Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Physiological effects of acute exposure to thermomechanical newsprint mill effluent on adult Atlantic… Linton, Elizabeth Dawn 2003

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2003-0498.pdf [ 10.37MB ]
Metadata
JSON: 831-1.0091388.json
JSON-LD: 831-1.0091388-ld.json
RDF/XML (Pretty): 831-1.0091388-rdf.xml
RDF/JSON: 831-1.0091388-rdf.json
Turtle: 831-1.0091388-turtle.txt
N-Triples: 831-1.0091388-rdf-ntriples.txt
Original Record: 831-1.0091388-source.json
Full Text
831-1.0091388-fulltext.txt
Citation
831-1.0091388.ris

Full Text

PHYSIOLOGICAL EFFECTS OF A C U T E EXPOSURE TO T H E R M O M E C H A N I C A L NEWSPRINT MILL EFFLUENT ON A D U L T ATLANTIC S A L M O N (SALMO SALAR L ) FROM THE EXPLOITS RIVER, N E W F O U N D L A N D C A N A D A by Elizabeth Dawn Linton B.Sc, University of Guelph, 2001 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE D E G R E E OF MASTER OF SCIENCE i n THE F A C U L T Y OF G R A D U A T E STUDIES (Faculty of Agricultural Science, Animal Science Program) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A May 2003 © Elizabeth Dawn Linton, 2003 In presenting this thesis in partial fulfillment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Agricultural Science The University of British Columbia Vancouver, Canada Abstract Adult Atlantic salmon (Salmo salar L.) returning to Exploits River to spawn encounter low concentrations (< 1%) of effluent as they migrate upstream. Laboratory experiments were designed to assess various endpoints of sublethal stress in adult Atlantic salmon acutely exposed to T M P m i l l effluent. Critical swimming speed (Ucrit) and hepatic E R O D induction were determined following 12-hour exposure to 0, 6, 12, and 25% (v/v) T M P effluent. Additionally, the effects of 6-hour exposure to 0, 6, 12, and 25% (v/v) T M P effluent on cardiac output (Q ), Ucrit, hematocrit, blood glucose, and plasma Cortisol, lactate, and osmolality were examined. The same physiological endpoints were examined in another experiment in which fish were exposed to a gradually increasing concentration gradient of effluent (0-25% v/v) while swimming at a steady velocity. Relative to other treatment groups, Q during routine and low-level activity was 7-10% higher (p < 0.05) in fish exposed to 12%, 25%, and an increasing concentration gradient of effluent. Effluent concentrations up to 12% had no effect on Ucrit, Q scope, and Q m a x . However, at the 25% exposure concentration, Ucrit following 12-hour exposure was significantly depressed relative to control fish. Fish exposed to 25% effluent for 6 hours had a distinctly lower Ucrit, Q scope, and Q m a x than fish in other treatment groups, although these changes only approached statistical significance. These results suggest that fish responded to T M P effluent by reducing metabolically costly activities such as swimming to allow for increased maintenance and repair costs associated with effluent exposure. There were few noteworthy changes in any of the other endpoints examined. The findings of this study suggest that free-swimming adult Atlantic salmon in the Exploits River probably do not suffer from sublethal physiological stress as a result of exposure to T M P effluent in the river. Riverine effluent concentrations are approximately 1% and effluent levels as high as 12% and 25% were needed to cause physiological changes in the present study. This is the first study to illustrate that Q is a sensitive indicator of sublethal xenobiotic-induced stress and has potential for use as a biological monitoring tool. Table of Contents i i Abstract i i i Table of Contents iv List o f Tables - ; v List o f Figures v i i Acknowledgements ix Dedication .' *" Chapter 1: Watershed and Theoretical Background Introduction 1 The Exploits River as a Model System for this Study. .2 Metabolic Demands of Pollutants 7 Indirect Methods of Determining Metabolism 12 Using Cardiac Output in Automated Biological Monitoring 18 Project Overview 24 Chapter 2: Physiological Effects of Acute Exposure to Thermomechanical Newsprint M i l l Effluent on Adult Atlantic Salmon (Salmo salar) from the Exploits River, Newfoundland Introduction.. . 26 Materials and Methods... 28 Results 44 Discussion 83 Conclusions 108 Chapter 3: New Information Supporting the use of Cardiac Output as a Biological Monitoring Tool 110 Bibliography. 115 Appendix I: Effluent Characteristics for Various Sample Locations and Dates 134 List of Tables Table 2.1 Characteristics of whole thermomechanical newsprint effluent following primary and secondary treatment 32 Table 2.2 Hematological profile of wi ld adult Atlantic salmon from the Exploits River, N L , prior to the commencement of surgery. 47 Table 2.3 Comparison of routine, pre-surgery and post-Experiment 1 and 2 hematological variables 49 Table 2.4 Comparison of critical swimming speed (Ucrit), maximum cardiac output (Q max), and post-Ucrit blood lactate concentrations in various fish species..95 iv List of Figures Figure 1.1 Overview of the Exploits River, N L , Canada, watershed 4 Figure 1.2 Adult Atlantic salmon returns to the Exploits River, N L . Enumerations performed at Bishops Falls and Grand Falls fishway's from 1960 to the present date , 5 Figure 2.1 Calibration curve for the 115 L Blazka-style respirometer used in the present study 36 Figure 2.2 Critical swimming speed (U c rit) o f adult Atlantic salmon exposed to 0, 6, 12, or 25% (v/v) T M P effluent following 12-hour exposure to matched effluent concentrations.: 45 Figure 2.3 Liver E R O D activity in adult Atlantic salmon following approximately 15-hour exposure to 0, 6, 12, or 25% (v/v) T M P effluent 46 Figure 2.4 Routine cardiac output (Q ) , heart rate ( /H), and stroke volume (Vs) in adult Atlantic salmon during 6-hour exposure to 0, 6, 12 or 25% (v/v) T M P effluent 54 Figure 2.5 Pooled exposure values of routine cardiac output (Q ), heart rate ( /H) , and stroke volume (Vs) in adult Atlantic salmon during 6-hour exposure to 0, 6, 12, or 25% (v/v) T M P effluent 56 Figure 2.6 Critical swimming speed (U c rit) o f adult Atlantic salmon exposed to 0, 6, 12, or 25% (v/v) T M P effluent following 6-hour exposure to matched effluent concentrations 57 Figure 2.7 Cardiac output ( Q ) , heart rate ( /H), and stroke volume (Vs) in adult Atlantic salmon during a U c r i t challenge in 0, 6, 12 or 25% (v/v) T M P effluent. Swimming challenges immediately followed a 6-hour exposure period to matched effluent concentrations. 58 Figure 2.8 Scope for increase in cardiac output (Q ) , heart rate ( /H) , and stroke volume (Vs) in adult Atlantic salmon during a critical swimming challenge in 0, 6, 12, or 25% (v/v) T M P effluent. Swimming challenges immediately followed a 6-hour exposure period to matched effluent concentrations 60 Figure 2.9 Plasma Cortisol concentrations in adult Atlantic salmon following 6-hour exposure and a UCrit challenge in 0, 6, 12 or 25% (v/v) T M P effluent 62 Figure 2.10 Blood glucose concentrations in adult Atlantic salmon following 6-hour exposure and a U c r i t challenge in 0, 6, 12 or 25% (v/v) T M P effluent. 63 v Figure 2.11 Plasma lactate concentrations in adult Atlantic salmon following 6-hour exposure and a Ucrit challenge in 0, 6, 12 or 25% (v/v) T M P effluent 64 Figure 2.12 Plasma ion profile in adult Atlantic salmon following 6-hour exposure and a U c r i t challenge in 0, 6, 12 or 25% (v/v) T M P effluent 65 Figure 2.13 Cardiac output (Q ), heart rate ( / H ) , and stroke volume (Vs) in adult Atlantic salmon during 7 hours in freshwater or 6-hour exposure to an increasing concentration gradient (0-25%; v/v) of T M P effluent followed by 1 hour o f recovery in freshwater ; . 68 Figure 2.14 Pooled exposure values of cardiac output ( Q ) , heart rate ( /H) , and stroke volume (Vs) in adult Atlantic salmon during 6-hour exposure in freshwater or an increasing concentration gradient (0-25%; v/v) o f T M P effluent 70 Figure 2.15 Plasma Cortisol concentrations in adult Atlantic salmon following 7 hours in freshwater or 6-hour exposure to an increasing concentration gradient (0-25%; v/v) of T M P effluent and 1 hour of recovery in freshwater.72 Figure 2.16 Blood glucose concentrations in adult Atlantic salmon following 7 hours in freshwater or 6-hour exposure to an increasing concentration gradient (0-25%; v/v) of T M P effluent and 1 hour of recovery in freshwater.73 Figure 2.17 Plasma lactate concentrations in adult Atlantic salmon following 7 hours in freshwater or 6 hour exposure to an increasing concentration gradient (0-25%; v/v) of T M P effluent and 1 hour of recovery in freshwater.74 Figure 2.18 Plasma ion profile in adult Atlantic salmon following 7 hours in freshwater or 6-hour exposure to an increasing concentration gradient (0-25%; v/v) of T M P effluentand 1 hour of recovery in freshwater. 75 Figure 2.19 Relationship between routine and maximum heart rate ( /H) and water temperature in adult Atlantic salmon affixed with a cardiac output flowprobe 78 Figure 2.20 Relationship between routine and maximum cardiac output (Q ) and heart rate (Ju) in adult Atlantic salmon affixed with a cardiac output flowprobe 80 Figure 2.21 Routine / H o f an adult Atlantic salmon affixed with a cardiac output flowprobe following the addition of 20 m L of green food colouring to a 115 L Blazka-style respirometer 82 v i Acknowledgements Thank you to the staff and the associates of the Department of Fisheries and Oceans, St. John's, N L . Particular thanks to the 'geek', Jeff Ryder, Craig Ke l ly , Curtis Pennell, Randy Decker, N e i l Ollerhead, Martha Hiscock, K i m Lawlor, Anne Mathieu, Catherine Andrews, and Dounia Hamoutene for their technical support, and of course, their ability to always keep a smile on my face. A s well , I am grateful to Chuck Bourgeois for the salmon returns data that he provided me with. Thank you as well the Ki tch Smith for his assistance with fish capture and transport, as wel l as to 'Knock ' , Art, and Doug, at the Bishops Falls fishway for being great fish-fetchers. A sincere thank you to Fred Parsons and Carl Newcombe at E R M A for allowing us to utilise the trout hatchery, fondly named 'The Hatch', as our laboratory. Graham Houze, Brian Locke, J i l l Kelland, Brian Little, Col in Rowe, and James Titus at Abi t ib i Consolidated Company of Canada were very helpful throughout this project. A s well , thank you to Kurt Gamperl and Jonathan Stecyk for sharing their time and surgical expertise. Members of my advisory committee, Tony Farrell, Trish Schulte, and Dave Scruton always provided invaluable advice and positive feedback. A particular thanks to Tony Farrell, who was dedicated to critiquing and improving this thesis. His expertise, professionalism, and commitment were very much appreciated. Fellow labmates, particularly Kristopher Hunter, Glenn Wagner, Jennifer Wilson, Jasmine Jarjour, and Jennifer Dougherty provided endless support through the duration of this study. A grammatical expert provided excellent assistance with revisions, editing, and printing. Thank you, sugar. This research would not have been possible without funding from Abi t ib i Consolidated Company of Canada, Grand Falls-Windsor, N L , division and the Natural Sciences and Engineering Research Council o f Canada (grant to R .S .M. ) . E . D . L . was supported by an Ontario Graduate scholarship for throughout a portion of this research as wel l as the Ol in fellowship from the Atlantic Salmon Federation. Dedication A s completing this work so often took me away from those who are most important, my family, it is appropriate that I can now bring something back to them. I lovingly dedicate this thesis to Rob and Jean Linton, Andrea and Geoff Jones, Elinor and Stanton Linton, Patches, and Sunny. ix Chapter 1: Watershed and Theoretical Background Introduction Salmon are of cultural, economic, commercial, and recreational importance worldwide. They can justly be termed the 'charismatic megafauna' of our freshwater fishes (Hinch, 2002). Anadramous salmon have a fascinating life history. Eggs hatch in freshwater and continue through the developmental stages of alevin, fry, parr, and smolt. Smolt migrate to sea and spend one to several years there, feeding and developing into adults. Adults return to their native river or stream to spawn and, in the case of most Pacific and some Atlantic salmon, subsequently die. During the upstream journey to their spawning grounds adult salmon do not feed, therefore energy reserves acquired at sea must be used for the energetically demanding processes of migration, sexual maturation, courtship, and spawning (Jonssen et al. 1991). Additional metabolic demands incurred through events such as predator avoidance or exposure to xenobiotics could upset the delicately balanced bioenergetics budget of the salmon. B y using the Exploits River, Newfoundland (NL) , Canada, as a model system, some of the energetic and physiological impacts of xenobiotic exposure on migrating adult salmon were able to be studied. Atlantic salmon returning to the Exploits River to spawn encounter low concentrations of fhermomechanical pulp and paper (TMP) effluent as the migrate upstream. A large-scale Atlantic salmon enhancement project was undertaken in the Exploits River, which increased the number of salmon in the river by an order of magnitude. Therefore, the Exploits River is an excellent model system for this study, as the protection of its Atlantic salmon population from anthropogenic pollutants is of the utmost importance. This salmon success story, as well as characteristics of the model system, w i l l be highlighted prior to discussing the scientific background of the present study. 1 The Exploits River as a Model System for this Study The Exploits River is by far the most noteworthy river in Newfoundland. It has even been recognized on a national scale as it was one of the various Canadian rivers featured in the Discovery Channel's 'Great Canadian River ' series (Discovery Channel, 2003). It is the longest river on the island, being approximately 246 km in length (Vavasour and Blair , 1955). Its width and depth vary from 150-450 m and 0.15-6.00 m, respectively, between tidewater and the dam at Red Indian Lake (O'Rei l ly , 1959). It also boasts the largest drainage basin on the island, being 11 300 k m 2 (Randall et al. 1989; Fig . 1.1). Due to its large size, it contains many suitable spawning and rearing habitats for Atlantic salmon. However, until the mid-1900's, salmon were largely unable access the majority of the Exploits River watershed. Prior to the initiation of watershed enhancement in the late 1950's, less than 10% of the Exploits River and its associated tributaries were accessible to Atlantic salmon. Migration throughout the watershed was blocked by a partial obstruction (a dam) at Bishops Falls and complete obstructions (impassable waterfalls) at Great Rattling Brook, a major tributary of the lower Exploits River, and Grand Falls (Fig. 1.1). A s there was great potential for Atlantic salmon proliferation in the Exploits River, one of the largest salmon enhancement projects ever undertaken in North America was initiated (Discovery Channel, 2003). Circa 1959, the area above the waterfall in Great Rattling Brook was made accessible to sea-run salmon through fishway construction and stream clearance. A s wel l , an additional fishway was constructed to facilitate passage beyond the dam at Bishops Falls. Encouraged by increased salmon returns in these areas, fisheries managers turned their attention towards the inaccessible area above Grand Falls where there were abundant spawning and rearing habitats. A stocking program was initiated in 1965 to populate the area o f the river above Grand Falls. W i l d adult Atlantic salmon from the Exploits River and its tributaries were used as broodstock. Eggs were incubated at a hatchery over the winter and tributaries of the Exploits River were stocked with fry the following spring. Stocking began in the lower tributaries of the river and progressed to the middle and upper reaches due to its continued success. However, to enable these stocked fish to return to the middle and upper reaches of the river to spawn, another fishway was needed to allow salmon to surpass both Grand Falls and the Abi t ib i Consolidated Company of Canada ( A C C C ) dam. Fishway construction began in the mid-1970's and was completed in 1992. In response to public interest in the fishway and Atlantic salmon ecology, a Salmonid Interpretation Centre was opened at the Grand Falls fishway. Additionally, a fish elevator was built at Red Indian Lake between 1985 and 1988, providing access to the uppermost reaches of the watershed. In 1993 the stocking program was terminated since the salmon population was self-sustaining, as indicated by increased salmon returns at Bishops and Grand Falls (Fig. 1.2; Fred Parsons, pers. comm.). Atlantic salmon are a significant icon in Newfoundland culture. Tourism and recreation of numerous forms involve interactions with salmon. For example, the community of Grand Falls-Windsor hosts an annual Salmon Festival complete with a concert, craft show, midway, dances, and o f course, a salmon dinner. Salmon angling in the Exploits River rivals that found throughout the rest of North America. Both independent and guided fishers congregate along riverbanks during the adult migration between mid-June and mid-September. The recreational fishery has experienced increased success since the 1992 moratorium on commercial salmon fishing in insular Newfoundland (Bourgeois et al. 1994). Nonetheless, catch limits are set at four per angler per year. A s suggested, salmon conservation is of importance in the Exploits River watershed, although such efforts face their share of challenges. Fisheries wardens report that a moderate amount of salmon poaching still occurs (unknown warden, pers. comm.). Furthermore, the watershed coexists with a golf course, urban centres, and industries. 