimilated inorganic Hg is most toxic to the kidneys whereas meHg is primarily a neurotoxicant (Wolfe et al, 1998). The biological half-life of meHg in birds has been estimated to be approximately 2 to 3 months in a variety of birds (Scheuhammer, 1987). Typical symptoms of acute meHg poisoning usually include: weight loss as a result of reduced food intake, uncoordinated muscle movements, and difficulty walking, flying and standing (Scheuhammer, 1987). Environmental meHg exposure has also been associated with behavioral and reproductive impairment in breeding common loons (Nocera and Taylor, 1998; Barr, 1986; Meyer et al, 1998). Decreased back-riding, a behavior which provides protection, saves energy, 7 and moderates thermoregulation of common loon chicks has been associated with increased blood-Hg levels (Nocera and Taylor, 1998). Mallard (Anas platyrhynchos) ducklings exposed to meHg in ovo were also shown to be less responsive to taped maternal calls, and more responsive to a frightening stimulus (Heinz, 1979). Overall, dietary meHg exposure insufficient to cause overt toxicity in adult birds can yet result in decreased reproductive success of 35 to 50% (Wolfe et al., 1998). One of the major routes of meHg elimination in birds is through deposition into growing feathers. Almost all of the Hg present in feathers is meHg (Thompson and Fumess, 1989), which has been incorporated into the structure of the feather by binding to sulphydryl groups in feather keratins (Burger and Gochfeld, 1997). MeHg in feathers represents both dietary intake and mobilization of meHg previously stored in other tissues such as the muscle and liver (Braune and Gaskin, 1987). Feathers have been shown to contain up to 93% of the body burden of Hg (Braune and Gaskin, 1987). Feathers typically contain some of the highest concentrations of meHg in bird tissues (Burger, 1993; Caldwell et al, 1999), and have been positively correlated to blood concentrations in many studies (Wood et al, 1996; DesGranges et al, 1998; Evers et al., 1998; Scheuhammer et al, 1998; Sepulveda et al, 1999). Deposition of meHg in eggs (Fimreite, 1971; Heinz, 1979) is regarded as another form of meHg excretion by adult female birds. However, since meHg is a potent embryo toxin (Wolfe et al, 1998), this can lead to significant detrimental effects on reproduction if concentrations of Hg deposited in the egg are sufficiently high. Concentrations of 0.5 to 1.5 pg/g Hg were found in unhatched eggs of pheasants (Phasianus colchicus) exposed to various dietary concentrations of meHg (from 3.4 to 13.5 pg/g meHg) for a maximum period of 12 weeks (Fimreite, 1971). Heinz (1974) found similar effects in breeding mallard ducks fed 3 pg/g meHg in their diet. Another study of mallard eggs exposed to externally applied meHg-chloride revealed dose-dependent effects on survival and growth, with the lowest dose affecting survival equivalent to approximately 0.5 pg/g ww (Hoffman and Moore, 1979). Based on a combination of past field data and a recent study of common loons raised in captivity, Evers et al. (in press) have concluded that 8 reproductive success of common loons can be categorized based on Hg concentrations in eggs as "low risk" (0-0.60 pg/g ww), "moderate risk" (0.60-1.30 pg/g ww) or "high risk" (>1.30 pg/g ww). The concentration of 0.5 pg/g ww continues to be used as the lowest observed adverse effect level (LOAEL) in egg monitoring studies (Thompson, 1996; Wolfe et al, 1998). MeHg toxicity may be reduced through demethylation, particularly in the liver. In vivo demethylation has been shown to occur in mammals such as the guinea pig (Komsta-Szumska et al., 1983), and likely occurs in avian species as well (Norheim and Froslie, 1978; Scheuhammer et al, 1998b; Henny et al, 2002). The resulting Hg may then be detoxified further through binding to Se in an equimolar Hg-Se-protein complex (Yoneda and Suzuki, 1997). Selenium has also been shown to have a protective effect on Hg toxicity in adult mallards fed both meHg and Se; however it should be noted that the toxic effects of these elements on developing mallard embryos from the same experiment were additive rather than protective (Heinz and Hoffman, 1998). Demethylation of meHg in the liver and subsequent association with Se has made developing a critical tissue concentration difficult. Concentrations of approximately 30 pg/g ww in liver and kidney have been associated with Hg intoxication in birds of prey in both laboratory- and field-based studies (Thompson, 1996). However, to accurately suggest a critical tissue concentration such as this, it should be known that the majority of Hg is present as meHg. As stated above, 0.5 pg/g meHg ww in eggs is currently considered the LOAEL for embryo toxicity in bird species. Mercury concentrations of 0.3 to 0.4 pg/g ww in prey of common loons was associated with reduced reproductive success, and decreased territory fidelity in adults (Barr, 1986). Nocera and Taylor (1998) suggested that exposure of 1.25 - 1.50 ug/ml Hg in the blood may be a critical level for behavioral and/or lethal effects in common loon chicks. Dietary concentrations known to cause reproductive impairment are typically one fifth of the dietary level of meHg that would be required to cause significant neurological defects in adult birds (Scheuhammer, 1988). Overall, critical concentrations are useful as guidelines, but are not likely applicable to all bird species. MeHg exposure in wild piscivorous birds is best 9 determined on a case-by-case basis, as we have attempted to do in the current study (see below). Mercury in the Pinchi Lake Region of British Columbia In British Columbia, several sources of Hg contamination exist which may potentially affect piscivorous wildlife; however, most are due to past uses of Hg. Two such examples of past Hg use in British Columbia are Hg used in chlor-alkali production, and mercurial slimicides used in pulp and paper manufacturing (Garrett et al., 1980). Increased meHg production, as a result of flooding due to reservoir creation, has also occurred, for example, in Williston Lake, central British Columbia (Watson, 1992). Another source of Hg release, which has recently been the focus of an environmental assessment, is the former Hg mine located on Pinchi Lake. The Pinchi fault is a source of natural geologic Hg in the form of cinnabar (HgS) to the Pinchi Lake region (Plouffe, 1995). As a result, a Hg mine was located on the north shore of Pinchi Lake, and was operational from 1940-1944 and 1968-1975 (EVS Environment Consultants etal, 1999). Prior to understanding the potentially harmful environmental effects of Hg, Hg wastes from the 1940-1944 mining operations at Pinchi were released directly into Pinchi Lake (EVS Environment Consultants et al., 1999). Estimates of close to 35,150 kg of Hg remained in the roasted ore, or calcine, that was deposited into Pinchi Lake (EVS Environment Consultants et al., 1999). When the mine was re-opened in 1968, mining operations had changed considerably, with the construction of a tailings pond, circulation of mill water in a closed circuit, and use of coarse tailings as backfill in the underground mine. With these improvements in place, it is believed that additional contamination of Pinchi Lake due to 1968-1975 mining operations was minimal, if any (EVS Environment Consultants etal., 1999). Fish from Pinchi Lake have significantly elevated concentrations of Hg in muscle tissue compared to fish from surrounding lakes, presumably due to the contamination of Pinchi Lake with Hg mining wastes (EVS Environment Consultants, 2001). Two lake trout (Salvelinus namaycush) collected in Pinchi Lake in 1969 had Hg concentrations of 1.07 10 and 10.5 ug/g wet weight (ww) and 4 rainbow trout (Oncorhynchus mykiss) collected had a mean Hg concentration of 0.38 pg/g ww (range - 0.25-0.68 pg/g ww) in dorsal muscle (Fimreite et al, 1971), compared to 3 lake trout (0.33 pg/g ww) and 6 rainbow trout (0.05 pg/g ww) collected from Tezzeron Lake in 1970 (Garrett et al, 1980). Tezzeron Lake is located immediately north of Pinchi Lake. A more comprehensive fish sampling study conducted by Reid and Morley (1975) in 1974 showed concentrations of 0.36-8.31 pg/g ww (mean = 5.23 pg/g ww, n = 11) in lake trout and 0.05-0.73 pg/g ww (mean = 0.40 mg/gm ww, n = 10) in rainbow trout from Pinchi Lake. Lake trout ranged in size from 23.9-83.0 cm (mean = 56.6 cm) and rainbow trout ranged from 13.6-38.4 cm (mean = 28.5 cm) in length (Reid and Morley, 1975). By 1986, Pinchi Lake was already beginning to show signs of reduced Hg concentrations in fish. Sixteen lake trout 24.5-71.5 cm in size (mean = 55.6 cm), were found to have much lower muscle Hg concentrations (mean = 1.06 pg/g ww, range = 0.38-3.05 pg/g ww) compared to lake trout collected in 1974 (Watson, 1992). Only 2 rainbow trout were sampled in this study, both 32 cm in length, with Hg concentrations of 0.15 and 0.24 pg/g ww in muscle. Since the 1986 study conducted by Watson (1992), concentrations of Hg in fish have not shown further decline. In 1995, a study conducted by EVS Environment Consultants et al. (1999) found mean Hg concentrations of 0.96 pg/g ww (range = 0.59-1.72 pg/g ww) in 10 lake trout averaging 48.1 cm in length (range = 35.1-68.4 cm). Two rainbow trout (length = 27.0 and 35.6 cm) collected at the same time had Hg concentrations of 0.15 and 0.89 pg/g ww in muscle. Finally, a comprehensive study of Hg in lake trout from Pinchi Lake and several surrounding lakes from the same area found that 31 lake trout collected over a wide range of sizes (41.7-83.1 cm, mean = 61.3 cm) had a mean Hg concentration of 1.33 pg/g ww (range = 0.34-3.03 pg/g ww) in muscle (EVS Environment Consultants, 2001). While it appears that Hg concentrations in fish have leveled off in Pinchi Lake, they are still significantly elevated compared to other lakes in the area. In 2000, when lake trout were found to have a mean Hg content of 1.33 pg/g ww in Pinchi Lake, mean Hg concentrations in lake trout from Tezzeron Lake were 0.50 pg/g ww (range = 0.26-0.78 11 pg/g ww) and from Stuart Lake (immediate south/downstream of Pinchi Lake) were 0.30 pg/g ww (range = 0.10-0.98 pg/g ww; EVS Environment Consultants, 2001). Based on the elevated Hg concentrations in fish from Pinchi Lake, and the presence of Hg-mining related contamination, Pinchi Lake was chosen as the focal lake in the current study. Lake trout from Tezzeron Lake were found to have significantly elevated muscle Hg concentrations compared to surrounding lakes (i.e. Stuart, Francois, and Trembleur Lakes; EVS Environment Consultants, 2001). It is suspected that mean Hg concentrations in lake trout from Tezzeron Lake are higher compared to surrounding lakes because, like Pinchi Lake, a portion of the Pinchi fault runs directly beneath Tezzeron Lake (Plouffe, 1995; EVS Environment Consultants, 2001). Because of the presumed additional influence of natural Hg in Tezzeron Lake, and its proximity to Pinchi Lake, Tezzeron was chosen as an additional focal lake in the current study. Smart Lake is southwest of the Pinchi fault, and therefore was not directly influenced by glacial transport and deposition of geologic Hg (Plouffe, 1995). It is, however, downstream of Pinchi Lake and the mine site. Lockhart et al. (2000) found that 2 sediment cores from Stuart Lake showed a distinct increase in Hg concentration associated with the 1940-1944 mining operations on Pinchi Lake. However, concentrations of Hg in sediments were shown to reach pre-mining levels in Smart Lake sediment cores sometime around 1980 (Lockhart et al, 2000). No increase in Hg concentrations in sediments from Smart Lake was found associated with 1968-1975 mining activities on Pinchi Lake (Lockhart et al, 2000). Stuart Lake was chosen as a study lake because of its proximity to Pinchi Lake, and because it had received Hg from Pinchi Lake in the past. While fish and sediments do not show elevated Hg concentrations compared to surrounding lakes, or as a result of Hg contamination from Pinchi Lake (Lockhart et al, 2000; EVS Environment Consultants, 2001), Hg concentrations in fish-eating wildlife might show a different result. Knowing the location of the fault and the direction of glacial transport of geologic Hg in the past, it was also possible to find lakes in the same area that were believed to have 12 been influenced little by natural Hg associated with the Pinchi fault. Great Beaver and Fraser Lakes are two such lakes. Both are hydrologically separated from Pinchi, Tezzeron and Stuart Lakes, and from each other. A portion of the Pinchi fault may reach Great Beaver Lake, but its influence on Hg concentrations is believed to be minor, if any (Plouffe, pers. comm.). Fraser Lake is west of the Pinchi fault, the opposite direction of glacial dispersion of Hg in the past (Plouffe, 1995). Both Great Beaver and Fraser Lakes were chosen as control lakes in the current study. Bioaccumulation of Hg released from natural and mining-related sources has not been well studied. (Chan et al., in press). The contamination of Pinchi Lake with Hg associated with past mining practices provided an excellent opportunity to study meHg bioaccumulation in the food chain as a result of Hg mining activities. In addition, the presence of a natural geologic source of Hg in the area provided the opportunity to investigate the ability of natural Hg (i.e. cinnabar) to become methylated and bioaccumulate in the food chain. Mercury from cinnabar is relatively resistant to oxidation in comparison to other metal sulphides (Engler and Patrick, 1975), and the strength of the Hg-S bond likely inhibits direct methylation (Jernelov, 1972; Jackson, 1987). Therefore, we hypothesized that Hg from cinnabar should not be readily available for methylation and uptake into the food chain; and that, consequently, Hg concentrations in fish and fish-eating wildlife might not be elevated. However, if Hg concentrations are elevated in fish and fish-eating birds, and the elevated concentrations could be shown to correlate with Hg concentrations in sediments and among fish and birds, then natural geologic Hg is, to some degree, a source of Hg for methylation and transfer into the food chain. To analyze Hg distribution in the food chain of the Pinchi Lake area, water, sediment, 4 species of forage fish (<40cm), blood and feathers from adult bald eagles and eaglets, and red-necked grebe eggs were sampled from Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes. We hypothesized that Hg concentrations would be lowest in the control lakes (Great Beaver and Fraser), higher in Tezzeron and possibly Stuart Lakes, and highest in Pinchi Lake. Productivity and nesting success of bald eagles were also 13 monitored over a 3-year period to determine whether or not Hg was having any adverse effects. Distribution of Hg in Adult Eagle Secondary Feathers Bird feathers are often used in studies of environmental Hg contamination because they can be sampled without harming the bird, and it is possible to obtain a sample without having the bird in hand (Bowerman et al, 1994; Cahill et al, 1998). Almost all of the Hg present in feathers is meHg (Thompson and Fumess, 1989), which has been incorporated into the structure of the feather by binding to sulphydryl groups in feather keratins (Burger and Gochfeld, 1997). MeHg in feathers represents both dietary intake and mobilization of meHg previously stored in other tissues such as the muscle and liver (Braune and Gaskin, 1987). Feathers have been shown to contain up to 93% of the body burden of Hg (Braune and Gaskin, 1987), and Hg concentrations in feathers are typically higher than in other tissues, making it possible to perform accurate analyses on relatively small sample masses. Because birds can eliminate much of their body burden of meHg through feather growth, it was hypothesized that the amount of Hg deposited into a growing feather would decrease over the duration of feather growth, and as the body burden of meHg in internal tissues declined. Results of this study would show if it is acceptable to digest only a small portion of larger feathers for Hg analysis or if the entire feather should be used in the analysis. Mercury in Bald Eagles throughout British Columbia In addition to Pinchi Lake, several areas throughout British Columbia are potential sources of Hg exposure for bald eagles and other fish-eating wildlife. As mentioned above, Hg used in chlor-alkali production, mercurial slimicides used in pulp and paper manufacturing (Garrett et al, 1980) and increased meHg production as a result of reservoir creation (Watson, 1992) may all be potential sources of Hg to bald eagles living in British Columbia. 14 Fish are the major dietary source of meHg in top predators in aquatic food chains (Clarkson, 1992). Mercury residues in fish of the same species, in the same system, often increase with increasing size (Wiener and Spry, 1996). As a result, bald eagles may be exposed to higher concentrations of meHg compared to other aquatic top predators such as common loons because their diet is typically comprised of large fish, as well as other fish-eating birds. Several studies have examined Hg concentrations in populations of bald eagles (Noble and Elliott, 1990; Kozie and Anderson, 1991; Bowerman et al, 1994; Wood et al, 1996), but few have examined Hg-exposure in bald eagles found dead. In the current study, post-mortem examinations were conducted on 82 bald eagles found dead or dying in British Columbia and total Hg concentrations were determined in the liver of each bird. MeHg and Se concentrations were also determined in the livers of 17 eagles. The main goal of this study was to document the degree of Hg exposure in bald eagles found dead or dying in British Columbia, and to investigate possible relationships between liver Hg accumulation and Se, age, body condition, and causes of death. Genotoxicity as a Potential Biomarker of meHg Exposure in Birds Based on previous research examining the ability of Hg and meHg to elicit genetic damage, it was suspected that DNA strand breakage might prove to be a useful biomarker of meHg exposure, particularly because the genetic effects of Hg have been shown to be both dose and time dependent (Kato, 1976; Cantoni et al, 1982; Robison et al, 1982; Cantoni and Costa, 1983; Betti et al, 1993). Past studies have shown that Hg can cause DNA damage in the form of strand breakage (Kato, 1976; Betti et al, 1993; Sugg et al, 1995). While DNA breakage is a common occurrence in cells, many toxic chemicals have been shown to intensify this effect to varying degrees within different tissues (Shugart, 1993; Theodorakis et al, 1994; Sugg et al, 1995; Martin Jr. and Black, 1998). Since strand breakage can be simply and effectively measured using agarose gel electrophoresis (Theodorakis et al, 1994), and previously frozen tissues can be used, this was the form of damage chosen for examination. 15 The general hypothesis of this research was that adult ring doves (Streptopelia risoria) and juvenile common loons (Gavia immer) exposed to environmentally-relevant concentrations of dietary meHg would experience greater DNA strand breakage in liver tissue compared to control birds, and that the degree of damage would be positively correlated to meHg exposure. If this could be shown, then DNA breakage might prove useful as a sensitive biomarker related to the degree of meHg exposure, especially for assessing the degree of meHg exposure in wild fish-eating birds that show no overt signs of meHg intoxication. Thesis Rationale and Organization The above studies were chosen for this thesis because they all pertain to Hg in fish and fish-eating birds. Since the field study in the Pinchi Lake region was the main focus of the thesis, chapters pertaining to that research are presented first. In particular, results of the fish investigation in the Pinchi Lake area are presented first because this was the initial part of the food chain to be analyzed. The information provided in this chapter gives relevant background information to aid in understanding Hg concentrations in the birds from the study region. Bald eagles and red-necked grebes are top-predators studied in the Pinchi Lake area, and thus results for these species are presented in the second chapter. Hg concentrations in these species were expected to be dependent on Hg concentrations in sediments and fish. The third chapter serves to combine all results obtained throughout the study (i.e. sediment, fish and bird results) to provide a summary of interactions among the different media; and also to discuss possible relationships between Hg concentrations and productivity/breeding success of fish-eating birds in the study area. A more detailed study on the Hg concentrations in adult bald eagle feathers from the study area was completed to show how Hg concentrations vary across secondary feathers. This was included as a separate chapter following the summary of the Pinchi Lake region results, rather than within chapter 2, because the in-depth analyses that were completed on these feathers justified a separate chapter for this part of the study. In addition, the outcome of the adult feather study may be directly applicable to future studies conducted by researchers examining Hg concentrations in larger (primary or 16 secondary) feathers, regardless of the fact that the samples were obtained from the Pinchi Lake region. To expand upon the toxicological assessment of Hg in bald eagles breeding in British Columbia, the results of Hg analyses of livers from bald eagles found dead or dying throughout British Columbia are presented in chapter 5. Unlike most other studies examining Hg residues in dead birds, this study also related Hg analytical results to post-mortem examination findings and causes of death. Finally, increased interest in developing a biomarker of environmentally-relevant meHg exposure in fish-eating birds led to a study of the potential use of genotoxicity, specifically DNA breakage, as a useful biomarker. The results of this laboratory work are presented in the final chapter, as they may be related to fish-eating birds in general. The results of the above studies contribute to the understanding of the ability of natural Hg to bioaccumulate in the food chain, how Hg-mining contamination can lead to increased Hg exposure in the food chain, and the effects of meHg exposure on the breeding success, productivity and general survivability of fish-eating birds. References Asmund, G. and Nielsen, S. P. (2000). Mercury in dated Greenland marine sediments. Sci. Total Environ. 245, 61-72. Atwell, L., Hobson, K. A., and Welch, H. E. (1998). Biomagnification and bioaccumulation of mercury in an arctic marine food web: Insights from stable nitrogen isotope analysis. Can. J. Fish. Aquat. Sci. 55, 1114-1121. Barr, J. F. (1986). Population dynamics of the Common loon (Gavia immer) associated with mercury-contaminated waters in northwestern Ontario. Can. Wildl. Serv. Occas. Pap. No. 56, pp. 25. Benoit, J. M . , Mason, R. P., and Gilmour, C. C. (1999). Estimation of mercury-sulfide speciation in sediment pore waters using octanol-water partitioning and implications for availability to methylating bacteria. Environ. Toxicol. Chem. 18, 2138-2141. Betti, C , Barale, C. B., and Pool-Zobel, B. L. (1993). Comparative studies on cytotoxic and genotoxic effects of two organic mercury compounds in lymphocytes and gastric mucosa cells of Sprague-Dawley rats. Environ. Mol. Mutagen. 22, 172-180. 17 Bjornberg, A., Hakanson, L, and Lundberg, K. (1988). A theory on the mechanisms regulating the bioavailability of mercury in natural waters. Environ. Pollut. 49, 53-61. Bloom, N . S., and Watras, C. J. (1989). Observations of methylmercury in precipitation. Sci. Total Environ. 87/88, 199-207. Bodaly, R. A., Rudd, J. W. M . , Fudge, R. J. P., and Kelly, C. A. (1993). Mercury concentrations in fish related to size of remote Canadian Shield lakes. Can. J. Fish. Aquat. Sci. 50, 980-987. Boening, D. W. (2000). Ecological effects, transport, and fate of mercury: A general review. Chemosphere 40, 1335-1351. Bowerman IV, W. W., Evans, E. D., Giesy, J. P., and Postupalsky, S. (1994). Using feathers to assess risk of mercury and selenium to bald eagle reproduction in the Great Lakes Region. Arch. Environ. Contam. Toxicol. 27, 294-298. Bowles, K. C , Apte, S. C , Maher, W. A., Kawei, M . , and Smith, R. (2001). Bioaccumulation and biomagnification of mercury in Lake Murray, Papua New Guinea. Can. J. Fish. Aquat. Sci. 58, 888-897. Braune, B. M . , and Gaskin, D. E. (1987). Mercury levels in Bonaparte's gulls (Lams Philadelphia) during autumn molt in the Quoddy Region, New Brunswick, Canada. Arch. Environ. Contam. Toxicol. 16, 539-549. Burger, J. (1993). Metals in avian feathers: Bioindicators of environmental pollution. Rev. Environ. Toxicol. 5, 203-311. Burger, J., and Gochfeld, M . (1997). Risk, mercury levels, and birds: Relating adverse laboratory effects to field biomonitoring. Environ. Res. 75, 160-172. Cahill, T. M . , Anderson, D. W., Elbert, R. A., Perley, B. P., and Johnson, D. R. (1998). Elemental profiles in feather samples from a mercury-contaminated lake in central California. Arch. Environ. Contam. Toxicol. 35, 75-81. Caldwell, C. A., Arnold, M . A., and Gould, W. R. (1999). Mercury distribution in blood, tissues, and feathers of double-crested cormorant nestlings from arid-lands reservoirs in south central New Mexico. Arch. Environ. Contam. Toxicol. 36, 456-461. Cantoni, O., Evans, R. M . , and Costa, M . (1982). Similarity in the acute cytotoxic response of mammalian cells to mercury (II) and x-rays: DNA damage and glutathione depletion. Biochem. Biophys. Res. Comm. 108, 614-619. Cantoni, O., and Costa, M . (1983). Correlations of DNA strand breaks and their repair with cell survival following acute exposure to mercury (II) and x-rays. Mol. Pharmacol. 24, 84-89. 18 Chan, H. M. , Scheuhammer, A. M. , Ferran, A., Loupelle, C , Holloway, J., and Weech, S. Impact of mercury on fish-eating birds and mammals in freshwater environments. Human Ecol. Risk Assess. In press. Chen, Y.-W., Belzile, N., and Gunn, J. M . (2001). Antagonistic effect of selenium on mercury assimilation by fish populations near Sudbury metal smelters? Limnol. Oceanogr. 46, 1814-1818. Clarkson, T. W. (1992). Mercury: Major issues in environmental health. Environ. Health Perspectives 100, 31-38. DesGranges, J.-L., Rodrigue, J., and Laperle, M . (1998). Mercury accumulation and biomagnification in Ospreys (Pandion haliaetus) in the James Bay and Hudson Bay Regions of Quebec. Arch. Environ. Contam. Toxicol. 35, 330-341. Dorea, J. G., Moreira, M . B., East, G., and Barbosa, A. C. (1998). Selenium and mercury concentrations in some fish species of the Madeira River, Amazon Basin, Brazil. Biol. Trace Elem. Res. 65, 211-220. Dryssen, D., and Wedborg, M . (1991). The sulphur-mercury (II) system in natural waters. Water Air Soil Pollut. 56, 507-519. Engler, R. M . , and Patrick Jr, W. H. (1975). Stability of sulfides of manganese, iron, zinc, copper, and mercury in flooded and nonflooded soil. Soil Sci. 119, 217-221. Evers, D. C , Kaplan, J. D., Meyer, M . W., Reaman, P. S., Braselton, W. E., Major, A., Burgess, N . , and Scheuhammer, A. M . (1998). Geographic trend in mercury measured in common loon feathers and blood. Environ. Toxicol. Chem. 17, 173-183. Evers, D. C , Taylor, K. M . , Major, A., Taylor, R. J., Poppenga, R. and Scheuhammer, A. M . Common loon eggs as indicators of methylmercury availability in North America. Ecotoxicol. In press. EVS Environment Consultants, Norecol Dames and Moore, and Frontier Geosciences. (1999). Pinchi Lake Mine Site and Lake Investigation: Environmental Assessment Report (Vancouver: Cominco Ltd.). EVS Environment Consultants. (2001). Regional survey of fish mercury concentrations -Pinchi Lake, B.C. (Vancouver: TeckCominco Ltd.), pp. 91. 19 Fimreite, N . (1971). Effects of dietary methylmercury on ring-necked pheasants. Can. Wildl. Serv. Occas. Paper 9, 39pp. Fimreite, N. , Holsworth, W. N. , Keith, J. A., Pearce, P. A., and Gruchy, I. M . (1971). Mercury in fish and fish-eating birds near sites of industrial contamination in Canada. Can. Field-Nat. 55,211-220. Fitzgerald, W. F., Engstrom, D. R., Mason, R. P., and Nater, E. A. (1998). The case for atmospheric mercury contamination in remote areas. Environ. Sci. Technol. 32, 1-7. Garrett, C. L., MacLeod, L. A., and Sneddon, H. J. (1980). Mercury in the British Columbia and Yukon environments - Summary of data to January 1, 1979: Environment Canada, Environmental Protection Service, Pacific Region, pp. 456. Giblin, F. J., and Massaro, E. J. (1973). Pharmacodynamics of methyl mercury in the rainbow trout (Salmo gairdneri): Tissue uptake, distribution, and excretion. Toxicol. Appl. Pharmacol. 24, 81-91. Grieb, T. M . , Driscoll, C. T., Gloss, S. P., Schofield, C. L., Bowie, G. L., and Porcella, D. B. (1990). Factors affecting mercury accumulation in fish in the upper Michigan Peninsula. Environ. Toxicol. Chem. 9, 919-930. Hall, B. D., Bodaly, R. A., Fudge, R. J., Rudd, J. W., and Rosenberg, D. M . (1997). Food as the dominant pathway of methylmercury uptake by fish. Water Air Soil Pollut. 100, 13-24. Heinz, G. (1974). Effects of low dietary levels of methylmercury on mallard reproduction. Bull. Environ. Contam. Toxicol. 11, 386-392. Heinz, G. H. (1979). Methylmercury: Reproductive and behavioral effects on three generations of mallard ducks. J. Wildl. Manage. 43, 394-401. Heinz, G. H., and Hoffman, D. J. (1998). Methylmercury chloride and selenomethionine interactions on health and reproduction in mallards. Environ. Toxicol. Chem. 17, 139-145. Henny, C. J., Hill , E. F., Hoffman, D. J., Spalding, M . G., and Grove, R. A. (2002). Nineteenth century mercury: Hazard to wading birds and cormorants of the Carson River, Nevada. Ecotoxicol. 11, 213-231. Hoffman, D. J., and Moore, J. M . (1979). Teratogenic effects of external egg applications of methyl mercury in the mallard, Anas platyrhynchos. Teratology 20, 453-462. Huckabee, J. W., Goldstein, R. A., Janzen, S. A., and Woock, S. E. (1978). Methylated mercury in brook trout (Salvelinus fontinalis): Absence of an in vivo methylating process. Trans. Am. Fish. Soc. 707, 848-852. 20 Iverfeldt, A. and Lindqvist, O. (1986). Atmospheric oxidation of elemental mercury by ozone in the aqueous phase. Atmos. Environ. 20, 1567-1573. Jackson, T. A. (1987). Methylation, demethylation, and bio-accumulation of mercury in lakes and reservoirs of northern Manitoba, with particular reference to effects of environmental changes caused by the Churchill-Nelson Rivers diversion. In Summary Report, Canada-Manitoba Agreement on the Study and Monitoring of Mercury in the Churchill River Diversion: Governments of Canada and Manitoba. Jackson, T. A. (1988). The mercury problem in recently formed reservoirs of northern Manitoba (Canada): Effects of impoundment and other factors on the production of methyl mercury by microorganisms in sediments. Can. J. Fish. Aquat. Sci. 45, 97-121. Jernelov, A. (1972). Factors in the transformation of mercury to methylmercury. In Environmental Mercury Contamination, R. Hartung and B. D. Dinman, eds. (Ann Arbor, Michigan: Ann Arbor Science Publishers Inc.), pp. 167-172. Johnels, A., Tyler, G., and Westermark, T. (1979). A history of mercury levels in Swedish fauna. Ambio 8, 160-168. Kato, R. (1976). Chromosome breakage associated with organic mercury in Chinese hamster cells in vitro. Mutat. Res. 38, 340-341. Komsta-Szumska, E., Czuba, M . , Reuhl, K. R., and Miller, D. R. (1983). Demethylation and excretion of methyl mercury by the Guinea pig. Environ. Res. 32, 247-257'. Kozie, K. D., and Anderson, R. K. (1991). Productivity, diet, and environmental contaminants in bald eagles nesting near the Wisconsin shoreline of Lake Superior. Arch. Environ. Contam. Toxicol. 20, 41-48. Lathrop, R. C , Rasmussen, P. W., and Knauer, D. R. (1991). Mercury concentrations in walleyes from Wisconsin (USA) lakes. Water Air Soil Pollut. 56, 295-307. Lee, Y. H., Hultberg, H., and Andersson, I. (1985). Catalytic effect of various metal ions on the methylation of mercury in the presence of humic substances. Water Air Soil Pollut. 25, 391-400. Lockhart, W. L., Macdonald, R. W., Outridge, P. M . , Wilkinson, P., DeLaronde, J. B., and Rudd, J. W. M . (2000). Tests of the fidelity of lake sediment core records of mercury deposition to known histories of mercury contamination. Sci. Total. Environ. 260, 171-180. Martin, J., L.K., and Black, M . C. (1998). Biomarker assessment of the effects of coal strip-mine contamination on channel catfish. Ecotoxicol. Environ. Safety 41, 307-320. 21 Mason, R. P., Laporte, J.-M., and Andres, S. (2000). Factors controlling the bioaccumulation of mercury, methylmercury, arsenic, selenium, and cadmium by freshwater invertebrates and fish. Arch. Environ. Contam. Toxicol. 5c?, 283-297. McMurtry, M . J., Wales, D. L., Scheider, W. A., Beggs, G. L., and Dimond, P. E. (1989). Relationship of mercury concentrations in lake trout (Salvelinus namaycush) and small-mouth bass (Micropterus dolomieui) to the physical and chemical characteristics of Ontario lakes. Can. J. Fish. Aquat. Sci. 46, 426-434. Meyer, M . W., Evers, D. C , Hartigan, J. J., and Rasmussen, P. S. (1998). Patterns of common loon (Gavia immef) mercury exposure, reproduction, and survival in Wisconsin, USA. Environ. Toxicol. Chem. 17, 184-190. Miskimmin, B. M . , Rudd, J. W. M. , and Kelly, C. A. (1992). Influence of dissolved organic carbon, pH, and microbial respiration rates on mercury methylation and demethylation in lake water. Can. J. Fish. Aquat. Sci. 49, 17-22. Noble, D. G., and Elliott, J. E. (1990). Levels of contaminants in Canadian raptors, 1966 to 1988; effects and temporal trends. Can. Field-Nat. 104, 222-243. Nocera, J. J., and Taylor, P. D. (1998). In situ behavioral response of common loons associated with elevated mercury (Hg) exposure. Conservation Ecology [online] 2, 1-16. Norheim, G., and Froslie, A. (1978). The degree of methylation and organ distribution of mercury in some birds of prey in Norway. Acta Pharmacol. Toxicol. 43, 196-204. Paquette, K , and Helz, G. (1995). Solubility of cinnabar (red HgS) and implications for mercury speciation in sulfidic waters. Water Air Soil Pollut. 80, 1053-1056. Paulsson, K , and Lundbergh, K. (1989). The selenium method for treatment of lakes for elevated levels of mercury in fish. Sci. Total Environ. 87/88, 495-507. Paulsson, K , and Lundbergh, K. (1991). Treatment of mercury contaminated fish by selenium addition. Water Air Soil Pollut. 56, 833-841. Pennacchioni, A., Marchetti, R., and Gaggino, G. F. (1976). Inability offish to methylate mercuric chloride in vivo. J. Environ. Qual. 5, 451-454. Phillips, G. R., and Buhler, D. R. (1978). The relative contributions of methylmercury from food or water to rainbow trout (Salmo gairdneri) in a controlled laboratory environment. Trans. Am. Fish. Soc. 107, 853-861. Plouffe, A. (1995). Glacial dispersal of mercury from bedrock mineralization along Pinchi fault, north central British Columbia. Water Air Soil Pollut. 80, 1109-1112. 22 Power, M . , Klein, G. M . , Guiguer, K. R. R. A., and Kwan, M . K. H. (2002). Mercury accumulation in the fish community of a sub-Arctic lake in relation to trophic position and carbon sources. J. Appl. Ecol. 39, 819-830. Reid, D. S., and Morley, R. L. (1975). Mercury Contamination of Fish from Pinchi Lake, B.C.: Fish and Wildlife Branch Report, B.C. Ministry of Environment, Lands and Parks, pp. 1-18. Riget, F., Asmund, G., and Aastrup, P. (2000). Mercury in Arctic char (Salvelinus alpinus) populations from Greenland. Sci. Total Environ. 245, 161-172. Robison, S. FL, Cantoni, O., and Costa, M . (1982). Strand breakage and decreased molecular weight of DNA induced by specific metal compounds. Carcinogenesis 3, 657-662. Rudd, J. W. M . , Furutani, A., and Turner, M . A. (1980). Mercury methylation by fish intestinal contents. Appl. Environ. Microbiol. 40, 777-782. Rudd, J. W. M . , Turner, M . A., Furutani, A., Swick, A. L., and Townsend, B. E. (1983). The English-Wabigoon River system: I. A synthesis of recent research with a view towards mercury amelioration. Can. J. Fish. Aquat. Sci. 40, 2206-2217. Scheuhammer, A. M . (1987). The chronic toxicity of aluminum, cadmium, mercury, and lead in birds: A review. Environ. Pollut. 46, 263-295. Scheuhammer, A. M . (1988). Chronic dietary toxicity of methylmercury in the zebra finch, Poephila guttata. Bull. Environ. Contam. Toxicol. 40, 123-130. Scheuhammer, A. M . , Atchison, C. M . , Wong, A. H. K., and Evers, D. C. (1998a). Mercury exposure in breeding common loons (Gavia immer) in central Ontario, Canada. Environ. Toxicol. Chem. 17, 191-196. Scheuhammer, A. M . , Wong, A. H. K., and Bond, D. (1998b). Mercury and selenium accumulation in common loons (Gavia immer) and common mergansers (Mergus merganser) from eastern Canada. Environ. Toxicol. Chem. 17, 197-201. Scheuhammer, A. M . , and Graham, J. E. (1999). The bioaccumulation of mercury in aquatic organisms from two similar lakes with differing pH. Ecotoxicol. 8, 49-56. Sepulveda, M . S., Frederick, P. C , Spalding, M . G., and Williams Jr., G. E. (1999). Mercury contamination in free-ranging great egret nestlings (Ardea albus) from southern Florida, USA. Environ. Toxicol. Chem. 18, 985-992. Shugart, L. (1993). Genotoxic responses in blood. In Nondestructive Biomarkers in Vertebrates, C. M . Fossi, ed.: Lewis Publishers, pp. 131-145. 23 Stafford, C. P., and Haines, T. A. (2001). Mercury contamination and growth rate in two piscivore populations. Environ. Toxicol. Chem. 20, 2099-2101. Sugg, D. W., Chesser, R. K., Brooks, J. A., and Grasman, B. T. (1995). The association of DNA damage to concentrations of mercury and radiocesium in largemouth bass. Environ. Toxicol. Chem. 14, 661-668. Suns, K., and Hitchin, G. (1990). Interrelationships between mercury levels in yearling yellow perch, fish condition and water quality. Water Air Soil Pollut. 50, 255-265. Theodorakis, C. W., D'Surney, S. J., and Shugart, L. R. (1994). Detection of genotoxic insult as DNA strand breaks in fish blood cells by agarose gel electrophoresis. Environ. Toxicol. Chem. 13, 1023-1031. Thompson, D. R., and Furness, R. W. (1989). Comparison of the levels of total and organic mercury in seabird feathers. Mar. Pollut. Bull. 20, 577-579. Thompson, D. R. (1996). Mercury in Birds and Terrestrial Mammals. In Environmental Contaminants in Wildlife - Interpreting Tissue Concentrations, W. N . Beyer, G. H. Heinz and A. W. Redmon-Norwood, eds. (Boca Raton: Lewis Publishers), pp. 341-356. Turner, M . A., and Rudd, J. W. M . (1983). The English-Wabigoon River system: III. Selenium in lake enclosures: It's geochemistry, bioaccumulation, and ability to reduce mercury bioaccumulation. Can. J. Fish. Aquat. Sci. 40, 2228-2240. Watras, C. J., Back, R. C , Halvorsen, S., Hudson, R. J. M . , Morrison, K. A., and Wente, S. P. (1998). Bioaccumulation of mercury in pelagic freshwater food webs. Sci. Total Environ. 219, 183-208. Watson, T. (1992). Evaluation of mercury concentration in selected environmental receptors in the Williston Lake and Peace River areas of British Columbia (Richmond, B.C., Canada: Triton Environmental Consultants Ltd.). Wiener, J. G., Martini, R. E., Sheffy, T. B., and Glass, G. E. (1990). Factors influencing mercury concentrations in walleyes in Northern Wisconsin lakes. Trans. Am. Fish. Soc. 119, 862-870. Wiener, J. G., and Spry, D. J. (1996). Toxicological Significance of Mercury in Freshwater Fish. In Environmental Contaminants in Wildlife - Interpreting Tissue Concentrations, W. N . Beyer, G. H. Heinz and A. W. Redmon-Norwood, eds. (Boca Raton: Lewis Publishers), pp. 297-339. Winfrey, M . R., and Rudd, J. W. M . (1990). Environmental factors affecting the formation of methylmercury in low pH lakes. Environ. Toxicol. Chem. 9, 853-869. 24 Wolfe, M. F., Schwarzbach, S., and Sulaiman, R. A. (1998). Effects of mercury on wildlife: A comprehensive review. Environ.Toxicol. Chem. 77, 146-160. Wood, P. B., White, J. H., Steffer, A., Wood, J. M . , Facemire, C. F., and Percival, H. F. (1996). Mercury concentrations in tissues of Florida bald eagles. J. Wildl. Manage. 60, 178-185. Xun, L., Campbell, N . E. R., and Rudd, J. W. M . (1987). Measurements of specific rates of net methyl mercury production in the water column and surface sediments of acidified and circumneutral lakes. Can. J. Fish. Aquat. Sci. 44, 750-757. Yoneda, S., and Suzuki, K. T. (1997). Detoxification of mercury by selenium by binding of equimolar Hg-Se complex to a specific plasma protein. Toxicol. Appl. Pharmacol. 143, 274-280. 25 CHAPTER 1 - MERCURY AND SELENIUM IN FISH FROM THE PINCHI LAKE REGION OF BRITISH COLUMBIA, CANADA: RELATION TO AGE, FORK LENGTH AND TROPHIC POSITION Introduction Methylmercury (meHg) in fish has long been recognized as a health concern, especially to humans subsisting primarily on a fish diet (Choi, 1989; Clarkson, 1992). Several studies have examined meHg accumulation by fish from industry-related mercury (Hg) releases such as pulp and paper manufacturing and reservoir creation (Fimreite, 1970; Norheim et al, 1986; Jackson, 1987; Hecky et al, 1991). In addition, increased atmospheric deposition of Hg during the industrial period has led to the ubiquitous enrichment of Hg in environments isolated from point sources (Lucotte et al, 1995; Fitzgerald etal, 1998). One source of environmental Hg that has received relatively little attention is that from natural geologic releases. An area along the Pinchi fault in central British Columbia is a known source of natural Hg in the form of cinnabar (Plouffe, 1995). A portion of the north shore of Pinchi Lake was also mined for Hg ore from 1940-1944, with wastes produced during operations deposited directly into Pinchi Lake (EVS Environment Consultants et al., 1999). This area is an ideal location to study the bioaccumulation of meHg in fish species from a natural/mining-related source because lakes with a range of Hg concentrations in sediments are all present in close proximity. Mercury becomes available to the aquatic food chain through bacterial methylation of its inorganic counterpart (Hg2 +) in surficial sediments and the water column (Winfrey and Rudd, 1990). Some portion of inorganic Hg from natural/anthropogenic sources is methylated, and taken up into the food chain (Barr, 1986; Evers et al., 1998; Meyer et al, 1998). Almost all of the Hg present in fish muscle (99%) is in the methylated form (Grieb et al, 1990; Scheuhammer et al, 1998). Mercury from cinnabar (HgS), however, is relatively resistant to oxidation in comparison to other metal sulphides (Engler and Patrick, 1975), and therefore, may not be readily available for methylation. 26 In 1997, EVS Environment Consultants et al. (1999) measured total Hg and meHg concentrations in surface sediments from 4 locations in Pinchi Lake. The samples taken from the eastern end of the lake, upstream of the mine, had 0.3 and 0.55% of total Hg present as meHg (total Hg = 3 and 0.2 pg/g dw, respectively) (EVS Environment Consultants et al., 1999). Samples taken from the western end of the lake, downstream of the mine, had 0.45 and 1.1% of total Hg present as meHg (total Hg = 6 and 11 pg/g dw, respectively), indicating a greater amount of meHg associated with the more contaminated sediments downstream of the mine (EVS Environment Consultants et al., 1999). In the sediments directly associated with mining wastes (calcine) near the mine, the percentage of total Hg present as meHg was much lower (approximately 0.01%); however the total concentration of Hg was greater, ranging from 206 to 493 pg/g dw, resulting in a similar or greater concentration of meHg in sediments (EVS Environment Consultants et al, 1999). Sediments associated with Hg mining wastes in Clear Lake, California were similarly low in the percentage of meHg, ranging from 0.007% closest to the mine to 0.3% further downstream (total Hg = 0.5 - 83 pg/g dw) (Suchanek et al., 1998). In contrast, the percentage of total Hg present as meHg in sediments from several seepage lakes in Wisconsin considered to be "pristine" was 1.6 - 7.9%, where total Hg concentrations in sediments were much lower (range = 0.001 - 0.14 pg/g dw) (Gilmour and Riedel, 1995). Therefore, it is possible that Hg associated with Hg-mining wastes from the processing of cinnabar (HgS), may be present primarily as HgS, which is considered to have very low bioavailablility (Jernelov, 1972; Jackson, 1987; Biester et al, 2000). The elevated Hg concentrations found by EVS Environment Consultants (2001) in large (i.e. 40-80 cm in length) lake trout (Salvelinus namaycush) from Tezzeron Lake, which is situated directly over the Pinchi fault, compared to other control lakes west of the Pinchi fault in the study area (Smart, Francois and Trembleur) suggest that some process must be occurring which produces a form of inorganic Hg from this natural geologic source that can be methylated and taken up into the food chain. One possibility is Hg degassing. This process primarily occurs along fault regions (Chang and Hu, 1989; Azzaria, 1992) 27 where the high vapour pressure of natural Hg allows a release of gas-phase Hg from rocks (Jonasson and Boyle, 1971). This Hg may then be released to the atmosphere, or is oxidized and accumulates bound to organic matter (Kromer et al, 1981; Trost and Bisque, 1972) and thus becomes available for methylation. Alternatively, the very slow equilibrium transfer of HgS to H g 2 + + S2" may produce enough free H g 2 + to allow subsequent formation of meHg. In order for this to occur, the sulphide group of HgS must be oxidized to sulphate. The slow rate of transformation of HgS is due its solubility product (K s p) (Jernelov, 1972; Paquette and Helz, 1995). The value used by Paquette and Helz (1995) in their experiments to determine the speciation of dissolved Hg over a wide range of pH values was K s p = 10"36 7 3 , indicating an extremely low solubility of HgS. While the conversion of HgS to meHg is very slow, it has been shown to occur under laboratory conditions (Jernelov, 1972). A somewhat different process was suggested by Benoit et al. (1999) who noted that methylation of Hg can proceed at low sulphide concentrations partly because HgS is a neutral complex with low volatility that has a high formation constant and relatively high diffusion rate (Dryssen and Wedborg, 1991), allowing low concentrations of HgS to enter bacterial cells. Benoit et al. (1999) suggested that, as environmental sulphide concentrations increase, the speciation of Hg shifts from predominantly neutral HgS 0 to charged HgHS2~ and HgS22" which cannot diffuse across microbial cell membranes and are thus unavailable for methylation. In any case, elevated Hg concentrations, particularly in lake trout from Tezzeron Lake (EVS Environment Consultants, 2001) indicate that some additional Hg is available for uptake into the food chain. The primary route of meHg uptake in fish is through the diet (Hall et al, 1997). Some additional meHg is also obtained through uptake from water passing over the gills, although this is minor in comparison to dietary uptake (Phillips and Buhler, 1978; Wiener and Spry, 1996). Several factors have been shown to affect meHg accumulation in fish. For example, fish from low pH/low alkalinity waters typically have higher concentrations of meHg in muscle tissue compared to those of the same species from higher pH/alkalinity waters (Suns and Hitchin, 1990; Wiener et al, 1990; Mason et al, 2000). 28 Lake temperature has also been positively correlated to Hg-methylation rates and inversely related to Hg-demethylation rates, indicating increased net Hg-methylation with increasing water temperature (Bodaly et al, 1993). Higher levels of dissolved organic carbon (DOC) have also been linked to increased Hg concentrations in lake trout from 91 lakes in Ontario (McMurtry et al, 1989); however Grieb et al. (1990) found a negative correlation between DOC and Hg in yellow perch (Perca flavescens) from 15 seepage lakes in the upper Michigan peninsula and Wisconsin, USA. Selenium availability has also been shown to affect Hg uptake by fish. Treatment of a Swedish lake with sodium selenite resulted in significantly decreased Hg concentrations in fish (Paulsson and Lundbergh, 1989). After a period of only one year, Hg concentrations in perch {Perca fluviatilis) decreased 65-77%, depending on size (Paulsson and Lundbergh, 1989). Similar decreases were also found in perch and northern pike (Esox lucius) from eleven different lakes in Sweden that were treated with Se (Paulsson and Lundbergh, 1991). Chen and Belzile (2001) found a significant inverse relationship between concentrations of Hg and Se in yellow perch and walleye (Stizosedion vitreum) from lakes near the metal smelters of Sudbury, Canada. It was suggested that Se had an antagonistic effect on the assimilation of Hg by fish (Chen and Belzile, 2001). Increasing Se concentrations in prey fish were also shown to reduce the assimilation of Hg (as Hg(N03)2) by northern pike (Turner and Swick, 1983). Analysis of stable nitrogen and/or carbon isotopes is increasingly being used as an analytical tool to verify an organism's trophic position, determine dietary sources of carbon, and to investigate the biomagnification of Hg through the food chain by showing that organisms feeding at a higher trophic position (5N) have higher Hg concentrations in tissues (Cabana and Rasmussen, 1994; Atwell et al, 1998; Bowles et al, 2001; Power et al, 2002). Stable isotopes are being used as an alternative to stomach content analyses because they are a measurement of assimilated food over an extended period of time (DeNiro and Epstein, 1981; Hobson and Clark, 1992; Atwell et al, 1998). Past studies have shown that stable isotope analyses are equally accurate, and sometimes preferable to, prediction of trophic positions of fish using more traditional dietary methods such as 29 examination of published dietary data (Vander Zanden et al., 1997; Vander Zanden et al, 2000). Stable nitrogen isotopes, or the ratio between l 5 N and l 4 N (5N), are now commonly used as an indication of an organism's position in the food chain (Minagawa and Wada, 1984; Estep and Vigg, 1985; Hesslein et al, 1991; Jarman et al, 1996; Gu et al, 1996). An increase of approximately +3.4%o in 5 N is expected during each complete trophic transfer (Minagawa and Wada, 1984). Stable carbon signatures give an indication of the dietary carbon source of heterotrophic 13 12 organisms (DeNiro and Epstein, 1978). The ratio between carbon isotopes C and C (8C) changes little (0.2-l%o) between trophic transfers (DeNiro and Epstein, 1978); however, there is a wide range of carbon isotopic signatures among different primary producers, and the 8 C of animals from freshwater environments fall within the range of 8 C values associated with freshwater primary producers (DeNiro and Epstein, 1978). The range of 8C signatures found in animals is a result of the consumption of different primary producers possessing either the C3 or C 4 photosynthetic pathway (DeNiro and Epstein, 1978). Similarities in 8 C signatures between an organism and a possible dietary carbon source can indicate the relative contribution of that source to the organism's diet (Gu et al, 1996). For example, Gu et al. (1996) used 8 C signatures to determine the relative contribution of various carbon sources to the diet of fish species in Lake Apopka, Florida. It was determined, based on similarities in 8 C , that fish in general (range = -15.5 to -7.9%o) primarily consumed phytoplankton (-13.4%o), and not the cyanobacterium, Microcystis (-3.0%o) or cattails (-27%o) (Gu et al, 1996). Whereas a similar 8C value in fish could also be obtained from a combination of cattails and Microcystis, the 8 N of fish capable of surviving on plants was >5%o which is greater than the 3.4%o enrichment in 8 N for consumers (Gu et al, 1996). Using this combination of data, the primary dietary carbon source of a consumer may be determined. Smaller changes in 8 C of more omnivorous organisms may indicate a preference towards certain typical dietary items; 30 however, in order to properly interpret 5C signatures of an organism it is necessary to know the 5C signatures of all dietary items. To assess the degree of Hg contamination and bioaccumulation in Pinchi Lake compared to surrounding lakes, water chemistry (pH, DOC, alkalinity); total Hg concentrations in sediments; and size, age, Hg, Se, 8N, and SC in several fish species were determined. This study was conducted as part of a larger study examining current Hg exposure in bald eagles and red-necked grebe eggs, higher trophic level predators in the food chain. Materials and Methods Fish Sampling The study area is shown in Figure 1.1. Fish were sampled in July of 2000 and 2001 from the main study lakes: Pinchi, Tezzeron, Stuart, Great Beaver and Fraser. Gill nets were used to capture approximately half of the fish from Pinchi and Tezzeron, and all fish from Stuart in 2000. The remainder of the fish from 2000 and all fish from 2001 were obtained by angling. Rainbow trout (Oncorhynchus mykiss), northern pikeminnow (Ptychocheilus oregonensis), kokanee salmon (Oncorhynchus nerka), and mountain whitefish (Prosopium williamsoni) were obtained in the 2000-sampling season. In 2001, fish sampling was confined to rainbow trout and northern pikeminnow. These two species were chosen because of their ease of capture, relative abundance in the 5 study lakes, differences in dietary habits, and because bald eagles on the study lakes are known to feed on these species (see chapter 3). Fish were killed immediately after capture using either lOOmg/L tricaine methanesulfonate (MS-222, Sigma Chemical Co., St. Louis, MO) dissolved in water, or by a sharp blow to the head. Fish were measured for fork length and stored individually in Ziploc® bags and frozen until return to Vancouver, British Columbia. In some cases, fork length measurements were taken and then only a small 31 portion of the fish was frozen and kept for analysis due to limited storage space in the field. Sediment Sampling A total of 27 sediment samples, 6 each from Tezzeron and Great Beaver, and 5 each from Pinchi, Stuart and Fraser, were collected with an Ekman dredge during May, 2001. Sampling locations were chosen based on presence of an adequate amount of sediment, proximity to fish sampling sites, and known bald eagle/red-necked grebe feeding areas. Sediment sampling locations were spaced across the entire study lake, or portion thereof, in the case of Stuart Lake. The top few centimeters were taken from the surface of each sample and stored in Ziploc® bags. Samples were stored frozen prior to analysis. Water Sampling and Analysis Water samples were collected from the five study lakes on September 20 and 21, 2001. Since these are large lakes, two samples were taken from each lake (or in the case of Staart, 2 samples were taken from the portion of the lake that was studied). Samples were taken from opposite ends of the study lakes, at least 1 foot below the surface of the water and a minimum of 100 metres from shore. Samples were kept cold, but not frozen, until analysis. Analysis of water chemisty [i.e. pH, alkalinity, DOC, etc.] was performed at the Great Lakes Forestry Centre in Sault Ste. Marie, Ontario. Mercury Analysis Fish muscle and sediment samples were analyzed for total Hg at the National Wildlife Research Centre (NWRC) in Hull, Quebec. A l l plastic and glassware was previously acid-washed in dilute (1.5%) nitric acid for a minimum of 8 hours, rinsed with double-deionized water and allowed to air-dry (completely). 