3 Figure 1.1. Overview of the Exploits River, N L , Canada, watershed. Bishops Falls, Grand Falls, Great Rattling Brook, and Red Indian Lake are indicated. 40 ^ 30 CO o c 4 20 CD a: c o E CD CO 10 D Bishops Falls I Grand Falls 1960 II 1970 1980 Year 1990 2000 Fig . 1.2. Adult Atlantic salmon (Salmo salaf) returns in the Exploits River, N L , Canada. Enumerations performed at the Bishops Falls fishway and the Grand Falls fishway from 1960 to the present date. 5 The primary point source of xenobiotic input to the river originates from a T M P newsprint mi l l that is located near the town of Grand Falls-Windsor. Process effluent that has undergone primary and secondary treatment is continuously discharged at a single river outfall immediately below Grand Falls. The effluent is discharged approximately 40 km inland from mouth of the Bay of Exploits. A s dictated by the Pulp and Paper Effluent Regulations that were constituted in the 1992 Fisheries Act, Canadian pulp and paper mills must perform bioassays with and characterize discharged effluent. Basic effluent characteristic information (biochemical oxygen demand, suspended solid levels, p H , and conductivity) and the results of acute lethality bioassays using rainbow trout (Oncorhynchus mykiss) and a species of water flea (Daphnia magna) must be submitted to environmental authorities on a monthly basis. Comprehensive effluent characterization reports are required four times annually, and Environmental Effects Monitoring ( E E M ) studies are to be executed every three or four years (Environment Canada, 1992). Although such regulations have been vital in protecting the fauna in receiving watercourses, they do not necessarily ensure that aquatic organisms are free from sublethal stress caused by discharged effluent. Supplementary effluent testing is often conducted by environmentally conscientious mills that are interested in examining additional endpoints of toxicity in fish or other aquatic organisms. The experiments performed in this study were conducted in addition to those required by the Fisheries Act . Several endpoints of sublethal stress in adult Atlantic salmon were investigated, one being the influence of T M P effluent on cardiac output (Q = ventral aorta blood flow), an indirect method o f assessing metabolism. 6 Metabolic Demands of Pollutants Alterations to aquatic ecosystems caused by the input of pollutants can have large impacts on fish health. Such impacts can be evaluated using the endpoint o f metabolism. Metabolism is a component of the overall bioenergetics budget offish, which takes the form of the following steady-state energy equation; C = M+G+W, where the total amount of energy consumed (C), is balanced by the total amount of energy expended through metabolism (M; standard, active, and specific dynamic action), growth (G; somatic and germ cell), and waste (W; egestion and excretion) (Adams and Breck, 1990). A s illustrated by this equation, metabolism is a component of an intricately balanced energetics budget. Therefore, assessing changes in metabolism can provide insight into the effects on other energetic parameters. Since metabolism is a function o f the molecular activity within cells (Fry, 1971), alterations at the cellular level can be translated into whole organism effects (Wedemeyer et al. 1990; Widdows and Donkin, 1991). Changes in metabolism can also act as an early warning system to identify sublethal stress before it manifests into lethal or irreversible effects on individuals or populations off ish (Widdows and Donkin, 1991). Moreover, metabolic responses to a toxic agent can be deemed a 'biomarker of effect'. Endpoints that elucidate effects have much more ecological relevance than commonly used biochemical 'biomarkers of exposure', such as mixed-function oxygenase ( M F O ) activity. The ultimate significance of M F O induction is largely unknown (Kloepper-Sams and Benton, 1994). Fish physiologists, who commonly provide tools to environmental managers interested in assessing how changes in the aquatic environment impact fish health, advocate for the evaluation o f metabolic disturbances (Jain etal . 1998). 7 There are two types of factors that govern metabolic rate. Controlling, or loading, factors act at the cellular level to elevate metabolism by increasing the state of molecular activation of components within the metabolic chain. Temperature is an example of an exceptional metabolic loading factor (Fry, 1971). Alternatively, limiting factors restrict the supply or removal of metabolites, food, water, and respiratory gases from the metabolic chain, causing a depression in metabolism. Secondary limiting factors influence the exchange rate of metabolites between the organism and the environment. L o w environmental oxygen levels are an example of a metabolic limiting factor (Fry, 1971). When there are concurrent metabolic limiting and loading effects, the full potential for metabolic rate to increase w i l l be suppressed by the effects of the limiting factor. In extreme cases, death can ensue when limiting and loading factors simultaneously interact (Fry, 1971). When metabolism is loaded or limited, metabolic scope and the energy available to support other bodily functions w i l l inevitably be altered. -Xenobiotics can also act to load or limit metabolic rate. The theory behind toxicant-induced metabolic loading is best conveyed through the description of an animal's stress response. It has been proposed that the mammalian response to stressors, including xenobiotics, follows the triphasic pattern of alarm, resistance, and exhaustion (Seyle, 1973). A n analogous pattern of response has been observed in fish (Davis, 1973; M c L e a y and Brown 1975, 1979; Johansen et al. 1994). The alarm reaction is characterized by the hypersecretion of corticosteroid and catecholamine stress hormones, which trigger a host o f secondary responses. During the stage o f resistance, defence mechanisms are initiated so that homeostasis can be maintained. However, the energetic costs involved can ultimately reduce performance. The stage of exhaustion follows i f the stressor is of appropriate magnitude and duration, during which time the organism is no longer able to maintain homeostasis (Seyle 1973; M c L e a y and Brown, 1979; Johansen et al. 1994). The second phase of this stress response has the greatest impact on metabolic rate. The energetic requirements of initiating and maintaining defence mechanisms 8 and of managing internalized toxicants (ie: catabolism, neutralization, transportation, and excretion) are generally predicted to cause metabolic loading. Metabolic costs can also be incurred during the repair of toxicant-induced tissue damages (Calow, 1991). A n elevation in metabolic rate is therefore expected with increasing magnitude or duration of toxicant exposure, until the onset of irreversible pathological effects that impair metabolism itself. However, metabolic responses to toxicants are not always this straightforward, as metabolic rate has been observed to both increase and decrease with varying levels of toxicant exposure (Calow, 1991; Brodeur et al. 2001b). Alterations in metabolic rate caused by the presence o f xenobiotics can impact the overall bioenergetics budgets of fish. A s such, the steady-state energy equation previously discussed could change to; C = [M±T] + G+W, where T represents the metabolic loading or limiting imposed by the toxicant(s). Measurements o f oxygen uptake (PO2) have frequently been used to monitor the metabolic demands of pollutants. The pioneer laboratory studies of Brett (1964) used sockeye salmon (Oncorhynchus nerka) to generate the paradigm that the rate of oxygen uptake is a suitable estimate o f overall metabolism when anaerobic contributions are insignificant (Cech, 1990). Therefore, the terms metabolic rate and VO2 are customarily used interchangeably. Exposure to xenobiotics does not always elicit a change in VO2 (Janz et al. 1991; Brauner et al. 1994; Yang and Randall, 1997; Alsop et al. 1999). This fact could indicate that not all toxicants influence metabolism, or that the magnitude and duration of exposure employed were not sufficient to have an impact on metabolic rate. Alternatively, toxicant exposure may not always induce changes in VO2 because there are trade-offs in the oxygen utilization of some tissues, such that oxygen is diverted away from tissues without immediate energetic demands and towards those in need. For example, the oxygen requirements of the liver may be greater during 9 xenobiotic exposure as detoxification enzymes are activated. To compensate, there could be a reduced oxygen demand in muscle tissues and as a result, VO2 would not need to increase. This explanation suggests that metabolism may not always be a sensitive index of sublethal toxicity. Metabolic disturbances are considered to be a tertiary stress response and generally follow more rapidly manifested primary (neuroendocrine) and secondary (biochemical, physiological, immunological) stress responses (Mazeaud et al. 1977; Wedemeyer et al. 1990; Janz et al. 1991; Johansen et al. 1994). Consequently, metabolic rate is not expected to be as sensitive an indicator of stress as primary and secondary stress responses. However, the consequence of altered metabolic rate is absolutely clear, unlike that of primary and secondary responses. Noteworthy increases in VO2 have been observed in fish exposed to various xenobiotics. Routine VO2 has been documented to increase in fish exposed to bleached kraft mi l l effluent ( B K M E ; Davis, 1973), dehydroabietic acid ( D H A , a common softwood extractive; Jain et al. 1998), tetrachloroguaiacol ( T C G , a chlorinated lignin compound) at its 96-hour LC50 concentration (Brodeur et al. 2001b), pentachlorophenol (PCP, a fungicide; Holmberg and Saunders, 1979; Farrell et al. 1998), selenium (Lemly, 1993), copper (O'Hara, 1971), and acidity (Butler et al. 1992). Holmberg and Saunders (1979) also discovered that 4-day pre-treatment with P C P increased VO2 during slow-speed swimming (35 cm s"1) in American eels (Anguilla rostrata). There is limited literature pertaining to the effects of xenobiotics on VO2 during progressively intensified exercise because researchers have largely focused on the metabolic state of fish performing at their maximum capabilities. Assessing metabolic changes by measuring VO2 at the point of peak performance is somewhat complicated. For example, when a fish is swimming at its critical swimming speed (UCnt), VO2 is at, or very near, its maximum potential ( F 0 2 m a x ) and cannot theoretically increase further. I f FC^ max is depressed in toxicant-exposed fish compared to controls, information pertaining to metabolic loading and limiting can be obtained. However, i f K)2max is similar in 10 control and toxicant-exposed fish, the mechanisms affecting metabolism must be inferred by measuring other components of the bioenergetics budget equation. Short-term metabolic changes are most commonly evaluated using the endpoint of swimming performance or repeat swimming performance, which can elucidate the energy available for active metabolism. A relative depression in \JCTii, or a recovery ratio ([Ucrjt(n+i)/Ucrjt(n)]; Farrell et al. 1998; Jain et al. 1998) less than unity, suggests that the energetic demands of active metabolism cannot be met when there are concurrent toxicant-imposed metabolic demands. Indeed, Farrell et al. (1998) and Jain et al. (1998) found that there was roughly a 50% increase in routine VO2 in mature sockeye salmon pre-exposed to P C P or D H A compared to controls. However, during Ucrit challenges VO2 was similar between control and pre-exposed fish, although the swimming performance of pre-exposed fish was impaired. These results indicate that although swimming performance was impaired, the maximum capacity of the oxygen transport system was not. Therefore, the mechanism(s) that caused routine metabolic rate to increase was not switched off during swimming. Additional studies have also noted that toxicant exposure causes reductions in Ucrit (Howard, 1975; Johansen et al. 1994; Alsop et al. 1999; Ytrestoyl, et al. 2001). Growth rate is another component of the bioenegergetics budget that can be used to infer toxicant-induced metabolic loading. For example, a reduced growth rate in fish exposed to acidic water and aluminium suggests that chronic metabolic loading was occurring (Brodeur et al. 2001a). Toxicants have also been found to limit metabolism, although such findings are not always easy to interpret. Brodeur et al. (2001b) reported that at most concentrations of P C P or T C G tested, VO2 in rainbow trout remained at or declined from routine levels in exposed relative to control fish. However, T C G acted as a metabolic loading agent at the highest concentration tested. Additionally, one would expect an increase in metabolic rate following P C P exposure, as it is a known uncoupler o f oxidative phosphorylation (Brodeur et al. 2001b) and has previously 11 been observed to increase the routine metabolic rate of fish (Holmberg and Saunders, 1979; Farrell et al. 1998). Further research in this area w i l l help explain these inconsistencies. A s indicated, i f fish are metabolically taxed, energy may be reallocated within the bioenergetics budget equation. In the case of adult Atlantic salmon returning to freshwater to spawn, pollution-induced metabolic loading could have significant impacts on fitness. Since salmon do not feed during this lifestage, energy reserves acquired at sea must be used for upstream migration, sexual maturation, courtship, and spawning (Jonssen et al. 1991). If insufficient energy is available for migration and gonadal development, the timing and location of spawning, as well as fecundity and egg size could potentially be affected. If somatic growth is impaired, there could be implications to the fish in terms of predation and mate and habitat selection. It is important to determine the metabolic requirements o f Atlantic salmon in the Exploits River, as they are exposed to low concentrations of T M P effluent during their spawning migration. The next section discusses methods of assessing metabolic rate, specifically using indirect measurement techniques. Indirect Methods of Determining Metabolism Developing accurate, sensitive, and reliable methods of appraising the metabolic demands of fish is an important area of scientific exploration. Directly measuring VO2 o f fish in a respirometer is a tried and proven method of assessing metabolism, although it often only has applications in laboratory research. Portable respirometers can be used so that respirometry measurements in fish can be made at remote field locations, in natal water, and at ambient temperature and photoperiod (Farrell et al. 2003). Using a portable respirometer also eliminates the transport of fish to laboratories that may be some distance away and enables fish to be directly moved from the watercourse to the respirometer, or, held for a short-duration at the field location prior to experimentation (Lee, 2002; Farrell et al. 2003). Regardless, measuring VO2 12 requires that the fish is confined in a chamber, and this is not always possible or desirable in field studies. Thus, many studies have explored indirect methods of measuring metabolism. Physiological endpoints such as repeat swimming performance, blood glucose and lactate concentrations, ventilatory frequency, and liver and muscle glycogen levels have been used to evaluate the metabolic changes in fish under laboratory conditions (McLeay and Brown, 1979; Larsson et al. 1988; Farrell et al. 1998; Jain et al. 1998). However, these endpoints are not considered indirect methods of estimating VO2. Physiological variables that have been used to indirectly assess VO2 include Q , swimming behaviour, ventilation rate, tailbeat frequency, locomotor muscle activity, and heart rate ( / H ) . The aforementioned variables, with the exception of Q , can be monitored using biotelemetry devices, which have applications in laboratory as well as field research. Telemetry involves affixing a device (commonly called a tag) on a subject to measure or detect a function, activity, or condition. Data are then transmitted to a remote receiver or stored within the tag until later downloaded (Cooke et al. 1999). Currently, Q flowprobes still need to be hardwired to the recording equipment, although a biotelemetry tag may eventually be developed. Before there is great investment in the development of a telemetred Q tag, the sensitivity of Q to toxicants and how wel l it can estimate VO2 need to be further explored. A reliable predictor of VO2 w i l l yield a calibration curve that describes its relationship with VO2 and the fish's scope for activity in a consistent manner. Hence, this interrelationship must be linear or log linear, contain limited variability (high correlation coefficient (r2)), and be stable under all environmentally relevant conditions (Thorarensen et al. 1996a). Some studies have employed a less accurate calibration method, in which the physiological variable is not directly correlated to VO2, but instead to swimming speed. The metabolic costs of such activity can then be calibrated (Stasko and Horrall, 1976; Ross et al. 1981; Johnstone et al. 1992; 13 Kaseloo et al. 1992; Demers et al. 1996). Laboratory techniques that minimize surgical stress and allow the fish to recover to standard resting conditions prior to calibration are essential for the accurate calibration of the variable to VO2 (Armstrong, 1998). It should be noted that many researchers have found that a simple relationship does not exist between the physiological variable and VO2 and that laboratory calibration procedures require rigorous testing under a range of conditions. For example, water temperature, feeding and digestion, excitement, and changes in water quality should also be taken into account so that energetic data can be more accurately and meaningfully interpolated (Scharold et al. 1989; Sureau and Lagardere, 1991). Although these challenges surround this general area of study, there are many unique limitations to using each physiological variable to predict metabolic rate. These are outlined below. Swimming Behaviour Positional telemetry enables researchers to estimate metabolic rate by systematically tracking a fish's location at certain time intervals. Swimming speed can be calculated from such data arid metabolic costs can be estimated from historical data or calibration curves of swimming speed versus VO2 (Young et al. 1972; Briggs and Post, 1997). However, this method of estimating metabolism is flawed because: (1) small movements that could account for a significant portion of daily activity metabolism are often unresolved, (2) swimming speed can be grossly underestimated as it is assumed that a fish swims in a straight line between two measurement points, and (3) the frequency with which reliable positional fixes can be obtained is often limited (Ross et al. 1981; Lucas et al. 1991; Boisclair and Tang, 1993; Demers et al. 1996). Ventilatory Frequency Attempts have been made to correlate telemetered bioelectric potentials from opercular muscle contractions with V02, as opercular movements help force water past the respiratory 14 surface of the gills. This method of estimating metabolism assumes that oxygen uptake at the gills is directly proportional to ventilation volume and that ventilation volume is proportional to ventilatory frequency and magnitude (Rogers and Weatherly, 1983). However, extraction efficiency of oxygen from the water can also vary significantly. Additional weaknesses to using ventilatory activity to estimate VO2 are that anaerobic metabolic events and cutaneous oxygen uptake cannot be accounted for. Furthermore, at swimming speeds above 50-80 cm s"1, fish commonly employ ram ventilation of the gills, during which time rhythmic breathing movements cease (Fry, 1971; Jones and Randall, 1978). Tailbeat Frequency Tailbeat frequency has also been used to indirectly estimate VO2. There is generally a positive linear relationship between tailbeat frequency and swimming speed (Stasko and Horrall, 1976; Scharold et al. 1989), therefore the relative energetic costs of swimming at the calculated speed can be used as an indicator of metabolic demands (Stasko and Horrall , 1976; Ross et al. 1981; Johnstone et al. 1992). The main shortcoming of using this tailbeat frequency to estimate metabolic rate is that the linear relationship between swimming speed and tailbeat frequency can break down at low and high swimming velocities due to changes in stride length (amplitude of the tail thrust; Webb, 1971; Scharold et al. 1989). Aerobic Locomotor Muscle Activity Since red muscle is fuelled aerobically, electromyogram signals from red axial swimming musculature can be correlated to metabolic costs. This method of predicting VO2 does not distinguish between the contributions of red and white muscles to swimming (Weatherly et al. 1982; M c K i n l e y and Power, 1992). It is also impossible for changes in metabolism associated with the maintenance of homeostatic mechanisms, recovery from oxygen debt, feeding, or 15 environmental stress to be determined (Hinch et al. 1996; Cooke et al. 1999; Brodeur et al. 2001c). Furthermore, it is questioned whether or not changes in stride length and swimming behaviour are accurately and precisely reflected in the intensity of the electromyogram signal. Studies conducted to date indicate that this technology is a reasonable method of estimating metabolism in rainbow trout, sockeye salmon, and lake sturgeon (Acipenser fulvescens; Weatherly et al. 1982; M c K i n l e y and Power, 1992; Hinch et al. 1996; Briggs and Post, 1997). However, it is not an accurate estimate of VO2 in largemouth (Micropterus salmoides) and smallmouth bass (M. dolomieu; Demers et al. 1996) Heart Rate Heart rate ( /H) is directly related to metabolic rate, as illustrated by the Fick equation; V02 = / H • V s • £ 0 2 , where VO2 is oxygen uptake (mg kg" 1 hr"1), / H (beats min"1) was defined above, V s is stroke volume (mL kg" 1), and £ 0 2 (mg mL" 1 ) is the difference in the arterial and venous blood oxygen contents (Priede and Tytler, 1977; Webber et al. 1998; Cooke et al. 1999). Assumptions that must be taken into account when using / H to estimate metabolism include: (1) cutaneous oxygen absorption is insignificant, (2) all metabolic events are aerobically fuelled, (3) hematocrit and hemoglobin levels remain constant (Gallaugher and Farrell, 1998), and (4) V s and EO2 either remain constant or vary systematically as / H changes (Lucas, 1994; Webber et al. 1998). Another weakness of using / H to estimate VO2 is that fish generally increase Q through elevations in V s rather than / H (Randall, 1968; Stevens and Randall 1967; Kiceniuk and Jones, 1977; Farrell, 1991; Farrell and Jones, 1992; Webber et al. 1998), indicating that / H is not the variable that is most sensitive to changes in blood flow regulation. Additionally, / H is highly sensitive to environmental change (Randall, 1970; Thorarensen et al. 1996a), particularly water temperature (Armstrong, 1986; Ko lok et al. 1993; Keen and Farrell, 1994; Claireaux et al. 1995a; 16 Webber et al. 1998). For further information, Thorarensen et al. (1996a) extensively reviewed the limitations of using / H as a predictor of metabolic rate. To date, the response o f / H to xenobiotics has been studied more extensively than any of the other physiological variables discussed here, as electrocardiogram signals are relatively easy to obtain (Klaverkamp, 1982). However, many such works have not attempted to correlate / H with VO2, probably because of the numerous limitations outlined above. There are mixed reports pertaining to the utility o f fu to evaluate VO2. In rainbow trout, brown trout (Salmo trutta), Atlantic cod (Gadus morhud), and individual leopard (Triakis semifasciatd) and lemon (Negaprion brevirostris) sharks, the variation in the regression line of VO2 versus fu is too great for any meaningful calibration curve to be constructed (Priede and Tytler, 1977; Scharold et al. 1989; Scharold and Gruber, 1991). In sea bass (Dicentrarchus labrax), fu is relatively stable regardless of activity level (Sureau and Lagardere, 1991). Nonetheless, / H has shown promise as a suitable indicator of VO2 in Atlantic salmon (Lucas 1994), sole (Solea solea; Sureau and Lagardere, 1991), pike (Esox lucius; Armstrong, 1986), and cod (Claireaux et al. 1995a, 1995b). For more information, Lucas et al. (1993) reviewed the literature pertaining to the use of fu telemetry as a method of estimating metabolism in free-ranging fish. Cardiac Output Cardiac output (mL m i n k g " 1 ) is equal to the product o f / H and V s . Using Q as a predictor of metabolic rate, and alleviating some of the issues raised when using fu, has not been explored to a great extent, largely because field studies are still technically out of reach. A n advantage to using this technique to estimate VO2 over fu telemetry is that Q enables both fu and 17 V s to be estimated. Sti l l , the assumptions that EO2 is constant or changes systematically with Q , hematocrit and hemoglobin levels remain stable, and cutaneous oxygen absorption is insignificant, hold fast. A n additional drawback is that Q does not yield information pertaining to anaerobic metabolism. Upon examination of the relationship between VO2, Q , and EO2 in rainbow trout, Brodeur et al. (2001c) concluded that Q offered few advantages over / H as a predictor of metabolic rate. The contributions of EO2 and Q to overall VO2 varied with temperature and swimming activity. Consequently, more than one regression line was needed to explain their interrelationship. Additionally, the correlation between Q and VO2 was often variable and weak. Nonetheless, Q is closely correlated to metabolic rate in Atlantic cod. Even low-frequency metabolic events such as spontaneous pectoral and caudal fin beats can be detected, allowing researchers to accurately determine overall metabolic costs (Webber et al. 1998). Despite the fact that concerns have been expressed regarding the suitability of Q technology to reliably predict metabolic rate, investigations in this area of study should continue. A s illustrated, there is merit in such an application of this technology in cod, thus it is possible that Q is the physiological correlate that w i l l permit the most accurate estimates of metabolic rate in other fish species. If Q proves to be a suitable estimate of VO2 in numerous fish species, its usefulness as a biological monitoring tool w i l l be enhanced. Using Cardiac Output in Automated Biological Monitoring Automated biological monitoring (biomonitoring) is defined as a self-acting or self-regulating means of assessing water quality or chemical toxicity using l iving organisms (Gruber, 1988). It offers many advantages over more traditional methods o f assessing toxicity, such as 18 laboratory-based bioassays. Although much useful information has been generated from bioassays, such methods are not amenable to real-time monitoring or pollution prevention. Results are not available immediately, thus irreversible environmental damage could occur by the time the test results are known. Additionally, water and wastewater quality often fluctuates over time; therefore bioassays are not adequate predictors of the risk or hazard to a receiving watercourse (Diamond et al. 1988). A s a result of these limitations, it was recognized that quick, reliable, continuous, and automated methods of evaluating changes in water quality were needed. A s a result, biological early-warning systems geared at monitoring the quality of drinking water, industrial discharges, and wastewater streams were developed in the 1980's (Diamond et al., 1988; Gruber, 1988). Today, automated biomonitoring systems assess changes in fish locomotor activity and ventilatory behaviour in automated, self-regulating monitoring chambers. Such chambers are land-based, and water from the area under assessment continuously flows through them. The biological response variable(s) of interest (for example, ventilatory frequency) is detected as or converted to an electrical signal that is transmitted to a computer. Using appropriate software, the computer recognizes i f the signals are representative of 'normal' conditions. If 'abnormal' conditions are detected, it is assumed that the fish are being exposed to polluted water. Subsequently, an alarm is activated or the appropriate corrective action is taken (Diamond et al. 1988; Gruber, 1988; Biological Monitoring Inc. 2003). This type of biomonitoring has focused on assessing changes in water quality and has been less concerned with the direct effects of xenobiotics on aquatic fauna. Detected changes in locomotion and ventilatory frequency observed in land-based monitoring chambers can be extrapolated to predict ecological effects, however the fish species used for biomonitoring are often not native to the threatened watercourse. This hinders the ability to accurately predict associated effects in resident fish species. In addition, alterations in locomotion and ventilatory frequency in captive fish are not necessarily the best predictors of changes in fish health. The current biomonitoring 19 protocols are clearly deficient in protecting the health of wi ld fish populations that reside in the waters under examination. Using a response variable such as Q in biomonitoring applications may allow for fluctuations in water quality as well as fish health to be concurrently monitored. A s discussed, changes in Q provide insight into the metabolic demands of cod, thus Q may be a good correlate of VO2 in other fish species as well . Relating alterations in Q to metabolism allows for effects on energetic parameters as well as associated repercussions at the individual and population levels to be predicted, since information is obtained in a common currency, energy (Widdows and Donkin, 1991). Accordingly, using Q as a biomonitoring tool of xenobiotic-induced stress is theoretically reasonable, but in practice, is it suitable? Cairns and van der Schalie (1980) proposed a list o f criteria that automated biological monitoring systems should possess. With some modifications, these w i l l be used as the backbone of the following discussion of the current knowledge regarding the suitability of Q technology as a biological monitoring tool. Criterion 1. The response variable in the chosen fish species changes over a wide range of developing toxic conditions and is sensitive to the pollutant(s) being screened for. Numerous databases exist that contain toxicity information for common endpoints in fish species that are regularly used in bioassays. Hence, it is relatively simple to determine the effects certain toxicants on such species. However, i f Q is used as the response variable, it could be challenging to determine a priori i f the chosen fish species is a reliable and sensitive indicator of the pollutants being screened for. There is limited literature pertaining to the effects of pollutants on Q , let alone for multiple fish species. Furthermore, ecologically relevant ] 20 biomonitoring should employ fish species native to the watercourse under investigation, although this could limit the number of species to choose from. Unfortunately, no single species or response variable is equally sensitive to all pollutants, and concurrent multiple species monitoring is complex and costly (Diamond et al. 1988). However, i f a known toxicant is being screened, its effects on other metabolic and energetic endpoints in the appropriate fish species (ie: VO2, ventilation rate, plasma lactate levels, swimming performance, growth) could shed light on the response of Q to the toxicant. Criterion 2. Fish restraint and the direct attachment of devices to it are not recommended. To measure Q , surgical techniques are required to affix the flowprobe around the ventral aorta and suture its lead wire to the body of the fish. Unt i l a Q tag is developed, fish must be somewhat restrained in a small holding chamber or respirometer so that they do not become tangled in the lead wire o f the flowprobe. Criterion 3. The operation system should be continuous, automatic, cost effective, and require little maintenance. Computer programs have been designed to collect Q data continuously and automatically. However, periodic downloading of the data would be necessary. Land-based operation systems need regular maintenance and the initial surgical attachment o f the flowprobe requires some labour. 