32 Approximately 2 grams of dorsal muscle tissue was dissected from each fish and freeze-dried for a minimum of 36 hours. Since it is known that almost all Hg present in fish muscle is meHg (Grieb et al, 1990), analyses of total Hg closely reflected meHg in fish muscle. Samples from 2000 were digested according to Scheuhammer and Bond (1991). Approximately 0.2 grams of freeze-dried fish tissue was first rehydrated using 1 ml of double-deionized water, and initial digestion was allowed to occur overnight at room temperature in 1 ml of nitric acid (Instra-analyzed 70%, JT Baker). The mixture was then heated to 70°C for 1 hour. After cooling, 1 ml sulfuric acid (95-97%, Merck) was added and test tubes were shaken gently to mix contents. Hydrochloric acid (0.5 ml; Instra-analyzed 36-38%o, JT Baker) was then added to the digest, which was heated to 70°C for approximately 2 hours to ensure complete digestion, and allowed to cool overnight. Samples were transferred to 30-ml test tubes and volumes were adjusted to 10 ml with 2 mM potassium dichromate in 3%> hydrochloric acid and mixed thoroughly; and finally, 9.9 ml of 1.5%) hydrochloric acid and 0.1 ml of octanol (an anti-foaming agent) were added. Digests were then set-aside until later analysis using continuous-flow Cold Vapour Atomic Absorption Spectrophotometry (CVAAS) on a Perkin-Elmer 3030B spectrophotometer (PerkinElmer Canada, Inc, Woodbridge, ON) with a Hg hollow cathode lamp, based on Scheuhammer and Bond (1991). A vapour generating accessory (VGA-76, Varian Instruments), an open quartz absorption cell (Varian) and an autosampler (PSC-55, Varian) were adapted for use with the AAS (Scheuhammer and Bond, 1991). In brief, the continuous-flow CVAAS uses a peristaltic pump to deliver sample and reductant, mixed with a pressurized inert gas, into a gas-liquid separator, where liquid is removed. Elemental Hg is then measured in the absorption cell of the spectrophotometer when a steady-state signal is achieved (Scheuhammer and Bond, 1991). Sample flow rate was 7 ml/min, and reductant (25%> SnCb.; Fisher Scientific) flow rate was 1 ml/min. Total sample introduction time was 64 seconds, including a delay time of 60 seconds to reach a steady-state signal, and 4 seconds to read the signal. This was followed by a rinse time of 90 seconds. Ultra high purity argon (43-45 psi) was used as the purge gas (Scheuhammer and Bond, 1991). 33 In 2001, NWRC acquired a dedicated Hg analyzer, so CVAAS was no longer used for Hg analyses. Sediments were freeze-dried and then homogenized using a mortar and pestle. Each sample was passed through a 1-mm sieve to remove stones and other foreign material. Sediment and fish samples from 2001 were analyzed directly using an automated mercury analyzer (AMA-254; Altec Ltd., Canalytical, Burlington, ON). The AMA-254 has a combustion/catalyst tube that decomposes the sample in an oxygen-rich environment and removes interfering elements. A gold amalgamator trap collects all Hg from the evolved gases and a dual-path length cuvette/spectrophotometer determines Hg content. Approximately 25 mg of sediments or 50 mg of freeze-dried fish muscle were loaded into a nickel sample boat (Canalytical, Burlington, ON) and transferred into the AMA-254 for total Hg analysis. Selenium Analysis Fish muscle was analyzed for selenium (Se) in 2002 using Graphite Furnace Atomic Absorption Spectrophotometry (GFAAS). A l l plastic and glassware used throughout the digestion process was previously acid-washed in dilute (1.5%) nitric acid for a minimum of 8 hours, rinsed with double-deionized water and allowed to air-dry completely. Sample digestion and analysis was conducted according to laboratory standard operating procedures. Approximately 0.1 gram of freeze-dried muscle tissue was placed in a plastic tube with 0.5 ml of deionized water and 0.5 ml of nitric acid (Instra-analyzed 70%), JT Baker). The tubes were loosely capped and allowed to sit overnight at room temperature to begin digestion. The following day, all samples were transferred to a dry bath and incubated at 70°C for 1 hour and then at 100°C for 2 hours. This ensured complete digestion of all tissue. Samples were allowed to cool and were then diluted to a total volume of 5 ml using double-deionized water and transferred to clean glass tubes. Selenium was analyzed by GFAAS using an electrodeless discharge lamp with deuterium background correction. The atomization program was based on that of Krynitsky (1987). Nickel (as the nitrate) was used as a matrix modifier to stabilize Se (Carnrick et al, 1983). Calibration blanks were run between each sample to auto-zero the machine and ensure no contamination carried over from previous samples. 34 Fish Aging Fish were aged at the Environment Canada Pacific Biological Station in Nanaimo, British Columbia. Otoliths were dissected from 74 fish (39 from northern pikeminnow, 14 rainbow trout, 10 from mountain whitefish, 11 kokanee salmon), and were the primary source of age data. Northern pikeminnow otoliths were sectioned using a slow-speed rotary saw while mountain whitefish otoliths were simply broken in half. The sectioned surface was then burnt over an ethanol burner, covered in oil and viewed under a microscope (Leica Microsystems Inc., Richmond Hill , ON). Age was determined according to methods described in MacLellan (1997). Twenty-five of the salmonids (11 kokanee and 14 rainbow trout) were aged using both otoliths and scales. Another 36 rainbow trout for which otoliths were not available (i.e. heads were discarded earlier in the study) were aged using scales (see below). Salmonid otoliths were aged by microscopic examination of growth rings on the intact surface (Chilton and Beamish, 1982). Aging by scales was determined by counting the number of compressed growth regions, or rings (Jearld, 1983). Five scales were taken from the area just dorsal of the lateral line and fixed into scale books. Acetate impressions were obtained for each scale book at 90 degrees Celsius and 5000 psi for 1.5 minutes. Scale impressions were viewed using a Neo Promar projection microscope (Wild Leitz GmbH, Wetzlar, Germany). Stable Nitrogen (57V) and Carbon (bC) Analysis Dorsal muscle tissue from 90 fish, including 42 northern pikeminnow, 28 rainbow trout, 11 kokanee and 9 mountain whitefish, was freeze-dried and stored in sterile 5 ml plastic tubes prior to analysis. Approximately 1 mg of freeze-dried fish muscle was sealed into a 5x9 mm tin capsule (Costech Analytical Technologies, Inc., Valencia, CA, USA) for stable isotope analyses. 35 Samples were analyzed for concurrent stable nitrogen and carbon isotopes at the Stable Isotope Facility, the University of California, Davis. Stable isotope ratios of carbon and nitrogen were measured by continuous flow isotope ratio mass spectrometry (CFIRMS) using a 20-20 mass spectrometer (PDZEuropa, Sandbach, UK). After sample combustion to C 0 2 and N 2 at 1000 °C in an on-line elemental analyzer (ANCA-GSL, PDZEuropa), the gases were separated on a Carbosieve G column (Supelco, Bellefonte, PA, USA) before introduction to the isotope ratio mass spectrometer (IRMS). Sample isotope ratios were compared to those of standard gases injected directly into the IRMS before and after the sample peaks, and 8 1 5 N and 8 1 3C values were calculated. The standards used were carbonate rock from the Pee Dee Belemnite (PDB) formation for 5C (Craig, 1957), and atmospheric nitrogen gas (AIR) for 5N (Mariotti, 1983). Both standards are assigned 8-values of 0.0%o by convention. For 13CPDB, the gas was calibrated against Hydrocarbon Oil - NBS 22 (National Institute of Standards and Technology - NIST 8539); for 1 5 N a i r the gas was calibrated against IAEAN1 (International Atomic Energy Agency - N l ; NIST 8547). A l l international standards were obtained from the National Bureau of Standards in Gaithersburg, MD. Stable isotopes of carbon and nitrogen were measured as parts per thousand differences (%o) between the isotope ratio of the sample under examination and a known international standard, and were expressed as either 8C or 8N, respectively. The formula used to calculate 8N or 8C is: 8N = [(Nsample-Nstandard)/Nstandard] X 1000 In the case of nitrogen, N in the above equation represents the ratio of 1 5 N / 1 4 N in the sample or the standard, whereas C would represent the ratio of 1 3 C to 1 2 C. 36 Quality Assurance Hg and Se analyses Blanks, Hg standards (0.01 and 0.1 ug/ml Hg), and certified reference materials from the National Research Council of Canada (NRCC) (Dogfish liver - Dolt-2 and Dogfish muscle - Dorm-2) were run prior to fish Hg sample analysis for calibration purposes and to check instrument sensitivity. Duplicate samples and additional standards and certified reference materials were also checked throughout the Hg analysis. Al l certified reference materials were recovered at +/- 10% of the certified value (93.9 - 103.8% for Dolt-2; 96.9 - 102.2%) for Dorm-2). Sediment standards used for Hg analysis were obtained from the Northern Contaminants Program Interlaboratory Study on the analysis of trace metals in sediments. Samples used were SC-1 - Lake St. Clair reference sediment (100.5% recovery), and WQB-3 - Lake Ontario sediment C R M (106.2% recovery). Dolt-2 and Dorm-2 certified reference materials from NRCC and and Oyster-15 66 from the National Bureau of Standards were digested and analyzed prior to and during Se analysis of fish tissues. Recoveries were 97.5 - 103.2% for Dolt-2, 95.0 - 101.8%) for Dorm-2, and 99.4 - 104.8% for Oyster-1566. Blanks, Se standards (0.25 and 0.5) and digest blanks were also run during Se analyses to ensure accuracy of data. Stable Isotopes Ammonium sulfate (5N = 1.33%o) and sucrose (5C = -23.83%o) were used as check standards for nitrogen and carbon, respectively. Check standards were analyzed twice, before and after each set of samples, as well as after every 12 samples, to determine the accuracy of 5 1 5 N and 5 1 3C measurements. Mean values ± one standard deviation for repeat measures of the standards were 5N = 1.33 ± 0.09%o and 5C = -23.83 ± 0.02%o. 37 Statistical Analyses All data were tested for normality prior to performing statistical analyses and were In-transformed when not normal. If data could not be normalized, non-parametric statistics were used. Water quality parameters (pH, DOC, alkalinity) were compared for relatedness using Pearson correlations to determine if lakes with a higher pH also had higher DOC and alkalinity. To test for a statistically significant difference in sediment Hg concentrations from the different study lakes, a one-way analysis of variance (ANOVA) followed by Tukey's test for mean separation was used. Fish were separated by species for analysis. To test whether northern pikeminnow and rainbow trout from Pinchi Lake had higher concentrations of Hg in muscle, least-squares analysis of covariance (ANCOVA) was used, with fork length as a covariate. Analysis of Hg in pikeminnow and rainbow trout in relation to age was also conducted using ANCOVA. To ensure consistency in trophic position and carbon/food source across the study lakes, a one-way ANOVA was conducted for each fish species from each lake. A one-way A N O V A was also performed on the length of fish sampled for stable isotopes to ensure no differences existed among lakes. Tukey's test was used to determine which lakes differed whenever a significant difference was noted in the ANOVA. Variation in Se concentrations across all lakes was tested in a similar manner. T-tests were used to test for differences in mean Hg concentration (one-tailed) and mean Se concentration (two-tailed) between northern pikeminnow and rainbow trout from the same lake. The relationship between Hg and Se in rainbow trout and northern pikeminnow was analyzed using Least Squares ANCOVA to standardize the between lake variation. Additional relationships between age and fork length, 5N and fork length, 5N and Hg, and 5C and Hg in rainbow trout and pikeminnow were tested using the same ANCOVA model. Forward stepwise regression was used to determine which parameters (pH, DOC, alkalinity, sediment Hg, length, age, Se, 8N, 5C) best-predicted Hg concentrations in northern pikeminnow and rainbow trout. T-tests were used to determine if mountain whitefish and kokanee from Pinchi Lake differed in length, Hg, Se, 5N or 8C compared to whitefish and kokanee collected from Stuart Lake. Al l statistical analyses were 38 performed using SigmaStat for Windows, version 2.03S (Jandel Scientific, 1995) and JMP, version 4.01 (SAS Institute Inc., 2001). Results Sediments and Water Chemistry Sediment sampling information and analytical results are shown in Table 1.1. A l l results are reported on a dry weight (dw) basis. Sediment sampling locations for Pinchi Lake are depicted in Figure 1.2. Pinchi Lake was found to have a significantly (P = 0.002) higher mean concentration of Hg in sediments (2315 ±2310 pg/kg dw) compared to all other study lakes. The highest concentration of Hg in sediments (5337 pg/kg dw) from Pinchi Lake was found at the sampling location furthest downstream of the mine (Figure 1.2). The sediment sample taken furthest upstream of the mine on Pinchi Lake had a similar Hg concentration to control lakes (48 pg/kg dw). No significant difference in Hg concentrations among reference lake sediments was found; however mean sediment Hg concentrations did show a decreasing trend, with Tezzeron (123 ± 76 pg/kg dw) > Smart (84 ± 97 pg/kg dw) ~ Great Beaver (82 ± 50 pg/kg dw) > Fraser (49 ± 35 pg/kg dw). Two water samples were collected from each study lake in 2001 and results for water quality parameters are presented on Table 1.2. Water chemistry measurements specifically applicable to the study of Hg accumulation in the food chain (pH, alkalinity, DOC) were all significantly correlated. The strongest correlation was found between alkalinity and DOC (r = 0.908, P < 0.001), followed by pH and alkalinity (r = 0.789, P = 0.007), and pH and DOC (r = 0.783, P = 0.007). Based on these results, lakes in the study area with higher pH also have higher alkalinity and DOC. The pH in all lakes was above neutral, with Pinchi Lake having the highest average pH at 7.89. Pinchi Lake also had the highest alkalinity at 1538.2 peq/L, and the second highest DOC (10.69 mg/L). Great Beaver Lake had a slightly higher concentration of DOC than Pinchi Lake (10.91 mg/L). 39 Fish During the summers of 2000 and 2001, 145 fish were caught from the study region including 55 northern pikeminnow, 62 rainbow trout, 15 kokanee salmon and 13 mountain whitefish. A summary of all fish caught, including size, age and total Hg and Se range for each species is listed in Table 1.3 (see appendix 1.1 for raw data). Al l Hg and Se concentrations in fish are reported on a pg/g dw basis. The average moisture content of fish muscle was 80.2% in northern pikeminnow, 76.8% in rainbow trout, 78.8%) in mountain whitefish and 11.1% in kokanee salmon. Age determination using non-regenerative scales proved to be accurate for aging rainbow trout and kokanee. When both otoliths and scales were available for age determination, the age established from the scales and the otoliths matched. Thus we believe age determination using scales alone to be equally accurate as aging using otoliths in the age range of rainbow trout and kokanee sampled in this study (up to 6 years). Mean Hg concentrations in northern pikeminnow from each of the 5 main study lakes are shown on Figure 1.3. ANCOVA revealed that Hg concentrations were significantly different among pikeminnow from the 5 study lakes (P < 0.0001). Specifically, pikeminnow from Pinchi Lake were found to have higher Hg concentrations at a given length than pikeminnow from all other study lakes. Pikeminnow from the reference lakes (Tezzeron, Stuart, Great Beaver and Fraser) did not differ significantly. For a given fork length, mean Hg concentrations in rainbow trout from Pinchi Lake were greater than all other study lakes (P < 0.0001). Rainbow trout from reference lakes also did not differ significantly in mean Hg concentration. Mean Hg concentrations in rainbow trout from each of the five study lakes are shown on Figure 1.4. Mercury concentrations in northern pikeminnow were positively correlated to fork length. The relationship for pikeminnow from Pinchi Lake is shown on Figure 1.5 (r = 0.901, P - 0.014). Positive regressions of Hg concentration on pikeminnow length were also found for Tezzeron (r = 0.843, P = 0.001), Fraser (r = 0.0.741, P < 0.001), Great Beaver (r = 0.880, P < 0.0001) and Stuart (r = 0.894, P = 0.04) (Figure 1.6). A 40 significant relationship was found between length and Hg concentration in rainbow trout from Pinchi Lake (r = 0.666, P = 0.005; Figure 1.5); however this same relationship was not found in rainbow trout from individual reference lakes. Since 'lake' was not a significant factor affecting length of reference lake rainbow trout, data from all reference lakes were pooled to determine if a significant relationship could be found between length and Hg concentration. No significant relationship was found between Hg and length in the pooled data set (P = 0.177). Pikeminnow from Pinchi Lake were also found to have higher Hg concentrations at a given age than pikeminnow from Tezzeron, Fraser and Great Beaver Lakes (P < 0.0001). Age data were not available for pikeminnow from Stuart Lake. Again, reference lakes did not differ significantly. Significant positive regressions of Hg on age were found for pikeminnow from Pinchi (r = 0.953, P = 0.003), Tezzeron (r = 0.930, P < 0.001), Fraser (V = 0.698, P = 0.012) and Great Beaver (r = 0.830, P < 0.001) Lakes. Within the same age, rainbow trout from Pinchi Lake also had a higher mean Hg content compared to reference lakes (P = 0.007). No significant difference was found in Hg concentrations among rainbow trout from reference lakes. Since 'lake' was not a significant factor affecting age of reference lake rainbow trout, data from all reference lakes were pooled to determine if a significant relationship could be found between age and Hg concentration. A significant relationship was found between Hg and age in rainbow trout from the reference lakes (r = 0.410, P = 0.007; Figure 1.7). No relationship was found between age and Hg concentration in rainbow trout from Pinchi Lake (Figure 1.7). Significant positive relationships were found between age and length for northern pikeminnow (r = 0.901, P < 0.0001) and rainbow trout (r = 0.733, P < 0.0001). ANCOVA revealed that the age-length relationship was significantly different among pikeminnow from the 5 study lakes (P = 0.003). Specifically, pikeminnow from Pinchi Lake were found to have a longer fork length at a given age than pikeminnow from Great Beaver Lake. No other significant differences were found among pikeminnow from the 5 study lakes, and 'lake' was not a significant factor in the rainbow trout age-length relationship. Age-length relationships for rainbow trout and northern pikeminnow are 41 shown on Figure 1.8. For graphic purposes, age and length values for northern pikeminnow from the different study lakes were combined. Trophic position (as measured by 8N) is another factor that may influence Hg concentrations in fish muscle. Nitrogen isotopic enrichment of approximately +3.4%o in an animal relative to its diet has been shown to occur independent of habitat, form of nitrogen excreted, growth rate and trophic level (Minigawa and Wada, 1984). Based on this, differences in 5 N among fish indicate a difference in trophic position, with fish having a higher 8 N occupying a higher trophic position. In the case of the northern pikeminnow, only fish from Great Beaver (mean 5 N = 10.39%o) were found to be feeding at a higher trophic position when compared to fish from Tezzeron (mean 5 N = 8.88%o, P = 0.007). No significant differences in trophic level were found among pikeminnow from any of the other study lakes. Multiple comparisons testing revealed that rainbow trout collected from Fraser Lake (mean 5 N = 9.85%o) were feeding at a higher trophic position when compared to rainbow trout from Pinchi (mean 5 N = 8.35%o, P = 0.003), Stuart (mean 5 N = 8.40%o, P = 0.032), and Tezzeron (mean 8 N = 8.68%o, P = 0.021) Lakes. No other differences were noted. A significant relationship was found between 8 N and fork length for northern pikeminnow (r = 0.60, P < 0.0001) and rainbow trout (r = 0.473, P = 0.018; Figure 1.9). A relationship between Hg and 8 N was also found for northern pikeminnow (r = 0.650, P < 0.0001) but not rainbow trout (Figure 1.10). It should be noted that the subset of fish measured for 8 N did not differ in length among the 5 study lakes for either northern pikeminnow or rainbow trout. Dietary carbon source (as measured by 8C) may also affect Hg concentrations. Different sources of organic matter (i.e. plants, phytoplankton and zooplankton) within a system have been shown to possess a broad range of carbon isotopes (Gu et al, 1996), and a consumer's 8C is known to closely resemble the 8 C of its diet within l%o (DeNiro and Epstein, 1978). Based on this, northern pikeminnow from Fraser Lake (8C = -25.59%o) 42 were found to be consuming a different (P < 0.001) diet than pikeminnow from Tezzeron (SC = -29.27%o), Pinchi (5C = -28.59%o) and Great Beaver (5C = -27.66%o) Lakes. No other significant differences were noted. In contrast, dietary carbon source (SC) did not differ significantly among rainbow trout from the different study lakes (P = 0.223). No significant relationships were found among Hg concentrations and 5 C in northern pikeminnow or rainbow trout. A significant negative relationship was found between Hg and Se concentrations in northern pikeminnow (r = -0.378, P = 0.014; Figure 1.11) but not in rainbow trout. Selenium concentrations in both northern pikeminnow and rainbow trout were significantly different (P < 0.001) across all study lakes. Multiple comparisons revealed that no differences existed among pikeminnow and rainbow trout from Stuart, Great Beaver and Pinchi Lakes with respect to Se concentration in muscle, whereas, pikeminnow and rainbow trout from Tezzeron Lake were, higher in Se compared to all other lakes and pikeminnow and rainbow trout from Fraser Lake had the lowest concentrations of Se compared to all other lakes. Mean Se concentrations in northern pikeminnow and rainbow trout from the five study lakes are shown on Figures 1.3 and 1.4, respectively. It is also worth noting that mean Se concentrations in northern pikeminnow and rainbow trout from the same lake were not significantly different (P > 0.05), whereas Hg concentrations were always significantly (P < 0.01) higher in northern pikeminnow than rainbow trout from the same lake. Northern pikeminnow and rainbow trout from all lakes were combined to determine which variables were the best predictors of Hg concentrations in dorsal muscle. In addition to fish measurements (Se, 8N , 5 C , fork length and age), sediment Hg and water chemistry parameters (pH, D O C , alkalinity) were added into the regression analyses. Results showed that for northern pikeminnow, sediment Hg, length and 8 C were all shown to add significantly to the regression (r2 = 0.767, P < 0.001). The relationship between Hg in northern pikeminnow and these variables is as follows: ln(Hg) = -8.839 + (0.213 * ln(sed Hg)) + (1.524 * ln(length)) - (0.108 * 8C). 43 For rainbow trout, a combination of sediment Hg and age best predicted total Hg concentrations in dorsal muscle (r2 = 0.349, P < 0.001). The resulting equation is: ln(Hg) = -2.552 + (0.346 * ln(age)) + (0.137 * ln(sed Hg)). Mercury in mountain whitefish collected from Pinchi Lake (mean = 0.48 pg/g dw) was significantly higher than from Stuart Lake (mean = 0.16 pg/g dw; P < 0.001). Whereas these fish from the two lakes did not differ in mean length, they did differ (P = 0.028) in 8 N (Pinchi = 9.05%o, Stuart = 8.14%o), with whitefish from Pinchi Lake feeding at a higher trophic position. Similar results were found comparing kokanee salmon collected from Pinchi and Stuart Lakes. Kokanee from Pinchi Lake had significantly higher (P = 0.006) Hg concentrations in muscle (0.41 pg/g dw) compared to kokanee from Stuart Lake (0.28 pg/g dw). Whereas neither differed in mean length, kokanee from Pinchi Lake (8N = 8.82%o) were feeding at a significantly higher (P = 0.006) trophic position compared to those from Stuart Lake (8N = 7.57%o). Whitefish (8C = -29.64%0) and kokanee (8C = -31.80%o) from Pinchi Lake also differed significantly in food carbon source compared to whitefish (8C = -24.01%o, P < 0.001) and kokanee (SC = -30.25%o, P = 0.005) from Stuart Lake. Discussion Sediments and Water Chemistry In the study region, where natural Hg is elevated in sediments and soils compared to areas removed from the Pinchi fault (Plouffe, 1995; Cook et al, 1996; EVS Environment Consultants et al, 1999), it is likely that much of the natural Hg present is in the form of HgS. HgS was the form of Hg mined at Pinchi Lake (EVS Environment Consultants et al, 1999). In Pinchi Lake specifically, high levels of inorganic Hg in sediments are mostly attributable to 1940-1944 mining activities (EVS Environment Consultants et al, 1999). After ores were roasted to extract Hg, much of the remaining ore rock (calcine) 44 was deposited into Pinchi Lake. Although the exact concentration of Hg remaining in the calcine was not known, it was estimated that approximately 35,150 kg of Hg was deposited in Pinchi Lake (EVS Environment Consultants et al, 1999). Overflowing water with suspended sediments from the ore processing operations was also released into Pinchi Lake and likely contributed to the Hg loading (EVS Environment Consultants etal, 1999). The effect of previous mining operations on Hg concentrations in sediments from Pinchi Lake is evident in the gradient of Hg concentrations in sediments across the lake. Over the 60 years since mining operations resulted in contamination of Pinchi Lake with increased Hg, it was found that significant mixing of contaminated with natural sediments throughout Pinchi Lake, including upstream of the mine, had occurred (EVS Environment Consultants et al, 1999). Mercury concentrations in sediments from Pinchi Lake varied considerably depending on where they were sampled (Figure 1.2). By comparing sediment-sampling (SS) locations across Pinchi Lake with the position of the mine, it is evident that samples collected upstream of the mine have lower Hg concentrations (SS3 and SS4) compared to those near (SS5) or downstream of the mine (SSI and SS2). Therefore, even though Pinchi Lake had significantly higher Hg in sediments compared to reference lakes, the majority of sediments with higher Hg compared to reference lakes are located in the smaller northwestern half of the lake. Water chemistry is another factor that contributes to the bioavailability of Hg for methylation. The amount of meHg produced by microbes in surficial sediments and the water column is inversely related to lake pH (Winfrey and Rudd, 1990). Several studies have shown that fish from low pH/low alkalinity waters have higher concentrations of Hg compared to those of the same species from higher pH waters (Suns and Hitchin, 1990; Wiener et al, 1990; Mason et al, 2000). Xun et al. (1987) showed that net methylation rates are lower in high pH lakes compared to lakes with low pH. While Pinchi Lake was shown to have the highest Hg concentrations in fish, it also had the highest pH. Therefore, the higher pH/alkalinity of Pinchi Lake may contribute to a reduction in net Hg-methylation and subsequent uptake into the food chain. 45 Higher levels of DOC have also been linked to increased fish-Hg concentrations in some species (McMurtry et al, 1989); however, this may not be the case in our study lakes. MeHg associated with DOC is proposed to come from the terrestrial ecosystem as opposed to within-lake methylation (Winfrey and Rudd, 1990). This is supported by the findings of Grieb et al. (1990), which showed a negative correlation between Hg in perch and DOC from seepage lakes, where precipitation was the major water source. Increased DOC has actually been shown to decrease within-lake Hg-methylation, particularly in higher pH lakes (Miskimmin et al, 1992). In lower pH lakes, it has been suggested that acidity causes DOC to release Hg 2 + to the water column, making it available for methylation (Winfrey and Rudd, 1990). Therefore, the high pH and high alkalinity in our study lakes (Pinchi Lake being the highest) may be inhibiting sediment Hg-methylation, and preventing DOC-bound Hg from becoming available for methylation. Fish Previous studies of several species and size ranges of fish sampled from Pinchi Lake and Tezzeron Lake indicated that Hg concentrations in fish were higher than in other lakes from the same region, with Pinchi Lake having the highest concentrations (Fimreite et al, 1971; Reid and Morley, 1975; Watson, 1992; EVS Environment Consultants et al, 1999; EVS Consultants, 2001) (Table 1.4). It should be noted that, with the exception of Pinchi Lake, regular fish sampling for Hg concentrations in muscle tissue has not been conducted in our study area over the past 30 years (particularly Great Beaver and Fraser Lakes). As a result, data (Table 1.4) are often available for only one or two individual fish per species, and may not be indicative of the actual average levels of Hg found in the different species of fish from this region. Apart from EVS Environment Consultants (2001) study of lake trout and lake whitefish from Pinchi, Tezzeron and Stuart Lakes, much of these data were collected over 20 years ago, and may not be representative of the current levels. 46 The results of the current study show that northern pikeminnow and rainbow trout < 40 cm in length from Pinchi Lake are currently higher in Hg compared to those from all other study lakes. For the same size fish, mean Hg concentrations were 2-4 fold greater in northern pikeminnow, and approximately 1.5 times greater in rainbow trout from Pinchi Lake compared to reference lakes. Northern pikeminnow and rainbow trout from all other lakes did not differ significantly in Hg concentration. This suggests that past Hg contamination of Pinchi Lake from mining operations is still causing significantly increased Hg concentrations in fish compared to other lakes in the study area. Fish from Tezzeron Lake had previously been shown to have higher Hg concentrations (Table 1.4; EVS Environment Consultants, 2001), presumably because, like Pinchi Lake, Tezzeron is also situated directly over a portion of the Pinchi fault, a source of natural Hg to the area (Plouffe, 1995). In the current study, no significant difference was found in mean Hg concentrations of northern pikeminnow from the reference lakes; however a decrease from lake to lake was noted with Tezzeron (0.93 pg/g dw) > Great Beaver (0.78 pg/g dw) > Fraser (0.75 pg/g dw) > Smart (0.58 pg/g dw). No decrease in mean Hg concentrations was found for rainbow trout from the reference lakes. This is comparable to EVS Environment Consultants (2001) findings, which showed that lake trout from Tezzeron have higher Hg concentrations compared to control lakes, but lake whitefish from Tezzeron did not differ from control lakes. This suggests that the influence of additional natural Hg in Tezzeron may only be seen in larger fish feeding at a higher trophic position. The lake trout examined by EVS Environment Consultants (2001) ranged from 48.0 to 81.0 cm in length, making them larger than the lake whitefish examined in their study (12.4 - 47.4 cm) and the northern pikeminnow (12.6 - 38.2 cm) and rainbow trout (11.7 - 37.2 cm) examined in our study. Lake trout are also typically piscivorous, with fish > 40 cm known to consume a diet almost entirely of fish (Coad et al, 1995). Northern pikeminnow are also mainly piscivorous (Coad et al, 1995), but unlike the lake trout examined by EVS Environment Consultants (2001), the pikeminnow examined in our study would only have been able to consume small fish given their own small size. Lake whitefish and rainbow trout are more omnivorous, consuming a diet (for example) of plankton, insects, other fish eggs and small fish (Coad et al, 1995). This, in 47 turn, would place them at a lower trophic position compared to piscivorous fish. As shown in this study, northern pikeminnow, which feed at a higher trophic position (see Figure 1.9) have significantly higher Hg concentrations compared to rainbow trout (see Figures 1.3 and 1.4). Similarly, lake trout had higher mean Hg concentrations compared to lake whitefish (EVS Environment Consultants, 2001). Therefore, the larger size and higher trophic position of the lake trout from Tezzeron may have lead to the significantly higher Hg concentrations due to increased natural Hg exposure. Based on this reasoning, similarly increased Hg concentrations in pikeminnow from Tezzeron Lake compared to control lakes might also have been found had larger fish been collected. Stuart Lake is downstream of Pinchi Lake, which may have influenced Hg concentrations in fish in the past, however increased Hg concentrations were not found in fish from this study or EVS Environment Consultants (2001) study of Hg in lake trout and lake whitefish. Lockhart et al. (2000) found that sediment cores from Stuart Lake showed a distinct increase in Hg concentration associated with the 1940-1944 mining operations on Pinchi Lake. However, concentrations of Hg in sediments were shown to reach pre-mining levels in the two sediment cores sometime around 1980 (Lockhart et al, 2000). A portion of the Pinchi fault may reach Great Beaver Lake, but its influence on Hg concentrations is believed to be minor, if any (Plouffe, pers. comm.). Fraser Lake is hydrologically separated from all other study lakes, and is west of the Pinchi fault, the opposite direction of past glacial dispersion of Hg (Plouffe, 1995). Concentrations of Hg in northern pikeminnow muscle were highly correlated to fork length. EVS Environment Consultants (2001) also found significant correlations of size with Hg concentrations in larger lake trout (approximate 40 - 80 cm in fork length) from Pinchi, Tezzeron and Stuart Lakes (Great Beaver and Fraser Lakes were not included in their study). This relationship has also been shown to occur in many other species of fish (Wiener et al, 1990; Lathrop et al, 1991; Scheuhammer and Graham, 1999; Riget et al, 2000; EVS Environment Consultants, 2001). While Hg concentrations in rainbow trout from Pinchi Lake were found to correlate with length, this was not the case for rainbow trout from the reference lakes. Even when all rainbow trout collected from the reference 48 lakes were pooled to produce a larger data set, no relationship between length and Hg concentration was found. The difference between rainbow trout from Pinchi Lake and the reference lakes may be due to growth biodilution of Hg concentrations in rainbow trout from reference lakes. Perhaps as rainbow trout from reference lakes grow larger, the additional Hg accumulated only serves to keep muscle Hg concentrations constant. In contrast, the elevated Hg concentrations in the diet of rainbow trout from Pinchi Lake may result in Hg concentrations in muscle increasing, despite a similar increase in size. Stafford and Haines (2001) have found that biodilution of Hg can occur in fish, and that biodilution can bias contaminant versus fish size relationships as a result. Elimination of meHg may also play a part in the lack of relationship between Hg and length in rainbow trout from reference lakes; however, the rate of meHg elimination is very slow compared to the rate of meHg uptake in fish (Wiener and Spry, 1996). As expected, Hg concentrations in northern pikeminnow from all study lakes were positively correlated to age. Mercury concentrations in rainbow trout from reference lakes were also significantly related to age, but only when fish from all reference lakes were pooled to produce a larger data set. Even then, the relationship only shows a marginal increase in Hg concentration with increasing fish age. No relationship was found between Hg concentration and age of rainbow trout from Pinchi Lake. Sample size, distribution of samples among the age groups, and variability in fish size associated with a single age are likely the reasons no relationship was found in rainbow trout from Pinchi Lake. Of the 14 rainbow trout collected from Pinchi Lake for which ages were obtained, almost all (10) were two or three years of age. Also, because fish age can only be determined to the closest year, a wide range of lengths is often associated with a single age (see raw data in appendix 1.1). For example, 3-year-old rainbow trout from Pinchi Lake ranged in size from 21.2 to 32 cm. Knowing the positive relationship between Hg concentration and length in rainbow trout from Pinchi Lake, this resulted in a wide range of Hg concentrations associated with a single age. Again, in the case of 3-year-old rainbow trout from Pinchi Lake, Hg concentrations ranged from 0.17 pg/g dw in the smallest fish to 0.60 pg/g dw in the largest. Therefore, the variability in data about a single age, the distribution of fish among the age groups, and the small sample size likely 49 contributed to the lack of relationship among Hg concentrations and age in rainbow trout from Pinchi Lake. In the case of rainbow trout from the reference lakes, where Hg concentrations were relatively low, discreet groupings by age illustrated increasing Hg where length did not. This may be because the average Hg concentration for each discreet age group can be shown to increase slightly, unlike with length, where variability was too great to produce a significant relationship. A significant positive relationship was found between length and age for both northern pikeminnow and rainbow trout. This was expected, and supports the relationships between Hg and length and Hg and age in each fish species. Northern pikeminnow from Pinchi Lake had a different age-length relationship compared to pikeminnow from Great Beaver Lake. Specifically, pikeminnow from Pinchi Lake were older at a given length. There are many possible reasons for this difference. Differences in food availability, water temperature, preferred habitat availability and water clarity among lakes could lead to variations in the age-length relationship (Callahan et al, 1995). The age-length relationship in rainbow trout from the various study lakes did not differ significantly, indicating that rainbow trout from all study lakes are a similar length at a given age. Evaluation of Hg concentrations in fish should also consider lake-specific trophic levels (as measured by 8N) for the studied species. Neither rainbow trout nor northern pikeminnow from Pinchi Lake were found to be feeding at a higher trophic position compared to fish from the other study lakes. This indicates that the elevated Hg concentrations found in fish from Pinchi Lake are not simply due to fish feeding at a higher trophic position. As with rainbow trout and northern pikeminnow from the current study, stable nitrogen isotopes have been positively related to the length of gizzard shad (Dorosoma cepedianum) and largemouth bass (Micropterus salmoides) (Gu et al, 1996). Therefore, given the observed correlation between Hg and fish length, particularly in northern pikeminnow, it is reasonable to suggest that 5N would also relate positively to Hg concentrations in fish. Indeed, Hg concentrations in northern pikeminnow were found to 50 positively correlate with 8N in the present study. Significant positive relationships between Hg in fish and 8N were also found in piscivorous fish species from Lake Murray, Papua New Guinea (Bowles et al, 2001), and several fish species from Stewart Lake in eastern Canada (Power et al, 2002), indicating that an increase in trophic position subsequently leads to consumption of prey with increased Hg concentrations by northern pikeminnow. Since length was also correlated to 8N in northern pikeminnow, the increase in fish size can be related to a consumption of larger fish prey, which in mm, have higher Hg concentrations. As discussed in Bowles et al. (2001), these results show that even slight changes in trophic position can be related to increased Hg concentrations. Species-specific trophic levels have also been correlated to Hg concentrations in past studies (Jarman et al, 1996; Atwell et al, 1998; Bearhop et al, 2000). Whereas rainbow trout length was positively correlated to 8N, indicating an increase in trophic position with increasing size, fish-Hg concentrations were not related to stable nitrogen isotopes in rainbow trout. The difference noted between rainbow trout and northern pikeminnow may be a function of both their growth and diet. For example, at a fork length of approximately 38cm, one of our pikeminnow was found to be 18 years of age, while a rainbow trout of the same fork length was 6 years old. Therefore, in the size range examined in the present study, rainbow trout of a given size were much younger than northern pikeminnow. As mentioned previously, a faster growth rate can result in biodilution of Hg concentrations in fish (Stafford and Haines, 2001). Based on this, a dilution of Hg concentrations in rainbow trout muscle relative to the higher Hg concentrations found in a pikeminnow of the same size, may result in a correlation of Hg with 8N in northern pikeminnow but not rainbow trout. The higher Hg concentrations in northern pikeminnow compared to rainbow trout are likely a result of feeding at a higher trophic position. Northern pikeminnow were shown to feed at a significantly higher trophic position when compared with rainbow trout of a similar size (Figure 1.9). This was expected given that northern pikeminnow are piscivorous, known to feed preferentially on smaller salmonids (Coad et al, 1995), whereas rainbow trout will feed on a variety of plankton, crustaceans and insects, and occasionally other fish eggs and 51 small fish (Coad et al, 1995). We suggest that both diet and biodilution resulted in a correlation of length with 5N, but not Hg with 5N in rainbow trout. As mentioned previously, it was expected that fish collected from Tezzeron Lake would have higher concentrations of Hg compared to control lakes due to the presence of the Pinchi fault. In addition, elevated Hg concentrations were found in lake trout collected from Tezzeron Lake compared to control lakes (EVS Environment Consultants, 2001). However, northern pikeminnow and rainbow trout from Tezzeron Lake did not differ significantly in Hg concentration compared to fish from control lakes. Another possible reason for the lack of significantly higher Hg concentrations in fish from Tezzeron Lake could be due to the greater presence of Se. Northern pikeminnow and rainbow trout from Tezzeron Lake had significantly higher concentrations of Se compared to fish from all other study lakes (P < 0.001). In addition, a significant negative relationship was found among Hg and Se concentrations in northern pikeminnow. Turner and Rudd (1983) found that fish Hg concentrations decreased as environmental Se increased. Southworth et al. (2000) found that Hg concentrations in muscle tissue of largemouth bass increased significantly following the elimination of Se-rich fly ash to Rogers Quarry in Tennessee. Over a period of 8 years, Hg concentrations were found to increase linearly from 0.02 to 0.61 pg/g ww in largemouth bass (Southworth et al, 2000). Based on a negative correlation between Hg and Se concentrations in perch and walleye from lakes near a Se source (metal smelter) in Sudbury, Canada, Chen and Belzile (2001) suggested that Se had an antagonistic effect on the assimilation of Hg by fish. A similar relationship was found for northern pikeminnow in the present study. It is not known, however, whether the reduction of Hg in fish exposed to higher levels of Se is a function of lowered absorption of Hg, increased depuration or a combination of the two. Based on previous studies, it is likely a combination of both factors. For example, Turner and Swick (1983) found that assimilation of Hg (as 2 0 3Hg) by northern pike declined, in comparison to controls, with increasing Se concentrations in prey. The findings of Paulsson and Lundbergh (1989) suggest that addition of Se results in increased depuration of Hg in fish because a reduction of 77% in Hg concentration was found in perch (Perca fluviatilis) > 9 years of age, one year after Se addition. Chen and Belzile (2001) also suggest that 52 higher concentrations of Se in water and sediments may inhibit the activity of sulfate-reducing bacteria, which are responsible for Hg-methylation. While a significant negative relationship was found between Hg and Se concentrations in northern pikeminnow examined in this study, no relationship was found between Hg and Se concentrations in rainbow trout. A relationship might not have been found in rainbow trout simply because a wide range in Hg concentrations was not obtained. For example, had pikeminnow having only up to 1.0 pg/g dw Hg been used in the regression (see Figure 1.11), no significant relationship would have been found. Mercury concentrations in rainbow trout ranged from 0.13 to 0.60 pg/g dw. Alternatively, fish that are exposed to higher concentrations of Hg in their diet may have some physiological difference that provides for use of Se in limiting Hg assimilation. To date, it is not known how widespread this relationship is, or what causes it to occur in some fish species and not others, as shown in the inconsistent findings of previous studies examining the relationship between Hg and Se in fish muscle. Dorea et al. (1998) found a significant positive relationship between Hg and Se in herbivorous fish species, but not omnivorous or piscivorous species. Of the 11 species of fish analyzed for total Hg and Se by Burger et al. (2001), a positive correlation between Hg and Se was found for yellow perch and red-breasted sunfish (Lepomis auritus), and a negative relationship between Hg and Se was found for channel catfish (Ictalurus punctatus). No relationship was found between Hg and Se concentrations in any of the other 8 species of fish studied. Chen and Belzile (2001) found significant negative exponential relationships between Hg and Se in perch and walleye. Based on these inconsistent findings, it is not necessarily surprising that no correlation was found between Hg and Se in the muscle of rainbow trout in the current study. In the present study, northern pikeminnow were found to be feeding at a higher trophic position when compared to rainbow trout; however, unlike Hg, concentrations of Se in pikeminnow were not statistically different compared to rainbow trout from the same lake, an observation that applied across all study lakes. Based on this information, it is surprising that both species of fish would have equivalent Se concentrations if Se levels 53 in fish tissue were dependent on dietary Se uptake, as suggested by Turner and Swick (1983). Turner and Swick (1983) found that Se concentrations in northern pike tissue depended almost entirely on the amount of Se present in their diet, as opposed to Se in water. In addition, Dorea et al. (1998) found that Se concentrations in fish increased with increasing trophic level. In contrast, Chen and Belzile (2001) found a significant correlation between total dissolved Se content in lake water and Se concentrations in muscle of perch and walleye. Selenium concentrations were similar in perch and walleye from the same lakes, despite differences in species, size and Hg concentrations (Chen and Belzile, 2001); thus Se concentrations in fish from the Pinchi Lake region may also be primarily a function of Se concentrations in water. However, since Se concentrations in fish prey and water were not determined, this remains speculative. Mountain whitefish and kokanee salmon from Pinchi Lake not only had higher concentrations of Hg compared to whitefish and kokanee from Stuart Lake, they were also feeding at a higher trophic position, on average. Diets of whitefish and kokanee from Pinchi Lake also differed significantly compared to whitefish and kokanee from Stuart Lake, which may have had an affect on Hg bioaccumulation. An obvious difference in diet was found between whitefish from Pinchi Lake (5C = -29.64%o) compared to Stuart Lake (8C = -24.0l%o), while smaller differences were found for kokanee from Pinchi and Stuart Lakes (8C = -31.80%o compared to -30.25%o). Whitefish and kokanee from Pinchi Lake may have higher Hg concentrations because of the concentrations of Hg in Pinchi Lake, in combination with the fact that fish from Pinchi Lake feed at a higher trophic position, reflected by differences in 8N. Taking into account all variables examined in this study, including water chemistry (DOC, pH, alkalinity), sediment-Hg, fish length, age, 8 C , 8 N and Se concentrations, the variables found to best predict Hg in northern pikeminnow included a combination of sediment Hg, fork length and 8C. This relationship explained 76.7% of the variability in Hg concentrations in northern pikeminnow. An almost equally strong relationship remained if 8 C was eliminated from the analysis (r2 = 0.703), indicating that the significance added by this variable is minimal. Since SC was not related to Hg 54 concentrations in northern pikeminnow, it may be coincidence that this variable added significantly to the relationship. For rainbow trout, Hg was best predicted using age and sediment Hg alone. While significant (P < 0.001), this relationship was not as strong as the relationship found for northern pikeminnow, explaining only 34.9% of the variability in rainbow trout Hg concentrations. The relationship for pikeminnow is a stronger relationship because of the strong relationship found between Hg concentration and fork length. In contrast, the Hg-age relationship in rainbow trout was weak, and non-existent in rainbow trout from Pinchi Lake, leading to a weak overall relationship between Hg, age and sediment-Hg for all rainbow trout. Conclusions and Implications Pinchi Lake is similar to Clear Lake, California, which was also contaminated with roasted ore from Hg-mining operations. Intermittent mining activities from 1872-1957 resulted in approximately 100 metric tons of Hg deposited into Clear Lake (Suchanek et al, 1998). Total Hg concentrations in Clear Lake sediments were found to range from approximately 450 pg/g dw close to the mine, to <1.0 pg/g dw over 20 km away (Suchanek et al., 2000). It was determined that the Hg present in Clear Lake was the result of mining activities, and not releases from natural sources such as geothermal springs (Suchanek et al, 2000). Mercury concentrations in fish tissues were also found to be relatively low considering the highly elevated concentrations in sediments (Slotton et al, 1997). For example, adult largemouth bass (Micropterus salmoides), a known piscivorous fish species, had mean muscle Hg concentrations of 0.71 pg/g ww, and carp (Cyprinus carpio) had mean Hg concentrations of 0.15 pg/g ww. It was concluded that while Hg in fish tissue from Clear Lake is elevated, it was significantly lower than might be expected based on Hg concentrations in sediments, and levels are comparable to other lakes, which had not experienced point-source Hg contamination (Slotton et al, 1997). The high pH and high alkalinity of the water characteristic of our study lakes (Pinchi Lake being the highest) may be inhibiting sediment Hg-methylation, and preventing DOC-bound Hg from becoming available for methylation. In conjunction with the fact 55 that much of the Hg in sediments may be present as highly insoluble HgS, this may help to explain why Hg concentrations in sediments from Pinchi Lake have not resulted in highly elevated fish-Hg concentrations. Relationships between Hg concentration and fork length and/or age in northern pikeminnow and rainbow trout indicated bioaccumulation of meHg with increasing size and age. Significant relationships among 5N and fish length, and 5N and Hg in northern pikeminnow specifically, indicated that an increase in trophic position occurs as fish get larger, possibly leading to consumption of larger prey with increased concentrations of Hg. Selenium concentrations in fish varied significantly among the study lakes, but were similar in rainbow trout and northern pikeminnow from the same lake, despite significantly higher Hg concentrations in northern pikeminnow compared to rainbow trout. A significant negative relationship between Hg and Se was also found in northern pikeminnow indicating that Se may have an antagonistic effect on the assimilation of Hg. This study focused mainly on smaller (< 40 cm) forage-type fish that might be consumed by fish-eating birds. Consumption of forage fish having greater than approximately 1.5 pg/g dw Hg (0.3 pg/g ww Hg) has been shown to cause detrimental effects on reproduction in common loons, including reduced egg-laying and decreased nest-site and territory fidelity (Barr, 1986). Dietary concentrations known to cause reproductive impairment are typically one fifth of the dietary level of Hg that would be required to cause significant neurological defects in adult birds (Scheuhammer, 1988). Mercury levels obtained from the smaller fish examined in this study indicate that birds which consume fish < 40cm in fork length are unlikely to encounter concentrations greater than 1.5 pg/g dw Hg in fish, including on Pinchi Lake. It is worth noting that EVS Environment Consultants (2001) also found significant correlations of size with Hg concentrations in larger lake trout (approximate 40 - 80 cm in fork length) from Pinchi, Tezzeron and Stuart Lakes (Great Beaver and Fraser Lakes were not included in their study). In addition, 29 of 31 lake trout (size range 41.7 - 83.1 cm) sampled on Pinchi Lake exceeded the federal human consumption guideline of 0.5 ppm of Hg (EVS Environment Consultants, 2001). Therefore, animals such as river otters or black bears, 56 which tend to consume mainly larger or older fish, may be exposed to dietary meHg concentrations sufficient to cause adverse biological effects. Acknowledgements Field assistance was provided by Lori Smith and Jennifer Young. TeckCominco Ltd., the Metals in the Environment (MITE) fund, and the Canadian Wildlife Service provided financial support for this study. Diana Little and Shayne McClellan at the Pacific Biological Research Station in Nanaimo, British Columbia provided guidance and assistance in aging the fish. Tony Scheuhammer provided laboratory support, and Delia Bond and Ewa Neugebauer provided guidance and assistance at NWRC for digestion and analysis of samples for Hg and Se. Randy Baker and EVS Environment Consultants, Inc. supplied some of the fish used in this study. References Atwell, L., Hobson, K. A., and Welch, H. E. (1998). Biomagnification and bioaccumulation of mercury in an arctic marine food web: Insights from stable nitrogen isotope analysis. Can. J. Fish. Aquat. Sci. 55, 1114-1121. Azzaria, L. M . (1992). Mercury in soil associated with faults, Charlevoix seismic zone, Quebec, Canada. Iraqi Geol. J. 25, 18-28. Barr, J. F. (1986). Population dynamics of the Common loon (Gavia immer) associated with mercury-contaminated waters in northwestern Ontario. Can. Wildl. Serv. Occas. Pap. No. 56, pp. 25. Bearhop, S., Waldron, S., Thompson, D., and Fumess, R. (2000). Bioamplification of mercury in great skua Catharacta skua chicks: The influence of trophic status as determined by stable isotope signatures of blood and feathers. Mar. Pollut. Bull. 40, 181-185. Benoit, J. M . , Mason, R. P., and Gilmour, C. C. (1999). Estimation of mercury-sulfide speciation in sediment pore waters using octanol-water partitioning and implications for availability to methylating bacteria. Environ. Toxicol. Chem. 18, 2138-2141. 