21 Criterion 4. Simple and appropriate methods of data analysis and results interpretation should be available. To date, no standardized methods of data analysis or results interpretation have been developed. Criterion 5. The response variable must be continuously expressed and change rapidly and reliably in response to low toxicant concentrations. Cardiac output is indeed a continuously expressed physiological variable in living organisms, provided that blood flow from the heart is not restricted. However, there is some uncertainty as to the sensitivity and reliability of this response variable to changes in water quality. Klaverkamp (1982) argued that powerful homeostatic mechanisms regulate cardiovascular function within relatively narrow limits. Although his discussion focused on / H and ventilatory frequency, it was suggested that cardiovascular responses are not sensitive to toxicant exposure. Indeed, this was the case in the study conducted by Brodeur et al. (2001a). They discovered that in Atlantic salmon, locomotive behaviour and blood glucose levels were more sensitive endpoints of exposure to acidic water and aluminium than Q . In another experiment, Brodeur et al. (2001b) observed that exposure to P C P and T C G did not alter Q in rainbow trout, although VO2 changed significantly following 24-hour exposure. Pentachlorophenol is a known uncoupler of oxidative phosphorylation (Brodeur et al. 2001b), therefore one would expect an increase in Q to follow P C P exposure. In a different study, Brodeur et al. (1999) found that Q was suitable for detecting lethal levels of aluminium and acidic water, as adult Atlantic salmon suffered from a depression in Q accompanied by concomitant elevations in / H and a lowering of V s approximately 4 hours prior to mortality. In some instances lethality may be an endpoint worthy of monitoring, however most biomonitoring 22 programs seek to detect sublethal stress responses so that irreversible effects are prevented. Based on the small body of current literature, it would be incorrect to assume that Q is an insensitive biological monitoring tool of pollution, because only a few toxicants, toxicant concentrations, and species have been studied to date. Furthermore, the sensitivity of a response variable also depends on the available technology and its detection limits. Sublethal effects may not have been observed in the experiments of Brodeur et al. (1999, 2001a, b) because Doppler flowprobes were used. These are inherently less accurate and precise than Transonic flowprobes (Farrell and Jones, 1992; Thorarensen et al. 1996b) and are technically more challenging to correctly surgically implant (A.P. Farrell, pers. comm.). Criterion 6. The normal range of the response variable under common conditions (diurnal fluctuations, temperature, changes in hardness, oxygen) should be determined prior to its use in biological monitoring to reduce the occurrence of 'false alarms'. To date, there is moderate body o f literature pertaining to the effects of fish activity, water temperature, hypoxia, hypercapnia, and p H fluctuations on Q in fish (Davie and Forester, 1980; Farrell et al. 1983, 1989, 1996; Barron et al. 1987; Farrell and Jones, 1992; Ko lok et al. 1993; Takeda, 1993; Keen and Farrell, 1994; Thorarensen et al. 1996b; Brodeur et al. 1999; Crocker et al. 2000; Perry and McKendry, 2001; Stecyk and Farrell, 2002). Nonetheless, before applying Q technology to a specific biological monitoring situation, extensive laboratory calibration over a wide range o f conditions would need to be performed with the chosen, fish species. A t a minimum, the normal range of Q in response to different levels of activity, water temperatures, and food rations would need to be evaluated. 23 The preceding discussion shed some light on the criteria that a suitable automated biomonitoring system should meet. It is evident that the use of Q technology as a biomonitoring tool is in its infancy. There has been relatively little study into the effects of toxicants on Q , thus no suitable biomonitoring applications for this technology have been discovered to date. Furthermore, a standardized protocol, monitoring chamber, recording equipment, and methods of statistical analysis and results interpretation have yet to be established. Nonetheless, further exploration into the use of Q technology in biomonitoring applications is warranted. Cardiac output may be a sensitive enough to detect changes in water quality and bring ecological relevance to automated biomonitoring by elucidating the metabolic and energetic effects of toxicants on fish. One of the objectives of the present study was to generate more information regarding the applicability and usefulness of using Q technology as a biological monitoring tool. In the final chapter of this thesis, the above criteria are revisited to highlight how the current study has supplemented the knowledge of this topic. Project Overview The experiments conducted for the requirements of this thesis examined several physiological effects of acute exposure to T M P effluent on wi ld adult Atlantic salmon from the Exploits River, N L . In the wi ld , the majority of adult salmon are exposed to low concentrations of T M P effluent at some point during their upstream migration. Roughly half of the adults that return to spawn in a given year swim directly past the effluent outfall during their upstream journey, thus they are likely exposed to higher effluent concentrations. The present experiments were conducted in the laboratory using a 115 L Blazka-style respirometer and were designed to mimic the conditions faced by adult salmon during their spawning migration in the Exploits River. The laboratory was situated beside the Exploits River, therefore transport distance of the 24 salmon to the laboratory was very short and Exploits River water of the ambient temperature was used. Two different experiments were completed, both of which examined numerous physiological indicators of sublethal stress during and following exposure to T M P effluent. These physiological endpoints included Q , swimming performance, as well as blood Cortisol, glucose, lactate, hematocrit, and osmolality. In the first experiment salmon were exposed to 0, 6, 12, or 25% (v/v) T M P effluent for 6 hours and were subsequently challenged to swim to Ucrit under identical conditions to that of exposure. The second experiment simulated an upstream swim past the outfall o f the T M P mi l l . Fish swam through an increasing concentration gradient (0-25%; v/v) of effluent at a constant velocity over a 6-hour period. This was followed by 1 hour of recovery in freshwater. These experiments tested several hypotheses (Hn): Hi: Salmon exposed to T M P effluent w i l l have a similar or elevated Q during routine and moderate activity relative to control fish H2: Salmon exposed to T M P effluent w i l l have a similar or lower UCrit than control fish H 3 : Salmon exposed to T M P effluent w i l l have similar or higher plasma lactate, Cortisol, and blood glucose levels relative to control fish H 4: The osmoregulatory status and the hematocrit levels of salmon exposed to T M P effluent may be different relative to control fish 25 Chapter 2: Physiological Effects of Acute Exposure to Thermomechanical Newsprint Mill Effluent on Adult Atlantic Salmon (Salmo salar) from the Exploits River, Newfoundland Introduction The pulp and paper industry is a widespread and important industry in Canada. Effluent from pulp and paper mills is discharged into surface waters and its detrimental effects on the receiving environment and biota have been demonstrated. Due to the high strength and yield of mechanical pulps, there was an explosive growth in thermomechanical pulping operations in North America during the early to mid-1970's (Wong et al. 1978). Today, approximately 42-46%, 9%, and 31-68% of the wood pulp produced in Canada, the United States, and Scandinavia, respectively, is generated using mechanical or thermomechanical processes (Food and Agriculture Organization, 1999; Canadian Pulp and Paper Association, 2001, 2002). Although many North American and Scandinavian mills produce large quantities of pulp using non-chemical processes,, less is known about the physiological responses of fish to mechanical pulping effluents compared to effluents, such as B K M E , from chemical pulping operations (Johnsen et al. 1995, 1998). Perhaps less research has focussed on assessing the toxicity of mechanical pulping effluents as they have proven to be less toxic than B K M E (Leach and Thakore, 1975; Kovacs et al. 1995). However, generalizations regarding the toxicity of these effluents should be made cautiously (Kovacs et al. 1995). Some of the physiological impacts observed in fish following exposure to chemical and mechanical pulps are similar. This similarity indicates that sublethal effects are partially induced by extractives in the wood, and not solely from bleaching chemicals or chlorinated pulping by-products (Johnsen et al. 1995, 1998). The toxicity of mechanical and chemimechanical pulps is largely due to naturally occurring wood extractives leached from the wood during the pulping process, such as resin acids, 26 juvabione and its derivatives, unsaturated fatty acids, alcohols, diterpene aldehydes, and manool (Leach and Thakore, 1976; Kovacs and Voss, 1992). There are roughly ten major resin acids, which account for 60-90% of the toxicity o f mechanical pulping effluents. Juvabione, a neutral compound, can account for up to 30% of the toxicity of mechanical pulping effluents (Leach and Thakore, 1976). Toxicity studies using mechanical or chemimechanical pulping effluents have been conducted by a handful o f scientists in North America and Scandinavia. Endpoints under examination in fish have included liver metabolism, function, histology, proximate analysis, and hepatic mixed function oxygenase ( M F O ) induction, resin acid bioaccumulation, growth, reproduction, hematology, and acute lethalty. Alga l growth, bacterial fluorescence (Microtox test ®), as well as Ceriodaphnia mortality and reproduction have also been explored (Gibbons et al. 1992; Kovacs and Voss, 1992; O'Connor et al. 1992; Martel et al. 1994; Kovacs et al. 1995; Johnsen et al. 1995, 1998; Martel et al. 1997; Martel and Kovacs, 1997). Other research has examined how D H A , one of the most abundant and persistent softwood extractives (Brownlee et al. 1977), and other wood extractives, alter fish hematology, liver and muscle enzyme activities, organ mass, and respiratory capabilities (Iwama et al. 1976; Kruzynski , 1979; Nikinmaa and Oikari, 1982; Oikari et al. 1983; Tana et al. 1988). The present work aimed to expand on the present literature concerning the acute toxicity of whole T M P effluent and to assess its toxicity using various endpoints of sublethal stress that are indicative of primary, secondary, and tertiary stress responses. Cortisol levels were examined as its secretion is a primary response to stress. Hematocrit, blood, glucose, plasma lactate, and plasma osmolality were the secondary stress endpoints inspected. Two tertiary stress responses that involve the interactions of many body systems, namely swimming performance and Q , were also measured (Seyle, 1973; Mazeaud et al. 1977; Wedemeyer et al. 1990; Johansen 27 et al. 1994). Three known studies have investigated the effects of xenobiotics on Q in fish. Cardiac output has not proven to be sensitive to aluminum and acidic conditions (Brodeur et al. 1999, 2001a), P C P , or T C G (Brodeur et al. 2001b). This is the first study to examine the effects of whole T M P effluent on Q in any fish species. Materials and Methods Preliminary Experiments (2001) Preliminary experiments were conducted during the summer of 2001. Methodology was identical to that of the core experiments that were conducted in 2002 (outlined below), with the following exceptions. Experiments were performed at the Fisheries and Oceans facility in St. John's, N L . Adult Atlantic salmon from the Exploits River were transported 430 km to St. John's in aerated, temperature-controlled, 400 L holding tanks. Average (± S E M ) body weight and fork length of the fish were 1.47 ± 0.06 kg and 54.3 ± 0.7 cm, respectively. Fish were held for at least 6 weeks before being used in experimentation. Effluent was shipped overnight from the mi l l in Grand Falls-Windsor, N L , and stored for up to 7 days at 4 °C until used. Fish were equipped with a Doppler flowprobe (Iowa Doppler Products, Iowa City, U S A ; 20 Hz) that was hardwired to a directional pulsed Doppler flowmeter (Department of Bioengineering, University of Iowa, Iowa C i t y , U S A ; No : 545C-4). Salmon were exposed to 0, 6, 12, and 25% (v/v) T M P effluent diluted with dechlorinated surface water for 12 hours, and subsequently challenged to swim to Ucrit under identical conditions to that of exposure. Dissolved oxygen was maintained at 8.90 ± 0.14 mg L" 1 and water temperature .was 14.8 ± 0.1 °C throughout experimentation (mean ± S E M ) . Due to the inherent inaccuracies of and challenges using Doppler flowprobes, Q data 28 from preliminary experiments are not reported. Flowprobes from Transonic Systems Inc. were used to measure Q in subsequent 2002 experiments. Immediately following the termination of the swimming challenge approximately 3 g of tissue from the anterior lobe of the liver was flash frozen in liquid nitrogen and stored at -65 °C until its ethoxyresorufin 0-deethylase (EROD) activity was measured. Liver E R O D activity was determined following the methods of Porter et al. (1989) with some minor modifications. The S9 fraction was obtained by partially thawing the tissue then homogenising it in (1:4 w/v) ice-cold Tris-sucrose buffer (50 m M , p H 7.5) using a glass Potter-Elvehjem homogenizer. The reaction mixture was incubated at 27 °C for 15 minutes and stopped by the addition of 2.5 m L room temperature HPLC-grade methanol. Protein was precipitated from the reaction tubes by centrifugation at 3600 x g for 5 minutes and then the supernatant was analysed for E R O D activity. The protein content of matched S9 fractions was measured according to Lowry et al. (1951), so that 1 unit of E R O D activity could be expressed as the production of 1 pmol resorufin min" 1 mg protein"1. Average length of each Ucrit challenge was approximately 3 hours, therefore E R O D activity was measured after about 15 hours of exposure to T M P effluent. Sample sizes in the 0, 6, 12, and 25% exposure groups were n = 8, 4, 4 ,4 for the Ucrit challenge and n = 6, 3, 3,4 for E R O D determination, respectively. Mill and Effluent Characteristics Effluent was obtained from a thermomechanical pulp and paper m i l l that is located near the town of Grand Falls-Windsor, N L . This mi l l uses 90-95% black spruce (Picea mariana) and 5-10% balsam fir (Abies balsamed) to produce roughly 557 finished tonnes of standard newsprint per day. High brightness and coloured speciality grade newsprint have also been manufactured (Portt and Associates, 1999; J. Titus, pers. comm.). Sodium hydrosulphite is used to adjust paper brightness. For specialty paper grades, high brightness clay or violet dye are used 29 as bleaching agents. Petroleum-based oi l products including lubricating, hydraulic, and Bunker " C " o i l are also used on machinery at this m i l l ( B . A . R . Environmental, 1996; Portt and Associates, 1999). A l l