57 Biester, H., Gosar, M. , and Covelli, S. (2000). Mercury speciation in sediments affected by dumped mining residues in the drainage area of the Idrija mercury mine. Environ. Sci. Technol. 34, 3330-3336. Bowles, K. C , Apte, S. C , Maher, W. A., Kawei, M . , and Smith, R. (2001). Bioaccumulation and biomagnification of mercury in Lake Murray, Papua New Guinea. Can. J. Fish. Aquat. Sci. 58, 888-897. Burger, J., Gaines, K. F., Boring, C. S., Stephens Jr, W. L., Snodgrass, J., and Gochfeld, M . (2001). Mercury and selenium in fish from the Savannah River: Species, trophic level and locational differences. Environ. Res. Sec. A 87, 108-118. Cabana, G., and Rasmussen, J. B. (1994). Modelling food chain structure and contaminant bioaccumulation using stable nitrogen isotopes. Nature 372, 255-257. Callahan, S. P., Wahl, D. H., and Pierce, C. L. (1995). Factors Affecting Growth of Reservoir Fish. INHS Reports May-June. Carnrick, G. R., Manning, D. C , and Slavin, W. (1983). Determination of selenium in biological materials with platform atomic-absorption spectroscopy and Zeeman background correction. Analyst 70c?, 1297-1312. Chang, F., and Hu, G. (1989). A case history of finding ore deposits by mercury gas survey. J. Geochem. Explor. 33, 145-153. Chen, Y.-W., Belzile, N . , and Gunn, J. M . (2001). Antagonistic effect of selenium on mercury assimilation by fish populations near Sudbury metal smelters? Limnol. Oceanogr. 4(5, 1814-1818. Chilton, D. E., and Beamish, R.J. (1982). Age determination methods for fishes studied by the groundfish program at the Pacific Biological Station. Can. Spec. Publ. Fish. Aquat. Sci. 60, 1-102. Choi, B. H. (1989). The effects of methylmercury on the developing brain. Prog. Neurobiol. 32, 447-470. Clarkson, T. W. (1992). Mercury: Major issues in environmental health. Environ. Health Perspectives 700, 31-38. Coad, B. W., Waszczuk, H., and Labignan, I. (1995). Encyclopedia of Canadian Fishes (Singapore: Canadian Museum of Nature and Canadian Sportfishing Productions, Inc.). Cook, S. J., Jackaman, W., McCurdy, M . W., Day, S. J., and Friske, P. W. (1996). Regional lake sediment and water geochemistry of part of the Fort Fraser map area, British Columbia (NTS 93K/9, 10, 15 and 16): Ministry of Employment and Investment, Energy and Mines Division, Geological Survey Branch. 58 Craig, H. (1957). Isotopic standards for carbon and oxygen and correction factors for mass-spectrometric analysis of carbon dioxide. Geochim. Cosmochim. Acta 12, 133-149. DeNiro, M . J., and Epstein, S. (1978). Influence of diet on the distribution of carbon isotopes in animals. Geochim. Cosmochim. Acta 42, 495-506. DeNiro, M . J., and Epstein, S. (1981). Influence of diet on the distribution of nitrogen isotopes in animals. Geochim. Cosmochim. Acta 45, 341-351. Dorea, J. G., Moreira, M . B., East, G , and Barbosa, A. C. (1998). Selenium and mercury concentrations in some fish species of the Madeira River, Amazon Basin, Brazil. Biol. Trace Elem. Res. 65, 211-220. Dryssen, D., and Wedborg, M . (1991). The sulphur-mercury (II) system in natural waters. Water Air Soil Pollut. 56, 507-519. Engler, R. M . , and Patrick Jr, W. H. (1975). Stability of sulfides of manganese, iron, zinc, copper, and mercury in flooded and nonflooded soil. Soil Sci. 119, 217-221. Estep, M . L. F., and Vigg, S. (1985). Stable carbon and nitrogen isotope tracers of trophic dynamics in natural populations and fisheries of the Lahontan Lake system, Nevada. Can. J. Fish. Aquat. Sci. 42, 1712-1719. Evers, D. C , Kaplan, J. D., Meyer, M . W., Reaman, P. S., Braselton, W. E., Major, A., Burgess, N . , and Scheuhammer, A. M . (1998). Geographic trend in mercury measured in common loon feathers and blood. Environ. Toxicol. Chem. 17, 173-183. EVS Environment Consultants. (2001). Regional survey of fish mercury concentrations -Pinchi Lake, B.C. (Vancouver: Cominco Ltd.) 91pp. EVS Environment Consultants, Norecol Dames and Moore, and Frontier Geosciences. (1999). Pinchi Lake Mine Site and Lake Investigation: Environmental Assessment Report (Vancouver: Cominco Ltd.). Fimreite, N . (1970). Mercury uses in Canada and their possible hazards as sources of mercury contamination. Environ. Pollut. 1, 119-131. Fimreite, N . , Holsworth, W. N . , Keith, J. A., Pearce, P. A., and Gruchy, I. M . (1971). Mercury in fish and fish-eating birds near sites of industrial contamination in Canada. Can. Field-Nat. §5,211-220. Fitzgerald, W. F., Engstrom, D. R., Mason, R. P., and Nater, E. A. (1998). The case for atmospheric mercury contamination in remote areas. Environ. Sci. Technol. 32, 1-7. 59 Gilmour, C. C , and Riedel, G.S. (1995). Measurement of Hg methylation in sediments using high specific-activity 2 0 3 Hg and ambient incubation. Water Air Soil Poll. 80, 747-756. Grieb, T. M . , Driscoll, C. T., Gloss, S. P., Schofield, C. L., Bowie, G. L., and Porcella, D. B. (1990). Factors affecting mercury accumulation in fish in the upper Michigan Peninsula. Environ. Toxicol. Chem. 9, 919-930. Gu, B., Schelske, C. L., and Hoyer, M . V. (1996). Stable isotopes of carbon and nitrogen as indicators of diet and trophic structure of the fish community in a shallow hypereutrophic lake. J. Fish Biol. 49, 1233-1243. Hall, B. D., Bodaly, R. A., Fudge, R. J., Rudd, J. W., and Rosenberg, D. M . (1997). Food as the dominant pathway of methylmercury uptake by fish. Water Air Soil Pollut. 100, 13-24. Hecky, R. E., Ramsey, D. J., Bodaly, R. A., and Strange, N . E. (1991). Increased methylmercury contamination in fish in newly formed freshwater reservoirs. In Advances in Mercury Toxicology, T. Suzuki, N . Imura and T. Clarkson, eds. (New York: Plenum Press), pp. 33-52. Hesslein, R. H., Capel, M . J., Fox, D. E., and Hallard, K. A. (1991). Stable isotopes of sulfur, carbon, and nitrogen as indicators of trophic level and fish migration in the lower Mackenzie River basin. Can. J. Fish. Aquat. Sci. 48, 2258-2265. Hobson, K. A., and Clark, R. G. (1992). Assessing avian diets using stable isotopes I: Turnover of 1 3 C in tissues. Condor 94, 181-188. Jackson, T. A. (1987). Methylation, demethylation, and bio-accumulation of mercury in lakes and reservoirs of northern Manitoba, with particular reference to effects of environmental changes caused by the Churchill-Nelson Rivers diversion. In Summary Report, Canada-Manitoba Agreement on the Study and Monitoring of Mercury in the Churchill River Diversion: Governments of Canada and Manitoba. Jandel Scientific. (1995). SigmaStat Statistical Software, Version 2.03S for Windows 95, NT & 3.1. Jandel Scientific, San Rafael, CA. Jarman, W. M . , Hobson, K. A., Sydeman, W. J., Bacon, C. E., and McLaren, E. B. (1996). Influence of trophic position and feeding location on contaminant levels in the Gulf of the Farallones food web revealed by stable isotope analysis. Environ. Sci. Technol. 30, 654-660. Jearld, A., Jr. (1983). Age Determination. In Fisheries Techniques, L. A. Nielsen and D. L. Johnson, eds. (Bethesda, Maryland: American Fisheries Society), pp. 301-324. 60 Jernelov, A. (1972). Factors in the transformation of mercury to methylmercury. In Environmental Mercury Contamination, R. Hartung and B. D. Dinman, eds. (Ann Arbor, Michigan: Ann Arbor Science Publishers Inc.), pp. 167-172. Jonasson, I. R., and Boyle, R. W. (1971). Geochemistry of mercury. In Proceedings of the Royal Society of Canada Symposium: Mercury in Man's Environment, J. E. Watkin, ed. (Ottawa, Canada), pp. 5-21. Koeman, J. H., Vink, J. A. J., and de Goeij, J. J. M . (1969). Causes of mortality in birds of prey and owls in the Netherlands in the winter of 1968-1969. Ardea 57, 67-76. Kromer, E., Friedrich, G., and Wallner, P. (1981). Mercury and mercury compounds in surface air, soil gas, soils and rocks. J. Geochem. Explor. 15, 51-62. Krynitsky, A. J. (1987). Preparation of biological tissue for determination of arsenic and selenium by graphite furnace atomic absorption spectrophotometry. Anal. Chem. 59, 1884-1886. Lathrop, R. C , Rasmussen, P. W., and Knauer, D. R. (1991). Mercury concentrations in walleyes from Wisconsin (USA) lakes. Water Air Soil Pollut. 56, 295-307. Lockhart, W. L., Macdonald, R. W., Outridge, P. M . , Wilkinson, P., DeLaronde, J. B., and Rudd, J. W. M . (2000). Tests of the fidelity of lake sediment core records of mercury deposition to known histories of mercury contamination. Sci. Total. Environ. 260, 171-180. Lucotte, M. , Mucci, A., Hillaire-Marcel, C , Pichet, P., and Grondin, A. (1995). Anthropogenic mercury enrichment in remote lakes of northern Quebec (Canada). Water Air Soil Pollut. 80, 467-476. MacLellan, S. E. (1997). How to age rockfish (Sebastes) using S. alutus as an example -The otolith burnt section technique. Can. Tech. Rep. Fish. Aquat. Sci. 2146, 1-42. Mariotti, A. (1983). Atmospheric nitrogen is a reliable standard for natural 1 5 N abundance measurements. Nature 303, 685-687. Mason, R. P., Laporte, J.-M., and Andres, S. (2000). Factors controlling the bioaccumulation of mercury, methylmercury, arsenic, selenium, and cadmium by freshwater invertebrates and fish. Arch. Environ. Contam. Toxicol. 38, 283-297. McMurtry, M . J., Wales, D. L., Scheider, W. A , Beggs, G. L., and Dimond, P. E. (1989). Relationship of mercury concentrations in lake trout (Salvelinus namaycush) and small-mouth bass (Micropterus dolomieui) to the physical and chemical characteristics of Ontario lakes. Can. J. Fish. Aquat. Sci. 46, 426-434. 61 Meyer, M . W., Evers, D. C , Hartigan, J. J., and Rasmussen, P. S. (1998). Patterns of common loon (Gavia immer) mercury exposure, reproduction, and survival in Wisconsin, USA. Environ. Toxicol. Chem. 17, 184-190. Minagawa, M . , and Wada, E. (1984). Stepwise enrichment of 1 5 N along food chains: Further evidence and the relation between 8 1 5 N and animal age. Geochim. Cosmochim. Acta 48, 1135-1140. Miskimmin, B. M . , Rudd, J. W. M . , and Kelly, C. A. (1992). Influence of dissolved organic carbon, pH, and microbial respiration rates on mercury methylation and demethylation in lake water. Can. J. Fish. Aquat. Sci. 49, 17-22. Norheim, G., Hastein, T., and Waasjo, E. (1986). Mercury levels in fish in a river habitat after cessation of mercury contamination. Nord. Vet.-Med. 38, 298-302. Paquette, K., and Helz, G. (1995). Solubility of cinnabar (red HgS) and implications for mercury speciation in sulfidic waters. Water Air Soil Pollut. 80, 1053-1056. Paulsson, K., and Lundbergh, K. (1989). The selenium method for treatment of lakes for elevated levels of mercury in fish. Sci. Total Environ. 87/88, 495-507. Paulsson, K., and Lundbergh, K. (1991). Treatment of mercury contaminated fish by selenium addition. Water Air Soil Pollut. 56, 833-841. Phillips, G. R., and Buhler, D. R. (1978). The relative contributions of methylmercury from food or water to rainbow trout (Salmo gairdneri) in a controlled laboratory environment. Trans. Am. Fish. Soc. 707, 853-861. Power, M . , Klein, G. M . , Guiguer, K. R. R. A., and Kwan, M . K. H. (2002). Mercury accumulation in the fish community of a sub-Arctic lake in relation to trophic position and carbon sources. J. Appl. Ecol. 39, 819-830. Plouffe, A. (1995). Glacial dispersal of mercury from bedrock mineralization along Pinchi fault, north central British Columbia. Water Air Soil Pollut. 80, 1109-1112. Reid, D. S., and Morley, R. L. (1975). Mercury Contamination of Fish from Pinchi Lake, B.C.: Fish and Wildlife Branch Report, B.C. Ministry of Environment, Lands and Parks, pp. 1-18. Riget, F., Asmund, G., and Aastrup, P. (2000). Mercury in Arctic char (Salvelinus alpinus) populations from Greenland. Sci. Total Environ. 245, 161-172. SAS Institute, Inc. (2001). JMP, Version 4.01. SAS Institute, Inc., Cary, NC. Scheuhammer, A. M . (1988). Chronic dietary toxicity of methylmercury in the zebra finch, Poephila guttata. Bull. Environ. Contam. Toxicol. 40, 123-130. 62 Scheuhammer, A. M . , and Bond, D. (1991). Factors affecting the determination of total mercury in biological samples by continuous-flow cold vapor atomic absorption spectrophotometry. Biol. Trace Elem. Res. 31, 119-129. Scheuhammer, A. M . , Atchison, C. M . , Wong, A. H. K., and Evers, D. C. (1998). Mercury exposure in breeding common loons (Gavia immer) in central Ontario, Canada. Environ. Toxicol. Chem. 17, 191-196. Scheuhammer, A. M. , and Graham, J. E. (1999). The bioaccumulation of mercury in aquatic organisms from two similar lakes with differing pH. Ecotoxicol. 8, 49-56. Slotton, D. G., Suchanek, T. H., Mullen, L. H. , and Richerson, P. J. (1997). Mercury trends in Clear Lake fishes (Abstract). In First Annual Clear Lake Science and Management Symposium (Lakeport, California). Southworth, G. R., Peterson, M . J., and Ryon, M . G. (2000). Long-term increased bioaccumulation of mercury in largemouth bass follows reduction of waterborne selenium. Chemosphere 41, 1101-1105. Stafford, C. P., and Haines, T. A. (2001). Mercury contamination and growth rate in two piscivore populations. Environ. Toxicol. Chem. 20, 2099-2101. Suchanek, T. H., Mullen, L. H., Lamphere, B. A., Richerson, P. J., Woodmansee, C. E., Slotton, D. G., Hamer, E. J., and Woodward, L. A. (1998). Redistribution of mercury from contaminanted lake sediments of Clear Lake, California. Water Air Soil Pollut. 104, 77-102. Suchanek, T. H., Richerson, P. J., Flanders, J. R., Nelson, D. C , Mullen, L. H., Brister, L. L., and Becker, J. C. (2000). Monitoring inter-annual variability reveals sources of mercury contamination in Clear Lake, California. Environ. Monit. Assess. 64, 299-310. Suns, K., and Hitchin, G. (1990). Interrelationships between mercury levels in yearling yellow perch, fish condition and water quality. Water Air Soil Pollut. 50, 255-265. Trost, P. B., and Bisque, R. E. (1972). Distribution of mercury in residual soils. In Environmental Mercury Contamination, R. Hartung and B. D. Dinman, eds. (Ann Arbor, Michigan: Ann Arbor Science Publishers Inc.), pp. 178-196. Turner, M . A., and Rudd, J. W. M . (1983). The English-Wabigoon River system: III. Selenium in lake enclosures: It's geochemistry, bioaccumulation, and ability to reduce mercury bioaccumulation. Can. J. Fish. Aquat. Sci. 40, 2228-2240. Turner, M . A., and Swick, A. L. (1983). The English-Wabigoon River system: IV. Interaction between mercury and selenium accumulated from waterborne and dietary sources by northern pike (Esox lucius). Can. J. Fish. Aquat. Sci. 40, 2241-2250. 63 Vander Zanden, M . J., Cabana, G., and Rasmussen, J. B. (1997). Comparing trophic position of freshwater fish calculated using stable nitrogen isotope ratios (5 1 5N) and literature data. Can. J. Fish. Aquat. Sci. 54, 1142-1158. Vander Zanden, M . J., Shuter, B. J., Lester, N . P., and Rasmussen, J. B. (2000). Within-and among-population variation in the trophic position of a pelagic predator, lake trout (Salvelinus namaycush). Can. J. Fish. Aquat. Sci. 57, 725-731. Watson, T. (1992). Evaluation of mercury concentration in selected environmental receptors in the Williston Lake and Peace River areas of British Columbia (Richmond, B.C., Canada: Triton Environmental Consultants Ltd.). Wiener, J. G., Martini, R. E., Sheffy, T. B., and Glass, G. E. (1990). Factors influencing mercury concentrations in walleyes in Northern Wisconsin lakes. Trans. Am. Fish. Soc. 119, 862-870. Wiener, J. G., and Spry, D. J. (1996). Toxicological significance of mercury in freshwater fish. In Environmental Contaminants in Wildlife - Interpreting Tissue Concentrations, W. N . Beyer, G. H. Heinz and A. W. Redmon-Norwood, eds. (Boca Raton: Lewis Publishers), pp. 297-339. Winfrey, M . R., and Rudd, J. W. M . (1990). Environmental factors affecting the formation of methylmercury in low pH lakes. Environ. Toxicol. Chem. 9, 853-869. Xun, L., Campbell, N . E. R., and Rudd, J. W. M . (1987). Measurements of specific rates of net methyl mercury production in the water column and surface sediments of acidified and circumneutral lakes. Can. J. Fish. Aquat. Sci. 44, 750-757. 64 Table 1.1 - Mercury concentrations in sediments from Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes, 2001 Lake Sample ID Sediment Depth (feet) Latitude Longitude Hg (ug/kg) dry Fraser SSI 21 54°03'36" N 124°52'57n W 29 Fraser SS2 8 54°04*46" N 124°49'48" W 81 Fraser SS3 6 54°05'02" N 124°48,51" W 34 Fraser SS4 16 54°05*54" N 124°50*29" W 92 Fraser SS5 4 54°03,44"N 124°51'07" W 10 Great Beaver SSI 19 54°23'49" N 123°32'24"W 158 Great Beaver SS2 7 54°23'12"N 123°33'53" W 94 Great Beaver SS3 9 54°25'56" N 123o42'10" W 92 Great Beaver SS4 9 54°25'45" N 123°43'06" W 21 Great Beaver SS5 7 54°25'06"N 123°45'15" W 97 Great Beaver SS6 10 54°27*14"N 123°43'31" W 30 Pinchi SSI 30 54°39'17" N 124°34'02" W 5337 Pinchi SS2 9 54°37'18"N 124°28'29" W 1838 Pinchi SS3 4 54°36'09" N 124°17'15" W 48 Pinchi SS4 14 54°36'09" N 124°19'20" W 330 Pinchi SS5 20 54°37'11"N 124°24'35" W 4022 Tezzeron SSI 14 54°45'50" N 124°32'17" W 72 Tezzeron SS2 10 54°43'38" N 124°33'17" W 87 Tezzeron SS3 14 54°41'33"N 124°27'12" W 116 Tezzeron SS4 9 54°42,50" N 124°24,56" W 56 Tezzeron SS5 10 54°45'28" N 124°35'59" W 142 Tezzeron SS6 5 54°40'04" N 124°21'22" W 264 Stuart SSI 9 54°26'57" N 124°16'56" W 253 Stuart SS2 6 54°26'17" N 124°1732"W 50 Stuart SS3 13 54°25'59"N 124° 19*54" W 34 Stuart SS4 5 54°25'38" N 124°23'46" W 11 Stuart SS5 24 54°28'23" N 124°21'01" W 75 65 CS OA CO u (3 o PC Q, "8 2 u o ID-X) 3 '5b § CU •a 3 cd cu at O xi 3 8 c I o cu At . CO «n «n r— r- >n »n >n >n o\ r-CNCN-—< <—• ~ H co co — — >n oo C N as as VO \ © OS OO OO vi in oo o\ r- oo OiOcococNcNcocN^O^O O N O O C O C O C O C O * © ' ' ' * -~ - H ' — " ~ co vq co co vo O CN co T J - r- o £ m m o <—i lO 00 t o\ oo oo oo oo oo co c© CN CN n O -—1 co ^ in O O CN CN C N V O C N C N O N O O r - C N O a s o o r - - r - ; r - r - \ o r - ~ o o o \ o o o o o o o o o o O \ O \ C ^ C ^ O \ O \ 0 \ 0 \ O \ 0 \ £ £ £ £ £ £ £ £ £ £ oo vo CN CN O O CN CN i n i n i n i n m i n m i n m i n CN CN CO CO O O OO 00 m fi a Oi in in r- r-o o in in CO CO CO CO OS ON. ^ O I U I CO CO VO vo m in co co •S 1? § 00 JS ,2 a* * X> O B, I s 3 ^ CO +5! CO 03 03 •-^ o3 03 5 6 J3 Xi o o c c a! 3 0) CU > > CO CO CU CU CQ CQ CU CU N N 2 (X H 7.8. 9.9 9.2 9.2 2.7 4.9 16.4 10.6 0.5 6.7 CO 51,1 40.9 40.7 38.7 32.5 32.7 422 88.2 25.4 15.0 CO o z CM o 0.005 0.006 0.002 0.008 0.011 0.004 0.017 0.005 0.009 TN 0.42 0.45 0.31 0.30 0.20 0.21 0.75 0.35 0.40 0.41 Colour 45.0 45.0 37.5 40.0 27.5 27.5 27.5 25.0 45.0 37.5 U 5 16.3 16.0 11.4 11.1 8.3 8.4 8.7 8.8 13.7 13.5 DOC 10.7 10.6 8.1 O 1 8.1 6.1 5.9 7.4 6.9 10.9 10.9 u 0.353 0.386 0.142 0.188 0.226 0.227 0.918 0.887 0.251 0.147 CM o 3.38 3.64 6.12 6.06 4.68 4.70 3.66 3.49 2.60 2.93 d 00 3.40 3.35 2.51 2.44 3.62 3.77 5.79 6.14 1.83 1.78 0.97 0.95 0.52 0.53 0.46 0.46 1.05 1.01 0.86 0.88 Location East West (by launch) East West (by lodge) East (town) West (park) Pact West East (by launch) West (middle) Lake Pinchi Pinchi Tezzeron Tezzeron Stuart Stuart Fraser Fraser Great Beaver Great Beaver cu -a ' I CU CO O oo > a c ° S •£ •e s co U .a 'S § _ ^ po S P § £ CD o fi ft c B oo .ti u u M S O y •a -a -o ia b *~ (U . >. > h Z « O O O 3 _j *; cn cn cn O d . .2 .22 .52 "3 -g .-5 Q Q Q U H Z o a, cd o 5 & H 6 < H 'oo ^ ^ " o b p "ob'oT)'^.'^. £ 6 6 6 5 6 6 3 3 y W • § ^ CD £ Q „ o y o oo U Q Q O OH H o Ci 2 CU Q _ < > > 5? •« « 8 g 3 C3 -a -a cu cu > _> "o 3 o cn co cn cn cn Xi & o to 8 o Xi o I cu to Xi a. "3 CO -o cu > ^ U H Q (5 Q "Q Q c a 3 *oi ~oo lo! 3 6 6 6 6 6 03 CO O S 2 ^ 66 Total Se range (ug/g dry wt.) 0.56-1.86 0.73-0.78 0.58-0.98 0.90-0.98 0.69-1.44 0.84-0.97 0.72-0.98 1.37-2.03 1.23-2.04 1.20-1.48 0.86-2.57 0.31-0.62 0.30-0.54 0.94-1.15 0.67 0.41-1.48 0.78 Total Hg range (ug/g dry wt.) 0.16-0.60 0.29-0.49 0.97-5.28 0.43-0.57 0.13-0.28 0.22-0.33 0.27-1.53 0.13-0.23 0.14-0.37 0.17-0.19 0.42-2.61 0.13-0.42 0.17-3.23 0.14-0.31 0.18 0.29-2.57 0.61 Age Range (years) in ^ - ™ rA SO n OS o "1 oo 2 « N a r i b (N OS OS os _ J O Os "3 „ tJ -a . oo 85 S & o at « § cp go fc" w w o n Os Os rt °> >; ^;Ss 2S "« « T3 T3 2S m m Os ^ OS © "1 f~ "1 © ?? ^ £ X 2 ""*„'» 2} a s « s * •4-* V ^ § "53 § fc 3 e 3 fc a u S cu « O Oi O m m o - i «i N n n ro n so so so d d s25(2 I o © rr r-T T o o m © CN t~-o o co oo I m CN o o o © _ o o vi in >o —i O >n VO m CN o oo © o m O V> t~~ CN •n n so d i n oo co © — CN rs o SO ro O r- r- rr —« CN co m —, so co co CN CN CN <—1 co co n ' Os m OS CO rr n >n CN rr rr rr CN CO ro CN rr CN rr ro co CN * 9», j S pJ 1-1 fe m CN co SO SO r- CO SO so >n oo — r ^ a o o > © s o r ^ r - - o o o s s o r ~ r > o o o s OS Os OS OS OS Os © O s O s O s O S O S O s O s O s O s O s a o « m o o m o vi o r ^ r - r - - r - r - t - ~ r - r - r ~ -O S O S O S O s O s O s O s O s O s 68 8 I to & £ S ° o o o o o o o o o 2 2 2 OsOsOsOsOsOsOsOsOs •a S 2< -r os —r _r _r _r _r _? o o o o o o 0 0 0 0 0 0 00 OO 0 0 U> t> ON 0 \ C> O fl (3 01 01 CO B g g t > t > t > t 8 « « ( N N l S C g i i f i t ! <5 CD 4> (B ^ B B B B B B W W O O O O O O CN vs so CN 1 © © o\ ' r~- o •4 o o © o o 1—1 o © o d © © © © © © o © © o o" CN 0 0 VS O CN SO CN O O* © co co v> m sq "1 o r-s © -^s r- r- _ oo —• © t © t~-© © ° © © © © sor^moorrr-rr-^-•*->n- tntNcNso o' © © o' © © CO —i © © co d oo i - i O CN © © CN £ v> os so vs Co n n tsi n g © © © © © § 2 g S vs *^ © © to © © § vs S CN 6 CO _ © © © CO © © CN i © © © vs © © vs so Os I 1 © © © m vs l i CN ^ <= ° co vs co so CN 0 0 CN CN Os —I 0 0 SO — i oo vs cn co co 0 0 ~ . CO rr © so Os co >n © © OS r-> r~ oo • i © © r~ oo o © " oo ° § CN CO CN © © Os SO Os © CN 0 0 rr I I V I so CO r-> cu J e s8 < r- os © —i so so S so vs —< © I*1 CO © © © Os so Os VS (N "8 CO co Os © _ , , , •a CO cu C S' 1 fa o i 60 CU I C c E C 1 1 CQ Q •a O O S O O s O O O \ 0 « O S O s O O s O O s B O s O « r ~ r ~ © r ^ r - r - - t - © r ~ r ^ r ~ t ~ r - - r - i - ~ t ~ r ~ r ~ r ^ OsOs©OsOsOsOs©OsOsOsOsOsOsOsOsOsOsOs s H o o o o o o o o o o o o o©r-r-r-~r-r~-r-r-~r^ ©©OsOsOsOSOsOsOsOs ( S l ( S l r t i H r - ^ < « r t r - « 69 8 o oo Os O 00 Os (N (si tsl fs| N OS Os Os CPs Os Os OS OS Os Os § s § § § CA CA CO VI cd cd 03 c3 cQ £ £ £ £ £ o Os OS - "3 8 * > I CN CN CN Os Os OS OS Os —4 OS —• —< o S § s „ g " " C/2 £2 > 5 a t-~ oo r- t-- © O c o © in in © © © © d rr o r~ i n « o t-i O o —< oo o - . so C N r--o d d • i i OS —« —< O fi —< CN 00 CN ~ CN r f d d i i o —• so CS d • C N o O O O O O o o o o o SO >n O V~l SO T f * n CN O CN CN d d d d r--cs — r-sooot-~ co — ro — r s o © c o d d d d d d d d •a. o o o o oo m r~ m ^ 8 8 O CN T C rr i i O O o o CO f. II A i n S 3 o o I © -a ft o x> 5 CQ -a Os Os Os oo Os r- r~ r- t--Os OS Os OS Os CO .S •s s O O s © s © O s O s © O s r - - c - - © r ~ r > r - ~ © r ~ O s O s O O s O s O S O O s — — CN f* —< fi CN — 70 71 72 • • C/0 § u cu N N CU oo CN 00 — i — i o o (Mp S/Sri) H O I j u . v n i o ) 74 o «> i n o r-H O t N I 1 1 1 — 1 1 (- o vo «n rj- m CN <—i o (Mp S / 3 t i ) S | | 75 IT) o CN § " 8 2 "n 8 3 ? © oo O I •a SO CN s 2 S cu OX) oo SO i < ^e]^ CN 4-* O 2 -_, o a o 1 i 2 o 9 CN " a bp cS eS cu T l 03 1 3 -a | a -ju, t t 1 e ef 1 g I p Jo |JOj 78 79 % Moisture 74.4 74.9 78.9 79.6 78.1 78.1 77.8 78.8 79.4 78.0 79.0 78.7 76.0 76.3 77.1 82.4 79.9 80.0 80.6 79.3 80.9 79.0 79.0 78.0 76.0 77.2 76.9 75.5 79.2 80.0 80.8 81.1 80.6 Selenium (ug/g) Dry 0.67 . 0.73 0.78 0.74 0.78 0.77 n/a n/a 0.97 0.84 0.87 n/a n/a 1.20 1.48 0.78 0.98 0.90 n/a 0.96 0.97 0.94 1.53 2.03 1.43 n/a n/a 1.37 0.32 0.38 0.40 n/a 0.31 o ,—. | j 0.18 0.34 0.43 0.29 0.46 0.47 0.49 0.27 0.32 0.25 0.31 0.33 0.22 0.19 0.17 0.61 0.50 0.46 0.43 0.48 0.57 0.46 0.14 0.14 0.16 0.13 0.16 0.23 0.23 3.23 0.36 0.20 0.38 Fork Length (cm) 27.3 18.4 19.0 17.3 18.8 20.9 22.2 18.8 20.6 21.5 21.5 21.0 18.5 17.7 15.9 24.3 12.0 13.4 13.7 12.8 11.8 14.0 13.5 14.4 11.3 18.8 18.8 23.5 9.8 37.0 14.1 13.2 15.4 on 3 < £ 03 03 03 03 03 0 3 0 3 0 3 0 3 O u -32.61 -31.62 -31.71 -31.23 -32.50 -31.93 n/a n/a -30.83 -30.18 -29.75 n/a n/a -31.19 -31.63 -31.28 -29.75 n/a n/a -29.53 -29.45 -29.83 -23.35 -23.43 -25.03 n/a n/a -24.22 -26.47 -26.91 -25.76 n/a -24.44 o \o § 11.32 8.68 8.33 9.31 8.79 8.97 n/a n/a 7.56 7.10 8.06 n/a n/a 9.37 9.50 10.19 9.24 n/a n/a 8.95 9.21 8.79 7.44 7.84 8.63 n/a n/a 8.63 8.13 11.01 8.24 n/a 9.31 Sample ID GBSl PINSl PINS2 PINS3 PINS4 PINS5 PINS6 STSl STS2 STS3 STS4 STS5 STS6 TEZSl TEZS2 GBMWl PINMWl PINMW2 PINMW3 PINMW4 PINMW5 PINMW6 STMWl STMW2 STMW3 STMW4 STMW5 STMW6 2FRSQ1 2FRSQ2 FRSQl FRSQIO FRSQl 1 Species Kokanee Kokanee Kokanee Kokanee Kokanee Kokanee Kokanee Kokanee Kokanee Kokanee Kokanee Kokanee Kokanee Kokanee Kokanee Mtn Whitefish Mtn Whitefish Mtn Whitefish Mtn Whitefish Mtn Whitefish Mtn Whitefish Mtn Whitefish Mtn Whitefish Mtn Whitefish Mtn Whitefish Mtn Whitefish Mtn Whitefish Mtn Whitefish Pikeminnow Pikeminnow Pikeminnow Pikeminnow Pikeminnow Lake Great Beaver Pinchi Pinchi Pinchi Pinchi Pinchi Pinchi Stuart Stuart Stuart Stuart Stuart Stuart Tezzeron Tezzeron Great Beaver Pinchi Pinchi Pinchi Pinchi Pinchi Pinchi Stuart Stuart Stuart Stuart Stuart Stuart Fraser Fraser Fraser Fraser Fraser Year 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2001-2001 2000 2000 2000 82 v i v i r ~ v s . o s o o — r - - C N © c N c s o o o o o v ) o o r - ~ r - -— o © — — os © o CN —' © — o o \ o \ o i a o > o i o \ o o o o o \ » a 9 i » a c i \ - H H o o o o o o o o o o ^ » o o o o o o o o o o » ^ ^ ^ • ^ ^ ^ ^ o o o o ^ ^ ^ ^ ^ ^ ^ ^ o o o o o o iO ( V rt - h co co rr rr v> co d © ©' © © © -5 8 C O C O d © —< rs CN r- rr o rr v> vs SO so © © © o d d — oo rr v> rr — rr O oo o o r~; © d —'< — d — — m so oo oo o cs OS O OO oo OS Os O O O O O " ^ O l n m M O " t ^ o o o ^ « ! r ^ ^ o o o o ^ o M ™ - • o o l f l ( N | ^ c ^ p ^ l l O M ^ t ^ ^ ^ C O .—< V } co sq rr vs, OO C O rs *^ CS CS rr V I vs SO 00 V I C O SO C O rr rj" C— CS C O C O CS 00 OS OS 0\ © © d © d © © © © © © © © © rs° — — O O O O O O O O O O V N . C N O O © I'll 13 § fc cs © vs. r- so r-rs CN cs rs rs CN rs N O oo — r» oo oo vs N O c o v i r s c o s o co—1 os r- oo oo oo CN CN CS CN rs o f- 00 so — rs O O SO OS o o rs CN CN C O C O CS rr o — cs cs oo - ~5 ~£ O C O (N N O oo O N CS 12 od O OS © V> SO C O O VS OS ^ — d -'< d d rr xo o o rs —; d d m _ 0 0 so rs so -5 . Os so vi os 7Z os' od od od Q "a, 6 cs 00 2 ! 2 2 , £ ; £ . c s c o r r v s s o r - - o o o s X r ^ — — r^l i^i / v » /^l ^vi S—' fc fc fc fc fc fc fc fc fc fc CN CS CN CN CN a o o o o o o o o o fc OH OH O a o c a 1 e g g is £ £ o o o c e c c c c g I E E "s E E E E E E E E E E E E S E E U U S U U U U U I D U I U U N U I U U I U U U U N U U U I D E E E E E E E E 6 M M M M M ^ M M M M J£ ^ J£ M >j ^ ^ ^ J£. M M ^ ^ M M p . o . D . i i L O . B . Q . a . t i i i L D . n . a . i H i n . a . D . t f c Q . f c C L t D . a . i H a i a . a . a> (U > > VH (_I CS CS CS > > CS CS > > > > > > > > c S c S A c S c S c S c S c S t> - H I — , i— P — , P — i n - , ^ QJ qj qj qj qj qj qj qj q ^ f - H i — I P - H P — I P — t o o o o o o o o o o o o o o o CS o o o o o o o o o o o o o — — — — — o o o o o o o o o o — — o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o C N C N C S r S C N C S C N C N C N C N r S r S r S r S r S C N C N C N C N C N C N C N r S C N C N C N C S C S C N C N C ^ 83 % Moisture 79.8 77.6 78.2 79.4 78.7 80.8 78.9 81.2 81.3 82.2 81.2 81.0 80.1 80.7 78.8 81.3 80.7 76.4 76.3 75.8 76.1 77.7 77.4 77.3 76.3 77.2 77.3 78.2 77.4 51.7 77.4 78.2 76.5 Selenium (ug/g) Dry 0.63 0.98 0.95 0.75 0.72 n/a 1.09 2.00 n/a 2.53 2.57 1.82 1.44 1.90 n/a 1.28 0.86 0.45 0.52 0.45 0.62 0.36 0.31 0.43 n/a n/a 0.40 1.04 0.94 n/a 1.15 1.00 0.83 Mercury (ug/g) Dry 1.60 1.16 0.94 1.53 0.93 0.27 2.61 0.88 0.42 0.65 0.54 0.51 0.60 0.77 1.22 0.96 0.87 0.19 0.22 0.17 0.22 0.42 0.13 0.19 0.27 0.38 0.13 0.16 0.20 0.14 0.31 0.27 0.36 Fork Length (cm) 27.2 34.0 31.5 32.8 13.5 38.2 16.9 22.7 15.0 15.1 13.2 13.6 12.6 24.6 27.3 22.0 29.0 30.5 25.7 29.0 33.0 24.5 31.0 27.2 28.1 22.7 25.2 29.7 33.0 31,1 29.2 33.5 Age (years) _ cd d cd cd cd oo ca cd ca est cct ^ c ^ ^ ^ " " S ^ a o O T f , ' ' " ' , 1 T f n n ' S ' S t s ' * 1 ' n ' S ' ' 1 ' , , ' n n O Oi « OO ON rr h O I N « (0 — CN O O rr 00 CN O NO ON 00 oo o \ d d d - > « ^ > » « ' « o \ > ^ CN CN CN CN CN cl CN N m n f l m CN CN CN ( N CN CN CN CN CN CN CN CN CN I I I I I I I I I I I I I I I I I I i I I I I I I 2 CO 9.74 9.25 9.52 9.01 10.46 n/a 10.89 9.66 n/a 8.24 8.37 8.46 8.23 8.31. n/a 9.76 7.97 n/a 9.02 10.53 9.74 10.36 n/a 10.43 n/a n/a 9.03 9.07 9.81 n/a 9.53 9.53 8.96 Sample ID PINSQ4 2STSQ1 2STSQ2 2STSQ3 2STSQ4 STSQl 2TEZSQ1 TEZSQl TEZSQIO TEZSQ2 TEZSQ3 TEZSQ4 TEZSQ5 TEZSQ6 TEZSQ7 TEZSQ8 TEZSQ9 2FRR1 2FRR2 2FRR3 2FRR4 FRRl FRR2 FRR3 FRR4 FRR5 FRR6 2GBR1 2GBR2 GBRl GBR2 GBR3 2PINR1 Species Pikeminnow Pikeminnow Pikeminnow Pikeminnow Pikeminnow Pikeminnow Pikeminnow Pikeminnow Pikeminnow Pikeminnow Pikeminnow Pikeminnow Pikeminnow Pikeminnow Pikeminnow Pikeminnow Pikeminnow Rainbow Rainbow Rainbow Rainbow Rainbow Rainbow Rainbow Rainbow Rainbow Rainbow Rainbow Rainbow Rainbow Rainbow Rainbow Rainbow Lake Pinchi Stuart Stuart Stuart Stuart Stuart Tezzeron Tezzeron Tezzeron Tezzeron Tezzeron Tezzeron Tezzeron Tezzeron Tezzeron Tezzeron Tezzeron Fraser Fraser Fraser Fraser Fraser Fraser Fraser Fraser Fraser Fraser Great Beaver Great Beaver Great Beaver Great Beaver Great Beaver Pinchi Year 2000 2001 2001 2001 2001 2000 2001 2000 2000 2000 2000 2000 2000 2000 2000 2000 2000 2001 2001 2001 2001 2000 2000 2000 2000 2000 2000 2001 2001 2000 2000 2000 2001 84 % Moisture 76.1 77.6 77.5 75.9 76.9 76.7 76.6 77.7 77.6 77.9 77.1 112 78.8 78.5 77.5 77.3 74.7 77.2 77.6 76.2 78.1 76.7 78.1 78.9 77.5 78.1 78.3 76.2 77.6 78.2 76.9 77.4 76.2 Selenium (Ug/g) Dry 1.06 0.88 0.56 0.78 0.79 1.03 0.65 0.72 1.86 0.86 1.04 0.72 0.44 n/a 1.51 1.29 1.01 1.04 1.29 1.15 1.38 0.69 0.93 0.78 0.95 0.86 1.44 1.70 1.23 1.62 2.02 1.76 2.03 u ,—. 0.35 0.30 0.40 0.32 0.60 0.43 0.16 0.17 0.48 0.28 0.29 0.33 0.18 0.23 0.33 0.27 0.14 0.23 0.27 0.26 0.17 0.28 0.23 0.24 0.15 0.13 0.22 0.23 0.18 0.37 0.22 0.21 0.16 Fork Length (cm) 30.2 26.3 28.3 28.9 22.4 17.8 21.2 27.5 25.5 24.5 17.5 16.1 16.5 27.8 30.5 32.8 33.5 26.3 38.5 30.0 20.0 23.2 18.8 19.4 19.2 30.2 31.5 30.0 34.4 33.2 23.0 37.2 Age (years) *\ o u C O n/a n/a n/a -25.52 n/a -29.14 n/a n/a n/a n/a -26.49 -26.81 -28.16 n/a -27.94 n/a n/a n/a n/a n/a n/a -29.53 n/a n/a n/a -27.55 -25.41 n/a n/a n/a n/a -28.78 -26.71 . s o 3? g n/a n/a n/a 7.87 n/a 8.08 n/a n/a n/a n/a 8.46 7.16 8.56 n/a 9.35 n/a n/a n/a n/a n/a n/a 7.28 n/a n/a n/a 9.25 8.66 n/a n/a n/a n/a 8.85 8.73 Sample ID 2PINR10 2PINR11 2PINR12 2PINR2 2PINR3 2PINR4 2PINR5 2PINR6 2PINR7 2PINR8 2PINR9 PINR2 PINR3 PINR4 PINR5 2STR1 2STR10 2STR11 2STR12 2STR2 2STR3 2STR4 2STR5 2STR6 2STR7 2STR8 2STR9 2TEZR1 2TEZR10 2TEZR11 2TEZR12 2TEZR13 2TEZR2 Species l l l i l l i l l i l i l i i l i l i l l i l l i l l l l l l i i .5 .£3 .5 .5 .S .£3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .3 .5 .3 Lake G C C CS C C Lake 'JS xi '£'M xi '£ M 'Jc 'Ja '3s 2 2 2 2 2 2 o o o T J o o o o o o o o o o c S g g g g g g M g g .3 .5 .3 .3 .3 .3 .3 .3 .3 .3 .3 .S .3 .3 .S 3 3 3 3 3 3 3 3 5 3 3 3 S S 3 N N N fcDHB^ftfcOHfclifcftDHfcftfcfcWWWCOWCZlc/lWK , "5 e & •** o a ll es "a o 86 CHAPTER 2 - MERCURY EXPOSURE IN BALD EAGLES (HALIAEETUS LEUCOCEPHALUS) AND RED-NECKED GREBES (PODICEPS GRISGENA) BREEDING NEAR PINCHI LAKE, BRITISH COLUMBIA, CANADA, 2000-2002 Introduction It is well-known that mercury (Hg) originating from specific anthropogenic sources (e.g. chlor-alkali plants) has, in the past, contributed significantly to the contamination of aquatic organisms, resulting in toxicity to piscivorous birds (Barr, 1986; Becker and Bigham, 1995; Facemire et al, 1995). Flooding, especially due to hydro-electric impoundment creation, has also been shown to mobilize Hg and increase methylation leading to elevated levels of Hg in fish and wildlife (Cox et al, 1979; Jackson, 1988; DesGranges et al, 1998). However, it is not yet known whether inorganic Hg from mineralized, natural sources is similarly methylated and biomagnified through the food chain at levels that are hazardous to fish-eating wildlife species. An area along the Pinchi fault in central British Columbia is a known source of geologic Hg in the form of cinnabar (HgS). Various piscivorous birds breed in the area including bald eagles (Haliaeetus leucocephalus), red-necked grebes (Podiceps grisgena), osprey (Pandion haliaetus) and common loons {Gavia immer). Previous studies (Fimreite et al, 1971; Reid and Morley, 1975; Watson, 1992; Cook, 1996; EVS Environment Consultants et al, 1999; EVS Environment Consultants, 2001) and data obtained in this study (see chapter 1) have shown that higher levels of Hg exist in fish and sediments from Pinchi Lake, while lower amounts exist in nearby lakes such as Tezzeron. While some of the Hg in Pinchi Lake can be attributed to past Hg mining operations, natural Hg is also a contributor (EVS Environment Consultants et al, 1999). This area provides an ideal location to study the possible toxic effects of natural and mining-related Hg releases on local piscivorous bird populations because lakes with a range of sediment Hg concentrations are all present in close proximity. Fimreite et al. (1971) found levels of Hg in 3 red-necked grebe livers from Pinchi Lake ranging from 0.45 - 17.40 pg/g ww with a mean of 10.32 pg/g ww. These levels exceeded all other species tested as part of their study of Hg in fish and fish-eating birds 87 near sites of industrial contamination in Canada. Mercury in this range has since been shown to be not uncommon in livers of fish-eating birds (see chapter 5; Auspurger et al, 1998; Pokras et al, 1998; Scheuhammer et al, 1998); however, it does show that there is a case for potentially elevated Hg concentrations in fish-eating birds from Pinchi Lake. Few studies have focused on the accumulation and effects of methylmercury (meHg) on free-living populations of bald eagles or red-necked grebes. Both red-necked grebes and bald eagles feed primarily on fish during the breeding season when an adequate supply is present (Barr, 1986; Stout and Nuechterlein, 1999), and almost all of the Hg found in fish muscle (94 - 99%) is in the methylated form (Grieb et al, 1990; Scheuhammer et al, 1998a). Therefore, piscivorous birds can be at high risk of meHg exposure because of their dependence on a fish diet. While selenium (Se) is known to aid in the detoxification of Hg (Yoneda and Suzuki, 1997), it is not often measured in the blood of birds being examined for Hg contamination. This study examines Hg and Se in blood, and Hg in feathers, of adult bald eagles and eaglets, and assesses the effects of increased Hg on nesting success and productivity in an area contaminated by natural and Hg-mining related wastes. Methods Study Area The study area (Figure 2.1) was chosen based on its proximity to a well-known source of natural Hg in the form of cinnabar. Pinchi Lake was specifically chosen as the treatment lake for this study because of past Hg-contamination associated with Hg-mining operations located adjacent to the lake. Stuart Lake was chosen because of its location downstream of Pinchi Lake and the mine site. However, given the size of Stuart Lake (approximately 7 times larger in total surface area than Pinchi), it is likely that any Hg received from Pinchi has been diluted to an inconsequential amount over the 60 years since Pinchi was contaminated with mining wastes (1940-1944 - EVS Environment Consultants, et al. 1999). Since Stuart Lake is considerably larger (approximately 35,000 hectares), it would have been difficult to accurately study the entire lake given the 88 manpower and time available to complete this project. Therefore, only the southeast portion of the lake was used in this study. Water flows out of Stuart Lake via the Stuart River, located at the southeast corner of Stuart Lake and any inflow from Pinchi Lake may likely have affected this portion of the lake to a greater degree than the portion not studied. Tezzeron Lake was chosen because the Pinchi fault - the source of natural Hg for the area - runs directly beneath it. Tezzeron is also a useful control lake in the sense that it may have an input of natural Hg comparable to Pinchi Lake. Great Beaver and Fraser Lakes were chosen as control lakes, and are hydrologically separated from all other study lakes and each other. A portion of the Pinchi fault may reach Great Beaver Lake, but its influence on Hg concentrations is believed to be minor (Plouffe, pers. comm.). Fraser Lake is hydrologically separated from all other study lakes, and is west of the Pinchi fault, the opposite direction of past glacial dispersion of Hg (Plouffe, 1995). Choice of Indicator Species While several species of fish-eating birds breed in the study area, bald eagles and red-necked grebes were chosen as focal species primarily because of their abundance on the study lakes (especially Pinchi Lake). In addition, bald eagle and red-necked grebe nests in the study area are relatively easy to find and observe in comparison to common loon and osprey nests. Red-necked grebes also lay a larger clutch of eggs compared to common loons (up to 6 eggs), providing a greater window of time for removal of a single egg during the laying process without causing abandonment of the nest by the adult. Close to 25 pairs of common loons were found in the study area, however only one pair consistently nested on Pinchi Lake. In addition, common loons must have chicks to enable capture of adults and young for blood and feather sampling (Evers, 1992). Very few loon pairs in the study area produced chicks, and of those that did, few survived depredation to reach a suitable capture age. Osprey nests were not as plentiful as bald eagle nests in the study area. In addition, nests were considerably further from the study lakes in comparison to bald eagle nests, which were very close to the shoreline. Unlike eagles in the study area, ospreys were not seen consistently feeding on the study lakes. 89 Therefore, osprey blood and feather Hg levels may not have been representative of the lake closest to their nest. Red-Necked Grebe Egg Collection Eggs were collected in 2000 and 2001, as soon as nesting began in late May/early June. An attempt was made to collect only one egg per nest from nests with more than one egg to prevent abandonment of nests by adults. In all cases, only those eggs with little to no nest staining were taken to ensure the eggs were not in an advanced embryonic stage. No eggs were collected from Stuart Lake, as we were unable to locate any nesting sites in 2000 or 2001. Sampling occurred over a period of two years because in 2000, we were unable to find an adequate number of grebe nests on Great Beaver and Tezzeron Lakes. Any loose debris from the nest was removed from the outside of the eggs, which were then packaged separately in Ziploc® bags and frozen until future analysis. In the laboratory, eggs were thawed and homogenized (without the shell) using a T-25 homogenizer (IKA Works Inc., Wilmington, NC). A portion of the homogenized egg was then weighed and freeze-dried for a minimum of 36 hours. Freeze-dried samples were weighed again to obtain the percentage of moisture in each egg. Adult Eagle Capture and Sampling Adult eagles were captured in mid-June of 2001 and 2002, when they would be actively hunting for food to feed growing chicks. Adult eagles were trapped using a floating fish set as described in Cain and Hodges (1989). A collection of small fishes native to the Pinchi Lake area, as well as larger store-bought herring (Clupea harengus) were prepared by cutting a slit lengthwise on the belly, removing the entrails, and replacing them with a piece of pipe foam insulation. The fish were then sewn up and two to four, 30-pound monofilament nooses were sewn into each fish. The fish was then attached to a floating buoy using 100-pound monofilament. 90 As soon as a captured eagle was pulled into the boat, the nooses were cut from the toes and the talons were wrapped securely in a tensor bandage. A hood was placed over the eagle's head and it was taken to shore for sampling. A double-sided Velcro® strap was used to secure the wings to the eagle's body. The bird was first weighed by suspending it from a portable scale (Pesola AG, Baar, Switzerland) attached to the tensor bandage enclosing the talons. The #2 secondary feather from each wing was cut below the base of the vane and placed in a Ziploc® bag for Hg analysis. Approximately 10 ml of blood were taken from the brachial vein using a one-inch, 22-gauge needle. The blood was immediately transferred to sterile heparinized vacutainers, and kept cool and dark. Measurements of hallux claw length, and bill depth, width and length were taken using calipers. Measurements of bill depth and hallux claw length were used to determine the sex of adult birds (Bortolotti, 1984). Positive values obtained using the formula (Bill Depth * 0.392) + (Hallux claw * 0.340) - 27.694 indicate females whereas negative values indicate males (Bortolotti, 1984). Wing chord and tail length were measured using a tape measure. Each bird was then fitted with a U.S. Fish and Wildlife Service number band. When sampling was complete, the eagle was released facing an unobstructed area. We ensured that all eagles were perching normally before leaving the area. At the end of each sampling day, approximately 3 ml of whole blood were transferred to acid-washed tubes and frozen for later Hg and Se analysis at the National Wildlife Research Centre (NWRC) in Hull, Quebec. Eaglet Sampling A l l eaglet sampling occurred within the last few days of June or first few days of July, 2000-2002, inclusive. This proved to be the most appropriate time as most eaglets were at least 8 weeks old and others were close to fledging. Most nests contained two eaglets, and in these cases, only one eaglet was sampled at a time. A qualified tree climber 91 experienced at handling eaglets climbed to each nest and lowered the eaglets to the ground for sampling. Blood samples were taken (using a 21-gauge needle), and birds were measured/banded in the same manner as used for adult eagles. Feathers for Hg analysis were plucked from the chest region and stored in Ziploc® bags. Fresh food remains found in eagle nests from 2001 and 2002 were also examined during eaglet sampling to determine prey items fed to chicks and eaten by adults. Prey items found in nests are discussed in depth in chapter 3. Additional Adult Eagle Blood Samples Several hundred plasma and blood samples from adult bald eagles from British Columbia have been banked at NWRC since 1990. For this study, we selected whole blood samples from bald eagles caught in the wild for previous research, and/or from eagles that were brought into rehabilitation centers. Case histories were examined to ensure that all eagles chosen were blood sampled immediately after capture or admission to a rehabilitation center. Any birds showing signs of Hg exposure were eliminated from the analysis. In total, 46 blood samples obtained over the period of 1993-2000 were analyzed for total Hg. A l l birds were captured or found in coastal areas of mainland British Columbia or Vancouver Island. Mercury Analysis A l l samples were analyzed for total Hg at NWRC in Hull, Quebec. Plastic and glassware used throughout the digestion was previously acid-washed in dilute (1.5%) nitric acid for a minimum of 8 hours, rinsed with double-deionized water and allowed to air-dry completely. Eggs and Blood Approximately 1 g of blood from all eagles and 1 g of homogenized red-necked grebe egg sampled in 2000 was digested and analyzed by continuous-flow cold vapor atomic 92 absorption spectrophotometry (CVAAS) using a Perkin-Elmer 3030B spectrophotometer (PerkinElmer Canada, Inc., Woodbridge, ON) according to Scheuhammer and Bond (1991). In brief, approximately 1 g of fresh tissue (blood or egg) was allowed to digest overnight at room temperature in 1 ml of nitric acid (Instra-analyzed 70%, JT Baker). The mixture was then heated to 70°C for 1 hour. After cooling, 1 ml of sulfuric acid (95-97%, Merck) and 0.5 ml of hydrochloric acid (Instra-analyzed 36-38%, JT Baker) were added to the digest, which was again heated to 70°C for approximately 2 hours to ensure complete digestion. After cooling overnight, samples were transferred to 30-ml test tubes and volumes were adjusted to 10 ml with 2mM potassium dichromate in 3% hydrochloric acid. Finally, 9.9 ml of 1.5% hydrochloric acid and 0.1 ml of octanol (an anti-foaming agent) was added to the solution. Digests were then set-aside until later analysis using CVAAS. In 2001, NWRC acquired a dedicated Hg analyzer, so CVAAS was no longer used for Hg analyses. Blood and eggs obtained during the 2001 and 2002 field seasons, as well as adult eagle blood samples retrieved from the specimen bank, were analyzed directly using an automated mercury analyzer (AMA-254; Altec Ltd., Canalytical, Burlington, ON). The AMA-254 has a combustion/catalyst tube that decomposes the sample in an oxygen-rich environment and removes interfering elements. A gold amalgamator trap collects all Hg from the evolved gases and a dual-path length cuvette/spectrophotometer determines Hg content. Approximately 50 mg of fresh blood or homogenized egg was loaded into a nickel sample boat (Canalytical, Burlington, ON) and transferred into the AMA-254 for total Hg analysis. Feathers Feathers were washed prior to analysis by shaking in acetone for one minute, in dilute (1%) Triton-X 100 for another minute, followed by thorough rinsing in double-deionized water, and allowed to air-dry overnight (Scheuhammer et al, 1998a). Each eaglet feather sample consisted of a pool of approximately 5 breast feathers that were digested together. To enable digestion of the entire adult feather, each was cut into 10 or 11, 1-inch pieces. 93 Washed feathers were placed in a plastic tube with 0.5 ml of deionized water and 1 ml of concentrated nitric acid (Instra-analyzed 70%, JT Baker). The tubes were loosely capped and allowed to sit overnight at room temperature. The next day, all samples were transferred to a dry bath and incubated at 70°C for 1 hour and then at 100°C for another 2 hours to ensure complete digestion. Samples were allowed to cool and were diluted to a total volume of 5 ml using deionized water and transferred to clean glass tubes. After digestion, samples from 2000 were analyzed by CVAAS according to Scheuhammer and Bond (1991). Total Hg analysis for digested feathers obtained in 2001 and 2002 was completed using an AMA-254. For the purposes of this paper, only average Hg concentrations for the entire adult feather are reported. Mercury concentrations throughout the different segments of the adult feathers are reported in chapter 4. Selenium Analysis Blood used for Se analysis was digested in nitric acid in the same manner as 2001 and 2002 eagle feather samples. Approximately 1 g of blood from each eagle was digested. A l l blood samples were analyzed for Se using graphite furnace atomic absorption spectrophotometry (GFAAS) using an electrodeless discharge lamp with deuterium background correction. The atomization program was based on that of Krynitsky (1987). Nickel (as the nitrate) was used as a matrix modifier to stabilize Se (Carnrick et al, 1983). Calibration blanks were run between each sample to auto-zero the machine and ensure no contamination carried over from previous samples. Eagle Reproductive Success A boat survey was conducted in the study region in August of 1999 to ensure the presence of nesting eagles on the study lakes. Just after ice-off (beginning of May) in 2000 and 2001, all known eagle nests were surveyed by boat to check for active territories. Nests were located by surveying shoreline as well as according to a 1996 British Columbia Ministry of the Environment helicopter survey of all active and old 94 eagle and osprey nests. In 2002, an aerial survey of nests in the study area was conducted by helicopter, again, to check for active nesting territories and to ensure that no nests were likely to have been missed in previous surveys. Unfortunately, we were unable to survey nests any earlier, therefore eagles that had attempted to nest prior to the start of May, but failed and were no longer occupying a specific breeding territory, would not have been considered in this study. A l l identified active nesting territories were watched continuously throughout the breeding season for activity. Productivity and overall reproductive success were determined by climbing all nest trees at the end of June/beginning of July when chicks were approximately 8 weeks old. Terminology reviewed by Postupalsky (1974) was used to describe reproductive success and productivity. Quality Assurance Blanks, Hg standards (0.01 and 0.1 ug/ml Hg), and certified reference materials from the National Research Council of Canada (NRCC) [Dogfish liver (Dolt-2) and Dogfish muscle (Dorm-2)] were run prior to sample analysis for calibration purposes and to check instrument accuracy and sensitivity. Duplicate samples and additional standards and certified reference materials were also checked throughout the Hg analysis. A l l certified reference materials were recovered at ± 10% of the certified value. Recoveries are shown on Table 2.1. Dolt-2 and Dorm-2 certified reference materials from the NRCC were also digested and analyzed prior to, and during, Se analysis of eagle blood. Average recoveries for Dolt-2 and Dorm-2 were 97.9% and 98.6%, respectively. Blanks, Se standards (0.25 and 0.5) and digest blanks were also run during Se analyses to ensure accuracy of data. 95 Statistical Analyses Prior to completing any statistical analyses, data were tested for normality and In-transformed to normalize, when required. If data could not be normalized, non-parametric statistical analyses were performed on untransformed data. Paired t-tests were used to determine whether there was a significant difference in blood and feather Hg and blood Se levels in eaglet siblings. To determine whether adult eagles, or eaglets raised on Pinchi Lake had higher blood and feather Hg levels than the other study lakes, a one-way analysis of variance (ANOVA) followed by Tukey's test for mean separation was used. This same analysis was also used to test for differences in Hg concentrations in red-necked grebe eggs from Pinchi Lake compared to the other study lakes. For adult eagles, a two-tailed t-test was performed to test for a difference between Hg concentration in sexes of reference lake birds, as well as between Hg concentrations in blood from reference lake eagles (i.e. adults from Pinchi Lake not included) and Hg results obtained from banked blood samples. Pearson correlations were used to determine whether or not a relationship between blood and feather Hg existed for both adult eagles and eaglets. To test whether Se concentrations in blood differed across the study lakes, a one-way A N O V A was used, followed by the Tukey's test for multiple comparisons whenever a significant difference was noted. The Mann-Whitney rank-sum test was used to test for a difference between adult eagle and eaglet Se concentrations in blood. Least squares analysis of covariance (ANCOVA) was used to test for a relationship between Hg and Se concentrations in adult eagle and eaglet blood. Pearson correlations were used to determine whether a correlation existed between adult eaglet and eaglet Hg or Se levels for birds from the same nests. A two-way A N O V A using year and lake as factors was used to test whether reproductive success and productivity were comparable across all study lakes. The relationship between average Hg levels in eaglet blood and eagle productivity for each lake and year was tested using Pearson correlation. Al l statistical analyses were performed using SigmaStat for Windows, version 2.03S (Jandel Scientific, 1995) and JMP, version 4.01 (SAS Institute Inc., 2001). 96 Results Mercury in Red-necked Grebe Eggs Over the 2000 and 2001 field seasons, 6 eggs were collected from Pinchi Lake, 5 each from Tezzeron and Great Beaver, and 8 from Fraser Lake. Mercury concentrations for all grebe eggs are presented on Table 2.2. Results are reported in pg/g dry weight (dw) for egg Hg concentrations. Average moisture content of all eggs was 78.7%. Mercury levels [mean ± standard deviation (SD)] in eggs obtained from Pinchi = 1.17 ± 0.48 pg/g dw, Tezzeron = 0.83 ±0.17 pg/g dw, Great Beaver = 0.70 ± 0.27 pg/g dw and Fraser = 0.47 ± 0.12 pg/g dw. A significant difference was found in Hg levels in eggs from the four study lakes (P = 0.002); however multiple comparison testing revealed that a statistically significant difference occurred only between eggs from Pinchi and Fraser Lakes (P = 0.001), with eggs from Pinchi Lake having higher mean Hg concentrations. Mercury concentrations in grebe eggs from the four study lakes are illustrated on Figure 2.2. Mercury in Bald Eagles From 2000 to 2002, 43 eaglets were sampled for blood and 41 for feathers, from a total of 25 different nests. Two eaglets were not sampled for feathers because we were unable to obtain an adequate feather sample at the time. Blood and feather Hg concentrations for all eaglets are presented in Tables 2.3 and 2.4, respectively. A l l blood and feather Hg results are reported on a pg/g wet weight (ww) basis. Feather results are reported as wet weight because they were not freeze-dried, only air-dried. Since sibling eaglets did not differ significantly with respect to Hg levels in blood (P = 0.447) or feathers (P = 0.892), sibling results were averaged. Mercury concentrations in eaglet blood from the different study lakes differed significantly (P < 0.001). Specifically, eaglets from Pinchi Lake had significantly higher blood Hg levels compared to eaglets from Stuart, Great Beaver and Fraser Lakes. Blood Hg levels in eaglets from Pinchi and Tezzeron Lakes did not differ significantly (P = 0.359). Eaglets from Pinchi Lake also had significantly higher Hg concentrations in feathers (P < 0.001) compared to all other study lakes, including 97 Tezzeron. Average Hg levels [± standard error (SE)] in eaglet blood from the 5 study lakes are shown on Figure 2.3. In total, thirteen adult bald eagles (5 in 2001 and 8 in 2002; including one bird from Stuart Lake that was caught in 2001 and recaptured in 2002) were captured and sampled. Statistical analyses were performed using the recaptured bird both as two separate samples, and an average for this bird over the two years, but the significance of the results did not change. Consequently, it was decided to keep results separate instead of averaging. Of the 13 birds caught, 3 were from Pinchi, 2 from Tezzeron, 3 from Fraser and 5 from Stuart. Adult eagles from Pinchi Lake had significantly higher concentrations of Hg in blood compared to eagles from Fraser and Stuart Lakes only (P = 0.022). Mercury concentrations in blood of adults from Pinchi and Tezzeron Lakes were not significantly different (P = 0.342). Examination of adult feathers showed that adults from Pinchi Lake had significantly higher Hg concentrations compared to adults from Fraser Lake only (P = 0.032). No other significant differences were noted among Hg levels in adult feathers. It was only possible to test for differences in Hg concentration due to sex by combining control lake adults, because only when combined, was an adequate sample size of both sexes obtained (4 male, 6 female). No statistically significant difference in Hg levels in blood (P = 0.687) or feathers (P = 0.636) was found between sexes. Mercury levels in blood of adult eagles from the four study lakes are shown on Figure 2.4. Mercury concentrations in individual adult eagle blood and feather samples are presented in Tables 2.5 and 2.6, respectively. Mercury concentrations in blood and feathers from eaglets were significantly correlated (r = 0.755, P < 0.0001). The relationship between Hg concentrations in eaglet blood and feathers is shown on Figure 2.5. No correlation existed between Hg levels in adult blood and secondary feathers (P = 0.087). Adult eagles and eaglets from the same nest (siblings averaged) were very highly correlated with one another (r = 0.913, P = 0.004), adults having approximately 14 times greater Hg in blood than chicks, on average (Figure 2.6). 98 Of the 46 previously banked adult blood samples analyzed for total Hg (mean ± SD), 8 were from healthy, wild-caught birds (0.98 ± 0.36 pg/g ww), and the remainder were birds admitted for rehabilitation. Causes for admission of rehab birds included: trauma, such as an eagle attack, electrocution or gun shot wound (1.01 ± 0.48 pg/g ww, n = 15), disease such as aspergillosis (0.91 ± 0.55 pg/g ww, n = 4), exposure to pesticides or lead (1.01 ± 0.70 pg/g ww, n = 13) or undetermined causes (0.92 ± 0.23 pg/g ww, n = 6). Analysis of variance revealed no significant differences in blood-Hg levels from all categories, so results were combined. In total, previously banked adult blood samples from various locations in British Columbia had an average of 0.99 ± 0.50 pg/g ww total Hg and ranged from 0.18 to 2.61 pg/g ww. These levels were significantly lower (P < 0.00]) than Hg concentrations found in the blood of eagles from all lakes in the study area. Selenium in Bald Eagles Eaglet siblings did not differ in Se concentrations (P = 0.790) and so were averaged. Comparison of eaglets across all study lakes revealed significant differences among lakes (P = 0.007). Multiple comparisons showed that eaglets from Stuart Lake had significantly higher Se levels in blood compared to eaglets from Pinchi, Fraser and Great Beaver Lakes. No other significant differences were noted. Average Se concentrations for eaglets from Stuart Lake (± SD) were 0.63 ±0.17 pg/g ww (range - 0.49-0.93 pg/g ww) and all other eaglets combined were 0.45 ± 0.09 pg/g ww (range - 0.26-0.64 pg/g ww). A similar analysis on Se concentrations in adult eagle blood showed no significant differences among study lakes (P = 0.147). Selenium concentrations in adult eagle blood samples were 1.28 ± 0.72 pg/g ww on average and ranged from 0.66 to 2.96 pg/g ww. Adult blood Se levels were significantly greater than chicks (P < 0.001). No correlations were found between blood Hg and Se concentrations in adults; however, a marginally insignificant positive relationship was found in eaglet blood when corrected for differences among lakes (r2 = 0.488, P = 0.0509; Figure 2.7). Average Se concentrations in blood (± SE) for eaglets and adult eagles from the study lakes are compared on Figures 99 2.3 and 2.4, respectively. Selenium concentrations in blood for individual eaglets and adults are presented in Tables 2.3 and 2.5, respectively. Eagle Productivity and Reproductive Success The locations of active bald eagle nests from 2000, 2001 and 2002, including those that failed prior to eaglet sampling, are shown on Figures 2.8, 2.9 and 2.10, respectively. Comparing reproductive success of bald eagles over the 3-year study period showed no significant difference among any of the study lakes, taking the year into account (P = 0.95). Reproductive success of adult eagles from Pinchi Lake compared to all other study lakes combined is shown on Table 2.7. On average, 62% of territories found active at the beginning of May on Pinchi Lake successfully produced 8-week old eaglets compared to 64% of territories on reference lakes. Productivity of eagles nesting on Pinchi Lake compared to eagles nesting on each of the other study lakes did not differ significantly, taking year into account (P = 0.483). Productivity, as measured by the number of eaglets produced per active territory found in May, is shown on Table 2.7 (reference lakes combined). Approximately 27% of all eaglets produced over the 3-year period were raised on Pinchi Lake. Fledging success was not determined, and no banded eaglets were re-sighted in subsequent years over the course of the study. Average Hg concentrations in blood of eaglets were compared to eagle productivity for each study lake and year that the study took place to determine whether or not Hg levels measured in blood had any relationship to eagle productivity. As shown in Figure 2.11, no relationship was found (r = -0.254, P = 0.403, n = 13). 100 Discussion Mercury in Red-Necked Grebe Eggs Environmental meHg exposure has been associated with behavioral and reproductive impairment in common loons (Nocera and Taylor, 1998; Barr, 1986; Meyer et al., 1998). However, little is known about the potential effects of increased meHg ingestion by bald eagles or red-necked grebes, especially in habitats impacted by a mineralized, natural source of Hg. One measure of reproductive impairment caused by ingestion of increased levels of meHg is a decrease in hatchability of eggs. Levels of 0.5 to 1.5 ug/g Hg were found in unhatched eggs of pheasants (Phasianus colchicus) exposed to various dietary levels of meHg (from 3.4 to 13.5 (ig/g meHg) for a maximum period of 12 weeks (Fimreite, 1971). Heinz (1974) found similar effects in breeding mallard ducks (Anas platyrhynchos) fed 3 |j,g/g meHg in their diet, including an increase in duckling mortality, reduced egg laying and increased embryonic mortality. Another study of mallard eggs exposed to externally applied meHg-chloride revealed dose-dependent effects on survival and growth, with the lowest dose affecting survival equivalent to approximately 0.5 ug/g ww (Hoffman and Moore, 1979). This same level (0.5 \ig/g ww) was suggested as a no observed adverse effect level (NOAEL) for Hg in bald eagle eggs (Wiemeyer et al, 1984), and 0.5 |ig/g ww or approximately 2.5 jig/g dw continues to be used as the lowest observed adverse effect level (LOAEL) in egg monitoring studies (Thompson, 1996; Wolfe et al, 1998). Based on a combination of past field data and a recent study of common loons raised in captivity, Evers et al. (in press) have concluded that reproductive success of common loons can be categorized based on Hg concentrations in eggs as "low risk" (0-0.60 ug/g ww), "moderate risk" (0.60-1.30 ug/g ww) or "high risk" (>1.30 ug/g ww). Red-necked grebes and common loons have similar dietary habits. Red-necked grebes feed almost entirely on fish when available, especially during the breeding season, according to available data (Stout and Nuechterlein, 1999). Loons are also known to subsist almost exclusively on fish when sufficient numbers of fish are available (Barr, 101 1986). Both species are known to feed by catching prey underwater and swallowing it whole (Stout and Nuechterlein, 1999; Mclntyre and Barr, 1997). Both species of bird also tend to feed almost entirely on their nesting lakes when adequate food is available (Barr, 1996; Stout and Nuechterlein, 1999), suggesting that Hg levels in eggs can be used as an indicator of Hg availability of the lakes on which they nest. In this study, red-necked grebes were chosen instead of common loons because they are more abundant in the study area and lay a larger clutch of eggs (4-5 on average), than do loons (1-2 eggs per clutch). They also nest semi-colonially in the study area, whereas loon territories are widely dispersed. In addition, only one breeding pair of loons consistently returned to Pinchi Lake to breed each of the 3 years. It should be noted that this pair of loons was the only pair within the whole study area to successfully raise at least one chick in 3 of 4 years of observations (see chapter 3). Overall, dietary meHg exposure insufficient to cause overt toxicity in adult birds can result in decreased reproductive success of 35 to 50% based on the potent embryotoxicity of meHg (Wolfe et al, 1998). Therefore, whereas it may also have been advantageous to examine bald eagle eggs for Hg as part of this study, the results may not have been indicative of Hg levels in the study area because eggs are laid before lake ice breaks-up. Prior to, and during egg laying, eagles are likely feeding in open rivers and stream inlets/outlets to lakes, as they would have been ice-free upon arrival of the eagles to the study area. Thus egg-Hg concentrations in eagles probably represent Hg burdens accumulated, at least in part, by females on wintering grounds and areas surrounding the study lakes that are ice-free before ice break-up on the study lakes. In addition, disturbance of eagle nests by removal of an egg almost always leads to abandonment of the nest, which would result in no chicks being produced and ultimately no reproductive/productivity data being available for the study area. Therefore, blood and feather Hg concentrations in eaglets are better indicators of local exposure than eggs would be, for the reasons mentioned above; however, it is recognized that eggs would be the best sample to determine if eagle reproduction is affected by meHg because the egg stage is probably the most sensitive life stage of birds to meHg (Wolfe et al, 1998). 102 This study showed a gradient in Hg exposure in the study area, as expected. Red-necked grebe eggs from Pinchi Lake are generally higher in Hg than surrounding lakes, however, they are only statistically greater than Hg levels found in eggs from Fraser Lake. Mercury concentrations in red-necked grebe eggs were shown to correlate with trophic position (see chapter 3, Figure 3.5); however trophic position is likely of lesser influence on Hg concentrations in eggs from Pinchi Lake than the increased Hg concentrations present in fish and sediments because grebes eggs from the various study lakes were not found to differ significantly in trophic position (see chapter 3). Mercury concentrations in red-necked grebe eggs from Pinchi (1.17 pg/g dw) were greater than Tezzeron (0.83 pg/g dw), which were greater than Great Beaver (0.70 pg/g dw). Grebe eggs from Fraser Lake had the lowest concentrations of Hg (0.47 pg/g dw). This trend in decreasing Hg concentrations was also shown in the average sediment Hg concentrations from the study lakes with Pinchi (2315 pg/g dw) > Tezzeron (123 pg/g dw) > Stuart (84 pg/g dw) ~ Great Beaver (82 pg/g dw) > Fraser (49 pg/g dw) (see chapter 1). This is a trend that has also been shown in Hg concentrations in lake trout, with fish from Tezzeron having intermediate concentrations of Hg compared to Pinchi and control lakes (EVS Environment Consultants, 2001). The concentrations of Hg found in our red-necked grebe eggs were well below the LOAEL of 0.5 pg/g ww and were within Evers et al. (in press) defined limit of "low risk" to reproductive success of wild common loons (0-0.60 pg/g ww). The highest concentration of Hg was 0.42 pg/g ww found in one egg collected from Pinchi Lake in 2001. A l l other eggs collected from Pinchi Lake or anywhere else in the study area were below 0.30 pg/g ww. Whereas mean Hg concentrations in eggs were higher from Pinchi Lake, they were below the LOAEL and would be considered to be at low-risk of reduced reproductive success. It is worth noting, however, that the sensitivity of grebe embryos to meHg is currently unknown. Mercury levels may be higher in common loon eggs because loons are larger than grebes and therefore may consume larger fish. However, as mentioned previously, the pair of loons nesting on Pinchi Lake was the only pair to successfully raise at least one large chick in 1999, 2001 and 2002. 103 Mercury in Bald Eagles Bald eagles are opportunistic feeders, but are known to prefer a fish diet, especially during the breeding season (Buehler, 2000). In a specific lake system, larger fish of the same species, such as those that might be consumed by bald eagles, will generally tend to have higher levels of Hg than smaller fish (Burger, 2001; EVS Environment Consultants, 2001). A study conducted by Jenkins and Jackman (1994) showed that during the breeding season, bald eagles preferentially capture larger fish (340-380mm) over smaller fish (230-275mm). No such preference was shown during the non-breeding season (Jenkins and Jackman, 1994). Fish were commonly observed being carried back to the nest by adult eagles in our study area, and examination of 34 obvious prey remains in nests from 2001 and 2002 showed that large fish (i.e. those that could not be swallowed whole) comprised approximately 85% of prey remains (see chapter 3). Data on prey remains found in and below bald eagle nests is normally biased toward prey such as birds (Vermeer et al, 1989). Therefore, the fact that mainly fish remains were found in nests is likely strong evidence that these eagles were eating primarily fish. Bald eagles were also observed to actively feed on the same lakes where they nest, which ultimately made it possible to capture specific adults for blood and feather sampling. The Hg concentrations found in the blood and feathers of the eaglets from Pinchi Lake do not appear to be unusually high. They are within the range that Welch (1994) found for eaglets sampled in lake habitats of Maine from 1991-1992. Eaglets from Pinchi Lake had an average Hg in blood concentration of 0.57 ug/g whereas eaglets sampled in Maine in 1992 had an average of 0.553 ug/g. Levels in feathers were also very similar to birds from Pinchi Lake (average of 18.2 ug/g Hg), with eaglets from Maine having an average of 19.1 ug/g Hg. Wiemeyer et al (1989) reported that nestling bald eagles from Oregon (1979-1981) had, on average 1.2 ug/g ww Hg in blood (range - non-detect to 4.2 ug/g ww). These levels are double those found in the blood of the eaglets from Pinchi Lake, and the range of values extends to those of the adults found in our study area. Frenzel (1984) measured Hg in 13 addled eggs from their bald eagle nests in Oregon and found low Hg concentrations (0.031 - 0.25 ug/g ww) suggesting that the higher blood-Hg levels 104 in their eaglets was not likely a result of increased maternal deposition of Hg into the egg. The eagles from Oregon were also not known to be feeding in areas of high Hg contamination. In fact, the prey of those eagles tended to have low Hg concentrations (Frenzel, 1984). Levels of Hg ranged from non-detect in mammalian tissues, to 0.013 ug/g ww in mountain whitefish from the Cascade lakes region to 0.132 ug/g ww in blue chub from Upper Klamath Lake (Frenzel, 1984). No information was provided on the size range of fish sampled. In contrast, rainbow trout and mountain whitefish collected from Pinchi Lake during the current study had an average Hg concentration of 0.059 and 0.097 ug/g ww, respectively, with higher levels in northern pikeminnow (0.215 ug/g ww) and one lake trout found half-eaten in an eagle nest on Pinchi Lake (1.13 ug/g ww) (see chapter 3). Fresh remains of rainbow trout, lake whitefish, northern pikeminnow, largescale sucker and lake trout were all found in bald eagle nests from our study area (see chapter 3). This suggests that even though eaglets from Pinchi Lake appear to be feeding on fish with higher levels of meHg than those from Oregon, Hg concentrations in blood are lower on average. No explanation is given by Frenzel (1984) for the wide range of Hg values found in their chicks from the different regions examined, given the relatively low levels of dietary-Hg exposure. Adult eagles had significantly higher concentrations of Hg in blood compared to eaglets, with adults from Pinchi Lake having significantly greater concentrations of Hg in blood compared to adults from Stuart and Fraser Lakes. While adults from Pinchi Lake had higher mean blood Hg concentrations compared to adults from Tezzeron Lake, the difference was not significant. Adult eagles from Pinchi Lake averaged 6.54 ug/g ww Hg in blood, with the highest concentration equal to 9.44 ug/g ww. These concentrations are high compared with eagles from most other studies, and are also higher than most levels reported for common loons (Evers et al, 1998). However, two sub-adult eagles captured in Montana had 7.0 and 9.5 ug/g ww Hg in blood (Wiemeyer et al, 1989) and common loons from the Canadian maritimes have been found with blood-Hg concentrations up to 7.80 ug/g ww (Evers et al, 1998). Mean Hg concentrations in blood of adult eagles (not including sub-adults) captured in Oregon and Montana were 2.3 and 2.0 ug/g ww, respectively with the overall range of blood Hg being from 0.85 to 5.4 ug/g 105 ww (Wiemeyer et al, 1989). These concentrations are comparable to those found in the eagles from our reference lakes (mean = 2.83 pg/g; range = 1.59 - 4.86 pg/g ww). Similarly, Hg concentrations in blood of 15 adult bald eagles wintering in the Klamath Basin of Oregon/California were 2.29 pg/g ww on average (Frenzel and Anthony, 1989). In addition, Anthony et al. (1993) found an average level of 3.07 pg/g ww Hg in blood of adult bald eagles from the Columbia River estuary (range = 1.30-4.10 pg/g ww) in a study conducted from 1984-1986. To date, no threshold levels have been established for blood-Hg concentrations that would lead to decreased reproductive success or survivability in bald eagles. Controlled dosing studies that examine overt and behavioral/reproductive toxicity in birds as a result of dietary Hg exposure (Scheuhammer, 1988; Heinz, 1979; Finley et al, 1979; Fimreite, 1971) do not typically examine Hg levels in blood. When whole birds are available, it is usually the liver, kidneys or muscle that are analyzed for Hg residues (Burger and Gochfeld, 1997). There are also many interspecific differences in birds with respect to Hg absorption/elimination rates (Serafin, 1984), molting patterns, dietary habits and life histories that may enable one species of bird to tolerate levels of Hg in the blood that would be toxic to another species. In this respect, applying toxic thresholds obtained for laboratory-dosed finches, as an example, to bald eagles in the field, would not be appropriate. Other environmentally relevant chemicals that interact with Hg (i.e. Se, which is known to decrease the toxicity of Hg; Yoneda and Suzuki, 1997) must also be considered in field situations but are not often applied in laboratory studies. Therefore, the Hg concentrations found in the blood of our eagles are best interpreted based on the reproductive findings of this study and other field studies examining Hg concentrations in bald eagle blood. The same gradation in Hg exposure found in red-necked grebe eggs was also found in the blood of adult bald eagles and eaglets. Statistically, mean Hg concentrations in birds from the different reference lakes (i.e. Tezzeron, Stuart, Fraser and Great Beaver) did not differ; however, birds from Tezzeron and Pinchi Lakes also did not differ, indicating a decreasing trend in Hg concentrations across the study lakes. Examination of average 106 blood Hg concentrations shows a distinct decrease from Pinchi Lake to Fraser Lake. For adults, eagles from Pinchi Lake had the highest Hg concentrations in blood (6.54 pg/g ww), followed by Tezzeron (4.07 pg/g ww), Stuart (2.85 pg/g ww) and Fraser (1.96 pg/g ww). For eaglets, the highest blood-Hg levels were again found in birds from Pinchi Lake (0.57 pg/g ww), followed by Tezzeron (0.42 pg/g ww), Stuart (0.27 pg/g ww), Great Beaver (0.20 pg/g ww) and Fraser (0.20 pg/g ww). As mentioned previously, this trend in Hg concentrations was also shown to exist in red-necked grebe eggs, sediments and some fish (see chapter 1; EVS Environment Consultants, 2001). There is also a large amount of variability in blood Hg concentrations, especially in adult bald eagles, and this is likely a reflection of their diet (see chapter 3). It is difficult to interpret how elevated blood Hg concentrations in adult eagles from our study area may be because data for wild-caught adult eagles are scarce. Therefore, we decided to analyze blood samples from adult eagles that had been banked since 1993. Concentrations of Hg in banked blood from other adult bald eagles captured in British Columbia (mean = 0.99 pg/g ww) were significantly lower than those of our reference lake birds, and those from Pinchi Lake. This is most likely because the majority of these birds were captured or found near coastal regions of British Columbia, unlike our birds, which were from inland lakes. Blood and feather Hg concentrations of eaglets from different habitats were significantly different, as noted by Welch (1994). In particular, nestlings from inland lakes were noted to have the highest blood Hg concentrations (mean = 0.425 pg/g in 1991 and 0.553 pg/g in 1992), with coastal marine eaglets having the lowest blood Hg levels (mean = 0.068 pg/g in 1991 and 0.094 pg/g in 1992), and eaglets from estuarine and riverine habitats having intermediate concentrations. Based on this pattern of blood Hg concentrations, eaglets from inland lakes had Hg levels approximately 6 times greater than coastal eaglets. If our banked blood samples are extrapolated in a similar manner, it would show that Hg concentrations in blood of adult eagles from our site, including Pinchi Lake, are not particularly elevated. An important aspect to note about the high levels of Hg found in two of the adults from Pinchi Lake in particular is that these birds did not exhibit abnormal behavior or poor 107 reproductive success. Despite their elevated Hg levels, both birds appeared to be in excellent condition, with no evidence of abnormal behavior or lack of coordination, and they successfully raised two broods each throughout the study period. The most highly exposed adult raised two chicks in both 2001 and 2002 (this nest was not active in 2000, and the adult was not banded until 2001), making it one of the most productive birds in the study area. This was also the nest where adults and chicks were feeding on several large lake trout (Hg =1.13 ug/g ww in one fish sampled from nest). In contrast, the adult with the lowest Hg concentration in blood from Pinchi Lake (4.25 ug/g ww) was unable to successfully raise any eaglets over the 3 year study period, assuming this was the same bird at the same nest in 2000 (it was not banded until 2001). This bird hatched a single chick in both 2001 and 2002, however the 2001 chick had disappeared without a trace prior to sampling, and the 2002 chick was found dead in the nest. Based on the observations from the present study, it does not appear that meHg in particular is affecting the behavior or nesting capabilities of adult eagles on Pinchi Lake. This may be due to a combination of factors. First, adult eagles on Pinchi Lake lay their eggs while the lake is still frozen, and therefore may be feeding on small mammals/carrion or fish from thawed streams that have lower Hg-concentrations than fish from Pinchi Lake. Because of this, Hg being transferred to the egg may be below threshold effects levels (0.5 \ig/g ww), which have been related to decreased hatching success (Fimreite, 1971). Unfortunately, as no unhatched eggs remained in any nests at the time of eaglet sampling, we were unable to determine Hg concentrations in eggs. Another factor assisting chicks in elimination of meHg is through deposition into growing feathers. Both factors mentioned above would contribute to the survivability of the embryo and the chick. As with the present study, Anthony et al. (1993) and Wiemeyer et al. (1989) also found higher blood Hg levels in adult eagles than in chicks. Wiemeyer et al. (1989) suggested that this is a result of greater bioaccumulation in adults compared to chicks. We believe it is more likely a combination of meHg bioaccumulation in adults, and eaglets eliminating much of their meHg through growth of all body feathers at once, unlike adult 108 eagles that replace feathers continuously over a six-month period (McCollough, 1989). It is well known that the deposition of meHg in growing feathers can act as an efficient system for removal of a substantial portion of the body-burden of meHg (Kenow et al, in press; Becker et al, 1994; Monteiro and Furness, 2001). Since eaglets are actively producing all of their body feathers over a relatively short period of a few weeks, there is a greater opportunity to eliminate meHg from their bodies. Adults on the other hand, have accumulated meHg in their tissues for many years, particularly in the months prior to each molt. Feathers have been shown to reflect tissue concentrations as well as blood (Braune and Gaskin, 1987). Based on this information, meHg bioaccumulation cannot fully explain the greater Hg concentrations in adults compared to chicks, and the majority of the difference is likely due to greater elimination of meHg by chicks through deposition into growing feathers. Second secondary feathers from adult eagles were chosen for analysis because they are typically used in studies of Hg in adult common loon feathers (Evers et al, 1998; Scheuhammer et al, 1998; Meyer et al, 1998). One distinct difference between these two bird species is that bald eagles undergo their yearly molt on their breeding territories, whereas secondaries obtained from loons during summer sampling would have been grown on their wintering grounds. Therefore, unlike loons, adult eagle feathers may be more indicative of meHg levels on breeding grounds. Bowerman et al. (1994) showed that no significant differences in Hg concentrations existed between adult eagle primaries, secondaries, retrices and body feathers. However, for larger feathers, such as primaries and secondaries, only the apical portion of the feather was digested and analyzed. This could produce a considerable bias if this concentration is extrapolated to the entire feather, as the, apical portion tends to have the highest concentration of Hg in adult secondaries (see chapter 4). Therefore, Hg concentrations reported for our secondary feathers are the result of digestion of the entire feather. Adult eagles from Pinchi Lake had significantly higher feather Hg concentrations (mean = 39.9 pg/g) compared to adults from Fraser Lake only (mean = 9.3 pg/g). Similar to blood Hg concentrations, adults from Pinchi and Tezzeron Lakes (mean = 12.7 pg/g) did 109 not differ significantly; however, unlike blood Hg concentrations, no statistically significant difference was found between feather Hg concentrations for adults from Pinchi and Stuart Lakes (mean = 14.0 pg/g). The lack of a statistically significant difference is likely due to the large variation in feather Hg concentrations from Pinchi Lake in addition to the small sample size. Mean concentrations of 21 pg/g (range = 3.6 -48 pg/g) Hg in adult eagle secondaries have been reported (Bowerman et al, 1994). Mercury concentrations in common loon secondaries are consistently lower than the bald eagles from Pinchi Lake. The highest mean concentration noted by Evers et al. (1998) was 13.1 ±5.3 pg/g for common loons from New England. Similar levels were found in adult loons from Wisconsin, with males having 12.3 ± 3.6 pg/g Hg and females having 10.1 ± 3.4 pg/g Hg in feathers (Meyer et al, 1998). This is not surprising as these loons also had lower blood Hg concentrations than did the bald eagles from Pinchi Lake. These levels are, however, in the range of the bald eagles from our reference lakes. No correlation was found between adult blood and feather Hg concentrations, most likely due to the small sample size, as the relationship was only marginally insignificant (P = 0.087). A significant positive correlation was found between chick blood and feather Hg concentrations, a relationship that has been shown in the past for eaglets (Welch, 1994; Wood et al, 1996) as well as other species of wild birds, including the common loon (Scheuhammer et al. 1998a), great egret (Sepulveda et al, 1999) and osprey (DesGranges et al, 1998). Although these studies have all shown a significant linear correlation between total Hg concentrations in feathers and blood, the correlations differ from study to study. While bald eagle chicks from central Florida had approximately 25 times more Hg in feathers than blood (Wood et al, 1996), eagle chicks from Maine had approximately 44 times greater Hg in feathers than blood (Welch, 1994) despite having similar mean blood Hg concentrations (0.13 pg/g ww in Florida eaglets vs. 0.12 pg/g ww in Maine eaglets). Eaglets from our study were similar to the eaglets from central Florida, having approximately 27 times more Hg in feathers than blood (mean blood Hg = 0.37 pg/g ww). In osprey chicks from the James Bay and Hudson Bay regions of Quebec, total Hg was approximately 20 times greater in feathers than blood (mean blood Hg = 1.9 pg/g ww; DesGranges et al, 1998). Great egret nestlings from Southern 110 Florida had approximately 12 times greater Hg in feathers than blood (average Hg in blood = 1 ug/g ww; Sepulveda et al, 1999) and common loon chicks from central Ontario had approximately 16 times greater Hg in feathers than blood (mean blood Hg = 0.142 ug/ml; Scheuhammer et al, 1998a). Therefore, while Hg concentrations in blood and feathers of nestlings have been shown to correlate in nestlings of varioius fish-eating bird species, the relationship varies considerably among species and geographical location. One of the more interesting relationships found in this study was the significant positive correlation (r = 0.913, P = 0.004) among blood Hg concentrations in adult eagles and eaglets from the same nest. Wood et al. (1996) found a similar correlation between adult eagle and chick feather Hg concentrations (r = 0.63, P = 0.02), but the relationship was not as strong as the blood relationship shown here. We did not test for the same relationship in feathers because no relationship was found between adult blood and feather Hg concentrations. In addition, eaglets from one of the nests were not sampled for feathers, thereby decreasing the sample size available for correlation. The relationship we found between adult and chick blood Hg concentrations could be very useful in predicting adult blood Hg during the breeding season, as adult bald eagles are very difficult to capture, especially in areas where competition is low and food is not limited. Future studies examining both adults and chicks, especially in different habitat types, would be interesting to further establish this relationship. Selenium in Bald Eagles Selenium is an essential micronutrient that can be bioaccumulated, and may be biomagnified through the food chain (Lemly, 1997). Selenium levels of 10 ug/g ww in liver or 3 ug/g ww in eggs have been suggested as LOAELs for reproductive effects in birds (Heinz, 1996; Lemly, 1997). Selenium concentrations in the blood of our eaglets (mean = 0.52 ug/g ww) and adult eagles (mean = 1.28 ug/g ww) were well below those found in surviving mallards (5 to 14 ug/g ww) that were fed the same dose of Se as 111 mallards that died as a result of exposure (Heinz and Fitzgerald, 1993). Combs and Combs (1986) indicated normal Se concentrations in the blood of captive chickens and turkeys of approximately 0.1-0.3 pg/g ww, on average. In a study examining Se accumulation in blood of adult ring doves (Streptopelia risoria) exposed to a Se-deficient (0.01 pg/g dw as selenomethionine), Se-adequate (0.1 pg/g dw) or Se-enriched (2 pg/g dw) diet (n = 14 birds in each group), mean concentrations of 0.3 ± 0.1 pg/ml, 1.4 ± 0.4 pg/ml, and 3.4 ± 0.4 pg/ml in blood were found, respectively (Scheuhammer, pers. comm.). Based on these concentrations, our adult eagles may be considered as having normal or adequate blood Se levels (mean = 1.28 pg/g ww). However, to date, no nutritionally adequate or toxic thresholds have been published for Se levels in blood of wild birds, including bald eagles. Selenium concentrations in the blood of nesting adult common eiders (Somateria mollissima) varied by location in coastal Finland, but averaged 1.26 to 2.86 pg/g ww (Franson et al, 2000). Franson et al. (1999) found means of 0.14 pg/g ww Se in blood of emperor goose (Chen canagica) goslings, 2.78 pg/g ww Se in blood of adult females during wing molt, and 5.60 pg/g ww Se in blood of adult females sampled during incubation. Nesting success was as high or higher compared to other populations of geese, indicating that even the higher levels of Se were not associated with lower reproductive success (Franson et al., 1999). It is unlikely that Se concentrations in the blood of eagles from our study area are having any kind of detrimental effects, especially since reproductive success and productivity were indicative of a healthy population producing a surplus of chicks (see below). In our study, adult bald eagles were found to have significantly higher Se concentrations in blood compared to chicks. Franson et al. (1999) found similar results in emperor goose goslings (mean blood Se = 0.14 pg/g ww) and adults (mean blood Se = 2.78 - 5.60 pg/g ww). In addition, adult female emperor geese were found to have significantly lower blood Se concentrations following the onset of molt (2.78 pg/g ww compared to 5.60 pg/g ww prior to molting) (Franson et al., 1999). Unlike adult eagles and emperor geese that molt different parts of their plumage over a longer period of time (Franson et 112 al, 1999; McCollough, 1989), chicks of both species are actively growing all body feathers at once. Bowerman et al. (1994), found that mean Se concentrations in feathers of adult eagles and chicks did not differ significantly (approximately 1.8 ug/g dw vs. 1.9 ug/g dw, respectively). Based on this finding, in order for eaglets to have feather Se concentrations similar to adults, a greater amount of Se must be being deposited into the plumage at any given time because eaglets are growing many more feathers at once. This is supported by the findings from the present study that chicks had significantly lower blood Se concentrations than adults. Bowerman et al. (1994) also found that Se concentrations in the various parts of the adult plumage (i.e. primaries, secondaries, retrices and body) did not differ significantly. Therefore, feather deposition of Se may cause blood levels to be lower in chicks, as was discussed previously for Hg. The toxicity of meHg can be significantly reduced through a poorly understood process of demethylation and the formation of an equimolar inorganic mercury-selenium-protein complex that alters the distribution of Hg among organs and is presumed to protect the organism against the toxicity of meHg (Yoneda and Suzuki, 1997). A marginally insignificant positive relationship between Hg and Se was found in eaglets from the study area, but no such relationship was found for adult bald eagles. Mercury and Se are known to be closely associated in other tissues (see chapter 5; Bischoff et al, 2002; Scheuhammer et al, 1998b; Cuvin-Aralar and Furness, 1991). Mercury in blood, however, is primarily meHg (Evers et al, 1998), which is likely why the association between Hg and Se concentrations in blood of eaglets was weak, and non-existent for adults. Small sample size was probably also a factor, particularly in the analysis of adult bald eagles. Franson et al. (1999) were unable to find a correlation between blood Hg and Se for Emperor geese from western Alaska. Bowerman et al. (1994) measured concentrations of Hg and Se in bald eagle adult and chick feathers from the Great Lakes region, but did not examine correlations between the two. They did, however, notice that where Hg concentrations in chick feathers varied significantly, there were no significant differences for Se. In addition, Bowerman et al. (1994) noted that no significant differences were found between adult and chick Se concentrations in feathers, even though adult feathers had significantly higher Hg concentrations. 113 Eagle Productivity and Reproductive Success Barr (1986) showed that dietary meHg levels of approximately 0.3-0.4 pg/g ww can lead to decreased reproductive success in wild common loons as shown by decreased hatching success of eggs, fewer eggs laid, and reduced nest and territory-site fidelity. Heinz (1974) also found reduced hatching success and fewer eggs laid by mallards exposed to a 3 pg/g diet of meHg. In addition, Heinz (1974) found increased mortality in ducklings whose parents were exposed to 3 pg/g meHg. Overall, ducks on the 3 pg/g meHg diet produced less than half the number of one-week old ducklings compared to controls (Heinz, 1974). Heinz (1979) found that female mallards fed 0.5 pg/g meHg in their diet laid fewer eggs than control birds, therefore producing fewer chicks, and also laid significantly more eggs outside their nest boxes. Decreased back-riding, a behavior which provides protection, saves energy, and moderates thermoregulation of common loon chicks has been associated with increased blood Hg concentrations (Nocera and Taylor, 1998). Based on this, Nocera and Taylor (1998) suggested that exposure of 1.25 - 1.50 pg/g Hg ww in the blood may be a critical level for behavioral and/or lethal effects in common loon chicks. Dietary concentrations known to cause reproductive impairment are typically one fifth of the dietary level of meHg that would be required to cause significant neurological defects in adult birds (Scheuhammer, 1988). No indication of reduced reproductive success or productivity was found among eagles nesting on Pinchi Lake and those on the reference lakes over the 3-year study period. According to Sprunt et al. (1973), productivity of the bald eagles from the study area (1.12 chicks/occupied territory on combined reference lakes and 0.98 on Pinchi Lake) was similar to that of a healthy population producing a surplus of chicks (i.e. greater than 1.0 chick/occupied territory), and greater than 0.7 young/occupied territory, the value associated with a sustaining population. Overall, reproductive success and productivity was similar to that of a population of reference bald eagles nesting on Adak and Tanaga Islands in the Western Andreanof Islands (67%) of the Aleutian Archipelago (Anthony et 114 al, 1999). Over the two-year period, productivity averaged 1.09 and 0.96 chicks/occupied territory on Adak and Tanaga Islands, respectively (Anthony et al, 1999). Similarly, productivity of reference populations of bald eagles nesting in the lower Fraser River Valley (1990-1996) and around southeast Vancouver Island (1991-1995) British Columbia averaged 1.2 and 0.95 young/occupied territory, respectively (Elliott et al, 1998). Mean productivity of bald eagles from our study area also exceeded that of a reference population of bald eagles nesting in Barkley Sound on the west coast of Vancouver Island from 1992-1998 (0.75 young/occupied territory; Gill and Elliott, 2003). Elliott and Harris (2001/2002) provides an excellent summary of nesting success and productivity of bald eagle populations from references regions and areas of poor recovery throughout Canada and the United States. In many cases, the mean productivity of bald eagles from the Pinchi Lake area exceeded that of other bald eagle populations (Elliott and Harris, 2001/2002). While the year was not found to be a significant factor affecting reproductive success on our various study lakes, it should be noted that had we conducted only a two-year study from 2000-2001, it might have appeared that reproductive success was depressed on Pinchi Lake compared to reference lakes (63% compared to 78%). Many factors beyond contaminant levels, such as weather, food availability, depredation and disease can negatively affect reproductive success, indicating the importance of conducting reproductive studies over several years, especially where smaller sample sizes can be a concern. Our study was similar to other studies relating bald eagle Hg concentrations to reproductive success and productivity in that no significant Hg effects were observed. Welch (1994) found no association between 5-year or 15-year mean productivity levels and Hg concentrations in blood (mean = 0.49 ug/g ww, range = 0.07 - 1.46 ug/g ww) or feathers (mean = 18.9 ug/g ww, range = 8.0 - 36.7 ug/g ww) of eaglets. Bowerman et al. (1994) found no significant correlations between feather Hg levels in adult eagles (mean = 21 ug/g, range = 0.20 - 66 ug/g) or chicks (mean = 9.0 ug/g, range = 1.5-27 ug/g) and 5-year productivity and reproductive success. Reproductive success also appeared to be normal for bald eagles breeding in the Oregon area (Frenzel, 1984) even though Hg exposure appeared high (up to 4.2 ug/g Hg in blood of eaglets; Wiemeyer et al, 1989). 115 Ospreys nesting near an area of increased fish-Hg concentrations compared to those of a control area also had significantly greater Hg concentrations in feathers (approximately 40 pg/g in chicks), but yet no difference was found in mean number of young fledged from either area (DesGranges et al, 1998). The levels of feather-Hg reported by DesGranges et al. (1998) are almost double those found in the eaglet feathers from Pinchi Lake. Based on previous studies and the results obtained in the present study, it does not appear that the Hg available in Pinchi Lake or the surrounding area is high enough to cause adverse effects on eagle reproduction. One interesting observation was made with regards to the nesting locations chosen by eagles on Pinchi Lake. Specifically, all nests (old and current) were located on the eastern half of the lake, upstream up the Hg mine (see Figures 2.8, 2.9 and 2.10). In addition, no evidence of previous bald eagle nesting was found around the western portion of the lake. As discussed in chapter 1, Hg concentrations in sediments are on the order of 100 times greater close to the Hg mine, and in the western half of Pinchi Lake, which is downstream of the Hg mine. Whether these increased Hg concentrations have influenced the choice of nesting sites by bald eagles on Pinchi Lake is unknown. Several other factors may be influencing nesting location. For example, the bathymetry of Pinchi Lake varies considerably between the eastern and western halves of the lake. The eastern half has a maximum depth of 67.5 metres whereas the western half has a maximum depth of 18 metres. Both areas have shallows around the perimeter of the lake. The larger deeper basin in the eastern half of the lake may attract bald eagles because some larger fish such as lake trout tend to prefer deeper cooler waters as the summer progresses, but may still be found in shallows (Coad et al, 1995). Based on this, it may be more likely that these larger fish would be found in shallows adjacent to deeper waters. Another factor affecting nest site location would be availability of suitable nesting trees. A small part of the southwestern end of Pinchi Lake was clear-cut in the past, and so would reduce available nesting habitat in the area. Bald eagles nesting in the study area appear to prefer larger trees, however, a few nests were found in relatively small trees suggesting that the size of the tree is not the only factor determining nest location. While a proper field assessment of available nesting habitat was not completed at Pinchi Lake, it did not 116 appear that a lack of suitable nesting trees in the western portion was preventing eagles from establishing nest sites. Therefore, several factors may have affected the choice of nesting locations by bald eagles on Pinchi Lake, including elevated Hg concentrations in the western half of the lake. During eaglet sampling in 2002, three noteworthy events occurred, which had not been previously observed during the study period. First, two of the nests we attempted to sample, one on Tezzeron Lake and one on Pinchi Lake, each had a single dead chick in the nest. The chicks were approximately the same size, and appeared to be no more than 2 weeks old. Adult eagles occupying these nesting sites gave no indication that the nest was no longer active anytime prior to sampling, and were very agitated when the nest tree was climbed. We did not determine the cause of death for these chicks or contaminant levels as they were very decomposed by the time they were found. These were the only times that dead chicks were found in nests over the 3-year study period. Second, one nest on Fraser Lake was obviously depredated by a black bear no more than two weeks before sampling. Claw marks were noted all the way up the tree, and the side of the nest where the bear climbed over was badly damaged. Again, this was the only time over the 3-year period where a nest was obviously depredated. In all other cases where nests were found inactive at the time of sampling, no obvious cause was noted. Nests were simply empty and adults may or may not have still been present in the territory. Finally, almost all eaglets had some degree of blow-fly (genus - Protocalliphora) larval parasitism in 2002, which was not encountered in previous years. Blow-fly larva infestations have been shown to infect nestling bald eagles in Saskatchewan in the past but were not associated with any pathological effects (Bortolotti, 1985). Conclusions Based on our three-year study of the bald eagles from Pinchi Lake and the surrounding region, Hg is having no obvious adverse effects on reproductive success. Mercury analysis of red-necked grebe eggs also showed concentrations below the LOAEL for reproductive effects from all study lakes. Significant positive relationships were shown 117 to occur between blood and feather Hg concentrations in eaglets, as well as between blood Hg concentrations in adults and eaglets from the same nest. A marginally insignificant relationship was found between Se and Hg concentrations in blood of eaglets, but not adult eagles. While meHg exposure, as measured by blood and feather Hg concentrations, was higher in eaglets from Pinchi Lake compared to reference lake birds, the concentrations were comparable to levels found in eaglets from other areas in North America not experiencing point source Hg contamination. Acknowledgements TeckCominco, Ltd., the Metals In The Environment Research Network, and Canadian Wildlife Service provided funding for this study. Field assistance was provided by Lori Smith, Jennifer Young and Sandi Lee. Additional assistance in eaglet sampling and adult bald eagle capture was provided by Darcy Haycock and Gregg Howald, respectively. Tony Scheuhammer provided laboratory facilities at NWRC for digestion and analysis of samples for Hg and Se. Delia Bond and Ewa Neugebauer (NWRC) provided guidance and assistance in the laboratory. References Anthony, R. G., Garrett, M . G., and Schuler, C. A. (1993). Environmental contaminants in bald eagles in the Columbia River Estuary. J. Wildl. Manage. 57, 10-19. Auspurger, T., Franson, J. C , Converse, K. A., Spitzer, P. R., and Miller, E. A. (1998). An epizootic of common loons in coastal waters of North Carolina: Concentrations of elemental contaminants and results of necropsies. Environ. Toxicol. Chem. 77, 205-209. Barr, J. F. (1986). Population dynamics of the Common loon (Gavia immer) associated with mercury-contaminated waters in northwestern Ontario. Can. Wildl. Serv. Occas. Pap. No. 56, 25pp. Barr, J. F. (1996). Aspects of Common loon (Gavia immer) feeding biology on its breeding ground. Hydrobiologia 321, 119-144. 118 Becker, P. FL, Henning, D., and Furness, R. W. (1994). Differences in mercury contamination and elimination during feather development in gull and tern broods. Arch. Environ. Contam. Toxicol. 27, 162-167. Becker, D. S., and Bigham, G. N . (1995). Distribution of mercury in the aquatic food web of Onondaga Lake, New York. Water Air Soil Pollut. 80, 563-571. Bischoff, K., Pichner, J., Braselton, W. E., Counard, C., Evers, D. C , and Edwards, W. C. (2002). Mercury and selenium concentrations in livers and eggs of common loons (Gavia immer) from Minnesota. Arch. Environ. Contam. Toxicol. 42, 71-76. Bortolotti, G. R. (1984). Sexual size dimorphism and age-related size variation in bald eagles. J. Wildl. Manage. 48, 72-81. Bortolotti, G. R. (1985). Frequency of Protocalliphora avium (Diptera: Calliphoraidae) infestations in bald eagles (Haliaeetus leucocephalus). Can. J. Zool. 63, 165-168. Bowerman IV, W. W., Evans, E. D., Giesy, J. P., and Postupalsky, S. (1994). Using feathers to assess risk of mercury and selenium to bald eagle reproduction in the Great Lakes Region. Arch. Environ. Contam. Toxicol. 27, 294-298. Braune, B. M . , and Gaskin, D. E. (1987). Mercury levels in Bonaparte's gulls (Larus philadephia) during autumn molt in the Quoddy Region, New Brunswick, Canada. Arch. Environ. Contam. Toxicol. 16, 539-549. Buehler, D. A. (2000). Bald Eagle. The Birds of North America 506, 39pp. Burger, J., and Gochfeld, M . (1997). Risk, mercury levels, and birds: Relating adverse laboratory effects to field biomonitoring. Environ. Res. 75, 160-172. Burger, J., Gaines, K. F., Boring, C. S., Stephens Jr, W. L., Snodgrass, J., and Gochfeld, M . (2001). Mercury and selenium in fish from the Savannah River: Species, trophic level and locational differences. Environ. Res. Sec. A. 87, 108-118. Cain, S. L., and Hodges, J. I. (1989). A floating-fish snare for capturing bald eagles. J. Raptor Res. 23, 10-13. Carnrick, G. R., Manning, D. C , and Slavin, W. (1983). Determination of selenium in biological materials with platform atomic-absorption spectroscopy and Zeeman background correction. Analyst 108, 1297-1312. Coad, B. W., Waszczuk, H., and Labignan, I. (1995). Encyclopedia of Canadian Fishes (Singapore: Canadian Museum of Nature and Canadian Sportfishing Productions, Inc.). Combs, G. F., and Combs, S. B. (1986). The Role of Selenium in Nutrition. (Orlando, Florida: Academic Press Inc.). 119 Cook, S. J., Jackaman, W., McCurdy, M . W., Day, S. J., and Friske, P. W. (1996). Regional lake sediment and water geochemistry of part of the Fort Fraser map area, British Columbia (NTS 93K/9, 10, 15 and 16): Ministry of Employment and Investment, Energy and Mines Division, Geological Survey Branch. 31pp. Cox, J. A., Carnahan, J., DiNunzio, J., McCoy, J., and Meister, J. (1979). Sources of mercury in fish in new impoundments. Bull. Environ. Contam. Toxicol. 23, 779-783. Cuvin-Aralar, M . L. A., and Furness, R. W. (1991). Mercury and selenium interaction: A review. Ecotoxicol. Environ. Safety 21, 348-364. DesGranges, J.-L., Rodrigue, J., and Laperle, M . (1998). Mercury accumulation and biomagnification in ospreys (Pandion haliaetus) in the James Bay and Hudson Bay regions of Quebec. Arch. Environ. Contam. Toxicol. 35, 330-341. Elliott, J. E., and Harris, M . L. (2001/2002). An ecotoxicological assessment of chlorinated hydrocarbon effects on bald eagle populations. Rev. Toxicol. 4, 1-60. Elliott, J. E., Moul, I. E., and Cheng, K. M . (1998). Variable reproductive success of bald eagles on the British Columbia coast. J. Wildl. Manage. 62, 518-529. Evers, D. C. (1992). A replicable capture method for adult and juvenile common loons on their nesting lakes. Proceedings, 1992 Conference on the Loon and It's Ecosystem: Status, Management, and Environmental Concerns. Bar Harbor, ME, USA, August 20-22, pp. 214-220. Evers, D. C , Kaplan, J. D., Meyer, M . W., Reaman, P. S., Braselton, W. E., Major, A., Burgess, N. , and Scheuhammer, A. M . (1998). Geographic trend in mercury measured in common loon feathers and blood. Environ. Toxicol. Chem. 77, 173-183. Evers, D. C , Taylor, K. M . , Major, A., Taylor, R. J., Poppenga, R. and Scheuhammer, A. M . Common loon eggs as indicators of methylmercury availability in North America. Ecotoxicol. In press. EVS Environment Consultants (2001). Regional survey offish mercury concentrations - Pinchi Lake, B.C. (Vancouver: TeckCominco Ltd.) 91pp. EVS Environment Consultants, Norecol Dames and Moore, and Frontier Geosciences. (1999). Pinchi Lake Mine Site and Lake Investigation: Environmental Assessment Report (Vancouver: Cominco Ltd.). 114 pp. Facemire, C , Augspurger, T., Bateman, D., Brim, M . , Conzelmann, P., Delchamps, S., Douglas, E., Inmon, L., Looney, K., Lopez, F., Masson, G., Morrison, D., Morse, N . , and Robison, A. (1995). Impacts of mercury contamination in the southeastern United States. Water Air Soil Pollut. 80, 923-926. 120 Fimreite, N . (1971). Effects of methylmercury on ring-necked pheasants, with special reference to reproduction. Can. Wildl. Serv. Occas. Pap. No 9, 39pp. Finley, M . T., Stickel, W. H., Christensen, R. E. (1979). Mercury residues in tissues of dead and surviving birds fed methylmercury. Bull. Environ. Contam. Toxicol. 21, 105-110. Franson, J. C , Hollmen, T., Poppenga, R. H., Hario, M . , Kilpi, M . , and Smith, M . R. (2000). Selected trace elements and organochlorines: Some findings in blood and eggs of nesting common eiders (Somateria mollissima) from Finland. Environ. Toxicol. Chem. 19, 1340-1347. Franson, J. C , Schmutz, J. A., Creekmore, L. H., and Fowler, A. C. (1999). Concentrations of selenium, mercury, and lead in blood of emperor geese in western Alaska. Environ. Toxicol. Chem. 18, 965-969. Frenzel, R. W. (1984). Environmental Contaminants and Ecology of Bald Eagles in Southcentral Oregon. In Department of Fisheries and Wildlife (Corvallis: Oregon State University), 143pp. Frenzel, R. W., and Anthony, R. G. (1989). Relationship of diets and environmental contaminants in wintering bald eagles. J. Wildl. Manage. 53, 792-802. Gill , C. E., and Elliott, J. E. (2003). Influence of food supply and chlorinated hydrocarbon contaminants on breeding success of bald eagles. Ecotoxicol. 12, 95-111. Grieb, T. M . , Driscoll, C. T., Gloss, S. P., Schofield, C. L., Bowie, G. L., and Porcella, D. B. (1990). Factors affecting mercury accumulation in fish in the upper Michigan peninsula. Environ. Toxicol. Chem. 9, 919-930. Heinz, G. H. (1996). Selenium in birds. In Environmental Contaminants in Wildlife -Interpreting Tissue Concentrations, W. N . Beyer, G. H. Heinz and A. W. Redmond-Norwood, eds. (Boca Raton: Lewis Publishers), pp. 447-458. Heinz, G. (1974). Effects of low dietary levels of methyl mercury on mallard reproduction. Bull. Environ. Contam. Toxicol. 11, 386-392. Heinz, G. H. (1979). Methylmercury: Reproductive and behavioral effects on three generations of mallard ducks. J. Wildl. Manage. 43, 394-401. Heinz, G. H., and Fitzgerald, M . A. (1993). Overwinter survival of mallards fed selenium. Arch. Environ. Contam. Toxicol. 25, 90-94. Hoffman, D. J., and Moore, J. M . (1979). Teratogenic effects of external egg applications of methyl mercury in the mallard, Anas platyrynchos. Teratology 20, 453-461. 121 Jackson, T. A. (1988). The mercury problem in recently formed reservoirs of northern Manitoba (Canada): Effects of impoundment and other factors on the production of methyl mercury by microorganisms in sediments. Can. J. Fish. Aquat. Sci. 45, 97-121. Jandel Scientific. (1995). SigmaStat Statistical Software, Version 2.03S for Windows 95, NT & 3.1. Jandel Scientific, San Rafael, CA. Jenkins, J. M . , and Jackman, R. E. (1994). Field experiments in prey selection by resident bald eagles in the breeding and non-breeding season. J. Field Ornithol. 