process effluent undergoes primary clarification and secondary treatment. Primary clarification removes settled sludge and suspended solids from the effluent. There are two primary clarifiers; one treats effluent from the paper m i l l and the other treats effluent from the wood chipping room. The average retention time of effluent in each primary clarifier is 4 hours. Following primary clarification, all effluent is combined, fortified with sodium hydroxide (caustic), urea, and diammonium phosphate, and deposited into the secondary treatment aerated stabilization basin (ASB) . B y adding these supplements and intensely aerating the effluent, the p H , nutrient content, and oxygen levels are increased. Additionally, bacteria that are naturally present in the A S B promote the breakdown of effluent constituents. The retention time of effluent in the A S B is approximately 10 days (J; Kelland, pers comm.). Effluent treatment facilities operate, and effluent is discharged at the river outfall following primary and secondary treatment, 24 hours a day, 365 days a year. The m i l l itself operates 24 hours a day, 360 days a year, with scheduled bi-annual shut-downs on Labour Day weekend and over Christmas. However, unscheduled shut-downs also occur. Effluent used in experimentation was identical to that discharged into the Exploits River by the T M P mi l l . The mi l l was operating under normal conditions and producing standard newsprint during the time of experiments. Grab samples of effluent from the outfall o f the A S B were collected roughly 1 hour before being used in experimentation in 20 L plastic pails. The length of time that effluent samples were held prior to use in each trial, and used during each trial, was closely regulated. This practice helped to minimize and standardize the potential breakdown o f effluent constituents between trials. The majority o f effluent samples were partially characterized using a Horiba U-22 multi-parameter water quality monitor (Enviro-30 Equipment, Inc. Pineville, N C , U S A ) . The chemical and biological oxygen demands ( C O D and B O D , respectively) of most effluent samples were analyzed by personnel at the mi l l . Complete characterization of effluent samples collected from the outfall o f the A S B on M a y 16 and September 26, 2002, was undertaken by an independent laboratory. Since experiments were conducted between these dates, effluent data obtained on both dates are reported when possible. Characteristics of effluent used in the present study are outlined in Table 2.1. See Appendix I for complete effluent characteristic data for various effluent samples taken in 2001 and 2002. Riverine Effluent Levels A dam at Grand falls regulates a relatively stable river discharge throughout the year 3 1 3 1 (166-200 m s"). Nonetheless, river discharge increases to 234-359 m s" during spring runoff between A p r i l and June. Average effluent discharge at the river outfall is 0.495 m 3 s"1 (Portt and Associates, 2000). Therefore, the initial dilution of effluent in the river ranges from 0.14-0.30% (v/v). Approximately 2.5 km downstream from the outfall, water conductivity measurements have verified that effluent is completely mixed with river water and that effluent concentrations are between 0.14-0.30% (v/v; Portt and Associates, 1999). Likewise, plume delineation using rhodamine W T tracer indicates that effluent entering the river is diluted to concentrations below 1% (v/v) within a few kilometres downstream of the outfall ( B . A . R . Environmental, 1996). Rapid mixing of the effluent is expected as the area of the river immediately downstream of the outfall is turbulent. 3 1 Table 2.1. Characteristics of whole thermomechanical newsprint m i l l effluent that has undergone primary (clarification) and secondary (aerated stabilization basin) treatment. Effluent Parameter Mean" (± SEM) or Value(s)b pH 7.03 ± 0.02 Conductivity (uS cm"') 463.4 ± 6 . 4 Turbidity (NTU) 108.2 ± 12:0 Dissolved Oxygen (mg L"') 6.52 ±0 .19 Total Dissolved Solids (g L"1) 2.05 ± 1.75 Total Suspended Solids (g L"1) 0.017 ±0.004 Oxidation Reduction Potential (mV) 208.4 ± 7.4 Biological Oxygen Demand (mg L"1) 86.6 ± 13.7 Chemical Oxygen Demand (mg L" 1) 404.6 ±53 .3 Colour (CTU) 190 Sodium (mg L"1) 82.9 Potassium (mg L"1) 14.2 Calcium (mg L"1) 11.0 Magnesium (mg L~') 2.2 Chloride (mg L" 1) 7.0 Alkalinity (mg L"1) 74.0 Sulphate (mg L" 1) 160 Reactive Silica (mg L"1) 1.8 Orthophosphate (mg L ' 1 ) 0.47 Phosphorus (mg L"1) 1.6 Hardness (CaC0 3 ; mg L"1) 36.5 Bicarbonate (mg L" 1) 74.0 Phenolics (mg L"1) 0.001 Sulphide (mg L"1) 0.06 Aluminum (fig L"1) 170-320 Barium (fig L"1) 99-170 Boron (fig L"1) 18.0 Chromium (fig L"1) 2.0 Copper (fig L"1) 8.0 Iron (fig L"1) 530-800 Lead (fig L"') 0.7 Manganese (jug L"1) 2000-3500 Strontium (fig L"1) 38-67 Thallium (fig L"1) 1.6 Titanium (fig L"1) 5.0 Uranium (fig L"1) 1.1-2.8 Vandium (jtig L"1) 3.0 Zinc (fig L"') 40-74 Abietic acid (mg L"') 0.042 Chlorodehydroabietic acid (mg L"1) 0.013 Isopimaric acid (mg L~') 0.048 Neoabietic acid (mg L"1) 0.015 Palmitric acid (mg L"') 0.012-0.355 Palmitoleic (mg L"1) 0.007 Pimaric acid (mg L ' 1 ) 0.007 n = 15 effluent samples that were used in the present study h tests conducted on M a y 15 and September 26, 2002 by an independent analytical laboratory. Results are not reported i f tests weren't attempted or i f values were below the detection limit of the instrument. See Appendix I for details. 32 Experimental Animals and Housing Adult Atlantic salmon were collected from the Exploits River, N L , (49 °N, 57 °W) at the holding chamber above the Bishops Falls fishway between June 25 and August 9, 2002 (Licence No: NF-0293-02). Fish were immediately transported 20 km in a 250 L holding tank (<6 per tank) to the laboratory in Grand Falls-Windsor, N L , in aerated Exploits River water. Transport water was fortified with 0.3% (w/w) deiodized salt to prophylactically treat Saprolegnia spp. infections. Fish were maintained at the laboratory in 4000 L holding tanks under flow-through conditions of aerated Exploits River water. Tanks were divided and no more than six fish were housed on either side. Water replenished each holding tank at a rate o f 90 L min" 1 , creating a gentle current that flowed at roughly 25 cm s"1 at the outflow. The intake pipe that supplied water to the entire laboratory was located upstream of the pulp and paper mi l l . Ambient temperature water was employed for holding (17.8 ± 0.2 °C; mean ± S E M ) and experimentation (see below). Migrating adult salmon in the Exploits River encounter similar temperatures and the facility in which the experiments were conducted was not equipped to control the temperature of large volumes of incoming water. Fish were held in the laboratory for 2-8 days before being used in experimentation and were not fed while in captivity. Mean (± S E M ) body weight and fork length were 1.66 ± 0.05 kg and 55.5 ± 0.5 cm, respectively. Respirometer Characteristics A 115 L Blazka-style respirometer (Blazka et al. 1960) was used in all trials. The respirometer consisted of concentric outer and inner tubes, 34 and 24 cm in cross-sectional diameter, respectively. The length of the swimming area was 100 cm. However, this length was reduced to 80 cm when an extended screen, used to keep the fish out of the turbulent area at the front of the swimming chamber, was inserted. External disturbances were minimized by surrounding all but 30 cm at the posterior end of the respirometer with black plastic sheeting. 33 The respirometer was connected in series with a 200 L Nalgene ® header bath (Nalgene Labware, Rochester, N Y , U S A ) and an electrical chilling unit. The water or the effluent solution in the header bath was continually aerated. Calibration of the tachometer' s digital display to water velocity was performed using a flowmeter (Marsh-McBirney Inc., Frederick, M D , U S A ; Flow-mate No : 2000). The head of the flowmeter was attached to a bracket made from 2.5 cm P V C pipe and this was tightly aligned along the vertical axis in the centre o f the swimming chamber. The lead cable from the flowmeter head was fed out the front of the respirometer so that the majority of the cable and the digital velocity display unit were outside of the resiprometer. Veloci ty was measured in the centre o f the vertical axis and 6 cm above and below this position. A t all flow settings, water velocity in the centre of the swimming chamber was 9-32 % lower than that at the outer positions. Water velocity at the upper and lower positions corresponded wel l (differed by 0-12%; F ig . 2.1). Overall, this flow profile is far from optimal. Uniform cross-sectional flow through the section of the respirometer that holds the fish is required for accurate performance and respirometry measurements (Fry, 1971). This heterogeneous flow profile can be attributed to a weakness in the design of the respirometer's removable end or cap (K. Gamperl, pers. comm.). The inside of the cap was shaped like a concave ring that gradually extended upwards in the centre to a circular point that had been levelled off. When the cap was secured to the respirometer, the raised circular centre area (approximately 6 cm in diameter) lay directly against the extended screen that was inserted to keep the fish out of the turbulent area at the front of the swimming chamber. This blocked water flow and caused the water velocity in the centre of the swimming chamber to be slower than that measured at the two outer positions. A s this weakness in the design o f the respirometer's cap was not amended, the calibration curve for the respirometer was created with this variable flow profile in mind. A n additional consideration in the creation of the calibration curve was that fish generally swam a few centimetres away from 34 the upper, but more often the lower, wall o f the swimming chamber. The maximum body height of experimental fish was 9.84 ± 0.16 cm (mean ± S E M ) and cross-sectional radius of the swimming chamber was 12 cm. Therefore, the body of the fish l ikely passed through, or at least came very close to, the centre of the swimming chamber and one outer position where velocity had been measured. A s a result, a calibration curve was created using the average of all three velocities that were measured along the vertical plane (Fig. 2.1). 35 • centre v 6 cm above centre 0 20 40 60 80 100 120 140 160 180 Tachometer digital display Fig . 2.1. Calibration curve for the 115 L Blazka-type respirometer used in this study. The digital readout of the respirometer's tachometer was calibrated to water velocity. Velocity was measured using a flowmeter that was attached to a bracket which was securely aligned along the vertical axis in the centre of the swimming chamber. Water velocity was measured at three locations along vertical plane (centre and 6 cm above and below centre). Equation of the line-of-best-fit is y = 0.67x, r 2 = 0.98. 36 Hematology A 3 m L blood sample was taken from each fish via caudal puncture prior to surgery and following each trial. A blood sample was taken before surgery so that information about the routine hematological profile of wi ld adult Atlantic salmon could be obtained. Blood was sampled using 3 m L Luer-Lok ™ syringes equipped with a 20 x 1XA gauge Precision Glide® needle (Becton Dickson; Franklin Lakes, N J , U S A ) . Syringes were inverted a minimum of twenty times immediately prior to blood analysis. Glucose was measured using an Advantage AccuSoft®blood glucose monitoring system (Roche Diagnostics, Laval , P Q , Can.; No : 8179254664). Hematocrit was analysed in heparinised micro-hematocrit capillary tubes ( V W R International Mississauga, O N , Can.) after centrifugation (International Equipment, Needham Heights, M A , U S A ; No : 42831081; 5 min., high speed). Whole blood was transferred to 1.5 m L Eppendorf vials and plasma was separated from cellular components using the same centrifuge (5 min., high speed). Following centrifugation, plasma was transferred using a pipette into new 1.5 m L Eppendorf vials and immediately frozen (-20 °C) until analysed. Plasma osmolality and lactate levels were determined using a Stat9 Profile blood analyser (No: 00038215; Nova Biomedicals, Waltham, M A , U S A ) . Using specific electrodes, this blood analyzer quantifies blood constituents by exploiting the capability of each to create an electric potential either in its initial state or following an enzymatic reaction (Brodeur et al. 2001a). Plasma Cortisol was measured by radioimmunoassay using a Diasorin Cortisol kit (Oxoid Inc., Napean, O N , Can.). A l l hematological variables were determined for every fish. Individual values of each variable were pooled within each treatment group for analysis purposes. Flowprobe Surgery and Data Acquisition Fish were dip netted from their holding tank and immediately anesthetised in an aqueous solution of 60 ppm clove oil until opercular movements were irregular and there was no 37 resistance to a caudal peduncle grab. Each fish was placed on its left side on a wetted sponge surface and its gills were irrigated with a solution of 30 ppm clove o i l for the duration of the surgery. A 3 m L blood sample was promptly taken and total and fork lengths as well as maximum body height and girth were measured. Subsequently, the gills and the operculum were gently retracted using lA inch umbilical tape and the thin layers of connective tissue covering the ventral aorta were teased away. A single flowprobe (Transonic Systems, Ithaca, N Y , U S A ; 2.5 mm P S L ) was used for all but one trial where the ventral aorta of the fish was quite small, then a 2.0 mm P S L flowprobe was employed. The flowprobe was secured around ventral aorta and its lead wire was sutured inside the opercular cavity, under the right pectoral fin, and on the right side of the dorsal fin using 3-0 silk suture (Johnson and Johnson Medical Products, Peterborough, O N , Can.). A l l surgeries were completed within 18.1 ± 0.7 minutes (mean ± S E M ) . , Following surgery fish were immediately transferred to the respirometer to begin an overnight recovery and acclimation period. During this time, water velocity was set at 0.2 b l s"1 and there was partial water replenishment (7 L min"1) to the header bath. The lead wire from the flowprobe was fed through a hole atop the respirometer then up through a 65 cm standpipe. It was then connected to a transit time perivascular flowmeter (Transonic Systems, Ithaca, N Y , U S A ; Model : T402 A20023), which was hardwired to a laptop ( I B M Thinkpad) equipped with a P C M C I A data acquisition (DAQ) card (National Instruments, Austin, T X , U S A ; No: 183569A-01) and Lab View® software (National Instruments, Austin, T X , U S A ) . This equipment was used to determine Q , / H , and V s . Preconditioning Swim Critical swimming challenges conducted in 2001 demonstrated the restless nature of wi ld adult Atlantic salmon in the confinement of the swimming chamber. Therefore, each morning 38 fish were subjected to a short-duration preconditioning Ucrit challenge (increment = 0.2 bl s"1; interval = 5 minutes) (Jain et al. 1998). The swimming challenge terminated at 1.4 b l s"1, 75% of the average Ucri t that was determined from preliminary experiments. Fish were recovered at 0.2 bl s"1 until Q stabilized at basal levels (1.0-3.5 hours). A t this time, experimentation commenced. During the preconditioning swim and the recovery period the respirometer operated as a partial flow-through system, as there was water replenishment (7 L min"1) to the header bath. Experiment 1 - Acute Exposure and Ucrit Challenge Fish were exposed to 0, 6, 12, or 25% (v/v) T M P effluent (n = 4, 6, 5, 5, respectively) for 6 hours. Six hours was estimated to be the length of time that migrating adult Atlantic salmon in the Exploits River are exposed to the greatest effluent concentration. This value was determined by taking the product of the following data; the average fork length of adult salmon in the Exploits River ( -55 cm; determined from preliminary studies), the distance the fish swim through the greatest concentration of effluent (2 km; B . A . R . Environmental, 1996), and the slowest sustained cruising speed of migrating fish (0.2 b l s"1; Beamish, 1978). A 1-hour uncertainty factor was incorporated, yielding an exposure duration of 6 hours. Water velocity was maintained at 0.2 b l s"1 for the entire exposure period. Following exposure, fish were individually challenged to swim to Ucrjt (increment = 0.2 bl s"1; interval = 20 min) until fatigued in identical conditions to that o f exposure. Critical swimming speed was calculated using the fork length of individual fish and the equation proposed by Brett (1964); Ucrit = V + [(ti/tii)-v], where V is the highest velocity maintained for the designated period (bl s"1), tj is the time elapsed at the final velocity (min), tjj is the time increment (20 min), and v is the velocity increment (0.2 39 bl s"1). The procedure used to determine fatigue involved stimulating individuals whose caudal fin was resting against the posterior retaining screen using rapid changes in water velocity. Fish were considered fatigued when they failed to swim away from the downstream retaining screen after three consecutive stimulation attempts. Swimming challenges for all fish were performed at approximately the same time each afternoon to circumvent possible diel differences in exercise performance. Brett (1964, 1965) found that there was little difference in swimming performance between male and female salmonids, therefore fish were not sexually differentiated. A l l swimming speeds and U c r i t values were corrected for the solid blocking effect due to flow acceleration along the animal's body. Velocity values that were used to calculate U c r j t were corrected using the method of Smit et al. (1971) that was recommended by Beamish (1978); V c = V m (1 + Afi sh/ACylinder), where V c is the corrected velocity (bl s"1), V m is the current velocity in the absence of the fish (bl s"1), Afish is the cross sectional area o f the fish (AfjS r i = [7T (0.5 max. body height)(0.5 max. body 2 ' 2 width)]; cm ), and Asunder is the cross sectional area of the of the swimming chamber (452 cm ). The mean (± S E M ) blocking effects caused by the body of the fish were 9.70 ± 0.32% and 10.5 ± 0.30% in 2001 and 2002, respectively. Cardiac variables were recorded for five minutes immediately before exposure and for the final five minutes o f every exposure hour and U c r i t interval. During all trials, freshwater replenishment to the header bath was terminated. Hence, the respirometer, header bath, and chilling unit operated as a recirculating system in which temperature and oxygen levels were controlled. Dissolved oxygen and ambient water temperature were maintained at 9.40 ± 0.19 mg L" 1 and 17.5 ± 0.3 °C (mean ± S E M ) , respectively, throughout all trials. 