65, 441-446. Kenow, K. P., Gutreuter, S., Hines, R. K., Meyer, M . W., Fournier, F., and Karasov, W. H. Effects of methyl mercury exposure on the growth of juvenile common loons. Ecotoxicol. In press. Krynitsky, A. J. (1987). Preparation of biological tissue for determination of arsenic and selenium by graphite furnace atomic absorption spectrophotometry. Anal. Chem. 59, 1884-1886. Lemly, A. D. (1997). Environmental implications of excessive selenium: a review. Biomed. Environ. Sci. 10, 415-435. McCollough, M . A. (1989). Molting sequence and aging of bald eagles. Wilson Bull. 101, 1-10. Mclntyre, J. W., and Barr, J. F. (1997). Common Loon. The Birds of North America 313, 31pp. Meyer, M . W., Evers, D. C , Hartigan, J. J., and Rasmussen, P. S. (1998). Patterns of common loon (Gavia immer) mercury exposure, reproduction, and survival in Wisconsin, USA. Environ. Toxicol. Chem. 17, 184-190. Monteiro, L. R., and Furness, R. W. (2001). Kinetics, dose-response, excretion, and toxicity of methylmercury in free-living Cory's shearwater chicks. Environ. Toxicol. Chem. 20, 1816-1823.. Nocera, J. J., and Taylor, P. D. (1998). In situ behavioral response of common loons associated with elevated mercury (Hg) exposure. Cons. Ecol. 2, 1-16. Plouffe, A. (1995). Glacial dispersal of mercury from bedrock mineralization along Pinchi fault, north central British Columbia. Water Air Soil Pollut. 80, 1109-1112. Pokras, M . A., Hanley, C , and Gordon, Z. (1998). Liver mercury and methylmercury concentrations in New England common loons (Gavia immer). Environ. Toxicol. Chem. 17, 202-204. 122 Postupalsky, S. (1974). Raptor reproductive success: Some problems with methods, criteria, and terminology. In Management of Raptors, F. N . Hamerstrom Jr, B. E. Harrell and R. R. Ohlendorff, eds. (Vermillion, SD: Raptor Research Foundation), pp. 21-31. Reid, D. S., and Morley, R. L. (1975). Mercury Contamination of Fish from Pinchi Lake, B.C.: Fish and Wildlife Branch Report, B.C. Ministry of Environment, Lands and Parks, pp. 1-18. SAS Institute, Inc. (2001). JMP, Version 4.01. SAS Institute, Inc., Cary, NC. Scheuhammer, A. M . (1988). Chronic dietary toxicity of methylmercury in the zebra finch, Poephila guttata. Bull. Environ. Contam. Toxicol. 40, 123-130. Scheuhammer, A. M . , and Bond, D. (1991). Factors affecting the determination of total mercury in biological samples by continuous-flow cold vapor atomic absorption spectrophotometry. Biol. Trace Elem. Res. 31, 119-129. Scheuhammer, A. M . , Atchison, C. M . , Wong, A. H. K., and Evers, D. C. (1998a). Mercury exposure in breeding common loons (Gavia immer) in central Ontario, Canada. Environ. Toxicol. Chem. 17, 191-196. Scheuhammer, A. M . , Wong, A. H. K., and Bond, D. (1998b). Mercury and selenium accumulation in common loons (Gavia immer) and common mergansers (Mergus merganser) from eastern Canada. Environ. Toxicol. Chem. 17, 197-201. Sepulveda, M . S., Frederick, P. C , Spalding, M . G., and Williams Jr., G. E. (1999). Mercury contamination in free-ranging great egret nestlings (Ardea albus) from southern Florida, USA. Environ. Toxicol. Chem. 18, 985-992. Serafm, J. A. (1984). Avian species differences in the intestinal absorption of xenobiotics (PCB, dieldrin, Hg 2 +). Comp. Biochem. Physiol. 78, 491-496. Sprunt, A. IV, Robertson, B. W. Jr, Postupalsky, S., Hensel, R. J., Knoder, C. E., and Ligas, F. J. (1973). Comparative productivity of six bald eagle populations. Trans. No. Amer. Wildl. Nat. Resour. Conf. 38, 96-106. Stout, B. E., and Nuechterlein, G. L. (1999). Red-Necked Grebe. The Birds of North America 465, 31pp. Thompson, D. R. (1996). Mercury in Birds and Terrestrial Mammals. In Environmental Contaminants in Wildlife - Interpreting Tissue Concentrations, W. N . Beyer, G. H. Heinz and A. W. Redmon-Norwood, eds. (Boca Raton: Lewis Publishers), pp. 341-356. 123 Vermeer, K., Morgan, K. H., Butler, R. W., and Smith, G. E. J. (1989). Population, nesting habitat, and food of bald eagles in the Gulf Islands. In The Ecology and Status of Marin and Shoreline Birds in the Strait of Georgia, British Columbia. K. Vermeer and R. W. Butler, eds. (Special Publ. Ottawa ON, Canada: Can. Wildl. Serv.), pp. 123-130. Watson, T. (1992). Evaluation of mercury concentration in selected environmental receptors in the Williston Lake and Peace River areas of British Columbia (Richmond, B.C., Canada: Triton Environmental Consultants Ltd.). Welch, L. J. (1994). Contaminant burdens and reproductive rates of bald eagles breeding in Maine. Unpub. M.S. Thesis. (Orono: University of Maine), 86pp. Wiemeyer, S. N . , Frenzel, R. W., Anthony, R. G., McClelland, B. R., and Knight, R. L. (1989). Environmental contaminants in blood of western bald eagles. J. Raptor Res. 23, 140-146. Wiemeyer, S .N., Lamont, T .G., Bunck, C .M. , Sindelar, C .R., Gramlich, F J. , Fraser, J.D., and Byrd, M . A. (1984). Organochlorine pesticide, polychlorobiphenyl, and mercury residues in bald eagle eggs - 1969-79 - and their relationship to shell thinning and reproduction. Arch. Environ. Contam. Toxicol. 13, 529-549. Wolfe, M . F., Schwarzbach, S., and Sulaiman, R. A. (1998). Effects of mercury on wildlife: A comprehensive review. Environ.Toxicol. Chem. 17, 146-160. Wood, P. B., White, J. PL, Steffer, A., Wood, J. M . , Facemire, C. F., and Percival, H. F. (1996). Mercury concentrations in tissues of Florida bald eagles. J. Wildl. Manage. 60, 178-185. Yoneda, S., and Suzuki, K. T. (1997). Detoxification of mercury by selenium by binding of equimolar Hg-Se complex to a specific plasma protein. Toxicol. Appl. Pharmacol. 143, 274-280. 124 Table 2.1 - Recovery data for certified reference materials run during Hg analysis of bald eagle blood and feathers and red-necked grebe eggs Tissue Analyzed Sample Year Average Average Dolt-2 (%) Dorn>2 (%) Blood 2000 101.1 96.4 Blood & Grebe Egg 2001 109.6 102 Blood 2002 106.3 95.4 Feather & Grebe Egg 2000 97.1 110 Feather 2001 96.5 95.6 Feather 2002 102.4 98.1 125 '—V 00 =L —J 3 C N C N — — — o — — o o — — — — — CN — C N O — — o c s r r d d d d o ' d d d d d d d d d d d d d d d d d d d — ' c o © c N o o c N c N c N s © v - i o r j - r r c o c N o o r - - s O — o «ri os r- os ' I t ^ ' ^ . ' ! | ! ^ ' < ! ' ' ! 1 t ^ ' ' 1 M * 0 0 O t v O , ' ' 0 v * n " H a , 0 — 0 0 0 0 0 0 0 0 0 0 0 0 0 — - d —- d d d © — — s© — 0 0 o c o < n s © s © o o o o m © o s r r © o o r - - i r s c N — m r r o r -O rr © O © Os Os CO r- oo oo oo r> P - p-r - 0 s 0 s P - 0 0 O s O > s O 0 0 0 s 0 s 0 s O O s P - 0 0 p ~ f - r - c - P - P - P ~ t - - r - - c - - P - P - o o r - - r - - P -5 a> ' a, 'op| I £ — Os — o o o s o r ~ o o o c o — m s o o o — OOOs — — ^ t ( N » o t | n » i ( S r t - > S O O N — < o o o o o \ o o r r ( s j C s i r j r . N N q r s j N N N r S O ' - O O O O O d d d d d d o' d d d o' © — — — —'< — — — p- rr >/•> M M SC O f © |C/3 I 00 ^ r ~ c o p - « " i s o © r r p - O s — c-» co os — N » o o h m t M n v i o \ c s v > . © c o © © r ~ - o o c s < / ~ i c N r r o s — i n o c m v i a N ^ M o e o o © C N — O S O C N O S C N © — © — — rr — so — c N O s r r o s o o o o s — — — © —< — © — — — — — v i v i i O T f i r l v i ' t v i T f i n ' * ^ 00 1% CN CM o o 4-t <4-» U-i U-i U-t <+H l+-t l* - l C M Cfc-t Ck-i CLH (fc* lk-t CfH l^H <4-* l» - l o o o o o o o o o o o o o o o o o o o o o o 1 "5 U Q o o „ „ I t s s 2 s I I o o co co o o o o _ o o o O o i> ^ e Co CO 03 03 _3 2 2 2 S I o o 2 52 — — as os CO CO CS CN CN o o CO CO o o © © © © © 0 0 .1 I I I I I i 1 ' SO so so so so so s i — CN co rr — CN - c N O O O O O S S S r i n O b 3 fe fe m m x s - a x f e f e f e f e f e f e - K - K J J J 8 ) « t « g " lH JB4 I—t JZi o o _ fe fe N N N N N N II Ii I> 0) Si Si 03 03 CO n in in >n m «n CN CN CN i i i ON Os Os CN CN CN « « » — CN CO n m >n CN CN CN • i i ON OS OS CN CN CN NO NO NO T}- in NO r- oo in in in in in in in m in n in m in CN CN CN i i i ON ON ON CN CN CN NO NO NO NO 00 ON NO NO NO m in m in cn m CN CN CN • i i ON ON ON CN CN CN NO NO NO O Cl C N r~ r-~ r~ r ->n m m in m m m m C N C N C N C N i i i i ON ON ON ON C N C N C N C N NO NO NO NO in NO r - o r - oo m m m m in in in m CN CN CN CN I I I I ON ON ON ON CN CN CN CN NO NO NO NO —< OO ON oo r-- r-noo 'ool o o m a o m m o o n m n o o M n M o e o o m i n m i n m i n i n i n i n IN; H q IN| n os - ; ^ c N O N ' * i n c N i n c N O ^ o o c » r ^ ' — o r ~ c - > i n i n m . ON m c c o t n o i o - i n c i i T t ^ r . o i j ' r f n l ' - ' - i N W ' t t T f D ^ ^ ^ i n ^ i n i t h i o o M t i n i -© d d d © © © © d © © o " © " H N O i n N O N O c N c o c N o r ^ i f l t N O i n i o i o ^ ^ m n o — o o o o © o o ' o o o o d d o o d 3 X u 00 1 1 1 m — co —' —' CN CN CN CN © ' © ' © O © C N r - C N O N oo — < c o © c o - ^ - o o O N © o o c o © N O N O N O © T f i n O N ( ~ - i n N O r r c i i n r ^ i n c N C N c N C N r r c o c N ^ — < c N C N C N C N < n N o c N c N N O N O N O i ^ © ' © ' d o © © © © ' © © © ' © * © ' © ' © © ' © © © © © © © © © © C N c o r r < n N o r - - o o o N © O N O N O N O N O N O N O N O N O N © I — CN ro -rr - - - - - _ © © © © m m i n i n i n i n i n m i n N O N O N O N O N O ^ c ; 2 < © © © © © © CNCNCNCNCNCNCNCNCNCNCNCNCNCN,,'—'CN — CN'— —' in NO ° ° — —i NO NO NO NO t~-—' —* CN —1 CN —' —« —* —' CN U U U O O O U U C J U —*cNCNcoco'<3-tnNoi>r-~ CN o O y y s o o S o S o o S o S s s S ' S i-i CN m i n o o o o o N O N C N C N © © ~ O O © © © © © — i — CN —' CN —* CN — CN —* CN C J O O O C J O O O O O O C J O oo 00 ON ON ~ ~ CD "3 O © © © © © © © © © © © © O O © © C S O C O O © © . © © © © © © © O O © ^ t t t f t | n | n i n i ' i | n * « , C , e » « CNCNCNCNCNCNCNCNCNCNCNCNCNCNCNCN \ O ^ O N 0 ^ O ' O ^ « ^ ^ ^ N 5 ^ ^ ^ ^ ^ © © © © © ^ © © © © © © © ^ © o © CD XL at -J f-< Ut «M »M CD CD CD CD CD > > > > > TO 0 3 0 3 0 3 0 3 t ] ) O 0 ) u u C Q C Q P Q C Q C Q cd c3 at si c\t CD CD CD CD CD w u, j-i w o o o o o J : J : i x CJ o o cj . S . S . 5 . s CL, a. pu c u- a. o- 0- CJH 127 cn •M o cu ^ T3 O <+H o o o o C N -»H CU 3 I-I SH I CU r2 t f .s C O "2 8 jg >n >n m in CN CS] CN i i i ON ON ON CN CN CN VO NO NO co CN rr in NO ON ON OS ON ON in in in in in in in in in in CN C N CN cs CN i i • i i ON ON ON ON ON C N CN C N CN CN NO NO NO NO NO o o r- oo o\ ON ON Oi «n m >n vo •n >n —> CN ON ^ CO • n ' c o o d r r ^ - r ^ ^ r r r r r r r c o c o c o c o oo r r oo r~ oo ON Nn co co r r r r r r r r v i r N l > C N r r r r N q o o O N > n O N t ~ ; — O N t ^ * N o ' o d i S c N v i r s O N O N C N C O r S r S r S C N C N C O C N c O C N C N C O £ 8 * „, « <* B <* rvj--cocN — co ^ co oco — — ONinrsNorso — rf inincor^tcorrincoNOOsinin dcicidddcicicidcici rrnvoNO ^ _ _ —H ici in CN CN CN i i • O N O N O N CN CN CN N O N O N O rr m C N NO NO OO •n m in m ^ ^ to to co to CO CO to -~2 cO cO CO c C2 c i to CO CO B ,11 ,lS M 43 W MS e s s 3 i i • n c o o o N q i > r ^ — :o,NnNONONO>n>n>nNONo 00 IS rr in >n oo O N O N co oq N O in N oo •n'cocNco — ONiococNincoNOiS • n i n i n i n i n i n i n i n i n «n m «n I f Q S " E (S NO OO NO O N O in rr in C N N O o co r-—" N O N O rr — co *n C N O N co — rr sn cococococococococococococo .SP obi i n i n i n i n < n i n i n * n i n i n o i n i n p o o O N p o o c o N o < n i n < n s q - - ; O N >n rf rf «n co in >n '§ Ml lew *J CNCNOONOmm—i-^rONOoo—<—i ONf-NONO — — r - C N 0 0 O \ C N c o © O O O O — ' - ' O — O C N C N C N 1 1 1 N O r r m N O O o o o r r m r - O N t - - O i n o o r r c N C N ] N O i > o o O N - r r > n c O N q o o rrONrrcNrf — — >nco — CNCNCN s 00 CN CO I I < < I* < in N O I I < < 00 O N I I < < —i C N P . < < < Hi Q — — _ * _ < _ C N | C N C N C N C N C N C N C N © ^ © © © ^ © © © < 5 < S © ^ $ » c O © C N C O C O N O O O O N O N O — ! • — I C O C O C O C N C N C N C N C N • - " — C N C N C N C N C N C N N O N O N O N O N O N O N O N O N O N O N O N O N O CD V. 4= 1 8 J3 5 d O a > N N (U H fcu [fa cu § , H ;I +-»a 6 ex o •a 8 <*> <2 co a 3 I 3 1 cn o > Vi •rj cu I -a o i cu O °* ~ CQ 2 -s ° 03 o H CU Vi O • S H Vi fa > II o « T3 .8 1 o cu 3 .59 to .1 cn _o 8 1 •a 3 •8 | 8 H c3 .2 ai o 1 1 3 cu C/3 H 8> « 2 H 5 °° " i 8 © _; —1 CN 00 oo ©' © © oo t— © © © NO oo \c r» vo co l— CN r~- co 'cf cu cu cu •s 5 -a •s g -s .S h .S 0 H «S P H cu cu J 03 tu-rn o cz> 03 • r-H 134 fc ium ium O c Mer JJ Mer 13 in • • in a o 3 I a, S o CN O aj o s § o o CN T3 O " 0 4> X) la .S J u. W S3 -H fc w "2 S § .2 1 p i s | « o .1U()3 CN o 135 136 o* o o CO o" 00 se w X C N O < > „ O s o « o a CN © in a o •N 0 <^ «« o ft CN O • 5 CN 1 € & -a 00 ^ II CU ca o 3 *2 a £ t ca Pi I CN CU fc ( A J O | L U 3 | paidn3so/s}a|§B3 # ) A^iAijanpojj 143 CHAPTER 3 - THE ECOLOGY OF FISH-EATING BIRDS IN THE PINCHI LAKE REGION OF BRITISH COLUMBIA IN RELATION TO TROPHIC TRANSFER OF MERCURY FROM SEDIMENTS TO FISH AND FISH-EATING BIRDS - DEVELOPMENT OF HYPOTHESES Introduction An area along the Pinchi fault in central British Columbia is a known source of geologic mercury (Hg) in the form of cinnabar (HgS) (Plouffe, 1995). Previous studies (Fimreite et al, 1971; Reid and Morley, 1975; Watson, 1992; Cook, 1996; EVS Environment Consultants et al, 1999; EVS Environment Consultants, 2001) and data obtained throughout this study (see chapters 1 and 2) have shown that higher levels of Hg exist in birds, fish and sediments from Pinchi Lake, while lower amounts exist in surrounding lakes. While some of the Hg in Pinchi Lake can be attributed to past Hg mining operations, natural Hg is also a contributor (EVS Environment Consultants et al, 1999). This area proved to be an ideal location to study bioaccumulation of methylmercury (meHg) from a natural/mining-related source because lakes with varying amounts of Hg in sediments are all present in close proximity (see chapter 1). The presence of lakes with varying amounts of Hg also provided the opportunity to study relationships among Hg concentrations in sediments, fish and fish-eating birds across the study area. Methylmercury bioaccumulates and biomagnifies through aquatic food chains leading to increased exposure in those organisms at the top of the food chain (Watras et al, 1998; Kim and Burggraaf, 1999; Bowles et al, 2001). This has been shown many times in past studies of Hg accumulation in common loons (Gavia immer) (Barr, 1986; Evers et al, 1998; Meyer et al, 1998; Scheuhammer et al, 1998). In the Pinchi Lake region, fish-eating birds such as bald eagles (Haliaeetus leucocephalus), red-necked grebes (Podiceps grisgena) and common loons often occupy the position of top-predator. Bald eagles are opportunistic feeders, but are known to prefer a fish diet, especially during the breeding season (Buehler, 2000). Red-necked grebes are also abundant in the study area, and according to available data, are known to focus primarily on a diet of fish during the breeding season when fish are available (Stout and Nuechterlein, 1999). Common loons 144 are also known to subsist almost exclusively on fish when sufficient numbers offish are available (Barr, 1986). Knowing this, Hg levels in birds breeding on each study lake should correlate with fish Hg concentrations, which in turn, have been shown to be related to sediment Hg concentrations (see chapter 1). Analysis of stable nitrogen and carbon isotopes is increasingly being used as an analytical tool to verify an organism's position in the food web, and to support the bioaccumulation and biomagnification of Hg and other contaminants such as organochlorines by showing that organisms feeding at a higher trophic position (8N) have higher contaminant concentrations in tissues (Cabana and Rasmussen, 1994; Atwell et al, 1998; Kidd et al, 1999; Bowles et al, 2001; Braune et al, 2001; Braune et al, 2002; Power et al, 2002). Stable isotopes are also being used as an alternative to stomach content analyses to provide a better method for assessing an organism's diet and position in the food web because they are a measurement of assimilated food over an extended period of time (DeNiro and Epstein, 1981; Hobson and Clark, 1992a; Atwell et al, 1998). Past studies have shown that stable isotope analyses are equally accurate and sometimes preferable to prediction of trophic level by traditional dietary methods such as examination of published dietary data (Vander Zanden et al, 1997; Vander Zanden et al, 2000). Stable nitrogen, or the ratio between 1 5 N and 1 4 N (8N), is now commonly used to estimate an organism's position in the food chain (Minagawa and Wada, 1984; Estep and Vigg, 1985; Hesslein et al, 1991; Gu et al, 1996; Jarman et al, 1996). An increase of approximately 3.4%0 in 5N is expected during each complete trophic transfer (Minagawa and Wada, 1984). Stable carbon signatures give an indication of the dietary carbon source of heterotrophic organisms (DeNiro and Epstein, 1978). The ratio between carbon isotopes 1 3 C and 1 2 C (8C) changes little (0.2-l%o) between trophic transfers (DeNiro and Epstein, 1978); however, there is a wide range of carbon isotopic signatures among different primary producers, and the 5C of animals from freshwater environments fall within the range of 5C values associated with freshwater primary producers (DeNiro and Epstein, 1978). The range of 8C signatures found in animals is a result of the consumption of different 145 primary producers possessing either the C 3 or C 4 photosynthetic pathway (DeNiro and Epstein, 1978). Similarities in 5C signatures between an organism and a possible dietary carbon source can indicate the relative contribution of that source to the organism's diet (Gu et al, 1996). For example, Gu et al. (1996) used 8C signatures to determine the relative contribution of various carbon sources to the diet of fish species in Lake Apopka, Florida. It was determined, based on similarities in 8C, that fish in general (range = -15.5 to -7.9%o) primarily consumed phytoplankton (-13.4%o), and not the cyanobacterium, Microcystis (-3.0%o) or cattails (-27%o) (Gu et al, 1996). Whereas a similar 8C value in fish could also be obtained from a combination of cattails and Microcystis, the 8N of fish capable of surviving on plants was >5%o which is greater than the 3.4%o enrichment in 8N for consumers (Gu et al, 1996). Using this combination of data, the primary dietary carbon source of a consumer may be determined. Smaller changes in 8C of more omnivorous organisms may indicate a preference towards certain typical dietary items; however, in order to properly interpret 8C signatures of an organism it is necessary to know the 8C signatures of all dietary items. Typically, 8C signatures of more omnivorous higher trophic level consumers are used in a more general fashion. For example, Hebert et al. (1999) used 8C measurments to determine how much of a herring gull's (Larus argentatus) diet comes from terrestrial versus aquatic foods. Stable carbon signatures were shown to increase in herring gull eggs proportionally with the amount of terrestrial food (i.e. garbage) consumed, and were lower in eggs produced by adults consuming fish (Hebert et al, 1999). In this study stable carbon and nitrogen measurements were used to facilitate the understanding of dietary relationships between 4 local fish species, including rainbow trout (Oncorhynchus mykiss), northern pikeminnow (Ptychocheilus oregonensis), kokanee salmon (Oncorhynchus nerka), and mountain whitefish (Prosopium williamsoni) (see chapter 1), and red-necked grebes and bald eagles. In addition, Hg concentrations in the fish and birds from the study area (see chapters 1 and 2, respectively) were compared 146 to one another, and to sediment Hg concentrations, to see if correlative relationships existed between the different media. Finally, field observations were conducted and interpreted together with analytical results, to examine if Hg might be a factor in nest location, breeding success/productivity, and the general ecology of fish-eating birds in the study area, including bald eagles, red-necked grebes and common loons. Methods Mercury Analyses Mercury analytical methods for sediments, and fish and bird tissues are presented in chapters 1 and 2. Stable Isotope Analyses Approximately 1 ml of blood from 5 adult eagles and 16 eaglets was freeze-dried (for a minimum of 36 hours) and stored in sterile 1.5 ml microcentrifuge tubes. Samples of whole homogenized egg (see chapter 2) from 12 red-necked grebe eggs were also freeze-dried and stored in sterile 5 ml plastic tubes prior to analysis. Dorsal muscle tissue from 90 fish, including 42 northern pikeminnow, 28 rainbow trout, 11 kokanee and 9 mountain whitefish, was freeze-dried and stored in sterile 5 ml plastic tubes. Approximately 1 mg of freeze-dried whole blood, homogenized egg, or fish muscle was sealed into a 5x9mm tin capsule (Costech Analytical Technologies, Inc., Valencia, CA, USA) for stable isotope analyses. The Stable Isotope Facility at the University of California, Davis, analyzed all samples for concurrent stable nitrogen and carbon isotopes. Details on stable isotope analyses and quality assurance are given in chapter 1. 147 Bald Eagle Feeding Habits In 2001 and 2002, obvious prey remains found in bald eagle nests were surveyed to ensure that bald eagles were feeding primarily on fish in the study area. Only larger prey remains that were readily identifiable in the field as fish, birds or mammals were accounted for. A small piece of four fish found in nests in 2001 was taken and analyzed for Hg as described in chapter 1. Fresh remains found in nests in 2002 were simply identified and left in the nest. Bald Eagle, Common Loon and Red-necked Grebe Productivity In August of 1999, a preliminary survey of total numbers of common loons (with and without chicks), bald eagles and red-necked grebes in the study area was conducted. During the summers of 2000-2002, bald eagles breeding in the study area were monitored for nesting success and productivity as described in chapter 2. In 2000, common loons were also monitored for nesting success and chick production. Nests were located by paddling the shoreline in a kayak. Loons were not monitored continuously throughout the 3-year study because it was determined at the end of the 2000 field season that blood sampling of loon adults and chicks would not be feasible given the low abundance and low level of reproductive success in the study area. While red-necked grebes were not actively monitored for nesting success and chick production, many nests were located during egg sampling. Based on these observations, approximate numbers of nesting pairs of red-necked grebes were determined for each lake. Possible causes of nest failure for both loons and grebes were determined for nests visited on more than one occasion that failed between visits. Statistics A l l data were tested for normality using the Kolmogorov-Smirnov test prior to performing statistical analyses, and log-transformed when required. Relationships among 148 Hg concentrations in sediments, fish, bald eagle (adult and eaglet) blood and red-necked grebe eggs were tested using Pearson correlations. The relationship between Hg concentrations in northern pikeminnow and red-necked grebe eggs was tested using only those fish <25 cm in fork length (approximately 180g). This limit was based on red-necked grebes requiring a diet of a maximum of 180g of food per day (Stout and Nuechterlein, 1999). Using these size restrictions, it was not possible to test for a relationship between Hg in rainbow trout and red-necked grebes eggs as the sample size of rainbow trout under 25cm was too small. Paired t-tests were used to test for a difference between trophic position (SN) and dietary carbon source (8C) of sibling ealgets. One-way analysis of variance (ANOVA) followed by Tukey's test for mean separation was used to test for significant differences in SN and 8C among eaglet blood samples and in 8N among grebe eggs from the different study lakes. Due to the presence of unequal variances, the Kruskal-Wallis (K-W) one-way A N O V A on ranks test was used to determine if a difference existed between 8C in grebe eggs from the different study lakes. Two-tailed t-tests were then used to see if grebe eggs and eaglet blood from the same study lake differed in 8N or SC. Analysis of covariance (ANCOVA) was used to determine if a relationship existed between 8N and Hg, or 8C and Hg concentrations in red-necked grebe eggs across the study lakes. A one-way A N O V A followed by Tukey's test was also used to examine general differences in 8N among the different fish and bird species, combined across all study lakes. A K-W one-way A N O V A on ranks followed by Dunn's method for multiple comparisons was used to determine if a difference existed between 8C among the different fish and bird species. For comparison to combined lake results, 8N or 8C values for all fish and bird species on Pinchi Lake alone were compared using one-way ANOVAs. A l l statistical analyses were performed using SigmaStat for Windows, version 2.03S (Jandel Scientific, 1995) and JMP, version 4.01 (SAS Institute Inc., 2001). 149 Results Mercury Relationships between Birds, Fish and Sediments A significant positive relationship was found between mean sediment Hg concentrations and mean adult bald eagle blood Hg levels (r = 0.97, P = 0.028). This same relationship was also found between sediment Hg and eaglet blood Hg concentrations (r = 0.91, P = 0.034) and sediment Hg and red-necked grebe egg Hg concentrations (r = 0.95, P = 0.0495). Relationships are shown on Figure 3.1. No relationship was found between mean Hg concentrations in rainbow trout and mean blood Hg concentrations in adult bald eagles or eaglets. A significant positive correlation was found between mean Hg concentrations in adult bald eagle blood and northern pikeminnow muscle tissue (r = 0.96, P = 0.045). A marginally significant relationship was also found between Hg levels in eaglet blood and northern pikeminnow muscle tissue (r = 0.88, P = 0.050). Relationships between bald eagle blood and northern pikeminnow muscle Hg concentrations are shown on Figure 3.2. A significant positive relationship was also found between Hg concentrations in red-necked grebe eggs and northern pikeminnow <25cm (r = 0.96, P = 0.045), as shown on Figure 3.3. Positive relationships were found between Hg concentratios in red-necked grebe eggs and adult bald eagle blood (r = 0.998, P = 0.039), and between Hg concentrations in the blood of adult bald eagles and eaglets (r = 0.989, P = 0.011). However, no significant relationship was found between Hg concentrations in grebe eggs and eaglet blood. Relationships are shown on Figure 3.4. Stable Isotopes Stable carbon and nitrogen isotope analytical results for adult bald eagles, eaglets and red-necked grebe eggs are presented in Tables 3.1, 3.2 and 3.3, respectively. No 150 significant difference was found between 5C and 8N for sibling eaglets, so sibling 5C and 5N measurements were averaged for subsequent analyses. No significant differences were found in 5N or 5C among grebe eggs from the different study lakes. No significant differences were found in 8N among eaglets from the different study lakes, however a significant difference was found in 8C (P = 0.013). Specifically, eaglets from Stuart Lake had a different dietary carbon source (-25.65%o) compared to eaglets from Great Beaver Lake (-29.52%o). A significant positive relationship (r = 0.854, P = 0.003) between Hg concentrations and 8N in red-necked grebe eggs was found after correcting for differences in Hg concentrations among study lakes (Figure 3.5). No relationship was found between Hg concentrations and 8C in red-necked grebe eggs. Examination of mean 8C from all fish and bird species combined across the five study lakes showed that only kokanee differed significantly in dietary carbon source (8C = -31.38%o) compared to all other fish and birds (Figure 3.6). A similar analysis of mean 8N in all fish and birds showed that adult bald eagles (8N = 11.17%o), eaglets (8N = 10.62%o) and red-necked grebes (8N = 10.67%o) occupied a significantly higher trophic position compared to all fish species (Figure 3.7). Adult bald eagles, eaglets and red-necked grebes did not differ significantly in trophic position. It should be noted that when comparisons were limited to the 4 fish species, northern pikeminnow were found to occupy a higher trophic position compared to mountain whitefish (P = 0.030), kokanee salmon (P = 0.033) and rainbow trout (P = 0.006). When analyses were limited to only those fish and birds sampled from Pinchi Lake, again a significant difference in dietary carbon source (8C) was found between kokanee and rainbow trout, pikeminnow, adult eagles and grebes, but not mountain whitefish or eaglets. A significant difference (P = 0.034) in 8C between rainbow trout and mountain whitefish from Pinchi Lake was also noted. A graphical representation of mean 8C ± SD in fish and birds from Pinchi Lake is shown on Figure 3.8. Eaglets, adult eagles and red-necked grebes also occupied a higher trophic position in Pinchi Lake, but only significantly compared to rainbow trout (Figure 3.9). Adult eagles were also found to 151 occupy a higher trophic position compared to kokanee salmon from Pinchi Lake (P = 0.045). The relationship between Hg and 5N for individual fish and birds from all study lakes is shown on Figure 3.10. Bald Eagle Feeding Habits Bald eagles are opportunistic feeders, but are known to prefer a fish diet, especially during the breeding season (Buehler, 2000). This was true in our study area as well. Fish were commonly observed being carried back to the nest by adult eagles, and examination of 34 obvious prey remains in nests from 2001 and 2002 showed that fish (i.e. those that could not be swallowed whole) comprised approximately 85% of the large prey in their diet at the time of sampling (Table 3.4). In 2001, 100% of large prey remains found in nests were from fish, whereas in 2002, 76% were of fish origin, 19% were bird and 5% were mammalian (bear). Data on prey remains found in and below bald eagle nests is normally biased toward prey such as birds (Vermeer et al., 1989). Therefore, the fact that mainly fish remains were found in nests is likely strong evidence that these eagles were eating primarily fish. Both of the main study species of fish (northern pikeminnow and rainbow trout) were found in bald eagle nests, in addition to several other species of fish (i.e. lake trout, lake whitefish and largescale sucker). Mercury concentrations in fish sampled from bald eagle nests ranged from very low in one rainbow trout from Great Beaver Lake [0.17 fxg/g dry weight (dw)] to high in one lake trout tail from Pinchi Lake (4.89 ug/g dw). Since bald eagles were observed to actively feed on the same lakes where they nest, this allowed for comparison of Hg concentrations in blood to fish and sediment Hg concentrations. Ecology of Bald Eagles, Common Loons and Red-necked Grebes in the study area Bald Eagles Nesting success and productivity of bald eagles from 2000-2002 is discussed in chapter 2. Overall, no significant differences in either reproductive success or productivity were 152 found between eagles nesting on Pinchi Lake compared to all other study lakes. Reproductive success (the total number of active territories found at the beginning of May that produced 8-week-old eaglets) was 62% on Pinchi Lake compared to 64% on all other study lakes combined (P = 0.95). Average productivity (the total number of eaglets produced per occupied territory) over the three-year study was 0.98 on Pinchi Lake compared to 1.17 on all other study lakes combined (P = 0.48). According to Sprunt et al. (1973), productivity of bald eagles from the study area was similar to that of a healthy population producing a surplus of chicks (i.e. greater than 1.0 chick/occupied territory), and greater than 0.7 young/occupied territory, the value associated with a sustaining population. Common Loons Approximate numbers of breeding pairs of common loons were determined for each study lake in 2000. These numbers are given as approximate because nests were only found for 9 loon pairs (Table 3.5). In total, 2 breeding pairs of loons were found on Pinchi Lake, 8 on Tezzeron, 6 on Fraser, 6 on Great Beaver, and possibly 2 on the portion of Stuart Lake involved in the study. Of the 9 nests located, 2 were obviously new, but did not contain eggs. Based on our observations, only one of the 9 nests found produced a chick, and this chick later disappeared (Table 3.5). A l l other nests likely failed at the egg or very early chick stage. In addition to the nests that were found, 6 families of loons that produced chicks were located after the nesting period (Table 3.6). Unfortunately, only 3 of these families were actually located on the main study lakes, and as noted in Table 3.6, 3 of these loon chicks also disappeared at a later date. Common loons appeared very territorial in the study area, with distances of several hundred metres between breeding territories. In no instances would more than one pair of loons occupy a given shallow weedy area (the observed preferred location for nests and/or breeding pairs) or island, regardless of the size of the area. They would, however, share the area with breeding pairs of red-necked grebes. Loons in the study area tended to build large nests consisting mostly of dead aquatic vegetation. In some cases, nests 153 were floating and anchored to large reeds or cattails. In other cases, nests were built on solid ground at the shoreline. Several of the nests found at the start of the nesting period had virtually disappeared by late June. Unlike the red-necked grebes from the study area, loons that lost their nests due to bad weather in June of 2000 did not appear to attempt to re-nest. Common loons must have chicks to enable capture of adults and young for blood and feather sampling (Evers, 1992). Very few loon pairs in the study area produced chicks, and of those that did, few survived to reach a suitable capture age. Based on the results of the 2000 study it was determined that loon sampling would not be feasible in the study area. In addition, very few loons were actively breeding on Pinchi Lake, which was the focal lake of this study. Therefore, detailed information on loon nesting habits in the study area was limited to 2000. Red-Necked Grebes Throughout the 3-year study period, red-necked grebes were found to be reasonably abundant on all study lakes except Stuart. The largest number of grebes was found on Fraser Lake, followed by Great Beaver, Pinchi and Tezzeron Lakes. An approximate number of breeding pairs per lake was determined based on observations of pairs of grebes continuously occupying a territory throughout the breeding season, or by location of nests during egg sampling. In total, approximately 27 breeding pairs of grebes were located on Fraser Lake, 12 on Great Beaver, 10 on Pinchi, and 7 on Tezzeron. No nests or territorial pairs of grebes were found on the portion of Stuart Lake involved in the study. Over the course of the study, more than 100 red-necked grebe nests were located in the study area. Grebes appeared much less territorial than common loons in the study area, often constructing nests only a few metres apart. Grebes in the study area built floating nests primarily using dead emergent aquatic vegetation (i.e. reeds) topped with submergent aquatic vegetation. The vegetation used to construct the nest was woven into 154 dead or live emergent aquatic vegetation anchored in the lake bottom, or the submerged base of small bushes or trees. Many of the shallow weedy areas preferred for nesting by red-necked grebes in the study region were on exposed shorelines, and in the case of Pinchi Lake, on a shallow point extending from the southern shore partway out towards the middle of the lake. As a result, grebe nests were often constructed in areas with little or no protection against waves produced during bad weather. In addition, a considerable amount of windy weather was experienced during the month of June when grebes were nesting, particularly in 2000 and 2001. The result was that many grebes had to re-nest several times as their nests were destroyed by wave action. For example, one pair of grebes was observed to re-nest 4 times on Pinchi Lake in June of 2001. On most study lakes, grebe chicks were not evident until the end of July. At this point, many were at least 3 weeks old, based on known incubation periods. Some adults were observed sitting on the nest well after the eggs should have hatched. It appears that, unlike loons, grebe chicks in the study area do not venture far from the nest or sheltered areas until they are older and capable of diving. Discussion Mercury Relationships between Birds, Fish and Sediments Based on Hg-analysis of sediments, fish and tissues from bald eagles (blood and feathers) and red-necked grebes (eggs), a gradient in Hg concentrations was found between the different study lakes (see chapters 1 and 2). This allowed for analysis of correlations between Hg concentrations in sediments, fish and fish-eating birds across all 5 study lakes. Mercury concentrations in sediments were found to add significantly to the prediction of Hg concentrations in both rainbow trout and northern pikeminnow from the study area (see chapter 1). In this study, it was also found that blood Hg concentrations 155 from adult bald eagles and eaglets, as well as Hg concentrations in red-necked grebe eggs, were significantly and positively correlated to Hg concentrations in sediments. Based on this, a positive relationship should also exist between Hg concentrations in fish and bald eagles or red-necked grebes. While a significant positive relationship was found between Hg concentrations in northern pikeminnow and bald eagles, this was not found with rainbow trout. Average Hg concentrations in rainbow trout were only significantly greater in fish from Pinchi Lake (mean = 0.36 ug/g dw), and no decreasing trend in tissue Hg concentrations was found for the other study lakes (mean rainbow trout Hg concentrations in Tezzeron and Stuart = 0.22 ug/g dw, Great Beaver and Fraser = 0.21 ug/g dw) (see chapter 1). Hence no significant relationship between Hg concentrations in eagle blood and rainbow trout could be found. It was not possible to test for a relationship between Hg concentrations in red-necked grebe eggs and rainbow trout <25cm in fork length because the sample size of rainbow trout <25cm was too small. Previous analysis of Hg concentrations in the blood of adult bald eagles and eaglets from the same nest showed a significant positive relationship (see chapter 2). The relationship of blood Hg concentrations from adult bald eagles to eaglets from the same lake was also positively correlated, with adult eagles and eaglets from Pinchi Lake having the highest blood Hg concentrations, followed by Tezzeron, Stuart and Fraser Lakes. The same trend was found between adult bald eagles and red-necked grebe eggs, however the significance is based on a very small sample size (n = 3). A positive relationship was also found between Hg concentrations in the blood of eaglets and red-necked grebe eggs, however this relationship was not statistically significant. Again, this could be due to the small sample size (n = 4) because the relationship was only marginally non-significant (P = 0.073) and significant relationships were found between adult eagles and grebe eggs, as well as between adult eagles and eaglets. Sampling of more lakes along the Hg-gradient in the study area may have provided correlations with greater significance between the different environmental media. It is not surprising that positive correlations were found between total Hg concentrations in sediments, fish and fish-eating birds given the ability of meHg to bioaccumulate and 156 biomagnify through aquatic food chains (Mason et al, 1995; Watras et al, 1998; Kim and Burggraaf, 1999; Bowles et al, 2001). Mercury becomes available to the aquatic food chain through net bacterial methylation of its inorganic counterpart (Hg2 +) in surficial sediments and the water column (Winfrey and Rudd, 1990; Zillioux et al, 1993). Several factors are known to affect Hg methylation and bioaccumulation, such as lake pH, DOC, temperature and Se availability (Turner and Swick, 1983; Suns and Hitchin, 1990; Wiener et al, 1990; Winfrey and Rudd, 1990; Bodaly et al, 1993). The effects of lake pH, DOC and Se availability on Hg accumulation in the study lakes are discussed in chapter 1. Dietary uptake is believed to be the dominant route for meHg bioaccumulation in fish (Hall et al, 1997). Birds then consume these fish, thereby accumulating meHg that originated in the study lake sediments and water. Even though a mean blood Hg concentration per lake was used to detect relationships between adult bald eagles and sediments, fish, or other birds, the variation in blood Hg concentrations of adult bald eagles from the same lakes was considerable. For example, blood Hg concentrations ranged from 4.245 to 9.435 (ig/g wet weight (ww) in adults from Pinchi Lake. These differences may have been partially influenced by differences in trophic position (8N). The adult with the highest blood Hg from Pinchi Lake had a 8N = 11.40%o, whereas the adult with the lowest blood Hg from Pinchi Lake had a 5N = 10.05%o. Stable isotopes were only measured in 5 of 13 adult eagle blood samples so this data can only partially explain the variation in blood Hg concentrations found. Examining prey remains found in the nests showed considerable variation in prey consumed by different eagles. Large lake trout, which are known to have some of the highest Hg concentrations in the study area (EVS Environment Consultants, 2001), were found in the nest of the eagle with the highest blood Hg on Pinchi Lake. Heads from two rainbow trout were found in the nest of the adult with the lowest blood Hg on Stuart Lake. Rainbow trout have been shown to have relatively low Hg concentrations in muscle compared to other fish species from the study area (see chapter 1, EVS Environment Consultants, 2001). Overall, a wide range of prey items (i.e. bear, duck, gull and several fish species) were found in bald eagle nests, which was likely the major reason for the wide range in adult bald eagle blood Hg concentrations. One other 157 possibility is that bald eagles obtained some prey from lakes other than the chosen nesting lake. For example, the adult eagle with the highest blood Hg concentration on Tezzeron Lake (4.676 ug/g ww) was nesting on the south side of the lake, immediately north of Pinchi Lake. The same was true for the adult eagle with the highest blood Hg from Stuart Lake (4.864 io.g/g ww). This bird was found nesting on Battleship Island, which is on the north side of Stuart Lake, immediately south of Pinchi Lake. It is possible that these birds had higher blood Hg concentrations because they occasionally obtained fish prey from Pinchi Lake. In a similar manner, the bald eagle with the lowest blood Hg on Pinchi Lake may have spent time fishing elsewhere. Variation in both diet and feeding location probably influenced Hg concentrations in adult eagle blood. Despite the variations in blood Hg concentrations, the positive relationships found between Hg concentrations in northern pikeminnow and the blood of adult bald eagles and eaglets, and red-necked grebe eggs provides an indication that these birds are primarily feeding on the study lake where they are breeding. Scheuhammer et al. (1998) found this same relationship between Hg concentrations in fish and blood from adult common loons and chicks in Ontario. The relationship is less pronounced in eaglets compared to adult eagles because they were actively eliminating blood Hg through feather growth over their entire bodies, in comparison to adults, which replace feathers slowly over a 6-month period (see chapter 2; McCollough, 1989). As a result, eaglet blood Hg concentrations were approximately a full order of magnitude lower than adult eagle blood Hg concentrations, even though they likely fed on the same dietary items. Stable Isotopes To better understand the trophic position occupied by the fish and birds examined in this study, and the bioaccumulation of Hg in the study area, measurements of stable carbon and nitrogen isotopes were undertaken in fish and bird tissues. Measurements were primarily used to determine if stable isotopes could explain variations in contaminant concentrations among fish and birds, in addition to providing information on the relationships between different species of fish and the birds that are feeding on them 158 (bald eagles and red-necked grebes). Previous studies have successfully employed stable isotope analyses to explain Hg concentrations in fish (Cabana and Rasmussen, 1994; Atwell et al, 1998; Bowles et al, 2001; Power et al, 2002) and contaminant concentration variation among bird species (Bearhop et al, 2000; Wayland et al, 2000; Braune et al, 2001; Braune et al, 2002). In the current study, Hg concentrations in red-necked grebe eggs were significantly related to trophic position (8N). Previous studies have shown that stable nitrogen and carbon isotopes in eggs reflect the diet of the female prior to and during egg-laying (Hobson, 1995; Hebert et al, 1999). It is also known that Hg concentrations in eggs are related to the dietary Hg concentrations of female birds (Wolfe et al, 1998). This suggests that even slight changes in the trophic position of the female adult grebe can result in increased meHg consumption, leading to increased deposition of meHg in freshly laid eggs. Combining samples across all 5 study lakes showed that, on average, northern pikeminnow occupied a higher trophic position in comparison to other study fish. This is as expected knowing that northern pikeminnow are typically piscivorous whereas mountain whitefish and rainbow trout are more omnivorous, feeding on plankton, other small fishes, crustaceans, aquatic insects and other invertebrates (Coad et al, 1995). Young kokanee salmon prefer to feed primarily on planktonic crustaceans (Coad et al, 1995). Bald eagles and red-necked grebes occupied a higher trophic position compared to all fish. Both bird species prefer a fish diet when available, particularly in the breeding season (Stout and Nuechterlein, 1999; Buehler, 2000), and would be able to consume larger fish than the northern pikeminnow captured in this study. Thus, measurement of stable nitrogen isotopes appears to correspond with known data regarding the trophic status of all species involved in this study. While stable nitrogen levels in bald eagles and red-necked grebes indicate that they do occupy a higher trophic position compared to the fish species from the study area, the increase of approximately 3.4%o in 5N expected during each complete trophic transfer 159 (Minagawa and Wada, 1984) was not found. There are a few possible reasons for this difference. The most likely factor is that the diets of the various fish species, especially northern pikeminnow, overlapped with bald eagle and red-necked grebe diets. This is further supported by a lack of significant difference in dietary carbon source (5C) among fish and birds (except kokanee salmon). Northern pikeminnow are mainly piscivorous (Coad et al, 1995), as are bald eagles and red-necked grebes during the breeding season (Stout and Nuechterlein, 1999; Buehler, 2000). Based on this, all three species could be consuming mainly a fish diet, and the significant difference in trophic position arose because bald eagles and red-necked grebes consumed larger fish than northern pikeminnow. Fish, including the northern pikeminnow and rainbow trout from the current study (see chapter 1), have been shown to increase in trophic position with increasing length (Gu et al, 1996). Therefore, consumption of larger fish could have lead to the higher 5N in eagles and grebes compared to pikeminnow, but not the 3.4%o difference in 5N expected during each complete trophic transfer. Use of different tissues is another factor that may affect 5N and 5C among the different species examined. Muscle was taken from fish, and blood or eggs were used to determine trophic status of birds. Different tissues have been shown to have different 8N and SC fractionation factors within a species (Hobson and Clark, 1992b; Pinnegar and Polunin, 1999). Fractionation factor refers to the difference in isotopic signature between tissues of the diet and consumer. For example, mean tissue 8N fractionation factors (± SD) for muscle, liver and feather of ring-billed gulls (Larus delawarensis) were + 1.4 ± 0.1%o, +2.7 ± 0.1%o and +3.0 ± 0.2%o, respectively (Hobson and Clark, 1992b). Similarly, SC was shown to vary from +0.2 ± 0.6%o in liver to +2.7 ± 0.4%o in collagen of Japanese quail (Coturnix japonica) (Hobson and Clark, 1992b). Also, different species of birds have been shown to vary significantly in diet-tissue fractionation factors (Hobson and Clark, 1992b). For example, mean blood 5N fractionation factors (± SD) for Japanese quail, ring-billed gulls and peregrine falcons (Falco peregrinus) were +2.2 ± 0.2%o, +3.1 ± 0.2%o and +3.3 ±0.4%o, respectively (Hobson and Clark, 1992b). Similarly, blood SC was shown to vary from -0.3 ± 0.8%o in ring-billed gulls to 160 +1.2 ± 0.8%o in Japanese quail (Hobson and Clark, 1992b). Based on this, the isotopic fractionation factor of +3.4%o in 5N typically used to describe one complete trophic transfer may be not be indicative of the species examined in our study. Stable nitrogen isotopes have also been shown to be specific to the food web being examined, and can vary considerably in baseline levels between systems (Cabana and Rasmussen, 1996). For this reason, researchers have begun to use stable isotope analysis of unionid mussel tissue to develop a baseline 5N in each study lake (Cabana and Rasmussen, 1996; Vander Zanden et al, 1997; Lake et al, 2001). Unionid mussels live for many years, feeding primarily on phytoplankton or bacteria (inorganic nitrogen users), making their average 5N signatures less sensitive, compared to plankonic organisms, to short-term or seasonal fluctuations in 8N (Cabana and Rasmussen, 1996). Since we did not develop a baseline for each study lake, 8N and SC in birds and fish from Pinchi Lake alone were compared in the same manner as those from all lakes combined to see if a large difference in baseline levels might exist. Limiting analyses to Pinchi Lake alone did increase variability and reduce the sample size considerably, however, based on the similarity in trends shown among figures 3.6 through 3.9, baseline stable isotope levels likely did not vary considerably between the study lakes. Among the different fish and bird species studied, SC was shown to be significantly different only in kokanee salmon across the combined study lakes. This suggests that the main dietary carbon source for kokanee is different from that of the other fish species involved in the study. According to published literature, young kokanee prefer to feed on planktonic crustaceans, unlike the other fish in the study which tend to consume a combined diet of insect larvae, crustaceans, molluscs, fish eggs and other fish species (Coad et al, 1995). Because of this difference, it is also likely that kokanee comprise a smaller portion (if any) of bald eagle or red-necked grebe diets given the mean fractionation factors between kokanee and grebes and eaglets (-3.0%o) or adult eagles (-4.7%o). These fractionation factors are considerably greater than any found in the various species of birds on known diets examined by Hobson and Clark (1992b). In addition, DeNiro and Epstein, (1978) found that the ratio between carbon isotopes 1 3 C 161 and , 2 C (5C) changes little (0.2-l%o) between trophic transfers. Kokanee were not found in any of the bald eagle nests examined, however species were not determined for fish skeletons found in nests, and fish small enough to be swallowed whole would not have been considered. A more thorough understanding of bald eagle and fish diets and their relationship to 5C may have been gained had 5C been determined in tissues of all prey species contributing to their diets (i.e. birds, mammals, and additional fish in bald eagle diets; plankton, insects and crustaceans, etc. in fish diets). A detailed study of prey brought to the nest would also have been required in the case of bald eagles. Unfortunately, it was beyond the scope of this study to do an intense examination of fish and bird diets. Ecology of Common Loons and Red-necked Grebes in the study area Bald Eagles Discussion relating to bald eagle nest locations, nesting success and productivity can be found in chapter 2. Common Loons Common loons nesting in the study area appeared to have a very low rate of nesting success and productivity. Almost all of the nests that were located in 2000 did not produce chicks. The locations and structure of the nests left them vulnerable to both predators (primarily nests on shorelines) and large waves produced during bad weather (both floating nests and nests on exposed shoreline). Several of the nests found at the start of the nesting period virtually disappeared later in June, suggesting that they were destroyed by large waves. The one pair of adults that did produce a chick from one of two eggs laid in a nest on Fraser Lake later lost their chick. Several other chicks that were found later in the 2000 study season also disappeared on subsequent visits. Given the large size of some of the loon chicks that disappeared, it was suspected that predators, particularly bald eagles, were the cause. Bald eagles are plentiful in the study area, and were seen actively hunting in areas where common loon chicks were found. Although 162 eagles were never observed to take a common loon chick, it has been documented in the past (Paruk et al, 1999). Bald eagles have also been documented attacking a nesting adult common loon (Vlietstra and Paruk, 1997). Carcasses of ducks and gulls were found in the nests of bald eagles in the study area, so it is possible that eagles were the major cause of loon chick loss in the study area. Overall, it did not appear that Hg levels affected the nesting success of common loons in the study area. While 2 breeding pairs of loons were found on Pinchi Lake in 2000, only one pair of loons appeared to consistently return to Pinchi Lake for breeding over the course of the study. Of all loon pairs observed throughout the study period, this pair of loons on Pinchi Lake was the only pair within the study area to successfully raise at least one chick in 3 of 4 years of observations (1999, 2001 and 2002). Mercury concentrations found in forage fish and red-necked grebe eggs from the study area (see chapters 1 and 2) suggest that current Hg levels would not likely affect common loons breeding on Pinchi Lake. It may be that fewer loons nest on Pinchi Lake because suitable nesting habitat is limited compared to other lakes in the study area. Shallow weedy/bog areas or small islands are documented as the preferred nesting habitat of common loons (Mclntyre and Barr, 1997) and appeared to be the preferred location for nests and/or breeding pairs in our study area as well. Since Pinchi Lake has only one small island and two weedy areas, it is possible that common loons have simply chosen to nest elsewhere. Red-necked Grebes Unlike common loons, the number of red-necked grebes breeding on Pinchi Lake was fairly consistent from year to year, and did not appear to be reduced compared to other lakes in the study area. Because of the necessity of creating floating nests, habitat appeared to be the limiting factor for red-necked grebes. Nesting was semi-colonial on the study lakes, with nests often no more than a few metres apart. These floating nests were very susceptible to destruction by wave action, which appeared to be the main cause of grebe nest failure. Even if the nest was not completely destroyed by waves, and had only come unanchored, the nest was often abandoned. Only two of the more than 100 163 nests found over the course of the study had obviously been depredated, with eggshell remnants still present in the nest. While total numbers of breeding pairs of grebes exceeded those of loons in the study area, it is uncertain as to whether nesting or fledging success was any greater for grebes. In fact, fewer grebe chicks were observed over the period of 2000-2002 than loon chicks. This may be partly due to the apparent tendency of adult grebes to confine chicks to protected areas when they are very young. Only chicks old enough to dive were apparent in open waters in the study area. If this was the case, fledging success of grebe hatchlings may be greater than that of loons from the study area. Mortality of grebe chicks over 1 month old was found to be minimal in past studies (Stout and Nuechterlein, 1999). Mercury concentrations in grebe eggs from all 5 study lakes were below the lowest observed adverse effect level found in previous egg-Hg studies (0.5 ug/g ww; see chapter 2), and forage fish of the size consumed by red-necked grebes were below the Hg threshold effects level determined for common loons (0.3 ug/g ww; see chapter 1). Grebe numbers on Pinchi Lake were comparable to both Tezzeron and Great Beaver Lakes. Fraser Lake, which had considerably more sheltered and shallow, weedy habitat compared to the other study lakes also had a much larger grebe population. Based on these observations, it did not appear that Hg was a factor in choice of nesting location or nesting success of red-necked grebes in the study area. Summary and Potential Future Work Relationships among Hg concentrations in sediments, northern pikeminnow, red-necked grebe eggs and bald eagles from the 5 study lakes were found. The relationships indicated that Hg concentrations were greatest in Pinchi Lake, followed by Tezzeron, Great Beaver and Stuart, and Fraser Lakes. A relationship between 8N and Hg concentrations in red-necked grebe eggs was found, indicating that female grebes feeding at a higher trophic position deposit a greater amount of Hg in their eggs. A l l bird species 164 were found to be feeding at a higher trophic position compared to the various fish species, however a possible overlap in diet, among other factors, resulted in lower than expected 5N fractionation factors between bird and fish tissues. Additional research examining stable carbon and nitrogen isotopes in all parts of the food web (i.e. insects, plankton, crustaceans, other fish and bird species, etc.) could provide valuable information on the dietary habits of the various fish and bird species examined. Based on a three-year investigation of bald eagles in the study area, it was determined that Hg does not appear to be a factor in nesting success and productivity. It is also doubtful that Hg is a factor in the nesting success and productivity of red-necked grebes and common loons, however additional work would be required to confirm this. In general, it appears that the loon and grebe populations breeding on the study lakes may be 'sink' populations. While many adults of both species appeared to be attempting to breed on the study lakes, very few offspring were observed. Based on this, there must be productive lakes elsewhere that are producing a surplus of recruits into the population and/or there must be years when reproduction is very good and enough birds are produced to maintain the population. Further detailed research would be required to accurately assess the population dynamics of the loons and grebes in the study area. Acknowledgements TeckCominco, Ltd., the Metals In The Environment Research Network, and the Canadian Wildlife Service provided funding for this study. Field assistance was provided by Lori Smith, Jennifer Young and Sandi Lee. Darcy Haycock and Gregg Howald provided additional assistance in eaglet sampling and adult bald eagle capture, respectively. Randy Baker and EVS Environment Consultants, Inc. supplied some of the fish used in this study. 