40 Experiment 2 - Upstream Simulation A n upstream riverine migration past the effluent outfall o f a T M P m i l l was simulated by increasing the concentration of effluent in the respirometer as the fish swam at a constant velocity. Beamish (1978) reviewed various sources of literature and deduced that the mean sustained cruising speed of migrating fish ranges from 0.2-0.5 b l s"1. Preliminary experiments illustrated that adult Atlantic salmon did not swim continuously in the respirometer until water velocity reached 0.6 b l s"1. Therefore, water velocity in the respirometer was set at 0.6 b l s"1 for the duration of the experiment. Fish in the exposure group (n= 4) swam in fresh water for 20 minutes and cardiac variables were recorded for the final 10 minutes. Subsequently, fish were exposed to 6, 12, and 25% (v/v).effluent for 2 hours each, followed by a 1-hour recovery period in freshwater. Cardiac variables were recorded for the final 10 minutes o f each hour of exposure and recovery. During exposure and recovery freshwater replenishment to the respirometer was terminated and the respirometer, header bath, and chilling unit operated as a recirculating system in which temperature and oxygen levels were controlled. Fish in the control group (n = 4) were subjected to an identical protocol, except freshwater was recirculated through the respirometer, header bath, and chilling unit. Dissolved oxygen and ambient water temperature were maintained between 8.82 ± 0 . 1 9 mg L" 1 and 18.2 ± 0.4 °C (mean ± S E M ) , respectively. Experiments were performed at approximately the same time each day. Post-Experimentation Following all trials, 60 ppm clove o i l was injected into the respirometer to anaesthetise the fish and minimize handling stress. Upon removal from the respirometer a 3 m L blood sample was taken, the flowprobe removed, and the fish was weighed. Control fish and those exposed to the lowest concentration of effluent were recovered for at least 48 hours in the 41 laboratory before being released into the Exploits River atop the Bishops Falls fishway. A l l other experimental fish were euthanized in an aqueous solution of 90 ppm clove oi l . Replication of Environmental Conditions and Justification of Exposure Concentrations These experiments were designed to somewhat mimic the conditions faced by adult Atlantic salmon migrating up the Exploits River, N L . For example, Exploits River water was used, fish were subjected to the natural Newfoundland summer photoperiod, and the exposure duration and the lowest effluent concentration tested (6% v/v) were comparable to what migrating salmon could be exposed to in the river. Although effluent concentrations in the river seldom exceed 1% ( B . A . R . Environmental, 1996), the 6% exposure concentration may have ecological relevance in the Exploits River at times of low river and/or high effluent discharges. Furthermore, effluent concentrations may be in the range of 6% within a few metres of the effluent outfall. Concentrations as high as 12 and 25% (v/v) were experimented with to assess the dose-response effects of acute exposure to T M P effluent as wel l as the potential for cardiac output telemetry to be used as a biomonitoring tool of pollution stress. Data Analysis For each experiment, routine, pre-surgery hematological variables were compared to post-experimental hematological data that were pooled over all treatment groups using a two-tailed, paired samples t-test. Comparisons were made between pooled post-Experiment 1 and post-Experiment 2 hematological data using a two-tailed t-test. Cardiac data from fish having a routine / H greater than 60 beats min" 1 were considered to be outliers (justification in results and discussions) and data from these fish were not included in analysis. Cardiac output, / H , and V s were examined for significance differences between treatment groups, exposure times, and swimming speeds using a G L M univariate test for repeated measures. During exposure at a 42 constant swimming speed, pre-exposure values of each cardiac variable were applied as the covariate i f a significant effect was present. For the analysis of cardiac data during the U c r i t challenges, the value of each cardiac variable at the sixth hour of exposure was used as the covariate when a significant effect was present. A s there were no significant changes in any cardiac variable over time in either experiment (when swimming speed was constant), data for each treatment group were pooled. For pooled values of each cardiac variable, treatment groups were examined for homogeneity o f variance then for significant differences using a one-way A N O V A followed by Sidak's adjustment for multiple comparisons when significant interaction occurred. This method o f analysis was also used to determine i f there were any treatment related effects on liver E R O D activity, U c r i t , scope for increase in Q , / H , and V s during exercise (max. -i routineo.2 bi s"1)* and maximum values of Q , / H , and V s . In the analysis of maximum values of, and the scope for increase in Q , / H , and V s , U c r i t was applied as a covariate i f a significant effect was present. For the analysis of U c rjt data, the length of time a fish was held prior to experimentation as well as fork length were applied as covariates i f significant effects were present. One U c r i t value for a 2001 control fish was eliminated from data analysis based on the positive result of a discordancy test for an upper outlier in a univariate sample (Barnett and Lewis, 1978). Hematological endpoints were examined for homogeneity of variance then for significant covariate effects over length, weight, and U c r j t and subsequently analyzed using a one-way A N O V A (Expt. 1), a one-tailed t-test (Expt. 2 Cortisol, glucose, and lactate levels), or a two-tailed t-test (Expt. 2 ion and hematocrit levels). Covariables were applied and noted when a significant effect was present. Critical swimming speeds in control 2001 and 2002 groups were also tested for homogeneity of variance then analyzed using a two-tailed t-test. Bivariate correlation analysis was used to determine i f there was a significant regression between routine / H and water temperature and routine and maximum / H and Q . Statistical power (1-/3) has been reported when applicable and values > 0.80 (80%) were used to indicate sufficient statistical power, which largely ruled out the probability of a type II error (failure to reject H 0 ) , had been achieved. Statistical analyses were performed using the SPSS 10.0 statistical package and a 0.05 level of significance and Sigma Plot 2000 was used to create all graphical figures (SPSS Science, Chicago IL, U S A ) . A l l values are presented within the text, graphically, or in table form are reported as mean ± standard error of the mean (SEM). Results Preliminary Experiments (2001) The critical swimming speed o f salmon exposed to 25% T M P effluent for 12 hours was significantly depressed relative to the control group (p = 0.009; 1-/3 = 0.87; Fig . 2.2). Fork length had a significant covariate effect on U c r i t (p = 0.011). There were no significant changes in liver E R O D induction over the range of exposure concentrations tested (p = 0.794; 1-/3 = 0.10; Fig . 2.3). Heart rate data aquired using Doppler flowprobes is reported below to illustrate the influenced of water temperature because experimental temperatures were different in 2001 and 2002. Pre-Surgery Hematological Profile of Atlantic Salmon The routine hematological profile of wi ld adult Atlantic salmon from the Exploits River was determined through the analysis of blood samples taken from anaesthetized fish prior to surgery (Table 2.2). 44 5 0% 6% 12% 25% Effluent concentration (v/v %) Figure 2.2. Critical swimming speed (Ucrit) of adult Atlantic salmon exposed to 0, 6,12, or 25% (v/v; n = 8, 4 ,4 , 4, respectively) primary and secondary treated T M P effluent immediately following 12-hour exposure to matched effluent concentrations. Comparisons between treatment groups were made using a one-way A N O V A (observed power (1-/3) = 0.87) followed by Sidak's adjustment for multiple comparisons, t indicates a significant difference from the control group (p = 0.009). Fork length exhibited a significant covariate effect on U c r i t (p = 0.011). Fork length was used to calculate relative swimming speeds in body lengths per second (bl s"1). Bars are ± S E M . 45 0% 6% 12% 25% Effluent concentration (v/v %) Figure 2.3. Liver E R O D activity of adult Atlantic salmon following approximately 15 hours of exposure to 0, 6, 12, or 25% (v/v; n = 6, 3, 3, 4, respectively) primary and secondary treated T M P effluent. During approximately the final 3 hours of exposure all fish were subjected to a critical swimming challenge. Comparisons in E R O D activity between treatment groups were made using a one-way A N O V A (p = 0.794). Observed power (1-/3) = 0.10. Bars are ± S E M . 46 Table. 2.2. Routine hematological profile of wi ld adult Atlantic salmon from the Exploits River, N L , Canada. Fish were captured from their holding tank using a dip net and transferred to an anesthetic bath (60 ppm clove oil). When fully anaesthetized (~ 4 min), yet before the commencement of surgery, a 3 m L blood sample was taken via caudal puncture. Hematological Var iab le M e a n (± S E M ) Hematocrit (%) 47.6 ± 1.07a Glucose (mmol L" 1 ) 5 . 8 2 ± 0 . 3 3 a Plasma Cortisol (ng mL" 1 ) 6 1 . 3 ± 7 . 9 6 a Plasma lactate (mmol L" 1 ) 3 . 4 4 ± 0 . 1 5 a Plasma N a + (mmol L" 1) 156.8 ± 0 . 6 8 a Plasma K + (mmol L" 1) 1 . 2 2 ± 0 . 0 3 5 b Plasma C a 2 + (mmol L" 1) 1 . 1 6 ± 0 . 0 4 0 a Plasma CI" (mmol L" 1 ) 135.3 ± 0 . 7 2 a a n = 33 samples b n = 10 samples (23 samples below detection limit o f instrument, < 1.0 mmol L" 1 ) 47 Comparison of Routine, Pre-Surgery and Post-Experimental Hematological Variables Hematocrit levels did not change from routine, pre-surgery levels following either experiment nor were there differences in post-Experiment 1 or post-Experiment 2 hematocrit levels (Table 2.3). Blood glucose increased significantly from routine, pre-surgery levels following both experiments (Table 2.3). Post-Experiment 1 glucose levels were higher than post-Experiment 2 levels, although this increase was not quite statistically significant (p = 0.099). In both experiments, post-experiment plasma Cortisol levels increased significantly from routine levels (Table 2.3). Cortisol levels post-Experiment 1 were significantly elevated from post-Experiment 2 levels. Lactate levels following Experiment 2 were not significantly different from pre-surgery levels (Table 2.3). However, post-Experiment 1 lactate levels were significantly higher than both pre-surgery and post-Experiment 2 levels. The changes between routine and post-Experiment 1 and 2 plasma ion levels were similar between the two experiments (Table 2.3). Relative to routine levels, N a + significantly decreased and C a 2 + remained unchanged following both experiments. Potassium levels increased significantly from routine, pre-surgery levels following Experiment 1. Following Experiment 2, K + levels also increased relative to pre-surgery concentrations, although this change was not quite statistically significant (p = 0.194). Similarly, C l " levels significantly decreased from routine levels following Experiment 1, while post-Experiment 2 C l " levels were almost significantly lower than routine, pre-surgery levels (p = 0.064). Post-Experiment 1 and 2 concentrations of N a + , K + , C a 2 + , and Cl" did not differ significantly. 48 Table 2.3. Comparison of routine, pre-surgery and post-experimental hematological variables measured in Experiments 1 and 2. A l l routine, pre-surgery blood samples were taken via cadaul puncture in aneasthetized fish prior to the commencement of flowprobe surgery. Post-Experiment 2 blood samples were taken via caudal puncture from aneasthetized fish that had previously been exposed to an increasing concentration gradient of T M P effluent (0-25% v/v) for 6 hours followed by 1 hour of recovery in freshwater. Water velocity was set at 0.6 bl s"1 for the duration of Experiment 2. Post-Experiment 1 blood samples were taken via caudal puncture from aneasthetized fish that had been exposed to 0, 6, 12, or 25% (v/v) T M P effluent for 6 hours and were swum to U c r i t in matched effluent concentrations. Data from all treatment groups in each experiment were pooled for the determination of post-experimental levels of each hematological variable. Hematological Variable Pre-Surgery Post-Expt. 2a Pre-Surgery Post-Expt. 1 Expt. 2 a Expt. l c (Post-Ucrit)c Hematocrit (%) 46.6 ± 1.61 45.8 ± 1.52 47.7 ± 1.73 46.0 ± 1.57 Glucose (mmol L"1) 5.65 ±0.70 10.0± 1.38* 5.72 ±0.48 12.5 ±0.76* Plasma Cortisol (ng mL"1) 64.2 ± 12.9 199.0±21.0* 65.2 ± 11.6 368.0 ±27 .4* t Plasma lactate (mmol L ' 1 ) 3.18 ±0.34 2.81 ±0.29 3.58 ±0.20 5.92 ±0.42*1 Plasma N a + (mmol L"1) 156.4 ± 1.02 151.8 ± 1.88* 156.6 ±0 .95 151.4 ± 1.42* Plasma K + (mmol L"1) 1.19 ±0.029" 2.72 ± 0.98b 1.23 ± 0.083d 2.80±0.20 d * Plasma C a 2 + (mmol L"1) 1.17 ± 0.059 1.15 ±0.093 1.13 ±0.052 1.21 ±0.048 Plasma Cl" (mmol L"1) 136.1 ±0.91 132.6 ± 1.62 134.4 ± 1.01 129.57 ± 1.51* a n=8 (treatment groups pooled) b n=5 (3 values below the detection l imit 'of instrument, < 1.0 mmol L" 1 ) c n=20 (all treatment groups pooled) d n=4 (16 values below the detection limit of instrument, < 1.0 mmol L" 1 ) * indicates a significant difference from routine levels of the same experiment (p <0.034) t indicates a significant difference between Post-Expt. 1 and Post-Expt. 2 levels (p <0.001) Experiment 1 - Acute Exposure to TMP Effluent and Critical Swimming Challenge in 2002 After adjusting for pre-exposure levels of each cardiac variable (Q , / H , and Vs , p < 0.001), Q almost varied significantly among treatment groups (p = 0.051; 1-/3 = 0.63), but / H (p = 0.357; 1-/3 = 0.25) and V s (p = 0.193; 1-/3 = 0.37) did not (Fig. 2.4). Compared to other treatment groups, average / H in the 12% exposure class was slightly depressed, yet V s was concurrently elevated, thus allowing Q to be regulated within a narrow range (Fig. 2.4). Within each treatment group, there were no significant changes in any cardiac variable over the 6-hour exposure period (p > 0.05). Therefore, exposure data for each treatment group were pooled. After correcting for each covariable (Q , / H , and V s , p < 0.001), pooled Q in the 12% (18.1 ± 0.2 m L min" 1 kg"1) and 25% (17.8 ± 0.3 m L min" 1 kg"1) exposure groups was significantly elevated over Q in the control (16.4 ± 0.2 m L min" 1 kg"1) and 6% groups (16.6 ± 0.2 m L min" 1 kg" 1; p < 0.003 for all significant differences; 1-/3 = 1.0; Fig . 2.5). This translates to a 7-10% increase in routine Q in fish exposed to 12 and 25% effluent relative to those exposed to 0 and 6% effluent. These elevations in Q were achieved through significant increases in V s in all exposure groups (0.41 ± 0.01, 0.42 ± 0.01, 0.44 ± 0.01 m L kg" 1 for 6, 12, and 25%, respectively) compared to control fish (0.37 ± 0.01 m L kg" 1; p <0.001 for all significant differences; 1-/3 = 1.0). There were also significant reductions in / H in the 6% (42.0 ± 1.0 beats min" 1; p = 0.015) and 25% (41.6 ± 1.1 beats min" 1; p = 0.010) exposure groups compared to control fish (46.68 ± 1.09 beats min" 1). Heart rate in the 12% group was similar to control fish (43.3 ± 1 . 2 beats min" 1; p = 0.178; 1-/3 = 0.86). There was nearly a significant decrease in U c r i t between the 25% exposure group relative to the other treatment groups (p = 0.067; 1-/3 = 0.58) after correcting for the covariable, fish 50 holding time (p = 0.017; Fig . 2.6). Fork length did not elicit a significant covariate effect on U c rjt (p = 0.603). Values of each cardiac variable at the end of the exposure period (prior to initiation of the swimming challenge) demonstrated significant covariate effects (Q , p = 0.003; fu, p = 0.001; V s , p = 0.008). After adjusting each cardiac variable for its covariable, no statistical differences were present between treatment groups at any swimming speed (p = 0.491, 0.351, 0.631 and 1-/3 = 0.19, 0.28, 0.14 for Q , fu, and V s , respectively; F ig . 2.7). Significant increases in each cardiac variable were observed in the majority of progressive increases in swimming speed. This was expected as fish increase Q through elevations in V s and fu during exercise (Randall, 1968; Farrell and Jones, 1992). N o significant differences were apparent between treatment groups for maximum values of Q (p = 0.069; 1-/3 = 0.57), / H ( p = 0.756; 1-/3 = 0.11), or V s (p = 0.342; 1-/3 = 0.26) that were achieved at, or very near, U c r j t (Fig. 2.7). Similarly, there were no statistically significant differences between treatment groups in the scope for increase in Q (p = 0.079; 1-/3 = 0.55), fu (p = 0.425; 1-/3 = 0.22), or V s (p = 0.643; 1-/3 = 0.14) during exercise (Fig. 2.8). However, with respect to both Q scope and Q max, the p-value's are near the 5% level of significance and the statistical power of each test is low, suggesting that a type II error may have occurred. Certainly, one would expect that the relative 29-36% reduction in Q s c o p e observed in the 25% exposure group has biological significance. Critical swimming speed was not included as a covariate in the analysis of cardiac scope data (p = 0.303, 0.225, and 0.331 for Q , fu, and Vs , respectively) or maximum Q and fu data (p = 0.176 and 0.265, respectively). There was a significant covariate effect of U c r i t on Vsm a x (p = 0.038). 51 Plasma Cortisol (p = 0.418; 1-/3 = 0.22; Fig . 2.9), blood glucose (p = 0.869; 1-/3 = 0.09; Fig. 2.10), and plasma lactate (p = 0.381; 1-/3 = 0.24; Fig . 2.11) levels did not differ significantly between treatment groups following 6-hour exposure and the U c r i t challenge. However, post-U c r i t plasma lactate levels increased subtly in direct proportion with effluent concentration. No changes in plasma K + (p = 0.245; 1-/3 = 0.33), C a 2 + (p = 0.726; l-B = 0.12), N a + (p = 0.293; 1-/3 = 0.29) or CT (p = 0.189; 1-/3 = 0.38) resulted from exposure to T M P effluent (Fig. 2.12). Weight and length exhibited significant covariate effects in the analysis of C a 2 + levels (p = 0.031 and 0.032, respectively). Furthermore, hematocrit was similar between the control (51.18 ± 1.94%), 6% (46.08 ± 1.77%), 12% (41.57 ± 3.50%) and 25% (46.36 ± 4.18%) treatment groups (p = 0.253; 1-/3 = 0.32). 52 Figure 2.4. Cardiac output ( Q ) , heart rate ( / H ) , and stroke volume (Vs) in adult Atlantic salmon during 6-hour exposure to 0, 6, 12 or 25% (v/v; n = 4, 6, 5, 5, respectively) primary and secondary treated T M P effluent. A Transonic flowprobe was affixed around the ventral aorta of individual fish. Within treatment comparisons over time and comparisons between treatment groups at each exposure hour were performed using a G L M univariate test with repeated measures. Between treatment effects for Q , / H , and V s were p = 0.051, 0.357, and 0.193, respectively. Observed power (1-/3) of between treatment comparisons for Q , / H , and V s = 0.63, 0.25, and 0.37, respectively. * indicates that pre-exposure levels exhibited a significant covariate effect (p < 0.001). Plots are ± S E M . 53 Exposure time (h) Figure 2.5. Pooled exposure values of cardiac output (Q ), heart rate ( / H ) , and stroke volume (Vs) in adult Atlantic salmon during 6-hour exposure to 0, 6, 12, or 25% (v/v; n = 4, 6, 5, 5, respectively) primary and secondary treated T M P effluent. A Transonic flowprobe was affixed around the ventral aorta of individual fish. For all treatment groups each cardiac variable was adjusted for the average pre-exposure value, which served as the covariable (p < 0.001 for each). Significant differences between treatment groups were examined using a one-way A N O V A , followed by Sidak's adjustment for multiple comparisons. | indicates a significant difference from the control group (p <0.015). J indicates a significant difference from the 6% group (p < 0.003). Observed power (1-/3) of between treatment comparisons for Q , / H , and V s = 1.0, 0.86, and 1.0, respectively. Bars are ± S E M . 55 ' y t j -X t X 0.2 V 0 % 6% 12% 25% Effluent concentration (v/v %) 0% 6% 12% 25% Effluent concentration (v/v %) Fig. 2.6. Critical swimming speed ( U c r i t ) of adult Atlantic salmon exposed to 0, 6, 12, or 25% (v/v; n = 4, 6, 5, 5, respectively) primary and secondary treated T M P effluent following 6-hour exposure to matched effluent concentrations. Values are adjusted for the covariable, fish holding time (p = 0.017). Comparisons between treatment groups were made using a one-way A N O V A (p = 0.067). Observed power (1-/3) = 0.58. Fork length was used to calculate relative swimming speeds in body lengths per second (bl s"1). Bars are ± S E M . 57 Figure 2.7. Cardiac output (Q ), heart rate ( / H ) , and stroke volume (Vs) in adult Atlantic salmon during a critical swimming challenge in 0, 6, 12 or 25% (v/v; n = 4, 6, 5, 5, respectively) primary and secondary treated T M P effluent. Swimming challenges followed a 6-hour exposure period to matched effluent concentrations. A Transonic flowprobe was affixed around the ventral aorta of individual fish. Within and between treatment comparisons were performed using a G L M univariate test with repeated measures. Between treatment effects were p = 0.491, 0.351, 0.631 and for Q , / H , and V s , respectively. Observed power (1-/3) of between treatment comparisons for Q , / H , and V s = 0.19, 0.28 and, 0.14 respectively. * indicates that the cardiac variable at 0.2 bl s"1 exhibited a significant covariate effect (p <0.008). Plots are of bidirectional error bars ± S E M . 58 Figure 2.8. Scope for increase (max. - routineo.2 bi s~') in cardiac output (Q scope), heart rate (/Hscope), and stroke volume (Vs S C Ope) in adult Atlantic salmon during a critical swimming challenge in 0, 6, 12 or 25% (v/v; n = 4, 6, 5, 5, respectively) primary and secondary treated T M P effluent. Swimming challenges immediately followed a 6-hour exposure period to matched effluent concentrations. A Transonic flowprobe was affixed around the ventral aorta of individual fish. Differences between treatment groups were assessed using a one-way A N O V A (p = 0.079, 0.425, and 0.643 for Q , fu, and V s , respectively). Observed power (1-/3) = 0.55, 0.22, and 0.14 for Q , fu, and V s , respectively. Bars are ± S E M . 60 50 40 . £ 30 e X 0% 6% 12% 25% Effluent concentration (v/v %) 500 H 400 H E O) 300 -O 5 2 0 0 " 100 H 0 % 6% 12% 2 5 % Effluent concentration (v/v %) Figure 2.9. Plasma Cortisol concentration in adult Atlantic salmon following 6-hour exposure and a U c rj t challenge in 0, 6, 12 or 25% (v/v; n = 4, 6, 5, 5, respectively) primary and secondary treated TMP effluent. Differences between treatment groups were assessed using a one-way A N O V A (p = 0.418). Observed power (1-/5) = 0.22. Bars are ± SEM. 62 0 % 6 % 1 2 % 2 5 % Effluent concentration (v/v %) Figure 2.10. Blood glucose concentration in adult Atlantic salmon following 6-hour exposure and a U c r i t challenge in 0, 6, 12 or 25% (v/v; n = 4, 6, 5, 5, respectively) primary and secondary treated T M P effluent. Differences between treatment groups were assessed using a one-way A N O V A (p = 0.869). Observed power (1-/3) = 0.09. Bars are ± S E M . 63 0% 6% 12% 25% Effluent concentration (v/v %) Figure 2.11. Plasma lactate concentration in adult Atlantic salmon following 6-hour exposure and a U c r j t challenge in 0, 6, 12 or 25% (v/v; n = 4, 6, 5, 5, respectively) primary and secondary treated T M P effluent. Differences between treatment groups were assessed using a one-way A N O V A (p = 0.381). Observed power (1-/3) = 0.24. Bars are ± S E M . 64 Figure 2.12. Plasma ion profile in adult Atlantic salmon following 6-hour exposure and a U c r i t challenge in 0, 6, 12 or 25% (v/v; n = 4, 6, 5, 5, respectively) primary and secondary treated T M P effluent. Differences between treatment groups were assessed using a one-way A N O V A (p = 0.245, 0.726, 0.293, and 0.189 for K + , C a 2 + , N a + and CI", respectively). In the analysis of C a 2 + levels weight and length exhibited significant covariate effects (p = 0.031 and 0.032, respectively). Observed power (1-/3) = 0.33, 0.12, 0.29, and 0.38 for K + , C a 2 + , N a + and CI", respectively. Bars are ± S E M . 65 00 1.4 H 0% 6% 12% 2 5 % Effluent concentration (v/v %) Experiment 2 - Upstream Simulation There were no significant changes in Q (p = 0.290; 1-/3 = 0.16), fu (p = 0.562; 1-/3 = 0.08), or V s (p = 0.238; 1-/3 = 0.20) between treatment groups after adjusting for pre-exposure levels of Q (p = 0.006) and / H (p = 0.005; Fig . 2.13). The pre-exposure value of V s did not exhibit a significant covariate effect (p = 0.089). Average Q in the control group was higher at the commencement of and during the first few hours of Experiment 2 relative to the exposure group, due to an increase in V s (Fig. 2.13). Control fish were probably stressed at the beginning of Experiment 2, however the cause is uncertain, as surgery time, water temperature, and overnight recovery times were similar between treatment groups. A s there were no significant changes in any cardiac variable over the time-course of the experiment (p > 0.05), all exposure data (or in the case o f controls, the matched 6 hours spent in freshwater) were pooled. After exposure data were adjusted for the pre-exposure level of each cardiac variable (p < 0.001 for Q , fu, and Vs) , Q in the exposure group (21.57 ± 0.47 m L min" 1 kg"1) was nearly 9% higher than Q in control fish (19.85 ± 0.47 m L min" 1 kg" 1; p = 0.017; 1-/3 = 0.68; F ig . 2.14). There were no significant changes in / H (p = 0.072; 1-/3 = 0.44) or V s (p = 0.595; 1-/3 = 0.08) between the control and exposure groups. Plasma Cortisol (p = 0.397; 1-/3 = 0.06; Fig . 2.15), blood glucose (p = 0.256; 1-/3 = 0.09; Fig . 2.16), and plasma lactate (p = 0.423; 1-/3 = 0.05; Fig. 2.17) levels did not differ significantly between treatment groups. None of the plasma ions significantly changed from control levels ( K + , p = 0.365, 1-/3 = 0.20; C a 2 + , p = 0.338, 1-/3 = 0.12; N a + , p = 0.473, 1-/3 = 0.10; CI", p = 0.623, 1-/3 = 0.07; F ig . 2.18). Similarly, there were no changes in hematocrit levels between the control (45.23 ± 2.51%) and exposure (46.31 ± 2.05%) groups (p = 0.750; 1-/3 = 0.06). 67 Figure 2.13. Cardiac output (Q ), heart rate ( / H ) , and stroke volume (Vs) in adult Atlantic salmon (n = 4) during 6-hour exposure to an increasing concentration gradient (0-25%; v/v) of primary and secondary treated T M P effluent and 1 hour of recovery in freshwater. A Transonic flowprobe was affixed around the ventral aorta of individual fish. Water velocity in the respirometer was maintained at 0.6 b l s"1 for the entire experiment. Within treatment comparisons over time and between treatment comparisons at each exposure hour were performed using a G L M univariate test with repeated measures. Between treatment comparisons for Q , / H , and V s were p = 0.290, 0.562, and 0.238, respectively. Observed power (1-/3) of between treatment comparisons for Q , / H , and V s = 0.16, 0.08, and 0.20, respectively. * indicates that pre-exposure levels exhibited a significant covariate effect (p <0.006). Plots are ± S E M . 68 Figure 2.14. Pooled exposure values of cardiac output (Q ), heart rate (JH), and stroke volume (Vs) in adult Atlantic salmon (n = 4) over a 6-hour exposure period to an increasing concentration gradient (0 - 25%; v/v) of primary and secondary treated T M P effluent. A Transonic flowprobe was affixed around the ventral aorta of individual fish. Water velocity in the respirometer was maintained at 0.6 bl s"1 for the entire experiment. Each cardiac variable was examined for significant differences between treatment groups using a one-way A N O V A , with pre-exposure levels of each cardiac variable serving as the covariable ( p < 0.001 for each), f indicates a significant difference from the control group (p = 0.017). Between treatment comparisons for / H and V s were p = 0.072 and 0.595, respectively. Observed power (1-/3) = 0.68, 0.44, and 0.08 for Q , / H , and V s , respectively. Bars are ± S E M . 70 25 20 A - X . 0.2 U Control Exposure Treatment group 71 250 A 200 A E D) 150 -c o cn O 100 -O 50 -0 J 1 , 1 1 , 1 Control Exposure Treatment group Fig. 2.15. Plasma Cortisol concentration in adult Atlantic salmon (n = 4) following 7 hours in freshwater or 6-hour exposure to an increasing concentration gradient (0 - 25%; v/v) o f primary and secondary treated TMP effluent and 1 hour o f recovery in freshwater. Water velocity in the respirometer was maintained at 0.6 bl s"1 for the entire experiment. A difference between the control and exposure groups was assessed using a one-tailed t-test (p = 0.397). Observed power (1-/3) = 0.06. Bars are ± S E M . 72 14 12 -_ 10 -1 8 - [ I fi 6 -o _ D O 4 -2 -0 -I 1 , 1 1 , 1 Control Exposure Treatment group Figure 2.16. Blood glucose concentration in adult Atlantic salmon (n = 4) following 7 hours in freshwater or 6-hour exposure to an increasing concentration gradient (0 - 25%; v/v) of primary and secondary treated TMP effluent and 1 hour of recovery in freshwater. Water velocity in the respirometer was maintained at 0.6 bl s"1 for the entire experiment. A difference between the control and exposure groups was assessed using a one-tailed t-test (p = 0.256). Observed power (1-/3) = 0.09. Bars are ± SEM. 73 3.5 -3.0 -1 _ l 2.5 -o E E 2.0 -& CO o 1.5 -ro _ i 1.0 -0.5 -0.0 -Control Exposure Treatment group Figure 2.17. Plasma lactate concentration in adult Atlantic salmon (n = 4) following 7 hours in freshwater or 6-hour exposure to an increasing concentration gradient (0 - 25%; v/v) of primary and secondary treated T M P effluent followed by 1 hour of recovery in freshwater. Water velocity in the respirometer was maintained at 0.6 bl s"1 for the entire experiment. A difference between the control and exposure groups was assessed using a one-tailed t-test (p = 0.423). Observed power (1-/3) = 0.05. Bars are ± S E M . 74 Figure 2.18. Plasma ion profile in adult Atlantic salmon (n = 4) following 7 hours in freshwater or 6-hour exposure to an increasing concentration gradient (0 - 25%; v/v) of primary and secondary treated T M P effluent followed by 1 hour of recovery in freshwater. Water velocity in the respirometer was maintained at 0.6 b l s"1 for the entire experiment. A difference between the control and exposure groups was assessed using a two-tailed t-test. Between treatment comparisons were p = 0.365, 0.338, 0.473, 0.623 for K + , C a 2 + , N a 2 + , and Cl" , respectively. Observed power (1-/3) = 0.20, 0.12, 0.10, and 0.07 for K + , C a 2 + , N a 2 + , and Cl" , respectively. Bars are ± S E M . 