165 References Atwell, L., Hobson, K. A., and Welch, H. E. (1998). Biomagnification and bioaccumulation of mercury in an arctic marine food web: Insights from stable nitrogen isotope analysis. Can. J. Fish. Aquat. Sci. 55, 1114-1121. Barr, J. F. (1986). Population dynamics of the Common loon (Gavia immer) associated with mercury-contaminated waters in northwestern Ontario. Can. Wildl. Serv. Occas. Pap. No. 56, pp. 25. Bearhop, S., Waldron, S., Thompson, D., and Furness, R. (2000). Bioamplification of mercury in great skua Catharacta skua chicks: The influence of trophic status as determined by stable isotope signatures of blood and feathers. Mar. Pollut. Bull. 40, 181-185. Bodaly, R. A., Rudd, J. W. M . , Fudge, R. J. P., and Kelly, C. A. (1993). Mercury concentrations in fish related to size of remote Canadian Shield lakes. Can. J. Fish. Aquat. Sci. 50, 980-987. Bowles, K. C , Apte, S. C , Maher, W. A., Kawei, M . , and Smith, R. (2001). Bioaccumulation and biomagnification of mercury in Lake Murray, Papua New Guinea. Can. J. Fish. Aquat. Sci. 58, 888-897. Braune, B. M . , Donaldson, G. M . , and Hobson, K. A. (2001). Contaminant residues in seabird eggs from the Canadian Arctic. Part I. Temporal trends 1975-1998. Environ. Pollut. 114, 39-54. Braune, B. M . , Donaldson, G. M . , and Hobson, K. A. (2002). Contaminant residues in seabird eggs from the Canadian Arctic. II. Spatial trends and evidence from stable isotopes for intercolony differences. Environ. Pollut. 777, 133-145. Buehler, D. A. (2000). Bald Eagle. The Birds of North America 506, 39pp. Cabana, G., and Rasmussen, J. B. (1994). Modelling food chain structure and contaminant bioaccumulation using stable nitrogen isotopes. Nature 372, 255-257. Cabana, G., and Rasmussen, J. B. (1996). Comparison of aquatic food chains using nitrogen isotopes. Proc. Natl. Acad. Sci. USA 93, 10844-10847. Coad, B. W., Waszczuk, H., and Labignan, I. (1995). Encyclopedia of Canadian Fishes (Singapore: Canadian Museum of Nature and Canadian Sportfishing Productions, Inc.). 166 Cook, S. J., Jackaman, W., McCurdy, M . W., Day, S. J., and Friske, P. W. (1996). Regional lake sediment and water geochemistry of part of the Fort Fraser map area, British Columbia (NTS 93K/9, 10, 15 and 16): Ministry of Employment and Investment, Energy and Mines Division, Geological Survey Branch. DeNiro, M . J., and Epstein, S. (1978). Influence of diet on the distribution of carbon isotopes in animals. Geochim. Cosmochim. Acta 42, 495-506. DeNiro, M . J., and Epstein, S. (1981). Influence of diet on the distribution of nitrogen isotopes in animals. Geochim. Cosmochim. Acta 45, 341-351. Estep, M . L. F., and Vigg, S. (1985). Stable carbon and nitrogen isotope tracers of trophic dynamics in natural populations and fisheries of the Lahontan Lake system, Nevada. Can. J. Fish. Aquat. Sci. 42, 1712-1719. Evers, D. C. (1992). A replicable capture method for adult and juvenile common loons on their nesting lakes. In 1992 Conference on the Loon and its Ecosystem: Status, Management, and Environmental Concerns, Bar Harbor, ME, USA, pp. 214-220. Evers, D. C , Kaplan, J. D., Meyer, M . W., Reaman, P. S., Braselton, W. E., Major, A., Burgess, N . , and Scheuhammer, A. M . (1998). Geographic trend in mercury measured in common loon feathers and blood. Environ. Toxicol. Chem. 77, 173-183. EVS Environment Consultants. (2001). Regional survey of fish mercury concentrations -Pinchi Lake, B.C. (Vancouver: TeckCominco Ltd.), pp. 91. EVS Environment Consultants, Norecol Dames and Moore, and Frontier Geosciences. (1999). Pinchi Lake Mine Site and Lake Investigation: Environmental Assessment Report (Vancouver: Cominco Ltd.). Fimreite, N . , Holsworth, W. N . , Keith, J. A., Pearce, P. A., and Gruchy, I. M . (1971). Mercury in fish and fish-eating birds near sites of industrial contamination in Canada. Can. Field-Nat. 55,211-220. Gu, B., Schelske, C. L., and Hoyer, M . V . (1996). Stable isotopes of carbon and nitrogen as indicators of diet and trophic structure of the fish community in a shallow hypereutrophic lake. J. Fish Biol. 49, 1233-1243. Hall, B. D., Bodaly, R. A., Fudge, R. J., Rudd, J. W., and Rosenberg, D. M . (1997). Food as the dominant pathway of methylmercury uptake by fish. Water Air Soil Pollut. 100, 13-24. Hebert, C. E., Shutt, J. L., Hobson, K. A., and Weseloh, D. V. C. (1999). Spatial and temporal differences in the diet of Great Lakes herring gulls (Larus argentatus): Evidence from stable isotope analysis. Can. J. Fish. Aquat. Sci. 56, 323-338. 167 Hesslein, R. H., Capel, M . J., Fox, D. E., and Hallard, K. A. (1991). Stable isotopes of sulfur, carbon, and nitrogen as indicators of trophic level and fish migration in the lower Mackenzie River basin. Can. J. Fish. Aquat. Sci. 48, 2258-2265. Hobson, K. A. (1995). Reconstructing avian diets using stable-carbon and nitrogen isotope analysis of egg components: patterns of isotopic fractionation and turnover. Condor 97, 752-762. Hobson, K. A., and Clark, R. G. (1992a). Assessing avian diets using stable isotopes I: Turnover of 1 3 C in tissues. Condor 94, 181-188. Hobson, K. A., and Clark, R. G. (1992b). Assessing avian diets using stable isotopes II: Factors influencing diet-tissue fractionation. Condor 94, 189-197. Jandel Scientific. (1995). SigmaStat Statistical Software, Version 2.03S for Windows 95, NT & 3.1. Jandel Scientific, San Rafael, CA. Jarman, W. M . , Hobson, K. A., Sydeman, W. J., Bacon, C. E., and McLaren, E. B. (1996). Influence of trophic position and feeding location on contaminant levels in the Gulf of the Farallones food web revealed by stable isotope analysis. Environ. Sci. Technol. 30, 654-660. Kidd, K. A., Paterson, M . J., Hesslein, R. H., Muir, D. C. G., and Hecky, R. E. (1999). Effects of northern pike (Esox lucius) additions on pollutant accumulation and food web 13 15 structure, as determined by 8 C and 8 N , in a eutrophic and an oligotrophic lake. Can. J. Fish. Aquat. Sci. 56, 2193-2202. Kim, J. P., and Burggraaf, S. (1999). Mercury bioaccumulation in rainbow trout (Oncorhynchus mykiss) and the trout food web in lakes Okareka, Okaro, Tarawera, Rotomahana and Rotorua, New Zealand. Water Air Soil Pollut. 115, 535-546. Lake, J. L., McKinney, R. A., Osterman, F. A., Pruell, R. J., Kiddon, J., Ryba, S. A., and Libby, A. D. (2001). Stable nitrogen isotopes as indicators of anthropogenic activities in small freshwater systems. Can. J. Fish. Aquat. Sci. 58, 870-878. Mason, R. P., Reinfelder, J. R., and Morel, F. M . M . (1995). Bioaccumulation of mercury and methylmercury. Water Air Soil Pollut. 80, 915-921. McCollough, M . A. (1989). Molting sequence and aging of bald eagles. Wilson Bull. 101, 1-10. Mclntyre, J. W., and Barr, J. F. (1997). Common Loon. The Birds of North America 313, 31pp. 168 Meyer, M . W., Evers, D. C , Hartigan, J. J., and Rasmussen, P. S. (1998). Patterns of common loon (Gavia immer) mercury exposure, reproduction, and survival in Wisconsin, USA. Environ. Toxicol. Chem. 77, 184-190. Minagawa, M . , and Wada, E. (1984). Stepwise enrichment of 1 5 N along food chains: Further evidence and the relation between 5 I 5 N and animal age. Geochim. Cosmochim. Acta 48, 1135-1140. Paruk, J. D., Seanfield, D., and Mack, T. (1999). Bald eagle predation on common loon chick. Wilson Bull. Ill, 115-116. Pinnegar, J. K., and Polunin, N . V. C. (1999). Differential fractionation of 5 1 3 C and 5 1 5 N among fish tissues: Implications for the study of trophic interactions. Functional Ecol. 13, 225-231. Plouffe, A. (1995). Glacial dispersal of mercury from bedrock mineralization along Pinchi fault, north central British Columbia. Water Air Soil Pollut. 80, 1109-1112. Power, M . , Klein, G. M . , Guiguer, K. R. R. A., and Kwan, M . K. H. (2002). Mercury accumulation in the fish community of a sub-Arctic lake in relation to trophic position and carbon sources. J. Appl. Ecol. 39, 819-830. Reid, D. S., and Morley, R. L. (1975). Mercury Contamination of Fish from Pinchi Lake, B.C.: Fish and Wildlife Branch Report, B.C. Ministry of Environment, Lands and Parks, pp.1-18. SAS Institute, Inc. (2001). JMP, Version 4.01. SAS Institute, Inc., Cary, NC. Scheuhammer, A. M . , Atchison, C. M . , Wong, A. H. K., and Evers, D. C. (1998). Mercury exposure in breeding common loons (Gavia immer) in central Ontario, Canada. Environ. Toxicol. Chem. 77, 191-196. Sprunt, A. IV, Robertson, B. W. Jr, Postupalsky, S., Hensel, R. J., Knoder, C. E., and Ligas, F. J. (1973). Comparative productivity of six bald eagle populations. Trans. No. Amer. Wildl. Nat. Resour. Conf. 38, 96-106. Stout, B. E., and Nuechterlein, G. L. (1999). Red-Necked Grebe. The Birds of North America 465, 31pp. Suns, K., and Hitchin, G. (1990). Interrelationships between mercury levels in yearling yellow perch, fish condition and water quality. Water Air Soil Pollut. 50, 255-265. Turner, M . A., and Swick, A. L. (1983). The English-Wabigoon River system: IV. Interaction between mercury and selenium accumulated from waterborne and dietary sources by northern pike (Esox lucius). Can. J. Fish. Aquat. Sci. 40, 2241-2250. 169 Vander Zanden, M . J., Cabana, G., and Rasmussen, J. B. (1997). Comparing trophic position of freshwater fish calculated using stable nitrogen isotope ratios (5 1 5N) and literature data. Can. J. Fish. Aquat. Sci. 54, 1142-1158. Vander Zanden, M . J., Shuter, B. J., Lester, N . P., and Rasmussen, J. B. (2000). Within-and among-population variation in the trophic position of a pelagic predator, lake trout (Salvelinus namaycush). Can. J. Fish. Aquat. Sci. 57, 725-731. Vermeer, K., Morgan, K. FL, Butler, R. W., and Smith, G. E. J. (1989). Population, nesting habitat, and food of bald eagles in the Gulf Islands. In The Ecology and Status of Marin and Shoreline Birds in the Strait of Georgia, British Columbia. K. Vermeer and R. W. Butler, eds. (Special Publ. Ottawa ON, Canada: Can. Wildl. Serv.), pp. 123-130. Vlietstra, L. S., and Paruk, J. D. (1997). Predation attempts on incubating common loons, Gavia immer, and the significance of shoreline nesting. Can. Field-Nat. Ill, 656-657. Watras, C. J., Back, R. C , Halvorsen, S., Hudson, R. J. M . , Morrison, K. A., and Wente, S. P. (1998). Bioaccumulation of mercury in pelagic freshwater food webs. Sci. Total Environ. 219, 183-208. Watson, T. (1992). Evaluation of mercury concentration in selected environmental receptors in the Williston Lake and Peace River areas of British Columbia (Richmond, B.C., Canada: Triton Environmental Consultants Ltd.). Wayland, M . , Hobson, K. A., and Sirois, J. (2000). Environmental contaminants in colonial waterbirds form Great Slave Lake, NWT: Spatial, temporal and food-chain considerations. Arctic 53, 221-233. Wiener, J. G., Martini, R. E., Sheffy, T. B., and Glass, G. E. (1990). Factors influencing mercury concentrations in walleyes in Northern Wisconsin lakes. Trans. Am. Fish. Soc. 119, 862-870. Winfrey, M . R., and Rudd, J. W. M . (1990). Environmental factors affecting the formation of methylmercury in low pH lakes. Environ. Toxicol. Chem. 9, 853-869. Wolfe, M . F., Schwarzbach, S., and Sulaiman, R. A. (1998). Effects of mercury on wildlife: A comprehensive review. Environ. Toxicol. Chem. 171, 146-160. Zillioux, E. J., Porcella, D. B., and Benoit, J. M . (1993). Mercury cycling and effects in freshwater wetland ecosystems. Environ. Toxicol. Chem. 12, 2245-2264. 170 Table 3.1 - Results of stable nitrogen (8N) and carbon (8C) analyses on adult bald eagle blood from Pinchi, Tezzeron, Stuart and Fraser Lakes(2001) Lake Sample ID 8N(%o) 5C(%o) Hg (Ug/g ww) Stuart 01A101 11.47 -24.86 4.864 Pinchi 01A102 11.40 -28.14 9.435 Pinchi 0TA103 10.05 -27.15 4.245 Fraser 0TA104 11.50 -26.57 2.264 Tezzeron 01A105 11.43 -26.69 4.676 ) 171 Table 3.2 - Results of stable nitrogen (8N) and carbon (8C) analyses on eaglet blood from Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes(2000) Lake Sample ID 8N(%o) 5C(%o) Hg (ug/g ww) Great Beaver BAEA00-1C1 11.41 -30.88 0.152 Great Beaver BAEA00-2C1 10.34 -28.49 0.208 Great Beaver BAEA00-2C2 10.71 -28.66 0.240 Great Beaver BAEA00-3C1 11.00 -28.90 0.227 Great Beaver BAEA00-3C2 10.44 -29.31 0.208 Tezzeron BAEA00-4C1 10.82 -29.42 0.419 Pinchi BAEA00-5C1 10.15 -30.21 0.365 Pinchi BAEA00-6C1 10.48 -28.87 0.515 Pinchi BAEA00-7C1 10.73 •^28.88 0.785 Pinchi BAEA00-7C2 10.63 -28.72 0.578 Fraser BAEA00-8C1 11.54 -26.90 0.211 Fraser BAEA00-8C2 11.60 -26.83 0.229 Stuart BAEA00-9C1 9.81 -25.33 0.202 Stuart BAEA00-9C2 9.65 -25.03 0.228 Stuart BAEA00-10C1 10.38 -26.10 0.442 Stuart BAEA00-10C2 10.28 -26.11 0.379 172 Table 3.3 - Results of stable nitrogen (5N) and carbon (8C) analyses on red-necked grebe eggs from Pinchi, Tezzeron, Great Beaver and Fraser Lakes(2001) Lake Sample ID SN(%o) 8C (%o) Hg (ug/g dw) Fraser FRRG2 10.57 -25.75 0.390 Fraser FRRG1 11.10 -24.94 0.653 Great Beaver GBRG3 11.78 -29.58 0.898 Great Beaver GBRG1 12.19 -30.43 1.056 Great Beaver GBRG2 10.80 -28.94 0.414 Pinchi PINRG2 11.11 -28.59 1.985 Pinchi PINRG1 9.75 -27.74 1.165 Tezzeron TEZRG1 10.38 -30.26 0.836 Tezzeron TEZRG4 10.28 -30.75 1.083 Tezzeron TEZRG3 10.77 -27.17 0.817 Tezzeron TEZRG2 9.69 -27.22 0.628 Tezzeron TEZRG5 9.66 -28.85 0.765 173 Table 3.4 - Food remains found in bald eagle nests from Pinchi, Tezzeron, Stuart, Great Beaver and Fraser Lakes (2001-2002) Eagle nest Year Lake Food Found In Nest 1 2001 Stuart Bones of 2 fish (-36 and 42 cm each) 2 2001 Pinchi Lake Trout (tail only) - 4.89 ug/g dw 2 2001 Pinchi Large Lake Trout Tail 2 2001 Pinchi Large fish bones (~65cm) 3 2001 Great Beaver Rainbow Trout- 0.17 ug/g dw 5 2001 Tezzeron Large fish bones (~52cm) 8 2001 Tezzeron 2 Northern Pikeminnow heads 9 2001 Fraser Bones of 2 fish (~36cm each) 10 2001 Pinchi Lake Whitefish (19.8 cm) - 1.07 ug/g dw 10 2001 Pinchi Lake Whitefish (large tail) - 1.86 ug/g dw 1 2002 Stuart 2 Rainbow Trout heads 2 2002 Fraser Large Rainbow Trout tail 2 2002 Fraser 5 Largescale Sucker tails 3 2002 Pinchi 1 Largescale Sucker 4 2002 Pinchi 2 Rainbow Trout 4 2002 Pinchi Gull skull 4 2002 Pinchi Duck skeleton 5 2002 Pinchi Fish head bones 6 2002 Stuart Bear leg bones 6 2002 Stuart Gull skeleton 6 2002 Stuart Bones from 2 fish 7 2002 Tezzeron Bones from 2 fish 7 2002 Tezzeron Duck skeleton 174 a >. o 43 I-I CL> o JS C O o 00 oo 00 oo cu 00 o C U c .s CS 8P T3 CU 44 O JS 0) CQ 1 T3 .8 I $ S -8 43 co o S fe 32 X O I-I —. oT » -O X> gj T J o >> OO O I-I co 00 £ 43 co o a fe 4=1 CO 44 a i 1 CU a -a • | i> CO 44 O cn 1/1 iy 91 M ^ ^ ^ CN CN CN l l cn 92 C C o o 83 S3 cu cu H H M CO 3 176 Os O N O 0 0 oo 3 o a oC c o l _ N N CU H cu CO nj fc • CU CU CU t : 3 +-» CO C/3 c/J U • • m CN D C S o fl s a cu J 3 t: o © vO *r> TJ- m CN 3 J fc S o * o o 3 tU 1» CU X) 13 I 1 -*-» 3 *o 03 T3 CN O o CN © o o CN I -a fl CU O , fc 6 ^ I S g I-I CU •S CQ .2 2 1 3 3 u o CO fl S 3 N cu cu B H fl cu Z co f J3 0< I CN cn § 00 fc 178 CN I WD COD a a w o a 8 a oo © 4» t: o fc © CN © o o o (M.p 3/3tf) 3 H 33a aqaio S o cn 0 0 0 0 CN V Q O O CN i O o o CN cn i a C3 CO 00 00 s O O H O 1 43 S3 Vi cu CN CN (°%) 39 tin 182 * * * * * 00 O J w "3 < 60 60 W u O o o x> c u -*-» c u 11 E 183 * * * * * * * * * * * * * * * * rr co CN r o © co oo CN SO CN r f CN (°%) 39 C/3 00 a W 3 V3 oo 00 W J -Q OH 1 2 4> 44 CN CN .SP UH 184 * * * * * * * * * GO SP 3 3 CO 00 00 W Q CM O s C N 00 v© C N (°%) M9 185 D • • + + o xx X o o • • D a * *3* to o • CO • o cn Cu) o Cu) cn "Ho ca o 0) m W cn B L J x> Q < Cu) X) aj Ra o Ea • • 4 X X • + n o o (up S/Sri) gf | 186 CHAPTER 4 - DISTRIBUTION OF MERCURY AND SELENIUM IN ADULT BALD EAGLE (HALIAEETUS LEUCOCEPHALUS) SECONDARY FEATHERS Introduction Bird feathers are often used in studies of environmental methylmercury (meHg) contamination because they can be sampled without harming the bird, and it is possible to obtain a sample without having the bird in hand. Previous studies have taken advantage of molted feathers that were simply collected in the vicinity of nesting birds for mercury (Hg) analyses (Bowerman et al, 1994; Cahill et al, 1998). Almost all of the Hg present in feathers is meHg (Thompson and Furness, 1989), which has been incorporated into the structure of the feather by binding to sulphydryl groups in feather keratins (Burger and Gochfeld, 1997). MeHg in feathers represents both dietary intake and mobilization of meHg previously stored in other tissues such as the muscle and liver (Braune and Gaskin, 1987). Feathers have been shown to contain up to 93% of the body burden of Hg (Braune and Gaskin, 1987). Feathers typically contain some of the highest levels of meHg in bird tissues (see chapter 2; Burger, 1993; Caldwell et al, 1999), and have been positively correlated to blood concentrations in many studies (see chapter 2; Wood et al, 1996; DesGranges et al, 1998; Evers etal, 1998; Scheuhammer etal, 1998; Sepulveda etal, 1999). Feathers are often analyzed according to the type of plumage, and significant differences between types (i.e. mantle/breast contour feathers, primaries, secondaries and/or rectrices) and specific feathers of that type (i.e. #2 or #3 primary) have been noted with regards to Hg levels (Furness et al, 1986; Braune, 1987; Altmeyer et al, 1991). This has been attributed to the sequence in which the feather is molted, with those replaced first having the highest Hg concentrations (Furness et al, 1986; Braune, 1987; Braune and Gaskin, 1987). In a similar manner, it has previously been suggested that the distal portion of a growing feather can have higher Hg concentrations compared to the proximal portion (Solonen and Lodenius, 1990). It is this hypothesis that was examined in this study of #2 secondary feathers of wild-caught adult bald eagles. 187 Materials and Methods Adult Eagle Capture and Sampling Adult eagles were trapped in mid-June of 2001 and 2002 using a floating fish set as described in Cain and Hodges (1989). The study area where all adult eagles were captured is described in chapter 2. A collection of small fish native to the study area, as well as larger store-bought herring were prepared by cutting a slit lengthwise on the belly, removing the entrails, and replacing them with a piece of pipe foam insulation. The fish were then sewn up and two to four, 30-pound monofilament nooses were sewn into each fish. The bait fish was then attached, using 100-pound monofilament, to a floating buoy of sufficient weight to tighten the nooses around the eagle's toes and bring the bird down to the water. As soon as a captured eagle was pulled into the boat, the nooses were cut from the toes and the talons were wrapped securely in a tensor bandage. A hood was placed over the eagle's head and it was taken to shore for sampling. A double-sided Velcro strap was used to secure the wings to the eagle's body. The #2 secondary from each wing was cut below the base of the vane and placed in a Ziploc® bag for later Hg/Se analysis. Each bird was then fitted with a U.S. Fish and Wildlife Service number band. When sampling was complete, the eagle was released facing an unobstructed area. A l l eagles were ensured to be perching normally before leaving the area. Mercury Analysis Feather samples were analyzed for total Hg at the National Wildlife Research Centre (NWRC) in Hull, Quebec. Plastic and glassware used throughout the digestion and analysis processes described below were previously acid-washed in dilute (1.5%) nitric 188 acid for a minimum of 8 hours, rinsed with double-deionized water and allowed to air-dry completely. Feathers were washed prior to analysis by shaking in acetone for one minute, in dilute (1%) Triton-X 100 for another minute, followed by thorough rinsing in double-deionized water, and allowed to air-dry overnight (Scheuhammer et al, 1998). To enable digestion of the entire adult #2 secondary, each feather was cut into 10 or 11, 1 -inch pieces starting from the tip of the feather to the portion of the shaft below the vanes where the feather was originally cut from the body, hereafter referred to as the feather 'base'. One additional adult secondary was also cut in this manner, and then further subdivided into vanes and shaft for each one-inch feather section (a total of 22 samples for this feather). Feathers were placed in a plastic tube with 0.5 ml of double-deionized water and 1 ml of 70% nitric acid (Instra-analyzed, JT Baker). The tubes were loosely capped and allowed to sit overnight at room temperature to begin digestion. The following day, all samples were transferred to a dry bath and incubated at 70°C for 1 hour and then 100°C for another 2 hours to ensure complete digestion. Samples were allowed to cool and were then diluted to a total volume of 5 ml using double-deionized water and transferred to clean glass tubes. Total Hg analysis of feathers was completed using an automated mercury analyzer (AMA-254; Altec Ltd, Canalytical, Burlington, ON) as described previously (chapter 2). Selenium Analysis One adult secondary digested for Hg analysis was also analyzed for Se using graphite furnace atomic absorption spectrophotometry (GFAAS) using an electrodeless discharge lamp with deuterium background correction. The atomization program was based on that of Krynitsky (1987). Nickel (as the nitrate) was used as a matrix modifier to stabilize Se (Carnrick et al, 1983). Calibration blanks were run between each sample to auto-zero the machine and ensure no contamination carried over from previous samples. 189 Quality Assurance Blanks, Hg standards (0.01 and 0.1 ug/ml Hg), and certified reference materials from the National Research Council of Canada (NRCC) [Dogfish liver (Dolt-2) and Dogfish muscle (Dorm-2)] were run prior to sample analysis for calibration purposes and to check instrument accuracy and sensitivity. Duplicate samples and additional standards and certified reference materials were also checked throughout the Hg analysis. A l l certified reference materials were recovered at ± 10% of the certified value. Average Dolt-2 recovery was 96.5 - 102.4% and average Dorm-2 recovery was 95.6 - 98.1%. Dolt-2 and Dorm-2 certified reference materials from the NRCC were analyzed prior to Se analysis of adult eagle feather sections. Recoveries for Dolt-2 and Dorm-2 were 100.2% and 100.5%, respectively. Blanks, Se standards (0.25 and 0.5) and digest blanks were also run during Se analyses to ensure accuracy of data. Statistical Analyses To determine whether a significant decreasing trend in Hg existed across the length of each feather, from tip to base, simple linear regression was used. It was necessary to analyze each feather individually because of the large differences in Hg concentrations between feather samples from different lakes. The relationship between Se concentration and distance along the feather was also determined using simple linear regression. Pearson correlation was used to determine whether a relationship existed between Hg and Se concentration along the feather length. A two-tailed, paired t-test was used to test for a difference in Hg concentration between the shaft and vanes of each feather section. Al l statistical analyses were performed using SigmaStat for Windows, version 2.03S (Jandel Scientific, 1995). 190 Results Mercury in Adult Eagle Feathers In total, 13 adult bald eagle #2 secondary feathers were analyzed for total Hg. Eight feathers were divided into 10 one-inch subsections (Figures 4.1 and 4.2) and 5 slightly larger feathers were divided into 11 one-inch subsections (Figure 4.3). Mercury concentrations for the individual feathers used in this experiment are shown in Table 4.1. Mercury concentrations in all adult eagle secondary feathers, except one, decreased significantly from the tip of the feather to the base (P = 0.022 to P < 0.001). Examination of individual feather Hg trends (Figures 4.1, 4.2 and 4.3) showed that the decreasing trend in feather Hg is best illustrated in those feathers with higher Hg concentrations in the tip of the feather. Selenium in Adult Eagle Feathers One of the feathers digested for Hg analysis was also used in Se analysis. The concentration of Se in each piece of the feather, and the significant decreasing trend (P = 0.028) in feather Se from tip to base are shown on Table 4.2 and Figure 4.4, respectively. A significant positive relationship was found between the concentrations of Se and Hg in this particular feather (r = 0.718, P = 0.019), with both Se and Hg concentrations decreasing from the tip to the base of the feather. The relationship between Se and Hg in this feather is shown on Figure 4.5. Mercury in Feather Vanes vs. Shaft One additional #2 secondary feather was similarly divided into 11 one-inch sections, which were then further subdivided into the vanes and shaft specific to that section. Mercury results for this feather are presented in Table 4.3. As shown on Figure 4.6, a statistically significant decrease in Hg concentration from feather tip to base was found in 191 both the shaft and the vanes of this feather (P < 0.001). When the concentrations of both the shaft and vanes were combined, the total Hg concentration for the feather also showed a statistically significant decline from tip to base (P < 0.001, Figure 4.6), similar to the other feathers tested (Figures 4.1, 4.2 and 4.3). Feather vane Hg concentrations showed more variation (r = 0.745), in the gradient than the feather shaft (r = 0.859). The slope of the regression line is smaller for the vanes compared to the shaft, indicating a greater decrease in Hg deposition in the shaft from the tip to the base. When the vanes and shafts are combined, the resulting regression is tighter (r2 = 0.915). Mercury concentrations in the vanes were statistically greater than those in the shaft (P = 0.009) as illustrated on Figure 4.6. The amount of Hg in each feather section resulting from the vanes and shaft is shown on Figure 4.7. Total Hg contributed by the shaft only surpassed that of the vanes starting at the 9th section and continuing to the base (Table 4.3). A graphic representation of the percentage of Hg contributed by the shaft versus the vanes, in comparison to the weight percentage of the shaft versus the vanes for each feather section is presented on Figure 4.8. As shown, the percentage of Hg contributed by the vanes was consistently higher than the percentage of weight contributed by the vanes to each feather section, except at the tip of the feather where levels were almost equal. The opposite was found for the feather shaft. Discussion Feathers have often been used as a reliable monitoring tissue for meHg exposure in birds (Applequist et al, 1985; Furness et al, 1986; Solonen and Lodenius, 1990; Monteiro and Furness, 1997). Once meHg has been deposited into a growing feather, it has been shown to be both physically and chemically stable when exposed to various treatments (Applequist et al, 1984). Elevated Hg levels in blood are typically reflected in elevated feather Hg levels, and correlations between the two have been shown in several species of birds (see chapter 2; Wood et al, 1996; DesGranges et al, 1998; Evers et al, 1998; Scheuhammer et al, 1998; Sepulveda et al, 1999). In this study, the deposition of Hg and Se along the length of adult bald eagle second secondary feathers were examined, as well as the proportion of Hg deposited in the vanes of the feather compared to the shaft. 192 The results discussed here show that large discrepancies in feather Hg concentration may be realized if only a portion of a large feather like the secondaries used in this experiment is used for Hg analysis. A significant decrease, from tip to base, in the level of Hg along the length of the 12 of the 13 #2 secondaries was noted. This is because when a feather first begins to grow, meHg levels in the body are higher in comparison to when the feather is finished growing and is cut off from the blood supply (Braune and Gaskin, 1987). As the feather grows, a gradual decrease in Hg concentration in the body occurs as meHg is deposited into the growing feather by bonding to sulphydryl groups in feather keratin (Braune and Gaskin, 1987; Burger and Gochfeld, 1997). This, in turn, results in decreased levels of meHg deposited in the feather as it is grown. Solonen and Lodenius (1990) found this same trend in 3 adult accipiters and 20 young osprey. In some cases, while the overall trend is a decrease from tip to base, Hg concentrations in a feather section closer to the base may be higher than those closer to the tip. Additional factors including daily or weekly changes in diet, and the timing of feather growth in relation to diet may also contribute to the trend found in the feathers, specifically in the degree of variation in Hg concentrations along each feather. The decreasing trend in feather Hg is best shown in those feathers that start with a higher Hg concentration in the tip of the feather. This observation is consistent with results of Becker et al. (1994), who found that chicks with the highest Hg exposures also had the greatest Hg-elimination rates through deposition into feathers. The eagles with the highest Hg concentrations in this study were captured on Pinchi Lake. The Pinchi Lake area is high in natural Hg due to the presence of the Pinchi fault (Plouffe, 1995). In addition, a Hg mine operated on the north shore of Pinchi Lake in the past, and during World War II operations (1940-1944), all roasted Hg ore wastes were dumped directly into Pinchi Lake (EVS Environment Consultants et al, 1999). It is likely because of this past contamination that Hg levels in bald eagles from Pinchi Lake are significantly greater than those of bald eagles captured on surrounding lakes (see chapter 2). As a result, eagles from Pinchi Lake have significantly higher Hg concentrations in blood 193 compared to eagles from surrounding lakes (see chapter 2), and this is reflected in a higher amount of Hg deposited in the feathers. By examining the Hg concentrations in the feathers from all birds in Table 4.1, it appears that a baseline level of Hg may exist, and that, once reached, feather Hg concentrations no longer decrease significantly as the feather grows. It appears that, on average, approximately 10 ug/g Hg may be the baseline level for eagles nesting in the study area, outside of Pinchi Lake. This baseline level may reflect a balance between the daily uptake of Hg from the diet and the elimination of this Hg during feather growth as opposed to elimination of Hg previously stored in other tissues of the body. If this is true, then a higher baseline would be expected for those eagles nesting on Pinchi Lake because their daily uptake of Hg may be higher, based on known fish Hg concentrations (see chapter 1; EVS Environment Consultants, 2001). For the 2 adults captured on Pinchi Lake in 2001, it appears that a baseline of approximately 20 ug/g Hg may be applicable. In general, a baseline of 20 ug/g Hg for Pinchi Lake eagles compared to approximately 10 |xg/g Hg for eagles from surrounding lakes would be in agreement with blood Hg levels which are, on average, two times greater in eagles from Pinchi Lake compared to eagles from surrounding lakes (see chapter 2). Previous researchers have suggested that the Hg levels in feathers reflect those of the blood at the time of feather formation (Westermarket'a/., 1975; Solonen and Lodenius, 1990). The adult captured on Pinchi Lake in 2002 had a considerably higher concentration of Hg in its secondary feather, and thus may have had a greater body burden of meHg. Several factors could have contributed to the high level of Hg deposited in this feather. Initially, diet would be the major factor in Hg accumulation, especially if this bird was feeding, during or immediately prior to feather formation, primarily on large piscivorous fish from Pinchi Lake which are known to contain high levels of Hg (see chapter 1; EVS Environment Consultants, 2001). Alternatively, the secondary sampled from this bird might have been the only feather growing at the time of its formation, and therefore a considerably larger amount of meHg was deposited in that feather compared to secondaries sampled from other eagles on Pinchi, which may have been growing more 194 than one feather simultaneously. Bearhop et al. (2000) noted this to be a factor in the study of blood Hg kinetics and meHg deposition in feathers of great skuas (Catharacta skua) dosed with varying levels of methylmercuric chloride. It was suggested that skuas with higher feather Hg levels were molting fewer feathers at once compared to skuas with lower feather Hg levels (Bearhop et al, 2000). This could also partly account for the variability in Hg deposition into feathers in eagles from the other study lakes. Bearhop et al. (2000) also found significant individual variation in the ability of great skuas to excrete meHg into growing feathers, and this may also contribute to the variation seen between eagles in our study. Another factor may be the amount of time between successive feather elimination and regrowth. Bald eagles molt over a period of approximately 6 months, beginning in spring with head and neck feathers, followed later by body and flight feathers, and finally tail feathers in late July/early August (McCollough, 1989). The full molt is usually complete by late fall (McCollough, 1989). Based on this, several days may have passed between active feather growth in the most contaminated eagle from Pinchi Lake, allowing the body burden to increase compared to the other adult eagles from Pinchi which may have been continuously replacing feathers. Bearhop et al. (2000) found that Hg concentrations in blood of great skuas gradually increased when feathers were not being molted. Ultimately, a combination of the factors mentioned above could have contributed to the higher levels of Hg found in the secondary of this eagle and be responsible for a baseline Hg-elimination rate not being reached. Selenium concentrations along the length of a single secondary feather were also shown to decrease from tip to base. This decrease, while having a greater amount of variation than the corresponding decreasing Hg concentrations in the same feather, was also shown to correlate positively with the Hg concentrations along the feather. Within the body, the toxicity of meHg can be significantly reduced through a poorly understood process of demethylation and the formation of an equimolar inorganic Hg-Se-protein complex that is presumed to aid in protection of the organism against the toxicity of meHg (Yoneda and Suzuki, 1997). It is unlikely that the correlation found between Hg and Se in this feather was the result of an association or binding of Hg to Se for a couple of reasons. 195 First, Hg in feathers is primarily meHg (Thompson and Furness, 1989), which likely does not form this same association with Se. This is evident in examining the concentrations of Se deposited in the feather compared to the levels of Hg in each corresponding section, which are significantly lower by more than a factor of 10 (molar ratio of Hg:Se = 13:1). In addition, no correlation was found between Hg and Se concentrations in the blood of these same adult eagles (see chapter 1). Therefore, Se may behave similarly to Hg, with reduced Se in the blood or tissues resulting in a decreasing amount of Se being deposited into the feather during growth. A significant difference in Hg concentration between the feather shaft and vanes was noted in this experiment. Goede and de Bruin (1984) also found that the vanes of juvenile knots (Calidris canutus) and bar-tailed godwits (Limosa lapponicd) had higher Hg concentrations compared to the shaft. While Braune and Gaskin (1987) did not find a significant difference between these feather parts in Bonaparte's gulls, the mean concentration of Hg in the vanes was slightly higher than both the rachis and quill. Doi and Fukuyama (1983) also found higher Hg concentrations in the vanes in comparison to the shaft. A possible reason for this higher level of Hg in the vanes could be because the vanes have a greater degree of mineralization, therefore providing more options for Hg-substitution for other feather elements such as zinc (Braune and Gaskin, 1987). It may also be possible that feather shafts have less Hg than vanes because shafts contain dried blood vessels and other materials that are not as high in Hg as feather keratin. Regardless of whether the vanes or shaft of the feather are examined, Hg concentrations appear to decrease consistently from the distal to the proximal regions. The major differences lie in the rate of Hg deposition and the amount of total Hg being deposited in each piece of the feather. The Hg elimination rate for the entire feather appears to be halfway between that of the vanes and the shaft. As the feather grows, the concentration of Hg deposited in the shaft compared to the vanes decreases at a greater rate. However, the percentage of total Hg deposited in the shaft still exceeds what is deposited in the vanes closer to the base of the feather. This is likely due to the shaft contributing to a larger amount of the feather weight as the feather grows. As shown in Figure 4.8, the 196 percentage of Hg deposited in the vanes of each feather section is consistently greater than the percentage of weight contributed by the vanes to each feather section (except at the tip of the feather where percentages are almost equal). Based on the above information, the vanes contribute a greater amount of Hg to the feather compared to the shaft, however both feather parts show a significant decrease in Hg concentration along the length of the feather (from tip to base). The results of this study are contrary to the results found in a study conducted by Dmowski (1999) on the content of various metals along the length of primary feathers and in the vanes versus the shaft of Polish white-tailed eagles (Haliaetus albicilla). In their study, each feather was divided into a maximum of 8 pieces and further subdivided into shaft and vanes for each section. No significant decrease in feather Hg during growth was noted, however, only 3 different primary feathers were analyzed in the experiment and each exhibited a different pattern of Hg deposition. The feather with the highest level of Hg did show a decrease in Hg from the tip to the base, as well as higher levels of Hg in the vanes compared to the shaft of each section, as found in our study. The other two feathers examined in their study had considerably less Hg and did not exhibit the decreasing trend. This was similar to the trend observed in secondary feathers from 2 eagles in our study, which exhibited the lowest Hg concentrations (i.e. adults #6 and #10). Becker et al. (1994) found that black-headed gull chicks, which had low levels of Hg in their tissues, also had low Hg elimination rates, which were virtually unchanged throughout the feather growth period. Therefore, as mentioned above, the reason that no decreasing trend was seen in their feathers may be because the body burden of Hg prior to molt was not sufficiently high to result in more Hg being deposited at the start of feather growth. Based on these findings, estimation of feather Hg and possibly Se content based only on analysis of the distal portion of the feather, or only the vanes of the feather could create a large bias in observed feather Hg concentrations, particularly if the feather samples were obtained near an area of known or suspected Hg contamination. Solonen and Lodenius (1990) came to the same conclusion that Hg analyses should be conducted on the entire 197 feather as opposed to a portion thereof. As shown in this study, eagles breeding in an area of known Hg contamination have greater Hg deposition in the distal portion of growing feathers, as well as in the feather vanes compared to the shaft. Knowing this, it is advisable that future studies examining Hg and possibly Se concentrations in larger feathers digest the entire feather for metals analyses rather than only a conveniently smaller portion. Acknowledgements TeckCominco, Ltd., the Metals In The Environment Research Network, and Canadian Wildlife Service provided funding for the field portion of this study. Gregg Howald and Lori Smith assisted in adult eagle capture. Tony Scheuhammer provided funding and laboratory facilities at NWRC for digestion and analysis of samples for Hg and Se. Delia Bond and Ewa Neugebauer (NWRC) provided guidance and assistance in the laboratory. References Altmeyer, M . , Dittmann, J., Dmowski, K., Wagner, G., and Miiller, P. (1991). Distribution of elements in flight feathers of a white-tailed eagle. Sci. Total Environ. 105, 157-164. Appelquist, H., Asbirk, S., and Drabask, I. (1984). Mercury monitoring: Mercury stability in bird feathers. Mar. Pollut. Bull. 15, 22-24. Appelquist, H., Drabeek, I., and Asbirk, S. (1985). Variation in mercury content of guillemot feathers over 150 years. Mar. Pollut. Bull. 16, 244-248. Bearhop, S., Ruxton, G. D., and Furness, R. W. (2000). Dynamics of mercury in blood and feathers of great skuas. Environ. Toxicol. Chem. 19, 1638-1643. Becker, P. H., Henning, D., and Furness, R. W. (1994). Differences in mercury contamination and elimination during feather development in gull and tern broods. Arch. Environ. Contam. Toxicol. 27, 162-167. Bowerman IV, W. W., Evans, E. D., Giesy, J. P., and Postupalsky, S. (1994). Using feathers to assess risk of mercury and selenium to bald eagle reproduction in the Great Lakes Region. Arch. Environ. Contam. Toxicol. 27, 294-298. 198 Braune, B. M . , and Gaskin, D. E. (1987). Mercury levels in Bonaparte's gulls (Larus Philadelphia) during autumn molt in the Quoddy Region, New Brunswick, Canada. Arch. Environ. Contam. Toxicol. 16, 539-549. Braune, B. M . (1987). Comparison of total mercury levels in relation to diet and molt for nine species of marine birds. Arch. Environ. Contam. Toxicol. 16, 217-224. Burger, J. (1993). Metals in avian feathers: Bioindicators of environmental pollution. Rev. Environ. Toxicol. 5, 203-311. Burger, J., and Gochfeld, M . (1997). Risk, mercury levels, and birds: Relating adverse laboratory effects to field biomonitoring. Environ. Res. 75, 160-172. Cahill, T. M . , Anderson, D. W., Elbert, R. A., Perley, B. P., and Johnson, D. R. (1998). Elemental profiles in feather samples from a mercury-contaminated lake in central California. Arch. Environ. Contam. Toxicol. 35, 75-81. Cain, S. L., and Hodges, J. I. (1989). A floating-fish snare for capturing bald eagles. J. Raptor Res. 23, 10-13. Carnrick, G. R., Manning, D. C , and Slavin, W. (1983). Determination of selenium in biological materials with platform atomic-absorption spectroscopy and Zeeman background correction. Analyst 108, 1297-1312. DesGranges, J.-L., Rodrigue, J., and Laperle, M . (1998). Mercury accumulation and biomagnification in ospreys (Pandion haliaetus) in the James Bay and Hudson Bay regions of Quebec. Arch. Environ. Contam. Toxicol. 35, 330-341. Dmowski, K. (1999). Birds as bioindicators of heavy metal pollution: Review and examples concerning European species. Acta Ornithologica 34, 1-25. Doi, R., and Fukuyama, Y. (1983). Metal content in feathers of wild and zoo-kept birds from Hokkaido, 1976-78. Bull. Environ. Contam. Toxicol. 31, 1-8. Evers, D. C , Kaplan, J. D., Meyer, M . W., Reaman, P. S., Braselton, W. E., Major, A., Burgess, N . , and Scheuhammer, A. M . (1998). Geographic trend in mercury measured in common loon feathers and blood. Environ. Toxicol. Chem. 17, 173-183. EVS Environment Consultants, Norecol Dames and Moore, and Frontier Geosciences. (1999). Pinchi Lake Mine Site and Lake Investigation: Environmental Assessment Report (Vancouver: Cominco Ltd.). EVS Environment Consultants. (2001). Regional survey offish mercury concentrations -Pinchi Lake, B.C. (Vancouver: TeckCominco Ltd.) 91pp. 199 Furness, R. W., Muirhead, S. J., and Woodburn, M . (1986). Using bird feathers to measure mercury in the environment: Relationships between mercury content and moult. Mar. Pollut. Bull. 77, 27-30. 6* Goede, A. A., and de Bruin, M . (1984). The use of bird feather parts as a monitor for metal pollution. Environ. Pollut. Ser. B. 8, 281-298. Jandel Scientific. (1995). SigmaStat Statistical Software, Version 2.03S for Windows 95, NT & 3.1. Jandel Scientific, San Rafael, CA. Krynitsky, A. J. (1987). Preparation of biological tissue for determination of arsenic and selenium by graphite furnace atomic absorption spectrophotometry. Anal. Chem. 59, 1884-1886. McCollough, M . A. (1989). Molting sequence and aging of bald eagles. Wilson Bull. 101, 1-10. Monteiro, L. R., and Furness, R. W. (1997). Accelerated increase in mercury contamination in North Atlantic mesopelagic food chains as indicated by time series of seabird feathers. Environ. Toxicol. Chem. 16, 2489-2493. Plouffe, A. (1995). Glacial dispersal of mercury from bedrock mineralization along Pinchi fault, north central British Columbia. Water Air Soil Pollut. 80, 1109-1112. Scheuhammer, A. M . , Atchison, C. M . , Wong, A. H. K., and Evers, D. C. (1998). Mercury exposure in breeding common loons (Gavia immer) in central Ontario, Canada. Environ. Toxicol. Chem. 77, 191-196. Sepulveda, M . S., Frederick, P. C , Spalding, M . G., and Williams Jr., G. E. (1999). Mercury contamination in free-ranging great egret nestlings (Ardea albus) from southern Florida, USA. Environ. Toxicol. Chem. 18, 985-992. Solonen, T., and Lodenius, M . (1990). Feathers of birds of prey as indicators of mercury contamination in southern Finland. Holarctic Ecol. 13, 229-237. Thompson, D. R., and Furness, R. W. (1989). Comparison of the levels of total and organic mercury in seabird feathers. Mar. Pollut. Bull. 20, 577-579. Westermark, T., Odsjo, T., and Johnels, A. G. (1975). Mercury content of bird feathers before and after Swedish ban on alkyl mercury in agriculture. Ambio 4, 87-92. Wood, P. B., White, J. H., Steffer, A., Wood, J. M . , Facemire, C. F., and Percival, H. F. (1996). Mercury concentrations in tissues of Florida bald eagles. J. Wildl. Manage. 60, 178-185. 200 Yoneda, S., and Suzuki, K. T. (1997). Detoxification of mercury by selenium by binding of equimolar Hg-Se complex to a specific plasma protein. Toxicol. Appl. Pharmacol. 143, 274-280. 201 ™ CM g 5 ° cd od o 2 ^ co c n .« Hi 0 0 O 0 0 0 0 CM CO O N O ) CO S S N N CO O ) 0 ) o CO r - 0 5 CM o CO CO CO CO CO CO co rr m CM rr T— CO o co to CO 0 0 o ^— x— CM o o r - CM CM CM d d d d d d d d d d d Q. E CO CO CD "E x: x: x: x: CO CO 0 0 CO o s CO o g co o 8 CO CO CO CU •TJ CO 8 fe CU * r j co g CN 3 cu Ii S P 5 § CO 1 C+H o G). o OH CU N N CU H co a a " P H M 3 c o CU GO co cu 00 - B .a c« 2 a ca ' 3 b £ 3 ° O c« OH OH CU 5> S 3 • CU r - H fa rf •s a-3 . JC v - CO m <1> ^ J £ £ « 3 CO o O J W t- LO 0 0 CO 0 0 in CM CM d LO 0 0 rr rr CO CM CM CM CM CM CO CD r- rr LO O 0 0 m 0 0 r-in r- 0 0 CO 0 0 r » CO CM r- co CM o § o> o CO LO co 0 0 C 0 CO d d d d d d d d d d o CM CD O. E .a.^'gx;x:x:x:x:x: w i— C \ I C O ? 5 5 C D N C O 3 ) C/J i i i i i i i i i C M C M C M C M C M C M C M C M C M O O O O O O O O O 3 CO O ) O o oi rr LO 03 i ; o L O co co co o o ^ 0 ) 2 N M I - I - T - » -CD CO CD cd s in "O < 0) b3 J £ £ CO CO L O L O rr r- CO o CM L O CO co rr 0 0 to o> CM o CM L O co 0 0 oo CO o O 5 r - ^— ^— ^— r - CM d d d d d d d d d d o o CM a> CL E CO co K C M C O r f l O C D r - - 0 0 O > C Q i i i • i i i i i i o o o o o o o o o o 3 •o < 3 _ 0 > T - c D O J 0 0 0 0 > C » ^ _ l f ) l O S f t c d t o r g j d d o i - d I 1 ! r? I r - Q § ® CM Q. E CO CO c o r ~ CO r ~ o c o CD o o c o o r r 0 0 LO LO CM l O in CO co in C D C D o o o d d d d d d d d d o_ X) cz "S .c £ H CM CO r f in S 0 0 LO in in in in in LO in in o o o o o o o o o rr g d CD co CO 3 O l CM 0 0 f - L O oi CM rr T — d co 2 ^ CD CO CM d ^ d u < CO 0 IB 03 0 0 LO co CO s? r^ rr CO CO CO s CM co 0 0 CO o CM rr CD CD CO 0 0 o o o —^ *— —^ — —^ T— CM d d d d d d d d d d C L X ) c "S sz x: x: CM CO rr 55 s ft 0 0 s 3 3 3 3 3 3 3 o (0 CO X ) 202 x> o .& B GO It 3 o o C N i o © w co s :cr) - i J •= o a t M t - T - » - « - » - T - ^ - T - T -•o 5 ~ < (0 85 2> 0) CL E 03 CO CO m 00 IO CD o CM CM 00 o CM if) CM s CM co CD CT> CM m r- 00 o o o *— *— X — ^— d d d d d d d d d d Q. TJ "S JO .c JC H- CM CO K CO 00 as co d> OJ co OJ cb cn d> O o o o o o o o o cu co CQ 3 o n « >t 00 Ifl IO N ( D S r a d ^ c o ' - p ^ c M C M ' d d < t - l O O l O ) OD N (0 3 4-* CO CD CO 0) a. E ca co s CO CM co CM r- CD CM co 00 CO CO cS T— CM 00 in h - co CO O O x— T— CM CM CM CM d d d d d d d d d d d CL "O c TJ JC JC. -CZ s ase CM in CO 00 CO X — CO CM CM CM CN CN CN CM CM CM CN CM CB (U 1 2 8 CN 53 =tfc tj § S P 5 3 « J 1 o 5 N N o 3 co a « .s u § • & I-H c3 c2 M £ j§ 3 SP S w o 11 g o O CO It l l • s •s I" 0) OJ CM r-- m CO in CM 00 OJ CO CM d CM in" 00 CO CM eb CO CO CO CO CO CO CO CO in 3 TJ < cu w LL * o CN in CO s CO in T — o O) o CO in o m o 8 t-~ T — CO 00 CN N- m o o 1— CM CN CM CN CM d d o d d d d d d d d CM _CD CD Q. E 03 CO Q. TJ "2 JC JC H CM CO ? 55 CO 00 O OO O 00 O 00 o s oo o _ CD •*= •«= « S s co cS T - m I I I I I OO 00 oo 00 00 o o o o o 3 CJ> CO O) S N — CU ZX-r - t- .r -T- . r -T-T- ' - " 3 ^ TJ < -i £ J= 52 5 N- CO ^ OJ o CM in T— 5 CO in o OJ o o • * O CM CO in T— m co 00 OJ T — OJ o O — T— ^— T— t— CM ^— d d d d d d d d d d d CM O CL E CO CO CU t— C M C O - * W C O ^ - O O O J ' « - C O i i i i i i i i i i i o o o o o o o o o o o 3 O) 2 TO 0J s . a 5 3 TJ < CO OJ T - OJ CO CM CM CM CM CN T - CO co d t C8 3 CO I CM i a> • co cu Q ® a. E ca to CO in 00 00 co CM CO 00 OJ o CO s 00 o 00 ^— CO o OJ 00 o o o o CN CN d d d d d d d d d d OJ B c u £ £ £ £ £ £ co i— C M C O ^ I B S N S S C D I • i i i i i i i i 3 cn co co CM i o o co T -^ d d oo d 3 TJ < 0) (0 -I r CO 3 • CM CM n> OJ X — CD CO o in Q a> CM OO 00 CO 00 OJ 00 CO OJ CO OJ T — to in co OJ co o o o d d d o d d d d d d CD Q. E o - ' S p r r r r r r w K c M c o j i o i p N - o p O J j a o o o o o o o o 203 JO o .& § S 0 — * 2 1 ^ 8 £ ON « -Si 2 *ei ° « >-< . 3 «> ^ H o oo n rj- m C N — i ( 8 / 8 t i ) 8 H 207 O m © in O «r> O m fN C N —< —< (8/Sri)3H 208 o o o o o o o o o (3/Sri)8H 209 CN o o CD co CS e E o a o u a> U 4 3 fcfa CN 43 •a o u cu co C N =tfc CU g? cu i » 3 cd C + H O t i 03 I-I S C3 > 4=! • • • C/5 O O O V PH ON OO o II rj-cu 03 X* a o C M c .o *•** o en b cu 43 •** e» cu fan CM s o 03 1 co 42 43 03 VH I - I CU S3 « « s §• I "a T3 O § « i f o s .2 •4-1 o cu co C N =8: 15b O j CU T3 1 03 G O o cu co CU I .s a o 1 1 a o o 80 ug/g dw). A designation of Hg poisoning was based on pathological findings attributable to toxemia as well as high meHg concentrations in the liver. In several laboratory and field studies of birds of prey, birds that succumbed to Hg toxicity had >30 jig/g Hg ww in the liver (Thompson, 1996). 217 All statistical analyses were performed using SigmaStat for Windows, version 2.03S (Jandel Scientific, 1995). Data were tested for normality prior to statistical analysis, and log-transformed if found not to be normally distributed. Relationships between total Hg and Se and meHg and Se were examined using Pearson (r) correlations. Analysis of variance on ranks was used to examine the relationship between body condition and total Hg in liver. Analysis of variance was also used to test relationships between Hg levels and eagle age. The Tukey test for multiple comparisons was used to distinguish age groups that were significantly different from one another. The Student's t-test was used to test for sex differences based on Hg levels. Significance for all statistical analyses was based on a P-value less than 0.05. Results The mean total Hg level [± standard deviation (SD)] for the 82 eagles, was 11.8 ± 15.8 \ig/g dw (range = 0.50 - 130 ug/g dw). Sixty-seven eagles had low Hg exposure; 14 eagles were classed as having moderate exposure to Hg; and one bird was judged to be Hg poisoned. A subset of livers from 18 birds (1 poisoned, 10 moderately exposed, 7 low exposure) was analyzed for Se and meHg. Total Hg, meHg, and Se data are presented in Table 5.1. The mean levels of total Hg, meHg and Se in this subset of birds were 27.7 ± 26.8 ug/g dw, 15.7 + 21.3 ug/g dw, and 9.82 ± 3.63 ug/g dw, respectively. The average liver moisture content was 70.1% (range = 56.0 - 82.9%). A statistically significant positive relationship (r = 0.96, P < 0.0001, n = 18) was found for total Hg versus meHg; however, the significance of this relationship was dependent on the single Hg poisoned individual in the correlation. When this bird was removed from the correlation, the relationship is no longer significant (r = 0.40, P = 0.15, n = 17). Conversely, when examining the relationship between Se and total Hg in the liver, if the most contaminated bird was removed, a significant positive correlation resulted (r = 0.80, P < 0.001, n = 17) (Figure 5.2). Including the poisoned bird resulted in a non-significant correlation (r = 0.26, P = 0.29, n = 18). For the purposes of this paper, the poisoned bird will be considered an outlier in the correlation analyses, as it is the only bird with greatly 218 elevated Hg and meHg levels. Additional unpublished data (A.M. Scheuhammer, person, comm.) on total Hg in bald eagle livers supports labeling this bird as an outlier. Body condition (BC) ranking revealed a highly significant difference (P < 0.001, n = 70) between Hg levels in birds with good BC, and those with poor BC (Figure 5.3). Eagles with good BC had significantly lower Hg concentrations than those with poor BC. The significance of this relationship remains whether or not the poisoned bird is included in the analysis. However, for illustrative purposes, data are presented both with and without the poisoned bird (Figure 5.3). Ages were determined for all but one of the eagles. The relationship between total Hg in liver and eagle age is shown in Figure 5.4. A significant difference (P = 0.003) was found between the levels of Hg in the different age groups. Specifically, adults, four year olds, and one year olds, had higher Hg concentrations than young-of-the-year. It should be noted that sample sizes of young-of-the-year, and two to four year olds are much smaller than one year olds and adults. The degree of Hg exposure was not related to sex (t = -0.404, P = 0.687) in the 80 eagles for which sex was determined (41 male and 39 female). In addition, no difference in Hg concentrations due to the sex of the bird was found taking age into account (P = 0.639). However, out of the 15 moderately/highly-exposed birds, 10 were female (including the poisoned bird) and 5 were male. Despite this difference in sex ratio, the degree of Hg exposure in the liver was unrelated to sex when examining only the 15 moderately and highly exposed birds (P = 1.00). Stomach contents were identified for 42 of the eagles, and in several cases, the stomach was empty. Thus, no associations could be found between the type of food eaten, and the concentration of Hg measured in the liver. No obvious link was noted between degree of Hg accumulation and cause of death in all birds except the bird judged to have died as a result of its exposure. The cause of death 219 for all 82 eagles is presented in Figure 5.5. While only one eagle died from Hg poisoning, metal toxicosis as a whole resulted in the death of 6% of all eagles examined (one from Hg poisoning; four from lead poisoning - see Elliott et al, 1992). A designation of Hg toxicosis was made by a veterinary pathologist, and was based on necropsy findings attributable to toxemia, as well as a high Hg concentration in the liver, (>30 ug/g Hg ww; Thompson, 1996). Most birds (83%) died of traumatic injuries (electrocution, vehicle/powerline collision, eagle attack, trap, gunshot, drowning and asphyxiation, and undetermined causes). The remaining eagles perished from starvation or infectious disease, including avian pox and aspergillosis (11%). Necropsies revealed that eagles often suffered from varying symptoms, which is not surprising as the causes of death varied among birds. The Hg poisoned eagle suffered from enteritis, enteric necrosis, myositis, myocardial sarcocystosis, and intraocular hemorrhage, which were designated as symptoms of toxemia by the veterinary pathologist. Typical symptoms of acute meHg poisoning usually include: weight loss as a result of reduced food intake, uncoordinated muscle movements, and difficulty walking, flying and standing (Scheuhammer, 1987). Other than weight loss (a non-specific symptom), typical symptoms of meHg poisoning (including sensory and motor deficits, and behavioural changes) are impossible to confirm in birds found dead. Discussion Based on the designations of low (<20 ug/g dw), moderate (20 to 80 ug/g dw), and high (>80 ug/g dw) Hg exposure mentioned previously, approximately 20% of eagles found as a part of this study during the period of 1987-1994 had a moderate or greater level of Hg exposure. In some cases the ultimate source of this Hg may be from natural geological sources; however it is believed that anthropogenic Hg contamination has increased relative to natural sources of Hg release since the start of the industrial period (Fitzgerald et al, 1998). It should be noted that in many cases, known anthropogenic sources of Hg emissions in British Columbia have decreased or ceased during the last 30 years. 220 The highly-exposed eagle from this study was found near Powell River, the location of British Columbia's oldest pulp and paper processing plant. This plant, among others in British Columbia, used mercurial slimicides from the late 1940's to early 1970's (Garrett et al, 1980). Approximately 5-20% of the Hg used for this purpose was released in effluent (Fimreite, 1970). Concerns over the environmental consequences of using Hg as a slimicide led to its replacement in the early 1970's (Garrett et al, 1980). The Powell River mill also used caustic soda generated by a Hg cell chlor-alkali plant in the past (Garrett et al, 1980). Pulp mill operations may still be releasing small quantities of Hg to the environment through consumption of caustic soda, lime rock (at Kraft mills), and fuel for power generation (Paavila, 1971). Since no information is available regarding the concentrations of Hg in the various environmental media (i.e. water, sediments, fish) around the Powell River mill, it is difficult to determine whether past Hg releases from this mill were the source of meHg to the poisoned bird. Indeed, 4 other eagles were found in the Powell River area, and of these 4, 3 were classed as having low exposure to Hg and one as moderately exposed. Perhaps the highly exposed eagle was a year-round resident in the Powell River area, consuming a diet of primarily large fish with high levels of meHg, whereas the other eagles found in the area were migrants. Indeed the eagles with low Hg exposure were found in the Powell River area in late winter/early spring, whereas the moderately exposed and Hg poisoned birds were found in late spring and summer, respectively. Alternatively, the highly exposed eagle migrated to the Powell River area from a wintering ground where it obtained the majority of its meHg. At this point it is impossible to determine whether the Powell River area was the source of Hg to the highly exposed bird without further research. The molar ratio of Hg to Se in 17 of our bald eagle livers was very close to 2:1 (Figure 5.2), unlike the 1:1 ratio noted for Hg-exposed common loons (Scheuhammer et al, 1998b) and some marine mammals such as the harbour seal and ringed seal (Cuvin-Aralar and Furness, 1991). The toxicity of meHg can be significantly reduced through a poorly understood process of demethylation and the formation of an equimolar inorganic Hg-Se-protein complex that alters the distribution of Hg among organs and is presumed to protect the organism against the toxicity of meHg (Yoneda and Suzuki, 1997). In the 221 case of the suspected Hg poisoned eagle in the present study, we observed a ratio of 12:1 Hg to Se, and the majority of the liver Hg was present as meHg, leaving little doubt that demethylation and association with Se did not occur to a substantial degree in this bird, and reinforcing the diagnosis of Hg toxicosis in this bald eagle. The higher ratio of Hg to Se in the livers of our bald eagles in general suggests that more of the Hg is in a free, unbound state, compared to species that exhibit a 1:1 molar ratio of Hg to Se. Therefore, protection from Hg toxicosis through demethylation and binding of Hg to Se may be less in bald eagles compared to other species. As an alternative, Se may have been limited in the diets of our bald eagles, leading to the 2:1 Hg to Se ratio. However, given the fact that these eagles were found in different areas of British Columbia, and likely had a variety of diets, a Se-limited diet in all cases would be very coincidental. Additional research would be required to verify the 2:1 molar ratio between Hg and Se in bald eagle livers. The bald eagles examined in this study showed a higher percentage of total Hg present as meHg in the liver, compared to common loons and common mergansers. On average, in common loon livers containing 11 - 40 ug/g total Hg, meHg accounted for approximately 30% of the total Hg (Scheuhammer et al, 1998b). In contrast, meHg constituted over 50%, on average, of the total Hg measured in bald eagle livers over this same range of total Hg. The highly exposed bird from this study had approximately 77% of its liver-Hg present as meHg (100 ug/g). Scheuhammer et al. (1998b) found that meHg concentrations did not exceed 10 ug/g dw in common loon and common merganser (Mergus merganser) livers regardless of total Hg content. Pokras et al. (1998) reported similar results in the livers of common loons from the eastern United States, with meHg concentrations being below 10 ug/g in all cases. Excluding the highly exposed eagle, concentrations of up to 15.7 ug/g dw meHg were found in the livers of our bald eagles. Two bald eagles from Eastern Canada (A.M. Scheuhammer person, comm.) having total Hg concentrations of 104 and 670 ug/g dw, had 29 ug/g and 6.7 ug/g of Hg present in the organic form, respectively. In the former case, the level of meHg is much higher than that recorded for common loons or mergansers. 222 While our bald eagles appear to have a higher percentage of total Hg present as organic Hg compared to previous studies examining common loons and common mergansers, Kim et al. (1996) found that Arctic terns (Sterna paradiseae) and brown boobies (Sula leucogaster) also had an average proportion of 50% meHg to total Hg in the liver. It should be noted that the average total Hg in Arctic tern livers was 4.9 u.g/g dw and brown booby livers was 7.2 u.g/g dw, compared to 21.6 |ig/g dw in our adult eagles (not including the highly exposed bird). Kim et al. (1996) further noted that the northern fulmar (Fulmarus glacialis), believed to have intermediate Hg exposure (mean = 14.2 ug/g dw) had 25% meHg to total Hg in the liver, and that as total Hg concentrations increased from species to species, the proportion of meHg decreased further. Based on their classification of low and intermediate Hg exposure, our bald eagles would have had an intermediate level of exposure. At this level of exposure it would be expected that our bald eagles should have a lower percentage of meHg in the liver because many other bird species experiencing similar total Hg concentrations in the liver have a lower percentage of total Hg present as meHg (Kim et al, 1996; Pokras et al, 1998; Scheuhammer et al, 1998b). However, this was not the case. Why bald eagles appear to have a higher percentage of organic Hg in their livers compared to other birds experiencing similar total Hg concentrations in the liver is unclear. It may be due to a slower demethylation rate. In vivo demethylation has been shown to occur in mammals such as the guinea pig (Komsta-Szumska et al, 1983), and likely occurs in avian species as well (Norheim and Froslie, 1978; Scheuhammer et al, 1998b; Henny et al, 2002). Another possible reason for the higher levels of meHg found in these eagles compared to other birds could be related to their moult pattern. Most bald eagles begin moulting in the spring and this continues until early fall (McCollough, 1989). The majority of the eagles examined in this study were found just before or during the moulting sequence. Given the timing, these birds had several months to accumulate meHg, but may not have had the opportunity to rid their bodies of much of it through the formation of new feathers. Also, unlike the single annual moult of bald eagles, common loons undergo two annual moults; a complete moult following the breeding season and a partial moult of body feathers prior to breeding (Mclntyre and 223 Barr, 1997). This extra moult may allow common loons and other birds that moult biannually to eliminate more of their body burden of meHg compared to bald eagles, resulting in a lower percentage of meHg in the liver. Furness et al. (1986) suggest that moulting might result in decreased tissue Hg concentrations as Hg is released from tissues and deposited into growing feathers. Feathers have been shown to contain up to 93% of the body burden of Hg of Bonaparte's gulls (Larus Philadelphia) (Braune and Gaskin, 1987). Since it is primarily meHg that is deposited in newly forming feathers (Thompson and Furness, 1989), moult may play a part in the higher percentage of meHg detected in the livers of eagles in the present study. Additional possible reasons for the higher percentage of total Hg present as meHg in bald eagle livers may include a lower rate of meHg excretion, or differences in tissue distribution of meHg compared to other bird species. Apart from the highly exposed bird in this study, the total Hg concentrations found in the livers of our bald eagles are comparable to those of bald eagles recovered in other regions of North America. For comparison purposes, the following values were converted from their reported wet weight values to dry weight values using an approximate moisture content of 75%. Wood et al. (1996), found similar Hg concentrations in the livers of 32 Florida bald eagles, with values ranging from 0.56 to 4.04 ug/g dw in nestlings [geometric mean (GM) = 1.32 ug/g dw], 1.40 to 21.7 in subadults (GM = 7.92 ug/g), and 2.52 - 48.8 ug/g in adults (GM = 10.0 ug/g). Frenzel and Anthony (1989) also found concentrations ranging from 3.04 to 32.0 ug/g dw (mean = 7.56 ug/g) in livers from bald eagle carcasses collected near the Klamath Basin region of Oregon and California in 1979-82. Neither of these studies included measurements for meHg or Se. It may be expected that adults would have higher concentrations of Hg in their livers compared to young, specifically within a group of eagles inhabiting the same area and consuming similar diets, because of the ability of Hg (primarily inorganic Hg bound to Se) to accumulate in the liver over time. In this case, adults do have the highest mean levels of Hg in the liver; however, one year old birds had a higher mean level of total Hg in liver than two to four year old birds, though this difference was not significant. Age 224 trends in Hg levels are difficult to show in a data set such as this where eagles were collected from a wide variety of habitats, likely have different diets, and are not equally sampled across all age groups. It is possible that an adult consuming mainly small mammals from a non-contaminated region would have lower Hg levels in the liver than a one year old eagle that has been feeding mostly on large, piscivorous fish from a Hg-contaminated area. Adults are also known to live up to 30 years (Buehler, 2000), however it is difficult to discern the age of an adult bald eagle after it has attained its definitive plumage. This could result in eagles that span approximately 25 years of age being grouped together in the same age category. Thus, imprecision of age data may result in an inability to accurately identify age-related trends in Hg accumulation in wild eagles. Weight loss, a symptom associated with acute Hg poisoning (Scheuhammer, 1987), could result in poorer body condition correlating with increased Hg concentrations in the liver. However, in this case, only one bird was suspected to have died of Hg toxicosis, and that bird was classed as having good overall body condition (BC = 3, on a scale of 0 -emaciated to 5 - excellent). While the concentrations of Hg were higher in eagles with poorer body condition in this study, this was probably not related to the effects of Hg as they were obviously not acutely poisoned. Scheuhammer et al. (1998b) suggest the wasting of organs as a result of weight loss might lead to what appears to be an increase in tissue Hg concentration. This may be the case here as Hg concentrations are not unusually high. Since overall body weights and organ weights were not determined for most of these eagles, it is difficult to rule out wasting as an influential factor. Experimental examination of the effect of wasting on Hg concentrations in tissues would be required to confirm this phenomenon. In conclusion, the present study has shown that bald eagles are capable of accumulating toxic doses of Hg from their environment. To our knowledge, this is the first reported case where a bald eagle was suspected to have died from Hg poisoning. The mean total Hg concentration for the 82 eagles examined in this study, was 11.8 ± 15.8 |ig/g dw (range = 0.50 - 130 p.g/g dw). Bald eagles were also shown to have a higher percentage 225 of total Hg present as meHg in the liver compared to other bird species with similar total Hg concentrations in the liver. The molar ratio of Hg to Se in our bald eagle livers was also higher compared to other birds and mammals. Bald eagles should continue to be monitored for Hg, especially in areas of known Hg contamination. Given the ongoing presence of industrial sources of Hg and the fact that atmospheric deposition of Hg is widespread and appears to be increasing, especially in remote areas (Fitzgerald et al, 1998), birds such as the bald eagle will continue to be at risk of Hg poisoning. Acknowledgements L. Wilson provided data and valuable assistance in the preparation of this manuscript. K. Langelier completed post-mortem examinations on bald eagles. P. Sinclair assisted with fieldwork. M . Kassera and R. McNeil supervised tissue preparation and archiving of samples. E. Neugebauer conducted chemical analyses at NWRC. B. Wakeford supervised the quality assurance program. References Barr, J. F. (1986). Population dynamics of the Common loon (Gavia immer) associated with mercury-contaminated waters in northwestern Ontario. Can. Wild. Serv. Occas. Pap. No. 56, pp. 25. Bortolotti, G. R. (1984). Sexual size dimorphism and age-related size variation in bald eagles. J. Wildl. Manage. 48, 72-81. Bowerman IV, W. W., Evans, E. D., Giesy, J. P., and Postupalsky, S. (1994). Using feathers to assess risk of mercury and selenium to bald eagle reproduction in the Great Lakes Region. Arch. Environ. Contam. Toxicol. 27, 294-298. Braune, B. M . , and Gaskin, D. E. (1987). Mercury levels in Bonaparte's gulls (Larus Philadelphia) during autumn molt in the Quoddy Region, New Brunswick, Canada. Arch. Environ. Contam. Toxicol. 16, 539-549. Buehler, D. A. (2000). Bald Eagle. The Birds of North America 506, 39pp. 226 Clarkson, T. W. (1992). Mercury: Major issues in environmental health. Environ. Health Perspectives 100, 31-38. Cuvin-Aralar, M . L. A., and Furness, R. W. (1991). Mercury and selenium interaction: A review. Ecotoxicol. Environ. Safety 21, 348-364. DesGranges, J.-L., Rodrigue, J., and Laperle, M . (1998). Mercury accumulation and biomagnification in ospreys (Pandion haliaetus) in the James Bay and Hudson Bay Regions of Quebec. Arch. Environ. Contam. Toxicol. 35, 330-341. Elliott, J. E., Langelier, K. M . , Scheuhammer, A. M . , Sinclair, P. H , and Whitehead, P. E. (1992). Incidence of lead poisoning in bald eagles and lead shot in waterfowl gizzards from British Columbia, 1988-91. Canadian Wildlife Service Progress Notes 200, 1-7. Elliott, J. E., Wilson, L. K., Langelier, K. W., and Norstrom, R. J. (1996). Bald eagle mortality and chlorinated hydrocarbon contaminants in livers from British Columbia, Canada, 1989-1994. Environ. Pollut. 94, 9-18. Evers, D. C , Kaplan, J. D., Meyer, M . W., Reaman, P. S., Braselton, W. E., Major, A., Burgess, N . , and Scheuhammer, A. M . (1998). Geographic trend in mercury measured in common loon feathers and blood. Environ. Toxicol. Chem. 17, 173-183. EVS Environment Consultants, Norecol Dames and Moore, and Frontier Geosciences. (1999). Pinchi Lake Mine Site and Lake Investigation: Environmental Assessment Report (Vancouver: Cominco Ltd.). Fimreite, N . (1970). Mercury uses in Canada and their possible hazards as sources of mercury contamination. Environ. Pollut. 1, 119-131. Fitzgerald, W. F., Engstrom, D. R., Mason, R. P., and Nater, E. A. (1998). The case for atmospheric mercury contamination in remote areas. Environ. Sci. Technol. 32, 1-7. Frenzel, R. W., and Anthony, R. G. (1989). Relationship of diets and environmental contaminants in wintering bald eagles. J. Wildl. Manage. 53, 792-802. Furness, R. W., Muirhead, S. J., and Woodburn, M . (1986). Using bird feathers to measure mercury in the environment: Relationships between mercury content and moult. Mar. Pollut. Bull. 17, 27-30. Garrett, C. L., MacLeod, L. A., and Sneddon, H. J. (1980). Mercury in the British Columbia and Yukon environments - summary of data to January 1, 1979: Environment Canada, Environmental Protection Service, Pacific Region, pp. 456. Henny, C. J., Hill , E. F., Hoffman, D. J., Spalding, M . G., and Grove, R. A. (2002). Nineteenth century mercury: Hazard to wading birds and cormorants of the Carson River, Nevada. Ecotoxicol. 11, 213-231. 227 Jackson, T. A. (1988). The mercury problem in recently formed reservoirs of northern Manitoba (Canada): Effects of impoundment and other factors on the production of methyl mercury by microorganisms in sediments. Can. J. Fish. Aquat. Sci. 45, 97-121. Jandel Scientific. (1995). SigmaStat Statistical Software, Version 2.03S for Windows 95, NT & 3.1. Jandel Scientific, San Rafael, CA. Komsta-Szumska, E., Czuba, M . , Reuhl, K. R., and Miller, D. R. (1983). Demethylation and excretion of methyl mercury by the Guinea pig. Environ. Res. 32, 247-257'. Kozie, K. D., and Anderson, R. K. (1991). Productivity, diet, and environmental contaminants in bald eagles nesting near the Wisconsin shoreline of Lake Superior. Arch. Environ. Contam. Toxicol. 20, 41-48. McCollough, M . A. (1989). Molting sequence and aging of bald eagles. Wilson Bull. 101, 1-10. Mclntyre, J. W., and Barr, J. F. (1997). Common Loon. The Birds of North America 313, 31pp. Meyer, M . W., Evers, D. C , Hartigan, J. J., and Rasmussen, P. S. (1998). Patterns of common loon (Gavia immer) mercury exposure, reproduction, and survival in Wisconsin, USA. Environ. Toxicol. Chem. 17, 184-190. Noble, D. G., and Elliott, J. E. (1990). Levels of contaminants in Canadian raptors, 1966 to 1988; effects and temporal trends. Can. Field-Nat. 104, 222-243. Norheim, G., and Froslie, A. (1978). The degree of methylation and organ distribution of mercury in some birds of prey in Norway. Acta Pharmacol. Toxicol. 43, 196-204. Paavila, H. W. (1971). Use of mercury in the Canadian pulp and paper industry. In Mercury in Man's Environment, Royal Society of Canada, ed., Ottawa, Ontario, Canada, pp. 40-43. Pokras, M . A., Hanley, C , and Gordon, Z. (1998). Liver mercury and methylmercury concentrations in New England common loons (Gavia immer). Environ. Toxicol. Chem. 17, 202-204. Reid, D. S., and Morley, R. L. (1975). Mercury Contamination of Fish from Pinchi Lake, B.C.: Fish and Wildlife Branch Report, B.C. Ministry of Environment, Lands and Parks, pp. 1-18. Scheuhammer, A. M . (1987). The chronic toxicity of aluminum, cadmium, mercury, and lead in birds: A review. Environ. Pollut. 46, 263-295. 228 Scheuhammer, A. M . , and Bond, D. (1991). Factors affecting the determination of total mercury in biological samples by continuous-flow cold vapor atomic absorption spectrophotometry. Biol. Trace Elem. Res. 31, 119-129. Scheuhammer, A. M . , Atchison, C. M . , Wong, A. H. K., and Evers, D. C. (1998). Mercury exposure in breeding common loons (Gavia immer) in central Ontario, Canada. Environ. Toxicol. Chem. 7 7, 191-196. Scheuhammer, A. M . , Wong, A. H. K., and Bond, D. (1998). Mercury and selenium accumulation in common loons (Gavia immer) and common mergansers (Mergus merganser) from eastern Canada. Environ. Toxicol. Chem. 77, 197-201. Thompson, D. R., and Furness, R. W. (1989). Comparison of the levels of total and organic mercury in seabird feathers. Mar. Pollut. Bull. 20, 577-579. Thompson, D. R. (1996). Mercury in Birds and Terrestrial Mammals. In Environmental Contaminants in Wildlife - Interpreting Tissue Concentrations, W. N . Beyer, G. H. Heinz and A. W. Redmon-Norwood, eds. (Boca Raton: Lewis Publishers), pp. 341-356. Watson, T. (1992). Evaluation of mercury concentration in selected environmental receptors in the Williston Lake and Peace River areas of British Columbia (Richmond, B.C., Canada: Triton Environmental Consultants Ltd.). Wood, P. B., White, J. H., Steffer, A., Wood, J. M . , Facemire, C. F., and Percival, H. F. (1996). Mercury concentrations in tissues of Florida bald eagles. J. Wildl. Manage. 60, 178-185. Yoneda, S., and Suzuki, K. T. (1997). Detoxification of mercury by selenium by binding of equimolar Hg-Se complex to a specific plasma protein. Toxicol. Appl. Pharmacol. 143, 274-280. 229 •a 1 t I S s 43 a 3 *o U 43 CQ I cs on 3 '1 T3 CO c -SB E 3 60 3 O I cO E—i CO H I DC "c3 15 H ii w - I "c» '-3 n c o U to W 00 • ^ ^ o ' ^ o d ^ f ^ n S o o ^ c K i Q S ; n t - m ' r i (N r i « vi \d 9 | 8 j> M ) » .2 o e 8 -s S « 3 > T 3 5 43 a a o -2 = . h „ 2 o d » 5 o J> o s g a ^ "§ II » -I w "3. » 1 § x< 5 a> cct a tS CO x i X ) a> cct § o o n tf | 0> •7? -o x> < < r r CN O CN en cn *o TJ ^3 ^3 ^3 A_3 T3 — < — • < < < < < < < < < < S f c f c S S S S f c S f c ^ f c f c S S f c f c f c o 2 - gj « «3 ^ in " i n rt CN 1 o o P\ ON 2 ^ F= ^ 5 3 3 3 C P- C CU CN ^ So X ) 73 . 3 O CS 0- 00 EX <3: a, £ Q U 5 " - £ P ^ C m cs <0 > 3 O o s c « 4^ > CO 4-1 « 3 c U - C t2 CO * ™ o cS > a W Q ? h « w 2 T t v o o r - ~ \ 0 " n \ o o c o c o c < ^ o o r r o o c N r - ~ c n v o o o o r r o v o r r O O ' — ' - i ' — — —. o O ~ —< r - m CN cn r~- CN v © — O O . CN — — i o o o o o o o o o o o o o o o o o o o O Oi CJ\ O i 3 4TI 1 CS cl c u X •g o 00 > 00 I IS CN 43 - 0 •4-* 2 '5 C S E V o CS 0 6 45" 230 Figure 5.1 - Locations where 82 bald eagles were collected around southern British Columbia (1987-1994). 231 232 233 234 Mercury Poisoning I j Fell from nest Aspergillosis Asphyxiation Avian Pox Trap (Snare/Leg-Hold) Drowning Lead Poisoning Eagle Attack Gun Shot Starvation/Disease Undetermined Collision ' '—j (Vehicle/Powerline) •• 1 1 — ' Electrocution Undetermined Trauma 0 5 10 15 20 Total number of individuals Figure 5.5 - Cause of death for 82 bald eagles examined in this study 235 CHAPTER 6 - ASSESSMENT OF DNA STRAND BREAKAGE AS A POTENTIAL BIOMARKER OF METHYLMERCURY EXPOSURE IN BIRDS Introduction Mercury (Hg) has been shown to affect DNA integrity (see reviews Leonard et al, 1983; Sharma and Talukder, 1987; De Flora et al., 1994). For example, variations in total cell DNA content were found in human lymphocytes exposed in vitro to either HgCi2 or CHsHgCl (Verschaeve et al, 1984; Verschaeve et al, 1985). The formation of micronuclei was reported in embryos of the killifish (Fundulus heteroclitus) exposed in vivo to 0.05 mg/L of methylmercury (meHg) (Perry et al, 1988). Zoll et al (1988) also found an increased incidence of micronuclei formation in larvae and embryos of the urodele amphibian, Pleurodeles waltl, following in vivo exposure to HgCl2 or CFbHgCl. Babich et al. (1990) found an increased incidence of micronuclei formation in BG/F epitheliod cells derived from fin tissue of bluegill sunfish (Lepomis macrochirus) exposed in vitro to various concentrations of CHsHgCl, but not HgC^. DNA strand breaks were observed in Chinese hamster cells exposed to various concentrations of CH 3 HgCl in vitro (Kato, 1976). Betti et al. (1993) also found increased DNA strand breakage in human lymphocytes, and hepatocytes, lymphocytes and gastric mucosa cells of Sprague-Dawley rats exposed in vitro to various concentrations of CH3HgCl. In several instances, these effects have been shown to be both dose and time dependent (Kato, 1976; Cantoni et al, 1982; Robison et al. 1982; Cantoni and Costa, 1983; Betti et al, 1993). The ability of genetic material to be affected by Hg raises the possibility of developing a genetic biomarker of meHg exposure. If successfully developed, such a biomarker could be of particular use for assessing subtle effects of meHg exposure in fish-eating wildlife. Sugg et al (1995) suggested that Hg exposure causes breaks in the DNA of both liver and blood cells of largemouth bass {Micropterus salmoides) from lakes containing mercury and radiocesium released from a nuclear production facility in South Carolina. However the observed effects were likely confounded by the additional presence of 236 radiocesium (Sugg et al, 1995). The majority of the information concerning mercury's ability to cause genetic breakage comes from in vitro, rather than in vivo studies. Of the few in vivo studies that have been completed, even fewer focus solely on the effects of Hg. Rather, the effects of exposure to combinations of contaminants from specific study areas have been tested, of which Hg was one of several pollutants (Meyers-Schone et al, 1993; Sugg et al, 1995). Zoll et al. (1988) were the first to study DNA breakage in vivo in a vertebrate exclusively based on inorganic Hg and meHg treatment. Their results showed a significant number of DNA aberrations in red blood cells of the amphibian urodele, Pleurodeles waltl after exposure to water containing 12 and 24 ppb CHsHgCl, or 12ppb HgCl 2. Low rates of DNA breakage occur normally in cells, however many toxic chemicals have been shown to intensify this effect to varying degrees in different tissues (Shugart, 1993; Theodorakis et al, 1994; Sugg et al, 1995; Martin Jr. and Black, 1998). Single-strand DNA breakage can be caused either directly or indirectly. Direct breakage is characteristic of ionizing radiation, and is believed to be the result of the formation of reactive oxygen intermediates such as OH" and H2O2 (Armel et al, 1977). It is hypothesized that H g 2 + acts in a similar fashion to ionizing radiation to cause DNA breakage (Cantoni et al, 1982). Depletion of protective cellular glutathione levels by H g 2 + can induce the increased formation of oxygen radicals and may result in increased DNA breakage through free radical attack on the DNA itself (Cantoni et al, 1982). Indirect breakage, which is typical of many genotoxic chemicals, involves alterations to DNA structure (i.e. adduct formation or base modifications) that require repair, and therefore the strand is broken as a part of the enzymatic repair process (Shugart et al, 1992). HgCl2 has been documented to form adducts with DNA in Chinese hamster ovary (CHO) cells (Cantoni et al, 1984), although this has not been proven to be the cause of DNA breakage. In the presence of meHg, it has been proposed that the DNA double helix is denatured to single-stranded state, which can then be easily hydrolyzed by nucleases (Gruenwedel and Cruikshank, 1990). 237 The present study examines DNA damage in livers of birds after chronic exposure to a range of environmentally relevant concentrations of dietary meHg. This is the first study to examine both adult and immature avian species (ring doves and common loons) for DNA breakage in response to dietary meHg exposure. Methods Doves and MeHg Exposure Forty adult ring doves (Streptopelia risoria), housed in breeding pairs at the National Wildlife Research Centre (NWRC) in Hull, Quebec, were assigned to one of four different dietary doses of meHg. Five pairs of doves each were fed, ad libitum, a standard pelletized diet (ICN Nutritional Biochemical - Cleveland, OH) containing 0, 1, 2.5 or 5 fig/g meHg, respectively. After 136 days of exposure, all doves were sacrificed using anesthesia (4% isoforaine) followed by decapitation. Samples of dissected liver were frozen at -80°C for genetic analyses. Plastics used for storage of tissues were previously acid-washed in dilute (1.5%) nitric acid for a minimum of 8 hours, rinsed with double-deionized water and allowed to air-dry completely. A l l experimental protocols were approved by The NWRC Animal Care Committee. Common Loons and MeHg Exposure Methods for the collection of common loon (Gavia immer) eggs, and details on the dosing regime for the hatched chicks are presented in Kenow et al. (in press). In brief, common loon eggs close to hatching were obtained from various nesting sites, and brought to the Upper Midwest Environmental Sciences Center (UMESC) in La Crosse, Wisconsin for the final few days of incubation. Eggs were collected from areas where loons were known to have elevated meHg exposure (lake pH < 6.3) and low meHg exposure (lake pH > 6.3). Hatched chicks were raised for a period of 105 days on one of three dietary treatments: an ad libitum diet of rainbow trout (Oncorhynchus mykiss) dosed 238 with 0 (control), 0.1, or 0.5 |ag meHg per g wet fish weight, respectively. The dose was administered by adding a gelatin capsule with the prescribed daily dose of meHg to the trout (Kenow et al. [in press]). Rainbow trout used for feeding were raised at UMESC, and contained only low background levels of meHg (less than 0.03 ng per g wet weight) (Kenow et al. [in press]). Chicks were randomly assigned to dose groups following hatching, with each dose group containing at least one bird of each sex from both a neutral and low pH lake. Following completion of the dosing study, loons were sacrificed and livers were dissected and frozen at -80°C. A small portion of each liver was shipped frozen, on dry ice, to NWRC for genetics analyses. A l l samples were kept frozen at -80°C until DNA isolation. DNA Isolation DNA was isolated from liver tissue using DNeasy Tissue kits supplied by QIAGEN Inc. (Mississauga, ON) with supplemental RNase (QIAGEN Inc.) treatment. After DNA elution was complete, all samples were tested for optical density at 260 and 280 nm to determine DNA concentration, and to ensure that samples were free of residual RNA. DNA Damage Assay Samples were analyzed for total strand breaks using an agarose gel electrophoresis protocol adapted from Theodorakis et al. (1994). Standard agarose gels (1%) were prepared using a solution of 50 mM NaCl and 4 mM EDTA. The 11 by 14 cm gels contained 14 wells each. This allowed 12 samples to be run with 2 standard lanes at either end of the gel. A DNA extension ladder (GibcoBRL, Life Technologies -Burlington, ON) was used for the standard to ensure that a broad range of molecular weights and migration distances were obtained. A solution of 30 mM NaOH and 2mM EDTA (pH 12) was prepared for use as a running buffer during electrophoresis. Samples and standards were loaded into the gels and 239 allowed to denature by soaking for 1 hour in the alkaline running buffer prior to electrophoresis. Gels were electrophosed for 1.5 hours at 125 volts. After electrophoresis, the gels were immersed in a neutralizing solution of 10 mM, 3-[N-Morpholino]propane-sulfonic acid (MOPS; Sigma Chemical Co. - St. Louis, MO), pH 7 for 45 minutes. To view the results, the gels were stained in a solution of SYBR Gold Nucleic Acid Gel Stain (Molecular Probes, Inc. - Eugene, OR) and lx TBE, diluted from a lOx TBE stock solution of 1.0M Tris, 0.9M Boric acid and 0.01M EDTA at pH 8.4 (GibcoBRL, Life Technologies) for a minimum of 30 minutes. The gels were then digitally photographed under UV light. Gel Analysis A bitmap image of each gel was imported into the National Institute of Health's (NIH) ImageJ (version 1.12) for analysis. A plot of intensity versus electrophoretic migration distance was created from each lane including the standards. The migration distance associated with the point at which the area under the curve could be divided into two equal halves by integration was then compared to values obtained from the standard curves. Standard curves were created by plotting the migration distance of the extension ladder DNA standards against their known base-pair lengths. In order to properly compare values across gels, duplicates, and in some cases triplicates of samples were run on separate gels. The final median DNA lengths for each sample were compared for consistency between gels, and then averaged. 240 Blood Mercury Analyses Ring Doves Approximately 0.5 ml of blood from each ring dove was digested and analyzed by continuous-flow cold vapor atomic absorption spectrophotometry (CVAAS) using a Perkin-Elmer 3030B spectrophotometer according to Scheuhammer and Bond (1991). Approximately 0.5 ml of blood was allowed to digest overnight at room temperature in 1 ml of nitric acid (Instra-analyzed 70%, JT Baker). The following day, the mixture was heated to 70°C for 1 hour. After cooling, 1 ml sulfuric acid (95-97%, Merck) was added, and test tube contents were shaken gently to mix. Hydrochloric acid (0.5 ml, Instra-analyzed 36-38%, JT Baker) was then added to the digest, which was again heated to 70°C for approximately 2 hours to ensure complete digestion, and allowed to cool overnight. Samples were transferred to 30-ml test tubes and volumes were adjusted to 10 ml with 2 mM potassium dichromate in 3% hydrochloric acid, and mixed thoroughly. Hydrochloric acid (9.9 ml of 1.5%) and 0.1 ml of octanol (anti-foaming agent) was added to the solution. Samples were then analyzed using CVAAS. Digest blanks, duplicate samples and certified reference materials (dogfish muscle [DORM-2] and dogfish liver [DOLT-2]) from the National Research Council of Canada (NRCC) were run during blood Hg analyses to ensure accuracy of Hg results. Recovery of digested certified reference materials was 91.7% for DORM-2 and 97.4% for DOLT-2. A l l digest blanks were below the Hg detection limit [0.1 (ig/g]. Common Loons Methods listed here were adapted from those reported by Kenow et al. (in press). Blood samples were collected from the jugular vein of each chick at the end of the 15-week dosing period and stored in a cryotube prior to Hg analysis. Mercury analyses were performed using standard USEPA methods for CVAAS at EnChem, Inc. (Madison, WI, USA). Blanks and certified reference materials were run with all samples to ensure analytical accuracy. 241 Statistics Prior to completing analyses, all data were tested for normality. Data that were not normal were ln-transformed. For the ring dove livers, a two-way analysis of variance (ANOVA) was used to determine if there was a difference between the median DNA length of birds on different diets, taking the sex of the bird into account. Least squares A N O V A was used to determine whether the median DNA length of male and female common loons raised on different diets varied between birds exposed to higher levels of Hg in ovo (i.e. those birds hatching from eggs obtained on a low pH lake) compared to those with minimal in ovo exposure (i.e. from neutral pH lakes). Analysis of covariance (ANCOVA) was further used to test if median DNA length was related to blood Hg concentrations. One-way A N O V A was used to test for a significant difference in blood Hg concentrations in ring doves and common loons from the various treatment groups. T-tests were used to examine differences in male and female blood Hg concentrations of ring doves and common loons from the same dose groups. A l l statistical analyses were performed using SigmaStat for Windows, version 2.03S and JMP (SAS Institute, North Carolina). Results Ring Doves Livers from all 40 ring doves were analyzed for total strand breakage using alkaline agarose gel electrophoresis. An example of a ring dove liver gel photograph is given on Figure 6.1. Details regarding each bird's diet, sex and resulting median DNA length are shown in Table 6.1. There was a significant (P = 0.044) sex by diet interaction on median DNA length. Specifically, females in the 1 ug/g dose group were found to have longer median DNA lengths compared to males from the 5 ug/g dose group. This significant difference was no longer found when blood Hg concentrations were added as 242 a covariate of median DNA length. Average median DNA lengths ± standard deviation (SD) for male and female birds from each dose group are shown on Figure 6.2. A l l ring doves fed the control diet had non-detectable levels of Hg in blood. Blood Hg concentrations increased with increasing dietary meHg dose (P < 0.001). Mean blood Hg concentrations (± SD) for doves in the 1 ug/g, 2.5 ug/g and 5 ug/g meHg dose groups were 2.85 ± 0.79, 6.43 ± 1.95 and 9.68 ± 2.48 ug/g wet weight (ww), respectively. As shown in Figure 6.3, average blood Hg concentrations were higher in females from each dose group compared to males. It should be noted that a significant difference was only found between males and females in the high (5 ug/g meHg) dose group (P = 0.047). Common Loons Thirty-one livers from common loon chicks were analyzed for total strand breakage (11 control birds, and 10 each from the 0.1 and 0.5 ug/g meHg dose groups). A characteristic loon liver gel photograph is shown on Figure 6.4. The median DNA lengths for each loon liver, as well as information regarding each bird's diet, sex and in ovo meHg exposure are presented in Table 6.2. No significant effect of diet was found for median DNA lengths regardless of sex or in ovo exposure. In addition, no significant sex by diet interaction was found, and blood Hg concentration was not a significant covariate of median DNA length. Average median DNA lengths (± SD) for the different dose groups are presented on Figure 6.5. Loons raised on the control diet had significantly lower mean (± SD) blood Hg levels (0.075 ± 0.019 ug/ml) compared to loons on the 0.1 ug/g meHg dose diet (0.60 ± 0.10 Ug/ml), which had significantly lower blood Hg levels than loons in the 0.5 \ig/g meHg dose group (3.2 ± 1.1 ug/ml, P < 0.001). Females from each dose group had higher blood Hg concentrations on average (see Figure 6.6), however the difference was not significant. Blood Hg concentrations for each of the 31 birds are presented in Table 6.2. 243 Discussion Chronic exposure to genotoxic chemicals may lead to an increased frequency of single-strand breaks, or production of double-strand breaks (Theodorakis et al, 1994). Alternatively, DNA strand breaks may accumulate in cells if the rate of repair is slow. Robison et al. (1984) showed that there was a notable decrease in DNA repair of cultured Chinese hamster ovary cells following exposure to HgCl2 at concentrations greater than 10 uM for 1 hour. In fact, H g 2 + directly inhibited repair when added to X-irradiated cells, which would normally undergo rapid repair following exposure (Cantoni and Costa, 1983). It should be noted that breakage caused by meHg was repaired in cultured human lymphocytes 90 minutes after termination of exposure (Betti et al, 1993); however, this was a single-dose, in vitro experiment. Miller et al. (1979) found that DNA repair was inhibited in the leucocytes of cats after chronic feeding on fish containing 0.14, 0.33, or 0.76 ug/g of meHg as 35% of their diet. In the case of fish-eating birds, dietary meHg levels can be similar or even higher than those fed to cats by Miller et al. (1979). When piscivorous birds consume fish contaminated with meHg, it distributes through the blood to various target organs such as the kidney, liver and central nervous system (Pokras et al, 1998). Virtually all of the Hg bound to red blood cells, and traveling in the blood stream is meHg (Evers et al, 1998). Methylmercury is often described as being significantly more toxic than inorganic Hg (Ramel, 1973; Sharma and Talukder, 1987); however the genotoxic effects of different Hg compounds are qualitatively comparable (De Flora et al, 1994). In a study conducted by Betti et al. (1993), hepatocytes, lymphocytes, and gastric mucosa cells of Sprague-Dawley rats, and human lymphocytes were treated in vitro with concentrations of meHg chloride (0.5 to 4.0 ug/ml) for 1 hour. Results showed a dose-dependent relationship between DNA breakage and meHg chloride concentration. In all cases except human lymphocytes, DNA breakage was significantly greater than controls, starting at concentrations as low as 0.5 (ig/ml meHg chloride. Significant damage to human lymphocytes was shown at concentrations of 1 Ug/ml or greater (Betti et al, 1993). Comparing these concentrations with the blood Hg levels observed in the current experiment, it appeared reasonable to assume that similar 244 DNA breakage should be occurring in the ring doves and common loons. It was also anticipated that the frequency of breakage would correlate positively with the degree of exposure to meHg. Robison et al. (1982) demonstrated that breaks induced by HgCh caused a reduction in average DNA molecular weight that increased as the level of exposure and time of exposure increased. Cantoni and Costa (1983) and Cantoni et al. (1982) also showed a dose and time-dependent response, following treatment of cultured CHO cells with HgCh. Methylmercuric chloride was also found to induce increasing chromosome breakage in cultured Chinese hamster cells at doses of 0.1-10ug/g (Kato, 1976). The results from the current study, however, do not indicate that increased dietary meHg exposure is causing greater strand breakage in the livers of either ring doves or common loons. Despite the high concentrations of Hg that are known to accumulate in the liver compared to blood (Bhatnagar et al, 1982), we were unable to find any significant DNA strand breakage that might be attributed to dietary meHg exposure. In a previous study where ring doves were fed meHg over a similar period of time at a dose of 4 ug/g, liver Hg concentrations (mean = 51.4 ug/g ww, range = 40.3 - 71.2 ug/g ww) were greater than those measured in the blood (4.1 ug/g ww, range = 3.8 - 4.5 ug/g ww; Canadian Wildlife Service, unpub. data). Based on this, the ring doves dosed with 5 ug/g meHg in the diet in the current study would have experienced similar or greater liver meHg accumulation, yet no significant increase in DNA strand breakage was observed compared to control doves. No significant differences were found in the median DNA lengths between all dose groups of common loons, regardless of sex, or whether the egg was produced on a low or neutral pH lake. In fact, the average median DNA lengths for all three dose groups were almost identical, and a similar amount of variation within each group was found (Figure 6.5). This suggests that at the low, albeit environmentally relevant meHg exposure experienced by these common loons, no increased amount of DNA breakage was found in the liver. 245 In a study conducted by Sugg et al. (1995), exposure to Hg and radiocesium by largemouth bass in lakes near a nuclear production facility in South Carolina resulted in considerable differences in DNA breakage between blood, liver and gill tissue. They observed greater DNA damage in blood than liver, and this damage was correlated with the amount of Hg in the tissue. The greater amount of damage found in the blood compared to the liver was attributed to a greater enzymatic repair capability of liver than of blood cells. This is contradictory to much of the literature which suggests that Hg inhibits DNA repair of strand breaks (Miller et al, 1979; Cantoni and Costa, 1983; Robison et al, 1984). However, since the results obtained in our study showed no significant difference in median DNA length between control birds and birds dosed with meHg, it appears unlikely that meHg is causing inhibition or a decreased rate of DNA repair in the livers of our birds. In fact, DNA repair rates may have increased in livers of meHg-exposed birds thereby compensating for damage caused by the presence of meHg. Another possible explanation for the lack of increased DNA strand breakage in the liver could be the presence of glutathione (GSH) in liver cells. The majority of GSH synthesis occurs the liver (Hoffman and Heinz, 1998). Glutathione contains a sulphydryl group that readily binds meHg, and once bound, this complex allows for excretion of meHg by the liver (Hoffman and Heinz, 1998). Hoffman and Heinz (1998) concluded that meHg forms a complex with GSH in mallard livers, effectively reducing the amount of free GSH. The same trend of decreased free GSH with increased Hg was found in diving ducks from the San Francisco Bay region (Hoffman et al, 1998). MeHg was also found to bind almost exclusively to GSH in rabbits given a subcutaneous injection of CHsHgCl, as well as in human and mouse blood incubated with CHsHgCl in vitro (Naganuma and Imura, 1979; Naganuma et al, 1980). Clausing et al. (1984) also speculated that meHg binds primarily to GSH in blood of Japanese quail. Woods and Ellis (1995) actually found that GSH synthesis was upregulated in kidneys of rats chronically exposed to 10 ug/ml meHg in drinking water, indicating increased production of GSH in response to meHg exposure. Mouse neuroblastoma cells exposed in vitro to 2.5 and 5 uM meHg for a period of 24 hours suffered severe alterations in microtubular architecture and also experienced decreased GSH activity; however simultaneous administration of 10 mM 246 GSH dramatically prevented cell injury due to meHg (Kromidas et al, 1990). Thus, it is possible that much of the meHg is being sequestered by GSH in the liver and is not freely available to damage the DNA. Holloway et al. (2003, in press) came to a similar conclusion with respect to the ability of meHg to induce immunotoxicity in whole blood incubated with meHg. It was speculated meHg-induced phagocytic activity was not occurring in white blood cells because proteins such as GSH and hemoglobin present in red blood cells may be sequestering meHg in whole blood (Holloway et al, 2003, in press). The present experiments provided information regarding what might be expected in different scenarios of environmental meHg exposure in fish-eating wild birds. In the worst-case scenario, adult ring doves were exposed to dietary levels of meHg up to 5 (ig/g dw. Concentrations of this magnitude are rare in fish, especially of the size consumed by common loons and many other piscivorous birds (up to 250 grams). In addition, adult ring doves exposed during this study did not undergo a molt. Feathers have been shown to contain up to 93% of the body burden of meHg (Braune and Gaskin, 1987). Therefore, the degree of meHg exposure in these ring doves could be considered to represent a worst-case scenario. In the common loon experiment, environmentally relevant dietary meHg doses were used (0.1 and 0.5 p.g/g meHg) and were fed to growing chicks. Common loon chicks are able to eliminate much of their body burden of Hg (26%, on average) during the first 105 days of life through feather growth (Kenow et al, in press). This is an important excretory route for meHg, and is believed to provide a protective effect against meHg toxicity in growing chicks (Montiero and Furness, 2001). The lack of elevated DNA breakage observed in the livers of the common loons from this experiment may mirror what is occurring in wild common loon chicks up to 105 days old. As Kenow et al. (in press) suggest, these results should not be extrapolated to common loons past the 105 day period, as tissue Hg concentrations will tend to increase more sharply due to the cessation of feather growth. However, if the results of the ring dove dosing are indicative of other adult birds exposed to elevated dietary levels of meHg, it is unlikely increased DNA breakage will be observed in the liver regardless of age. 247 DNA strand breakage resulting from exposure to a genotoxicant has been shown to be both simple and quick to measure. Many analytical techniques, such as the alkaline unwinding assay (Shugart, 1988) and DNA precipitation assay (Olive, 1988) are proven to accurately measure single-strand breaks in DNA. More recently, agarose gel electrophoresis has proven successful in measuring both double and single strand breaks in DNA (Theodorakis et al, 1992; Theodorakis et al, 1994; Martin and Black, 1998). In this study, we were unable to find evidence of increased DNA strand breakage using alkaline agarose gel electrophoresis. However, in some cases, this assay has been described as being somewhat insensitive (Husby and McBee, 1999), which may be part of the reason no significant relationship between meHg exposure and DNA breakage was shown. Slight variations in DNA concentration loaded into the wells may affect the calculation of the median DNA length (Theodorakis et al, 1994). In addition, an average variation of 0.9 kbp was found between duplicates run on different gels. This level of variation may affect the overall ability of this technique to accurately detect relationships between meHg dose and median DNA length in cases where length changes are relatively small. The sensitivity of this assay should be determined as part of future experiments. Based on the results of the present study, it appears that DNA breakage would not be useful as a biomarker of meHg-caused genetic damage in wild birds. Whereas several in vitro experiments have shown DNA damage as a result of Hg and/or meHg exposure (see reviews Leonard et al, 1983; Sharma and Talukder, 1987; De Flora et al, 1994), such effects might not be present in vivo, because there are many factors, such as DNA repair rates, and the binding of meHg to GSH, which may offer protection to an organism, and affect the ability of meHg to cause damage in vivo. In addition, a more sensitive technique may be required if a correlative relationship between Hg concentration and DNA breakage within a tissue is to be established. It would also be worthwhile to combine DNA breakage data with analysis of other forms of DNA damage known to occur as a result of toxicant exposure (i.e. sister-chromatid exchanges, micronuclei formation and/or variations in total cellular DNA content) to obtain a broader view of genetic damage caused by meHg exposure in vivo. 248 Acknowledgements Tony Scheuhammer, Angela Lorenzen, Sean Kennedy and Stephanie Jones provided assistance, laboratory space, equipment and reagents for the completion of this study. Jennifer Holloway completed Hg analyses on the ring dove blood, and provided information on the dosing of ring doves. Kevin Kenow and associates at UMESC provided liver tissues and blood-Hg analytical results for the common loons raised and dosed at their facilities. References Armel, P. R., Strniste, G. F., and Wallace, S. S. (1977). Studies on Escherichia coli x-ray endonuclease specificity. Roles of hydroxyl and reducing radicals in the production of DNA lesions. Radiat. Res. 69, 328-338. Babich, H., Goldstein, S. H , and Borenfreund, E. (1990). In vitro cyto- and genotoxicity of organomercurials to cells in culture. Toxicol. Lett. 50, 143-149. Betti, C., Barale, C. B., and Pool-Zobel, B. L. (1993). Comparative studies on cytotoxic and genotoxic effects of two organic mercury compounds in lymphocytes and gastric mucosa cells of Sprague-Dawley rats. Environ. Mol. Mutagen. 22, 172-180. Bhatnagar, M . K., Vrablic, O. E., and Yamashiro, S. (1982). Ultrastructural alterations of the liver of Pekin ducks fed methyl mercury-containing diets. J. Toxicol. Environ. Health 10, 981-1003. Braune, B. M . , and Gaskin, D. E. (1987). Mercury levels in Bonaparte's gulls (Larus Philadelphia) during autumn molt in the Quoddy Region, New Brunswick, Canada. Arch. Environ. Contam. Toxicol. 16, 539-549. Cantoni, O., Evans, R. M . , and Costa, M . (1982). Similarity in the acute cytotoxic response of mammalian cells to mercury (II) and x-rays: DNA damage and glutathione depletion. Biochem. Biophys. Res. Comm. 108, 614-619. Cantoni, O., and Costa, M . (1983). Correlations of DNA strand breaks and their repair with cell survival following acute exposure to mercury (II) and x-rays. Mol. Pharmacol. 24, 84-89. Cantoni, O., Christie, N . T., Swann, A., Drath, D. B., and Costa, M . (1984). Mechanism of HgCL cytotoxicity in cultured mammalian cells. Mol. Pharmacol. 26, 360-368. 249 Clausing, P., Riedel, B., Gericke, S., Griin, G., and Miiller, L. (1984). Differences in the distribution of methyl mercury in erythrocytes, plasma, and brain of Japanese quails and rats after a single oral dose. Arch. Toxicol. 56, 132-135. De Flora, S., Bennicelli, C , and Bagnasco, M . (1994). Genotoxicity of mercury compounds. A review. Mutat. Res. 317, 57-79. Evers, D. C , Kaplan, J. D., Meyer, M . W., Reaman, P. S., Braselton, W. E., Major, A., Burgess, N . , and Scheuhammer, A. M . (1998). Geographic trend in mercury measured in common loon feathers and blood. Environ. Toxicol. Chem. 17, 173-183. Gruenwedel, D. W., and Cruikshank, M . K. (1990). Mercury-induced DNA polymorphism probing the conformation of mercury-II DNA via staphylococcal nuclease digestion and circular dichroism measurements. Biochemistry. 29, 2110-2116. Hoffman, D. J. and Heinz, G. H. (1998). Effects of mercury and selenium on glutathione metabolism and oxidative stress in mallard ducks. Environ. Toxicol. Chem. 17, 161-166. Hoffman, D. J., Ohlendorf, H. M . , Marn, C. M . , and Pendleton, G. W. (1998). Association of mercury and selenium with altered glutathione metabolism and oxidative stress in diving ducks form the San Francisco Bay Region, USA. Environ. Toxicol. Chem. 17, 167-172. Holloway, J., Scheuhammer, A. M . , and Chan, H. M . (2003). Assessment of white blood cell phagocytosis as an immunological indicator of methylmercury exposure in birds. Arch. Environ. Contam. Toxicol, in press. Husby, M . P., and McBee, K. (1999). Nuclear DNA content variation and double-strand DNA breakage in white-footed mice (peromyscus leucopus) collected from abandoned strip mines, Oklahoma, USA. Environ. Toxicol. Chem. 18, 926-931. Kato, R. (1976). Chromosome breakage associated with organic mercury in Chinese hamster cells in vitro. Mutat. Res. 38, 340-341. Kenow, K. P., Gutreuter, S., Hines, R. K., Meyer, M . W., Fournier, F., and Karasov, W. H. Effects of methyl mercury exposure on the growth of juvenile common loons. Ecotoxicol. In press. Kromidas, L., Trombetta, L. D., and Jamall, I. S. (1990). The protective effects of glutathione against methylmercury cytotoxicity. Toxicol. Lett. 57, 67-80. Leonard, A., Jacquet, P., and Lauwerys, R. R. (1983). Mutagenicity and teratogenicity of mercury compounds. Mutat. Res. 114, 1-18. Martin, J., L.K., and Black, M . C. (1998). Biomarker assessment of the effects of coal strip-mine contamination on channel catfish. Ecotoxicol. Environ. Safety 41, 307-320. 250 Meyers-Schone, L., Shugart, L. R., and Beauchamp, J. J. (1993). Comparison of two freshwater turtle species as monitors of radionuclide and chemical contamination: DNA damage and residue analysis. Environ. Toxicol. Chem. 12, 1487-1496. Miller, C. T., Zawidzka, Z., Nagy, E., and Charbonneau, S. M . (1979). Indicators of genetic toxicity in leucocytes and granulocytic precursors after chronic methylmercury ingestion by cats. Bull. Environ. Contam. Toxicol. 21, 296-303. Monteiro, L. R., and Furness, R. W. (2001). Kinetics, dose-response, excretion, and toxicity of methylmercury in free-living Cory's.shearwater chicks. Environ Toxicol Chem 20, 1816-1823. Naganuma, A., and Imura, N . (1979). Methylmercury binds to a low molecular weight substance in rabbit and human erythrocytes. Toxicol. Appl. Pharmacol. 47, 613-616. Naganuma, A., Koyama, Y. , and Imura, N . (1980). Behaviour of methylmercury in mammalian erythrocytes. Toxicol. Appl. Pharmacol. 54, 405-410. Perry, D. M . , Weis, J. S., and Weis, P. (1988). Cytogenetic effects of methylmercury in embryos of the killifish, Fundulus heterclitus. Arch. Environ. Contam. Toxicol. 17, 569-574. Pokras, M . A., Hanley, C , and Gordon, Z. (1998). Liver mercury and methylmercury concentrations in New England common loons (Gavia immer). Environ. Toxicol. Chem. 17, 202-204. Ramel, C. (1973). The effect of metal compounds on chromosome segregation. Mutat. Res. 21, 45-46. Robison, S. FL, Cantoni, O., and Costa, M . (1982). Strand breakage and decreased molecular weight of DNA induced by specific metal compounds. Carcinogenesis 3, 657-662. Robison, S. FL, Cantoni, O., and Costa, M . (1984). Analysis of metal-induced DNA lesions and DNA-repair replication in mammalian cells. Mutat. Res. 131, 173-181. Scheuhammer, A. M . , and Bond, D. (1991). Factors affecting the determination of total mercury in biological samples by continuous-flow cold vapor atomic absorption spectrophotometry. Biol. Trace Elem. Res. 31, 119-129. Sharma, A., and Talukder, G. (1987). Effects of metals on chromosomes of higher organisms. Environ. Mutagen. 9, 191-226. Shugart, L. R. (1988). Quantitation of chemically induced damage to DNA of aquatic organisms by alkaline unwinding assay. Aquat. Toxicol. 13, 43-52. 251 Shugart, L. R., Bickham, J., Jackim, G., McMahon, G., Ridley, W., Stein, J., and Steinert, S. (1992). DNA Alterations. In Biomarkers: Biochemical, Physiological, and Histological Markers of Anthropogenic Stress, R. J. Huggett, R. A. Kimerle, P. M . Mehrle and H. L. Bergman, eds. (Boca Raton, FL: Lewis), pp. 125-153. Shugart, L. (1993). Genotoxic responses in blood. In Nondestructive Biomarkers in Vertebrates, C. M . Fossi, ed.: Lewis Publishers, pp. 131-145. Sugg, D. W., Chesser, R. K., Brooks, J. A., and Grasman, B. T. (1995). The association of DNA damage to concentrations of mercury and radiocesium in largemouth bass. Environ. Toxicol. Chem. 14, 661-668. Theodorakis, C. W., D'Surney, S. J., Bickham, J. W., Lyne, T. B., Bradley, B. P., Hawkins, W. E., Farkas, W. L., McCarthy, J. F., and Shugart, L. R. (1992). Sequential expression of biomarkers in bluegill sunfish exposed to contaminated sediment. Ecotoxicol. 1, 45-73. Theodorakis, C. W., D'Surney, S. J., and Shugart, L. R. (1994). Detection of genotoxic insult as DNA strand breaks in fish blood cells by agarose gel electrophoresis. Environ. Toxicol. Chem. 13, 1023-1031. Verschaeve, L., Kirsch-Volders, M . , and Susanne, C. (1984). Mercury-induced segregational errors of chromosomes in human lymphocytes and Indian muntjac cells. Toxicol. Lett. 21, 247-253. Verschaeve, L., Kirsch-Volders, M . , Hens, L., and Susanne, C. (1985). Comparative in vitro cytogenetic studies in mercury-exposed human lymphocytes. Mutat. Res. 157, 221-226. Woods, J. S., and Ellis, M . E. (1995). Up-regulation of glutathione synthesis in rat kidney by methyl mercury. Biochem. Pharmacol. 50, 1719-1724. Zoll, C , Saouter, E., Boudou, A., Ribeyre, F., and Jaylet, A. (1988). Genotoxicity and bioaccumulation of methyl mercury and mercuric chloride in vivo in the newt Pleurodeles waltl. Mutagenesis 3, 337-343. 252 Table 6.1- Median liver DNA lengths and blood Hg concentrations for ring doves exposed to 0, 1,2.5 or 5 pg/g meHg. Diet Sample ID sex Blood Hg (ug/g ww) Median DNA length (kbp) control OIF F ~0 9.06 control 01M M ~0 9.21 control 02F F ~0 7.71 control 02M M ~0 6.91 control 03F F ~0 6.38 control 03M M ~0 8.56 control 04F F N/A 6.8 control 04M M ~0 8.97 control 05F F ~0 6.31 control 05M M ~0 6.9 1 ug/g meHg 06F F 1.94 12.82 1 ug/g meHg 06M M 2.60 8.77 1 ug/g meHg 07F F 3.29 11.43 1 ug/g meHg 07M M 3.15 7.77 1 ug/g meHg 08F F 3.68 8.26 1 ug/g meHg 08M M 2.34 7.94 1 ug/g meHg 09F F 2.16 9.42 1 ug/g meHg 09M M 2.28 8.38 1 ug/g meHg 10F F 4.48 8.53 1 ug/g meHg 10M M 2.59 6.77 2.5 ug/g meHg 1 IF F 11.59 9.87 2.5 ug/g meHg 11M M 5.74 11.49 2.5 ug/g meHg 12F F 7.18 7.63 2.5 ug/g meHg 12M M 6.06 6.4 2.5 ug/g meHg 13F F 5.56 8.05 2.5 ug/g meHg 13M M 5.52 6.17 2.5 ug/g meHg 14F F 5.91 7.37 2.5 ug/g meHg 14M M 4.65 8 2.5 ug/g meHg 15F F 6.76 7.22. 2.5 ug/g meHg 15M M 5.29 9.6 5 ug/g meHg 16F F 11.73 8.13 5 ug/g meHg 16M M 6.38 7.75 5 ug/g meHg 17F F 9.54 10.28 5 ug/g meHg 17M M 7.66 6.34 5 ug/g meHg 18F F 9.16 8.15 5 ug/g meHg 18M M 9.64 3.7 5 ug/g meHg 19F F 12.83 9.61 5 ug/g meHg 19M M 11.34 9.2 5 ug/g meHg 20F F 12.63 8.6 5 ug/g meHg 20M M 5.88 5.83 Notes: N/A - Not analyzed 253 Table 6.2 - Median liver DNA lengths and blood Hg concentrations for common loons exposed to 0,0.1 or 0.5 ug/g meHg. In ovo meHg Blood Hg Median DNA Diet Sample ID Sex exposure (ugAnl) length (kbp) control 1438 F L 0.11 6.06 control 1441 F L 0.086 6.06 control 1443 F H 0.087 6.71 control 1449 F H 0.063 7.79 control 1453 F L 0.081 6.28 control 1459 F L 0.062 6.57 control 1435 M H 0.067 4.79 control 1448 M L 0.057 4.88 control 1461 M H 0.041 6.37 control 1466 M L 0.079 4.6 control 1469 M L 0.094 6.58 0.1 ug/g meHg 1437 F L 0.83 5.3 0.1 fig/g meHg 1440 F L 0.66 6.42 0.1 ug/g meHg 1444 F L 0.56 4.9 0.1 ug/g meHg 1447 F H 0.53 6.42 0.1 ug/g meHg 1457 F H 0.64 5.74 0.1 ug/g meHg 1465 F L 0.51 7.39 0.1 ug/g meHg 1434 M H 0.6 6.46 0.1 ug/g meHg 1446 M H 0.46 5.96 0.1 ug/g meHg 1462 M H 0.62 5.11 0.1 ug/g meHg 1470 M L 0.56 4.22 0.5 ug/g meHg 1436 F H 3.9 6.38 0.5 ug/g meHg 1439 F L 3.1 7.01 0.5 ug/g meHg 1442 F L 2.8 7.15 0.5 ug/g meHg 1460 F H 2 6.4 0.5 ug/g meHg 1463 F L 5.4 6.08 0.5 ug/g meHg 1464 F H 1.9 3.78 0.5 ug/g meHg 1467 F H 4.4 7.3 0.5 ug/g meHg 1468 F L 3 4.89 0.5 ug/g meHg 1452 M L 2 5.04 0.5 ug/g meHg 1455 M H 3.2 6.86 Notes: L=chicks from neutral pH lakes H=chicks from low pH lakes 254 Figure 6.1 - Characteristic ring dove liver gel photographed under UV light. Lanes (numbered starting from left) 1 and 14 are standards, lanes 2-4 are controls, lanes 5-7, 8-10, and 11-13 are from the 1,2.5 and 5 pg/g meHg dose groups, respectively 255 (dqpf) qiSaaf VNQ n*!I»W[ 00 "So o CN o a> o -o a) O .3° i i 1 •3 s i i p < § I-I 1 3 CN 00 256 OO VO (MM §/3ri) § n poo[g 00 -S? bt) =L >r> o CN T3 1.5 ug/g dw Hg in the study area. Al l grebe eggs collected (including those from Pinchi Lake) were below the 0.5 ug/g ww Hg concentration often cited as the lowest observed adverse effect level (LOAEL) of embryo toxicity in birds. A relationship between trophic position (8N) and Hg concentrations in red-necked grebe eggs was found, indicating that female grebes feeding at a higher trophic position deposit a greater amount of Hg in their eggs. A l l bird species were found to be feeding at a higher trophic position compared to the various fish species; however a possible overlap in diet, among other factors, resulted in lower than expected 8N fractionation factors between bird and fish tissues. Over the three seasons, bald eagle reproductive success (the total number of active territories found at the beginning of May that produced 8-week-old eaglets) was 62% on Pinchi compared to 64% on all other study lakes combined. Average productivity (the total number of eaglets produced per active territory) over the three-year study was 0.98 on Pinchi compared to 1.17 on all other study lakes combined. According to Sprunt et al. (1973), productivity of the bald eagles from the study area was similar to that of a healthy population producing a surplus of chicks (i.e. greater than 1.0 chick/occupied territory), and greater than 0.7 young/occupied territory, the value associated with a sustaining population. Despite the apparently elevated Hg exposure of adult eagles from Pinchi Lake (mean blood-Hg = 6.54 ug/g ww, n = 3), the birds appeared to be in excellent condition, with no evidence of abnormal behavior or lack of coordination. The adult eagle with the highest concentration of Hg in blood (9.44 ug/g ww) successfully raised two eaglets in each of the summers of 2001 and 2002, making it one of the most productive eagles in the study area. Based on our three-year study of the bald eagles from Pinchi Lake and the surrounding region, Hg does not appear to have any obvious adverse effects on overall reproductive success. While meHg exposure, as measured by blood and feather Hg concentrations, was higher in eaglets from Pinchi Lake compared to reference lake birds, the concentrations were comparable to concentrations found in 263 eaglets from lake habitats in Maine (1991-1992; Welch, 1994) and Oregon (1979-1981; Wiemeyer et al, 1989). A review of the general ecology and breeding success of bald eagles, red-necked grebes and common loons from the study area suggests that Hg is not likely impairing reproduction, and that weather/wave action, habitat availability and predation are the most important determinants of breeding success. Based on the suspected and observed gradient in sediment Hg concentrations among the 5 study lakes, it was hypothesized that birds sampled on each study lake would reflect the Hg concentrations found in the fish, which in turn would be related to Hg concentrations in sediments. This was shown to be the case, with significant correlations occurring between sediment, fish, red-necked grebe egg- and bald eagle blood-Hg concentrations. Distribution of Mercury and Selenium in Adult Bald Eagle Secondary Feathers Previous studies have quantified Hg exposure in adult birds through analysis of only one portion of a larger feather, most often the tip of a primary (Bowerman et al, 1994; Cahill et al, 1998). Knowing that birds can eliminate a large portion of their body burden of meHg through deposition into newly forming feathers, it was hypothesized that Hg concentrations in larger feathers would decrease along the length as the feather is grown and the body burden of meHg decreases. In our study, both Hg and Se concentrations were found to decrease from the tip to the base of the feather, and a positive correlation was found between Hg and Se. Overall, the decreasing trend in feather Hg was best shown in feathers that started with a higher Hg concentration in the tip (i.e. feathers from eagles exposed to elevated concentrations of Hg from Pinchi Lake). In addition, higher Hg concentrations were found in the feather vanes compared to the shaft, with Hg concentrations decreasing in the shaft at a greater rate in comparison to the vanes. Based on the results of this study, estimation of Hg and possibly Se content in feathers based only on analysis of the distal portion of the feather, or only the vanes of the feather could create a large bias (error) in reported feather Hg 264 concentrations, particularly if the feather samples were obtained near an area of known.or suspected Hg contamination. Mercury Exposure in Bald Eagles found Dead or Dying in British Columbia Over the period of 1987-1994, 82 adult bald eagles found dead or dying in British Columbia were collected. Post-mortem examinations and total Hg analysis of liver tissue was conducted on all birds. A smaller subset of birds was also tested for meHg and Se content in the liver. Based on the results found, approximately 20% of eagles found as a part of this study were exposed to moderate or greater concentrations of Hg In British Columbia, there are many possible sources of environmental Hg contamination. Pulp and paper mills were significant contributors of Hg to the environment up to about 1970; however emissions were significantly reduced when mercurial slimicides were banned in the early 1970s (Garrett et al, 1980). Reservoir creation for hydroelectricity production resulted in increased meHg production and elevated Hg concentrations in fish in Williston Lake, British Columbia (Watson, 1992). Additional reservoirs throughout British Columbia may also have experienced similar increases in Hg concentrations in fish as a result of flooding, however detailed studies of fish-Hg concentrations are only available for Williston Lake. In addition, chlor-alkali production, known to contribute to increased meHg loadings in other areas of Canada (Barr, 1986), occurred in the Howe Sound district of British Columbia. And, as discussed above, past mining activities in the Pinchi Lake region also resulted in elevated Hg concentrations in sediments, fish and fish-eating birds, including bald eagles. The eagle suspected to have died as a result of meHg exposure and several other moderately exposed birds were found in the vicinity of pulp mill operations in British Columbia. It is suspected that these mills may still be releasing small quantities of Hg to the environment through consumption of caustic soda, lime rock (at Kraft mills), and fuel for power generation (Paavila, 1971). While present releases of Hg to the British Columbian environment may be low, pollution resulting from past uses of Hg may 265 continue to be a source. As discussed above, elevated Hg concentrations are still found in sediments and biota as a result of 1940-1944 Hg mining practices at Pinchi Lake. Because past Hg contamination continues to have consequences today, and given that long-range atmospheric transport and deposition of Hg is widespread and appears to be increasing (Fitzgerald et al, 1998), Hg concentrations in fish-eating birds such as bald eagles should continue to be monitored. The evidence of a higher percentage of total Hg present as meHg, and the higher molar ratio of total Hg to Se (2:1) in the livers of bald eagles, compared to other species of piscivorous birds experiencing similar total Hg concentrations in the liver, suggests that bald eagles may be more susceptible to meHg toxicity compared to other fish-eating birds. As the results of this study showed, bald eagles in British Columbia are sometimes exposed to potentially toxic doses of meHg from their environment. Potential use of DNA breakage as a Biomarker of Dietary meHg Exposure in Birds Previous in vitro studies had shown that inorganic Hg and meHg are capable of causing DNA strand breakage and that these Hg effects were both dose and time dependent (Kato, 1976; Robison et al. 1982; Cantoni and Costa, 1983; Betti et al, 1993). Thus DNA might be a useful biomarker of meHg exposure in wild birds if such damage could be shown to occur in a dose-dependent manner in vivo in tissues of birds dietarily exposed to meHg under controlled conditions. In our study using common loon and ring dove (Streptopelia risoria) livers, however, no evidence of increased DNA breakage was found in dosed birds compared to controls. This is despite the fact that liver tissue undergoes comparatively less cell replacement compared to blood, and is also known to accumulate higher concentrations of Hg over time compared to blood (Bhatnagar et al, 1982; Canadian Wildlife Service, unpub. data). While liver Hg concentrations were not measured, the blood Hg concentrations measured in this study were similar to concentrations known to result in significant DNA strand breakage in vitro. The lack of increased DNA breakage found in the liver of meHg dosed birds suggests that factors, such as DNA repair rates, and the binding of meHg to glutathione, may offer protection to an organism exposed in vivo. Based on these results, chronic consumption of diets 266 containing environmentally relevant concentrations of meHg does not appear to result in significantly increased liver DNA breakage in birds. At this point, it does not appear that measurement of DNA breakage would be useful as a biomarker of meHg exposure in wild birds. Future Research Studies on the chemical speciation of Hg, such as determination of the amount of Hg present as HgS, meHg, Hg 2 + , Hg°, and dimethylmercury, within the sediments of Pinchi Lake compared to lakes from the surrounding area may provide additional information on the bioavailability of Hg to the food chain. Similarly, determination of Hg species within the water column may be valuable. Long-term studies of Hg concentrations in bald eagle populations breeding in the eastern United States have suggested that increased dietary meHg exposure in bald eagles may lead to a greater turnover rate in breeding adult bald eagles due to reduced adult survivorship (C. Todd, Maine Department of Inland Fisheries and Wildlife, pers. comm.). This is still speculative in nature at this time, but may warrant additional long-term monitoring of adult eagle survivorship in known areas of Hg contamination, such as the Pinchi Lake region. Fish and surface sediment sampling in several areas of known or suspected environmental Hg contamination, such as the area where the highly-exposed eagle was found in Powell River, may help to delineate possible areas of concern to breeding bald eagle populations in British Columbia. Future work comparable to the current study of bald eagles in the Pinchi Lake region may be necessary to monitor bald eagle populations, (or other appropriate fish-eating wildlife species) in close proximity to known Hg sources throughout British Columbia. A similar practice of monitoring breeding success and productivity, and trapping adults and young to obtain blood and feather samples for Hg analyses, will assist in determining the variability in the current levels of Hg exposure in bald eagles in various areas across British Columbia. This type 267 of monitoring may also help to develop a better database of Hg concentrations in bald eagle tissues from various habitats, as well as provide more detailed information on bald eagle nesting success/productivity rates in relation to Hg concentrations in tissues. Whereas no significantly increased incidence of DNA breakage was found in the livers of common loons and ring doves exposed to dietary meHg compared to controls, DNA damage in general should not be ruled out as a potential biomarker. Other modes of DNA damage such as variations in total cell DNA content and the formation of micronuclei have been shown to occur as a result of meHg exposure. Future in vivo studies examining different types of genetic damage should be undertaken. If the damage can be correlated to the concentration of meHg in the tissue or diet, then development of a genetic biomarker of meHg exposure may yet be possible. References Barr, J. F. (1986). Population dynamics of the Common loon (Gavia immer) associated with mercury-contaminated waters in northwestern Ontario. Can. Wildl. Serv. Occas. Pap. No. 56, pp. 25. Betti, C , Barale, C. B., and Pool-Zobel, B. L. (1993). Comparative studies on cytotoxic and genotoxic effects of two organic mercury compounds in lymphocytes and gastric mucosa cells of Sprague-Dawley rats. Environ. Mol. Mutagen. 22, 172-180. Bhatnagar, M . K., Vrablic, O. E., and Yamashiro, S. (1982). Ultrastructural alterations of the liver of Pekin ducks fed methyl mercury-containing diets. J. Toxicol. Environ. Health 10, 981-1003. Bowerman IV, W. W., Evans, E. D., Giesy, J. P., and Postupalsky, S. (1994). Using feathers to assess risk of mercury and selenium to bald eagle reproduction in the Great Lakes Region. Arch. Environ. Contam. Toxicol. 27, 294-298. Cahill, T. M . , Anderson, D. W., Elbert, R. A., Perley, B. P., and Johnson, D. R. (1998). Elemental profiles in feather samples from a mercury-contaminated lake in central California. Arch. Environ. Contam. Toxicol. 35, 75-81. Cantoni, O., and Costa, M . (1983). Correlations of DNA strand breaks and their repair with cell survival following acute exposure to mercury (II) and x-rays. Mol. Pharmacol. 24, 84-89. 268 EVS Environment Consultants. (2001). Regional survey of fish mercury concentrations -Pinchi Lake, B.C. (Vancouver: Cominco Ltd.) 91pp. EVS Environment Consultants, Norecol Dames and Moore, and Frontier Geosciences. (1999). Pinchi Lake Mine Site and Lake Investigation: Environmental Assessment Report (Vancouver: Cominco Ltd.). Fitzgerald, W. F., Engstrom, D. R., Mason, R. P., and Nater, E. A. (1998). The case for atmospheric mercury contamination in remote areas. Environ. Sci. Technol. 32, 1-7. Garrett, C. L., MacLeod, L. A., and Sneddon, H. J. (1980). Mercury in the British Columbia and Yukon environments - summary of data to January 1, 1979: Environment Canada, Environmental Protection Service, Pacific Region, pp. 456. Kato, R. (1976). Chromosome breakage associated with organic mercury in Chinese hamster cells in vitro. Mutat. Res. 38, 340-341. Lockhart, W. L., Macdonald, R. W., Outridge, P. M . , Wilkinson, P., DeLaronde, J. B., and Rudd, J. W. M . (2000). Tests of the fidelity of lake sediment core records of mercury deposition to known histories of mercury contamination. Sci. Total Environ. 260, 171-180. Paavila, H. W. (1971). Use of mercury in the Canadian pulp and paper industry. In Mercury in Man's Environment, Royal Society of Canada, ed. (Ottawa, Ontario, Canada), pp. 40-43. Plouffe, A. (1995). Glacial dispersal of mercury from bedrock mineralization along Pinchi fault, north central British Columbia. Water Air Soil Pollut. 80, 1109-1112. Robison, S. H., Cantoni, O., and Costa, M . (1982). Strand breakage and decreased molecular weight of DNA induced by specific metal compounds. Carcinogenesis 3, 657-662. Sprunt, A. IV, Robertson, B. W. Jr, Postupalsky, S., Hensel, R. J., Knoder, C. E., and Ligas, F. J. (1973). Comparative productivity of six bald eagle populations. Trans. No. Amer. Wildl. Nat. Resour. Conf. 38, 96-106. Watson, T. (1992). Evaluation of mercury concentration in selected environmental receptors in the Williston Lake and Peace River areas of British Columbia (Richmond, B.C., Canada: Triton Environmental Consultants Ltd.). Welch, L. J. (1994). Contaminant burdens and reproductive rates of bald eagles breeding in Maine. Unpub. M.S. Thesis. (Orono: University of Maine), 86pp. 269 Wiemeyer, S. N. , Frenzel, R. W., Anthony, R. G., McClelland, B. R., and Knight, R. L. (1989). Environmental contaminants in blood of western bald eagles. J. Raptor Res. 23, 140-146. 270 APPENDIX A THE NEUROTOXICOLOGY OF MERCURY: OXIDATIVE STRESS AND EXCITOTOXICITY INTRODUCTION Mercury (Hg) is a ubiquitous metal in the environment with the potential to be very neurotoxic at acute concentrations. This is in part due to the specific ability of methylmercury (meHg) to bioaccumulate and biomagnify within the food chain. In areas where a significant input of anthropogenically-derived meHg has been added to the system, the resulting neurological disorders arising in the people from these regions were drastic. The first and most well known of these incidents occurred in Minimata, Japan beginning in 1956. This is also where the neurological disorder associated with meHg poisoning got its name: Minimata disease. The production of acetaldehyde and vinyl chloride used meHg as a catalyst, and this lead to a considerable build-up of meHg in Minimata Bay and its fish due to the direct discharge of untreated water. By 1975, 899 individuals were officially diagnosed with the disease, 143 of these people dying as a result (Takizawa, 1979). Another mass epidemic occurred in Iraq in 1971-1972. This incident differed from that of Minimata because the source of meHg in the diet of these people was wheat seed treated with methyl and phenyl Hg fungicides. The treated seed was ground up to make flour for bread, and thus was consumed directly. Food-chain biomagnification was not required for meHg to reach highly toxic levels in this epidemic. It is believed up to 50,000 people were exposed in this incident, with 6530 requiring hospitalization and 459 dying as a result of their exposure (Bakir et al, 1973; Greenwood, 1985). Studies conducted during the Iraq outbreak made it possible to identify symptoms and clinical findings associated with varying meHg exposure levels. In almost all cases, parethesias were the first recorded symptom, and ataxia was the first clinical finding 271 (Bakir et al, 1973). Additional symptoms and findings were also recorded and later plotted for an individual from the incident in Iraq (Figure 1). Figure 1: Onset of clinical symptoms of a victim of the meHg poisoning incident in Iraq (1971 - 1972) Nov Dec Jan Feb Mar Apr May Jun 1972 Source: Clarkson, 1987 These incidents also brought to light the teratogenicity of meHg. In many of cases, an infant would be born with severe neurological disorders while the mother exhibited no signs or symptoms of meHg poisoning (Tsubaki and Irukayama, 1977; Amin-Zaki et al, 1974). MeHg is believed to have a greater affinity for the fetal central nervous system (CNS) over the adult (Kelman and Sasser, 1977), although the reason for this is as yet, unknown. Once meHg enters the brain, it can inflict damage in many ways. Two current hypotheses on how this damage is inflicted will be discussed here, namely oxidative stress and excitotoxicity. 272 Mercury and the Blood-Brain Barrier In order for Hg to induce neurotoxicity in the brain, it must first cross the blood-brain barrier (BBB). Inorganic mercury (Hg2 +) appears to be mostly unable to cross this barrier (Moller-Madsen, 1990), but it can be an important factor in the neurotoxicity of elemental (Hg°) mercury. Elemental Hg is able to diffuse across the BBB where it then becomes oxidized to inorganic Hg (Magos, 1997). Inorganic Hg can cause many of the same effects that meHg does, but it often appears to require higher concentrations to inflict the same degree of damage (Sharma and Talukder, 1987). Methylmercury is highly neurotoxic, and is believed to cross the BBB through use of the large neutral amino acid carrier in endothelial cells of brain capillaries (Kerper et al, 1992). When meHg is bound to cysteine, its structure very closely resembles that of methionine (see Figure 2). As a result, it is able to cross the BBB using the same carrier that enables methionine to cross. This resemblance to methionine may also be responsible for the apparent inhibition of protein synthesis within certain regions of the brain (Clarkson, 1987). Figure 2 - A comparison of the structure of methionine to the meHg cysteine complex C H 3 - Hg - S - C H 2 - CH - COO- MeHg - Cysteine Complex N H 3 + C H 3 - S - C H 2 - C H 2 - CH - COO Methionine I N H , A3 273 OXIDATIVE STRESS IN T H E BRAIN Oxidative stress is regarded as being a very important aspect of meHg-induced neurotoxicity. A substantial amount of evidence exists for the generation of reactive oxygen species by Hg compounds, however much of this data was acquired from in vitro studies. Several in vivo studies have also provided support for this theory, but some have contradicted in vitro studies of the same mechanism. In this section, the mechanisms by which the various Hg species cause oxidative stress will be discussed, as well as current theories associated with these mechanisms. Mitochondria The mitochondria may be the first target in meHg-induced neurotoxicity (Yee and Choi, 1996). It is also a major site of superoxide anion and hydrogen peroxide (H2O2) production under the course of normal cellular respiration in the electron transport chain (Halliwell and Gutteridge, 1985). Mercury is suspected of generating reactive oxygen species by blocking electron transport (Sarafian, 1999), which results in the leakage of electrons from the electron transport chain directly onto oxygen, forming superoxide anion (Halliwell and Gutteridge, 1985). In purified cultures of oligodendrocytes, astrocytes, cerebral cortical and cerebellar granular neurons, meHg inhibited mitochondrial respiration and caused increased production of reactive oxygen species (Yee and Choi, 1996). This is thought to occur specifically at the ubiquinone-cytochrome b5 step in complex 3 of the electron transport chain (Yee and Choi, 1996). MeHg has also been shown to decrease ATP production in mitochondria (Verity et at, 1975). This is believed to occur through a process called uncoupling (Verity et al., 1975). Uncoupling occurs when the permeability of the mitochondrial membrane to hydrogen ions is increased, allowing a route for the dissipation of the electrochemical gradient within, without requiring ATP synthesis (Voet et al, 1999). Studies have shown that Hg disrupts the inner mitochondrial membrane allowing an influx of Mg and K ions, thereby uncoupling oxidative phosphorylation (Nath et al, 1996; Stohs and Bagchi, 1995). Therefore, Hg appears to cause the formation of reactive oxygen species through 274 two mechanisms in mitochondria: inhibition of electron transport, and uncoupling of oxidative phosphorylation. Iron and Copper The reaction of H2O2 with transition metals such as copper (Cu+) or iron (Fe2+) results in the formation of the highly reactive hydroxyl radical (Halliwell and Gutteridge, 1985). These reactions are well known as 'Fenton reactions'. The brain, which is high in iron content, may be particularly susceptible to the formation of reactive oxygen species as a result. Typically, hydroxyl radical formation through the Fenton reaction is prevented by the sequestering of iron and copper into various metalloproteins (i.e. ferritin) in the brain (Sarafian, 1999). Distortion of the tertiary structure of these proteins can result in the release of their metal cofactors, thus providing free metal ions for the Fenton reaction (Sarafian, 1999). MeHg has demonstrated the ability to bind and distort the structure of proteins (Wang and Horisberger, 1996). LeBel et al. (1992) have shown meHg/iron-mediated reactive oxygen species generation to occur in vivo in rats. Following a single intraperitoneal injection of 5 ug/g meHg, the formation of reactive oxygen species in brain regions known to be susceptible to meHg poisoning was significantly increased over controls. Pretreatment of the rats with deferoxamine, an iron-chelator, prevented increases in reactive oxygen species generation completely (LeBel et al, 1992). Sarafian and Verity (1991) also provided evidence supporting meHg/iron-mediated neurotoxicity in vitro. This study also illustrated that deferoxamine prevented neuronal damage in suspensions of cerebellar granule neurons incubated with meHg. In addition, this study showed complete prevention of lipid peroxidation, another mechanism known to generate reactive oxygen species formation from meHg poisoning (Sarafian and Verity, 1991). 275 L i p i d Peroxidation Free radicals generated by meHg can cause lipid peroxidation, a chain reaction formation of free radicals from peroxidation of unsaturated lipids that are very prevalent in the nervous system (Halliwell and Gutteridge, 1985). Vitamin E, an antioxidant used to protect cell membranes, has been shown to protect against meHg toxicity in vivo (Chang et al, 1978). Increases in thiobarbituric acid-reactive substances (TBARS), which are indicative of lipid peroxidation, were also shown in vitro by Sarafian and Verity (1991). Lipid peroxidation is not believed to be responsible for cell death, but could be a factor in the chronic neurotoxicity of meHg (Sarafian, 1999). Glutathione Glutathione (GSH) functions in cellular defense against xenobiotic compounds such as meHg by binding with them to prevent damage to cells. It also serves to maintain protein sulphydryl groups in a reduced state, to ensure proper functioning of proteins and enzymes within the cell. Glutathione peroxidase further functions to scavenge hydrogen peroxide, thereby decreasing the potential for the formation of hydroxyl radicals (Sarafian, 1999). Mercury binding to GSH could decrease the amount of freely circulating GSH. In humans, meHg preferentially binds to GSH during circulation (Rabenstein et al., 1982). Hussain et al. (1997) demonstrated a dose-dependent reduction in glutathione peroxidase levels in the cerebellum and brain stem of rats dosed with varying levels of mercuric chloride. It should be noted that the majority of studies looking at GSH levels as a result of Hg exposure were either in vitro studies, or studies that focused on an organ other than the brain (i.e. kidney or liver). Additional studies examining the in vivo effects of meHg on GSH levels are warranted, especially since neuronal levels of GSH are generally low, and neurons appear to be reliant on astrocytes for GSH production (Yudkoff et al, 1990; Aschner et al, 2000). 276 Phospholipase A2 Phospholipase A2 is an important enzyme in the repair of damaged cell membranes (Voet et al, 1999). MeHg has been shown to increase the activity of this enzyme in vitro in cerebellar granule neurons and liposomes (Verity, 1994). However, it should be noted that a cause and effect relationship has not been shown for meHg-induced neurotoxicity and phopholipase A2 induction (Verity et al, 1994). Cytoxicity resulting from meHg exposure was not mitigated when mepacrine, a phospholipase A2 inhibitor, was added to the cells. When this pathway is activated, reaction of phospholipase A2 with membrane lipids generates arachidonic acid. Arachidonic acid is further acted upon by cyclooxygenase and lipoxygenase enzymes, which generate superoxide anion during the breakdown process (Sarafian, 1999). Therefore, this pathway could play an important role in further reactive oxygen species generation following the onset of lipid peroxidation. In addition, since increased calcium levels activate this pathway, phospholipase A2 stimulation could play an important role in reactive oxygen species formation when linked to meHg-induced excitotoxicity. Calcium Increased levels of calcium resulting from meHg exposure have been implicated in decreased cellular viability (Marty and Atchison, 1998). Increased calcium has also been shown to augment oxidative stress (Pascoe and Reed, 1989). As already mentioned, calcium can activate phospholipase A2, which in turn, results in superoxide anion production during membrane repair. Another potentially very important aspect of increased calcium levels is the conversion of xanthine dehydrogenase to xanthine oxidase, which is involved in purine catabolism (Nishino, 1994). As shown in Figure 3, this enzyme produces H2O2 at each step involved in the breakdown of hypoxanthine to uric acid. Production of reactive oxygen species through this pathway may be of particular importance with decreased ATP production in mitochondria resulting from 277 meHg exposure. An increase in AMP may result with decreased ATP production, with a subsequent increase in the use of this pathway. Figure 3: Purine breakdown and the involvement of xanthine oxidase in the production of H2O2 (source: Voet et al, 1999) Nli, H 20 -N R i b — ® AMP nucleotidase AMP deaminase H,0 NUT-Adenosine adenosine deaminase Rib—(P) IMP nucleotidase Inosine .0 y H,0 P, O' N H H.,0 P, N R i b — © , XMP nucleotidase NH j P, -Ribose-l=P —A punne nucleoside Xanthpsine P; phosphorylase Ribo.,^.? (PNP) xanthine oxidase Hypoxanthine —-=^ - • Xanthine purine nucleoside phosphorylase (PNP) 0 2 + H 2 0 H 2 0 2 0 2 + H 2 0 H 2 0 2 H . N xantliihe_ oxidase ^ = 0 H Uric acid N \ H Another important enzyme activated by increased calcium is nitric oxide synthase (NOS) (Dawson, 1995). NOS converts arginine to citrulline and NO, and NO is then used as a neurotransmitter in the central nervous system (CNS) (Voet et al, 1999). The breakdown of NO occurs as follows: 278 NO + 0 2" + H + > ONOOH > OH + N 0 2 » As shown, the breakdown of NO results in the formation of peroxynitrite, which then further degrades to produce hydroxyl radicals. Chronic exposure to meHg in rats has been shown to increase NOS stimulation greater than 160% of controls (Himi et al, 1996). There are several mechanisms by which meHg may be increasing cytoplasmic calcium concentrations (Kauppinen et al, 1989). One specific mechanism, the stimulation of ionotropic glutamate receptors, may be particularly important in meHg-induced neurotoxicity. This mechanism will be discussed further in the following section. EXCITOTOXICITY While oxidative stress appears to have many causes as a result of meHg-poisoning, the primary cause of excitotoxicity is excessive glutamate present at the synaptic cleft. Glutamate is described as an excitatory amino acid (EAA), and is one of the most prominent excitatory neurotransmitters in the central nervous system (Aschner et al, 2000). Glutamate stimulates N-methyl-D-aspartate (NMDA), kainic acid (KA), and alpha-amino-3-hydroxy-5-methyl-4-isoxasole-proprionic acid (AMPA) receptors on the receiving neuron, which mitigate their effects through the opening of ion channels (Coyle and Puttfarcken, 1993). These receptors are thus termed ionotropic receptors. Metabotropic receptors also exist, which exert their effects through G-proteins, but these receptors do not appear to be involved in the neurotoxicity of excess glutamate (Coyle and Puttfarcken, 1993). As mentioned earlier, calcium activation of certain enzymes such as xanthine oxidase, can result in the formation of reactive oxygen species. This can be a particularly important link between oxidative stress and excitotoxicity. NMDA receptor stimulation results in an influx of calcium ions into the neuron as shown in Figure 4. Heteromers of 279 AMPA and K A receptors are also able to form calcium channels upon glutamate stimulation (Coyle and Puttfarcken, 1993). If overstimulation of these receptors were to occur from Hg intoxication, neuronal cytosolic calcium concentrations could increase to levels sufficient to induce reactive oxygen species formation through the many calcium-dependent pathways (Coyle and Puttfarcken, 1993). This theory is supported by evidence that cells which are typically undamaged after meHg exposure (i.e. Purkinje cells) express few N M D A receptors whereas those that are prone to meHg damage (i.e. cerebellar granular neurons) have a much higher density of N M D A receptors (Monaghan and Cotman, 1985). Figure 4: Glutamate receptors and their possible connection to the promotion of oxidative stress in neurons KA/AMPA Voltage sensitive ' NMDA receptor Na + /K + ATPase Receptor channels Source: Coyle and Puttfarcken, 1993. 280 When synaptic transmission is complete, glutamate is removed from the synaptic cleft by transporters on astrocytes and neurons rather than being metabolized by extracellular enzymes (Trotti et al, 1998). There are currently 5 known glutamate transporters: GLT1 and GLAST, which are primarily located on astrocytes, and EAAT3, EAAT4 and EAAT5, which are predominantly neuronal (Rothstein et al, 1994). Mercury appears to prevent proper removal of glutamate from the synaptic cleft, and several mechanisms have been proposed. Each of these possible mechanisms will be covered in detail in this section. Mercury and Glutamate Uptake by Astrocytes Mercury tends to accumulate in astrocytes following in vivo exposure (Oyake et al, 1966). Hg has further been shown to prevent glutamate uptake by astrocytes in vitro (Albrecht et al, 1993). Chronic in vivo exposure to meHg has also been associated with astrocyte swelling (Oyake et al, 1966). When astrocytes swell, they undergo regulatory volume decrease to re-establish a normal volume. By doing so, they release intracellular glutamate, in addition to other EAAs and ions (Pasantes-Morales and Schousboe, 1988), thereby increasing the concentration of glutamate outside the astrocyte, which could lead to excitotoxicity. MeHg has been shown to cause astrocyte swelling and glutamate release in vitro (Aschner et al, 1998a; 1998b). In addition to causing a release of excess glutamate through regulatory volume decrease of astrocytes, Hg may interfere with the initial uptake of glutamate by astrocytes. Although the mechanism is unknown, it is well known that Hg has a high affinity for sulfhydryl groups (Hughes, 1957). GLT1 and GLAST contain several cysteine residues that could be targets for Hg (Trotti et al, 1998). This may in turn, distort the tertiary structure of these transporters and cause the loss of their ability to transport glutamate. Mercury may indirectly cause a loss of transporter function through the induction of reactive oxygen species formation, which then act on the transporters. EAAT3, GLT1 281 and GLAST are all redox sensitive (Trotti et al, 1998). SDS-page experiments have shown that covalent bonds can form between transporters, which would alter their ability to transport glutamate (Trotti et al, 1998). In addition, oxidation of sulfydryl groups to disulphide bonds in transporters could cause a conformational change that reduces glutamate transport to a minimum. This was shown when dithionitrobenzoic acid (DTNB) and dithiothreitol (DTT) were added to cultured rat cortical astrocytes, and their ability to take up glutamate was measured. In the oxidized state (DTNB), very little glutamate was taken up by the astrocytes, whereas transporters worked at maximal capacity in the reduced state (DTT - free sulfydryl groups) (Trotti et al, 1998). Thus, reactive oxygen species generated by Hg may play a role in transporter dysfunction. Astrocytes and Glutathione Synthesis Glutathione was previously mentioned as an important factor in the protection against reactive oxygen species and other xenobiotic compounds. It is believed that GSH synthesis in neurons is dependent upon cysteine or cysteine-glycine generation by astrocytes (Aschner et al, 2000). This is because cysteine is oxidized to cystine in the extracellular fluid of the CNS (Wade and Brady, 1981), and neurons are unable to transport cystine across their membranes for GSH synthesis (Bannai and Tateishi, 1986). Therefore, the presence of GSH in neurons is completely dependent on cysteine of astrocytic origin. Cystine transfer into astrocytes occurs through a transporter, which uses intracellular glutamate as a counter-transport agent (Bannai, 1986). Therefore, increases in extracellular glutamate levels may inhibit cystine transport into astrocytes (Aschner et al, 2000). This could then lead to lowered glutathione levels in neurons due to a lack of astrocyte-generated cysteine. Aschner et al. (2000) also suggest that meHg may interact directly with the transporter itself, altering tertiary structure and preventing proper function. 282 CONCLUSIONS There are several ways by which Hg can cause neurotoxicity. In this report, I have attempted to combine two mainstream theories into one. It does not make sense to explore oxidative stress and excitotoxicity as two separate mechanisms of Hg-induced neurotoxicity when there could be much overlap between the two. This is especially true given that many of the above-mentioned mechanisms of both oxidative stress and excitotoxicity have been shown to occur in vivo, which further supports the notion of generating a single unifying theory. At this point, no single mechanism can be used to explain Hg-induced neurotoxicity. In addition, the delay between Hg intoxication and neuronal cell death has yet to be accounted for. It is obvious that further studies will be required to elucidate additional mechanisms involved in the neurotoxicology of Hg. Future studies should focus specifically on those mechanisms occurring as a result of chronic in vivo meHg exposure as this is the most common route of Hg exposure in organisms today. REFERENCES Amin-Zaki, L., Elhassani, S., Majeed, M . A., Clarkson, T. W., Doherty, R. A., and Greenwood, M . R. (1974). Intra-uterine methylmercury poisoning in Iraq. Pediatrics 54, 587-595. Albrecht, J., Talbot, M . , Kimelberg, H. K., and Aschner, M . (1993). The role of sulfhydryl groups and calcium in the mercuric chloride-induced inhibition of glutamate uptake in rat primary astrocyte cultures. Brain Res. 607, 249-254. Aschner, M . , Vitarella, D., Allen, J. W., Conklin, D. R., and Cowan, K. S. (1998a). Methylmercury-induced astrocytic swelling is associated with activation of the Na+/H+ antiporter, and is fully reversed by amiloride. Brain Res. 799, 207-214. Aschner, M . , Vitarella, D., Allen, J. W., Conklin, D. R., and Cowan, K. S. (1998b). Methylmercury-induced inhibition of regulatory volume decrease in astrocytes: Characterization of osmoregulator efflux and its reversal by amiloride. Brain Res. 811, 133-142. Aschner, M . , Yao, C. P., Allen, J. W., and Tan, K. H. (2000). Methylmercury alters glutamate transport in astrocytes. Neurochem. Int. 37, 199-206. 283 Bakir, F., Damluji, S. F., Amin-Zaki, L., Murtadha, M . , Khalidi, A., Al-Rawi, N . Y., Tikriti, S., Dahir, H. I., Clarkson, T. W., Smith, J. C , and Doherty, R. A. (1973). Methylmercury poisoning in Iraq. Science 181, 230-241. Bannai, S. (1986). Exchange of cystine and glutamate across plasma membrane of human fibroblasts. J. Biol. Chem. 267, 2256-2263. Bannai, S., and Tateishi, N . (1986). Role of membrane transport in metabolism and function of glutathione in mammals. J. Membrane Biol. 89, 1-8. Chang, L. W., Gilbert, M . , and Sprecher, J. (1978). Modification of methylmercury neurotoxicity by vitamin E. Environ. Res. 77, 356-366. Clarkson, T. W. (1987). Metal toxicity in the central nervous system. Environ. Health Perspect. 75, 59-64. Coyle, J. T., and Puttfarcken, P. (1993). Oxidative stress, glutamate, and neurodegenerative diseases. Science. 262, 689-695. Dawson, V. L. (1995). Nitric oxide: Role in neurotoxicity. Clin. Exp. Pharmacol. Physiol. 22, 305-308. Greenwood, M . R. (1985). Methylmercury poisoning in Iraq. An epidemiological study of the 1971-1972 outbreak. J. Appl. Toxicol. 5, 148-159. Halliwell, B., and Gutteridge, J. M . C. (1985). Oxygen radicals and the nervous system. Trends Neurosci. 8, 22-26. Himi, T., Ikeda, M . , Sato, I., Yuasa, T., and Murota, S. (1996). Purkinje cells express neuronal nitric oxide synthase after methylmercury administration. Brain Res. 718, 189-192. Hughes, W. H. (1957). A physicochemical rationale for the biological activity of mercury and its compound. Ann. N.Y. Acad. Sci. 65, 454-460. Hussain, S., Rodgers, D. A., Duhart, H. M . , and Al i , S.F. (1997). Mercuric chloride-induced reactive oxygen species and its effect on antioxidant enzymes in different regions of rat brain. J. Environ. Sci. Health. B32, 395-409. Kauppinen, R. A., Komulainen, H., and Taipale, H. (1989). Cellular mechanisms underlying the increase in cytosolic free calcium concentration induced by methylmercury in cerebrocortical synaptosomes from guinea pig. J. Pharmacol. Exp. Therapeutics 248, 1248-1254. 284 Kelman, B. J., and Sasser, L. B. (1977). Methylmercury movements across the perfused Guinea pig placenta in late gestation. Toxicol. Appl. Pharmacol. 39, 119-127. Kerper, L. E., Ballatori, N. , and Clarkson, T. W. (1992). Methylmercury transport across the blood-brain barrier by an amino acid carrier. Am. J. Physiol. 262, R761-765. LeBel, C. P., Al i , S. F., and Bondy, S. C. (1992). Deferoxamine inhibits methyl mercury-induced increases in reactive oxygen species formation in rat brain. Toxicol. Appl. Pharmacol. 112, 161-165. Magos, L. (1997). Physiology and Toxicology of Mercury. Metal Ions Biol. Syst. 34, 321-370. Marty, M . S., and Atchison, W. D. (1998). Elevations of intracellular Ca 2 + as a probable contributor to decreased viability in cerebellar granule cells following acute exposure to methylmercury. Toxicol. Appl. Pharmacol. 150, 98-105. Moller-Madsen, B. (1990). Localization of mercury in the CNS of the rat: Intraperitoneal injection of methylmercuric chloride (CH 3HgCl) and mercuric chloride (HgC^). Toxicol. Appl. Pharmacol. 103, 303-323. Monaghan, D. T., and Cotman, C. W. (1985). Distribution of NMDA-sensitive L - 3 H -glutamate binding sites in rat brain as determined by quantitative autoradiography. J. Neurosci. 5, 2909-2919. Nath, K. A., Croatt, A. J., Likely, S., Behrens, T. W., and Warden, D. (1996). Renal oxidant injury and oxidant response induced by mercury. Kidney Int. 50, 1032-1043. Nishino, T. (1994). The conversion of xanthine dehydrogenase to xanthine oxidase and the role of the enzyme in reperfusion injury. J. Biochem. 116, 1-6. Oyake, Y. , Tanaka, M . , Kubo, FL, and Cichibu, H. (1966). Neuropathological studies on organic mercury poisoning with special reference to the staining and distribution of mercury granules. Adv. Neurol. Sci. 10, 744-750. Pasantes-Morales, FL, and Schousboe, A. (1988). Volume regulation in astrocytes: A role for taurine as an osmoeffector. J. Neurosci. Res. 20, 503-509. Pascoe, G. A., and Reed, D. J. (1989). Cell calcium, vitamin E, and the thiol redox system in cytotoxicity. Free Rad. Biol. Med. 6, 209-224. Rabenstein, D. L., Isab, A. A., and Reid, R. S. (1982). A proton nuclear magnetic resonance study of the binding of methylmercury in human erythrocytes. Biochimica Biophysica Acta. 696, 53-64. 285 Rothstein, J. D., Martin, L., Levey, A. I., Dykes-Hoberg, M . , Jin, L., and Wu, D. (1994). Localization of neuronal and glial glutamate transporters. Neuron 13, 713-725. Sarafian, T. A. (1999). Methylmercury-induced generation of free radicals: Biological implications. Metal Ions Biol. Sys. 36, 415-444. Sarafian, T., and Verity, M . A. (1991). Oxidative mechanisms underlying methyl mercury neurotoxicity. Int. J. Devi. Neurosci. 9, 147-153. Sharma, A., and Talukder, G. (1987). Effects of metals on chromosomes of higher organisms. Env. Mutagenesis 9, 191-226. Stohs, S. J., and Bagchi, D. (1995). Oxidative mechanisms in the toxicity of metal ions. Free Rad. Biol. Med. 18, 321-336. Takizawa, Y. (1979). The Biogeochemistry of Mercury in the Environment. J. O. Nriagu, ed., Elsevier, Amsterdam, pp. 161-173. Trotti, D., Danbolt, N . C., and Volterra, A. (1998). Glutamate transporters are oxidant-vulnerable: A molecular link between oxidative and excitotoxic neurodegeneration? TiPS 19, 328-334. Tsubaki, T., and Irukayama, K. (1977). Minimata Disease. Elsevier, New York. Verity, M . A., Brown, W. J., and Cheung, M . (1975). Organic mercurial encephalopathy: in vivo and in vitro effects of methyl mercury on synaptosomal respiration. J. Neurochem. 25, 759-766. Verity, M . A., Sarafian, T., Pacifici, E. H., and Sevanian, A. (1994). Phospholipase A2 stimulation by methyl mercury in neuron culture. J. Neurochem. 62, 705-714. Verity, M . A. (1994). Oxidative damage and repair in the developing nervous system. Neurotoxicol. 75, 81-92. Voet, D., Voet, J. G., and Pratt, C. W. (1999). Fundamentals of Biochemistry. John Wiley & Sons, Inc., New York. Wade, L. A., and Brady, H. M . (1981). Cysteine and cystine transport at the blood-brain barrier. J. Neurochem. 37, 730-734. Wang, X. , and Horisberger, J. D. (1996). Mercury binding site on Na+/K+ - ATPase: A cysteine in the first transmembrane segment. Mol. Pharmacol. 50, 687-691. Yee, S., and Choi, B. H. (1996). Oxidative stress in neurotoxic effects of methylmercury poisoning. Neurotoxicol. 77, 17-26. 286 Yudcoff, M . , Pleasure, D., Cregar, L., Lin, Z.-P., Stern, J., and Nissim, I. (1990) Glutathione turnover in cultured astrocytes: Studies with [15N]glutamate. J. Neurochem 55, 137-145. 287