75 o E 3 E 2 0 1.4 H —. 1.0 o E 0.8 To o.6 H o 0 2 0.0 160 -150 -jr- 140 _ i o £ 130 E, ra 1 2 0 -Z 110 • 100 o E £ 110 H b 100 90 C o n t r o l E x p o s u r e Treatment group 76 Comparison of2001 and 2002 Ucrit Values for Control Fish There was almost a significant difference in U c r i t between control fish swum at 15 °C in 2001 (1.90 ± 0 . 1 1 b l s"1) and those swum at 18°C in 2002 (1.55 ± 0.15 b l s"';p = 0.094; 1-/3 = 0.39). The average fork length of control fish in 2002 (55.8 ± 0.9 cm) was slightly greater than that off ish in 2001 (53.1 ± 1 . 3 cm). Nonetheless, fork length did not have a significant covariate effect on U c r j t (p = 0.124) and when absolute U c r i t values (swimming speed reported in cm s"1) were analyzed, there was no significant difference in swimming performance between the two years (p = 0.230; 1-/3 = 0.21). Effect of Water Temperature on Routine and Maximum fn In 2001, the average water temperature was 13.7 ± 0.29 °C and the mean routine fu was 43.3 ± 3.04 beats min" 1. In 2002, the mean temperature was 17.7 ± 0.21 °C and the average routine fu was higher at 55.6 ± 2.69 beats min" 1. Individual values o f routine fu and the water temperature at which they were measured are weakly (r 2 = 0.12) and positively correlated (p = 0.002; Fig . 2.19). Maximum fu (achieved at or near U c r i t ) was also significantly correlated with water temperature (p = 0.05; Fig . 2.19). 77 c "E CD CD 100 80 A 60 40 A 20 • 2001 Routine fH V 2002 Routine fH Routine fH vs temperature • 2002 Maximum fH Maximum fH vs temperature V v v A 0 0 10 20 22 12 14 16 18 Water temperature (°C) Fig . 2.19. Relationship between routine and maximum / H and water temperature in adult Atlantic salmon. Experiments performed in 2001 and 2002 utilized Doppler and Transonic cardiac output flowprobes, respectively. Fish had recovered and acclimated overnight in a 115 L Blazka-style respirometer. Fish were oriented against a water velocity of 0.2 b l s"1 during routine measurements and were swimming at or near U c r i t when maximum fu was achieved. Only maximum fu data for fish in the 0, 6, and 12% exposure groups are included and all routine fu data was measured prior to effluent exposure. Significant correlations exist between routine fu and temperature (p = 0.002) and maximum fu and temperature (p = 0.05). Equation of the line-of-best-fit for the curve describing the relationship between routine fu and temperature is y = 2.61x + 8.42, r = 0.13. Equation of the line-of-best-fit for the curve describing the relationship between maximum fu and temperature is y = 4 .9 lx -12.0, r 2 = 0.26. 78 Large Range in Routine fu Values and Effects of High JHon Q Following overnight recovery, the pre-exposure, routine fu o f experimental subjects in 2002 fell within the wide range of approximately 30 to 90 beats min" 1 at temperatures near 18 °C (Fig. 2.19). It is evident in Figure 2.19 that some of the routine fu values were approaching maximum fu values. It was also determined that there was a weak (r 2 = 0.21) yet significantly con-elated (p = 0.009) relationship between routine fu and routine Q (Fig. 2.20). Maximum fu and maximum Q were not significantly correlated (p = 0.147; F ig . 2.20). Sudden and Extreme Increases in Routine fn There were sudden and extreme changes / H values in some individual fish during routine and low-level activity. These changes are somewhat reflected by the inconsistency in the widths of the standard error bars in the graphs depicting fu (Figures 2.4, 2.7, 2.13). For example, in one fish, fu increased from 43 to 80 beats min" 1 when water velocity increased from 0.4 to 0.6 b l s"1. In a different fish, fu increased from 42 to 88 beats min" 1 midway through the exposure period. 79 0 30 40 50 60 70 80 90 100 110 fH (beats min"1) Fig. 2.20. Relationship between routine and maximum Q and / H in adult Atlantic salmon following overnight recovery and acclimation in a 115 L Blazka-style respirometer. Fish were oriented against a water velocity of 0.2 b l s"1 during routine measurements and were swimming at or near U c r i t when maximum / H and Q were achieved. A Transonic flowprobe was affixed around the ventral aorta of individual fish. Only maximum / H and Q data for fish in the 0, 6, and 12% exposure groups are included and all routine / H data was measured prior to effluent exposure. Water temperature was 17.7 ± 0.21 °C (mean ± S E M ) . A significant correlation exists between routine / H and Q (p = 0.009). Maximum / H and Q were not significantly correlated (p = 0.147). Equation of the line-of-best-fit for the curve describing the relationship between routine / H and Q is y = 0.15x + 12.3, r 2 = 0.21. Equation of the line-of-best-fit for the curve r describing the relationship between maximum / H and Q is y = 0.25x + 29.6, r 2 = 0.15. 80 Decreases in Routine fu During Exposure to Dark Coloured Solutions Bathing a fish in a dark coloured solution was observed to cause bradycardia. When routine / H was high following overnight recovery, adding the dark coloured effluent to the respirometer caused / H to decline (Fig. 2.4; note the 6 and 25% treatment groups). This bradycardia was attributed to the dark colour and turbid nature of the effluent. This conclusion was given weight by an experiment in which 20 m L of dark green food colouring was allowed to fully mix inside the respirometer (5 min) when it housed a fish with a routine / H o f 67 beats min" 1 following overnight recovery (Fig. 2.21). In this fish, bradycardia was clearly observed within seconds, thus repeated trials with food colouring or other darkly coloured substances were not deemed necessary. The inflow and outflow valves of the respirometer were closed during the 10-minute duration of food colouring exposure, but the dissolved oxygen concentration within the respirometer did not fall below approximately 8 mg L" 1 (Grottum and Sigholt, 1998). 81 0 0.5 1 .1.5 2 2.5 3 3.5 4 Exposure time (min) Fig . 2.21. Routine / H o f an adult Atlantic salmon (n = 1) following the addition of 20 m L of dark green food colouring to a 115 L Blazka-style respirometer. A Transonic flowprobe was affixed around the ventral aorta of the fish. Pre-exposure / H (t = 0) was measured following overnight recovery and acclimation in the respirometer. Water velocity in the respirometer was set at 0.2 b l s"1. 82 Discussion At routine or low state of metabolism, doses of T M P effluent of 12% and greater caused Q to increase by 7-10% in adult Atlantic salmon, indicating a significant metabolic loading effect due to T M P effluent. Effluent concentrations up to and including 12% had little effect on U c r i t» Q scope, and Q m a x , thus, no metabolic limiting was evident. However, at the 25% exposure concentration, UCrit following 12-hour exposure to T M P effluent was significantly depressed relative to control fish, suggesting that sufficient metabolic resources were not available to fully meet the energetic requirements of active metabolism. Following 6-hour exposure to 25% T M P effluent there were relatively large numeric decreases in U c r i t , Q scope, and Q m a x . Although these changes were not statistically significant, p-values were extremely close to the chosen 0.05 level of significance. Given that the power of the statistical tests was only between 55-58%, these negative findings should be interpreted with caution. Comparisons of the results of the present study to existing literature are somewhat limited due to the inherent difficulties associated with capturing, holding, and experimenting with large, mature salmon. A s well , few published studies have examined similar endpoints of physiological stress in fish following exposure to mechanical pulping effluents or isolated wood extractives. It should also be noted that mechanical pulping effluents vary with pulp processing conditions, within individual mills, according to wood furnish, the length of time since felling and chipping, and the presence or absence of primary and secondary treatment processes (Wong et al. 1978; Gibbons et al. 1992; O'Connor et al. 1992; Kovacs and Voss, 1992; Martel et al. 1994; Johnsen et al. 1995; Martel and Kovacs, 1997; Martel et al. 1997). A s a result, comparisons of the present results to other studies must be made with a hint of caution. 83 Liver EROD Induction Liver E R O D activity following 12-hour exposure to T M P effluent did not change significantly in exposure groups compared to controls. Similarly, no E R O D induction was reported in adult Atlantic salmon exposed for 48 hours to 3 and 9% (v/v) primary and secondary treated T M P effluent (Sinnott and Mackey, 1997). The effluent was obtained from the same mi l l as the effluent used in the current experiments. Pre-Surgery, Routine Hematological Profile Adult Atlantic salmon from the Exploits River had a pre-surgery, routine hematological profile that was largely in agreement with that reported in the literature. However, it should be noted that the values of the hematological variables measured in these experiments may be slightly higher than those reported in other studies (for example, MilHgan, 1996). Blood samples in the present study were taken from anaesthetized fish that, depending on the ease of capture from the holding tank, may be been stressed. To obtain true routine hematological data, blood samples should be taken from unanaesthetized, cannulated fish. The chloride level of salmon in the present study (c. 135 mmol L" 1 ) parallels the concentrations (129-148 mmol L" 1 ) reported in farmed conspecfics (Brodeur et al. 1999,2001a; Wagner et al. 2003a, b). The routine N a + (c. 157 mmol L" 1 ) and C a 2 + (c. 1.2 mmol L" 1 ) levels in the current work are also very similar to those reported by Brodeur et al. (2001a; c. 152 and 1.4 mmol L " 1 , respectively). The routine plasma K + concentration o f wi ld adult Atlantic salmon (c. 1.2 mmol L" 1 ) was slightly lower than those noted by Wagner et al. (2003b; c. 2.3 mmol L" 1) and Brodeur et al. (2001a; c. 2.9 mmol L" 1 ) , which would not be the case i f the fish in the present study were severely stressed or had exercised heavily. 84 The routine hematocrit level of the salmon in the present study was approximately 48%, which coincides wel l with that reported for some farmed Atlantic salmon (44-49%; Sandnes et al. 1988). Conversely, the range in hematocrit levels reported for other hatchery reared Atlantic salmon is between roughly 28-43% (Brodeur et al. 1999, 2001a, Wagner et al. 2003a, b). The salmon used by Brodeur (1999, 2001a) and Wagner (2003a, b) were held at temperatures between 9-14 °C, while those used by Sandnes et al. (1998) were held at temperatures between 2.0-7.7 °C. Although the following theory is not consistent with hematocrit levels reported by Sandnes et al. (1998), the relatively high pre-surgery, routine hematocrit level observed in the current work may be a result of the warm ambient temperature (14.8-20.0 °C) of the housing water, which could have caused a release of stored red blood cells from the spleen during anaesthesia (Gallaugher and Farrell, 1998). The salmon in the present study may have had a higher routine metabolic rate than the salmon used by Brodeur et al. (1999, 2001a) and Wagner et al. (2003a, b), as metabolic rate varies directly with water temperature (Fry, 1971). Routine plasma lactate levels were approximately 3.4 mmol L" 1 in the present study, which compares well to the routine lactate concentration reported by Brodeur et al. (2001a; c. 3.3 mmol L" 1 ) for Atlantic salmon. Wagner et al. (2003a, b) noted routine lactate levels in Atlantic salmon to be as low as approximately 2.6 mmol L" ' and as high as 6.8 mmol L " 1 , respectively. A s there is tremendous variability around the mean routine lactate concentration of 6.8 mmol L" 1 (Wagner et al. 2003a), there may have been inconsistencies in the sampling procedures and this value should be interpreted cautiously. The blood glucose concentration of salmon in the present study (c. 5.8 mmol L" 1 ) falls between that reported for farmed Atlantic salmon by Wagner et al. (2003a, b; c. 3 mmol L" 1) and Brodeur et al. (1999, 2001a; c. 6-7 mmol L" 1 ) . This similarity in routine glucose levels is interesting, as salmon used by Brodeur et al. (1999, 2001a) and Wagner et al. (2003a, b) were fed 85 in the laboratory, whereas the fish used in the present study had not fed for many weeks. In a corresponding fashion, other studies have found that routine plasma glucose levels are comparable in fasted versus fed salmon (Sheridan and Mommsen, 1991; Duan and Plisetskaya, 1993). Differences in sampling techniques are the most likely explanation for the high routine Cortisol concentration found here (c. 61.3 ng mL" 1 ) compared with values reported in the literature for well rested fish (c. 5-19 ng mL" 1 ; Wagner et al. 2003a, b). Atlantic salmon in the present study were dip netted from a large holding tank and transferred to an anaesthetic bath prior to blood sampling. Some individuals were difficult to capture and almost all fish thrashed in the dip during transfer to the anaesthetic bath. Plasma Cortisol levels begin to increase within minutes of exposure to an acute stressor, though they may take up to 2 hours to reach peak levels (Mill igan, 1996). Hyperglycemia frequently follows, although this response is still slower (Soivio and Oikari , 1976; Wendelaar Bonga, 1997). This slower hyperglycaemic response is probably why glucose levels coincided well with values reported in other studies. Pre-Surgery and Post-Experimental Hematocrit, Lactate, Cortisol, and Glucose Levels The influence of the experimental procedures on blood chemistry can be assessed by comparing routine and post-experimental concentrations of each hematological variable as well as post-experimental hematologica data between both experiments. Post-experimental changes in the volume of packed blood cells relative to pre-surgery levels can indicate that surgery caused significant blood loss (Gallaugher and Farrell, 1998). In both experiments, pre-surgery and post-experimental hematocrit levels were consistent, signifying that the surgical affixation of the Q flowprobe around the ventral aorta did not cause substantial blood loss. Furthermore, post-Experiment 1 and post-Experiment 2 hematocrit levels were similar, implying that swimming to UCnt does not induce changes in hematocrit. This finding coincides wel l with the results of various other studies (reviewed in Gallaugher and Farrell, 1998; Ytrestoyl et. al. 2001). The 86 present post-Ucrit hematocrit levels are comparable to the post-Ucrit hematocrit levels reported by Ytrestoyl et. al. (2001; c. 38-44%) for adult Atlantic salmon. Post-Experiment 1 plasma lactate increased significantly from pre-surgery and post-Experiment 2 levels in all treatment groups. This was expected, since exhaustive exercise is partially powered by white muscle fibres and fuelled by anaerobic metabolism, which causes an increase in lactic acid (Black, 1955; Mil l igan , 1996; Lee, 2002). The average post-Ucrit lactate level for all treatment groups (c. 5.9 mmol L" 1) is slightly higher, but still comparable to those of Atlantic salmon equipped with a Doppler flowprobe and swam to Ucrjt (Wagner et al. 2003a, b; c. 3-5.5 mmol L" 1 ) . Elevations in Cortisol were mirrored by increases in blood glucose levels, since Cortisol is implicated in the modulation of gluconeogenesis and glycogen metabolism (Vijayan et al. 1991; Vijayan and Letherland, 1992; Wendelaar Bonga, 1997). Relative to pre-surgery, routine levels (c. Cortisol = 64 ng mL" 1 ; glucose = 5.7 mmol L" 1 ) , the increase in Cortisol (c. 199 ng mL" 1 ) and glucose (c. 10 mmol L" 1 ) following Experiment 2 represents the cumulative stresses imparted by surgery, confinement to the swimming chamber, and the experimental protocol. Relative to post-Experiment 2 values, there was a significant increase in Cortisol (c. 368 ng mL" 1 ) and an insignificant increase in glucose (c. 13 mmol L" 1 ) in all treatment groups following Experiment 1, which represents the additional stress of exhaustive exercise. However, these concentrations are much higher than those reported in other experiments, indicating that fish in the present study were highly stressed following the swimming challenge. For example, Wagner et al. (2003a, b) reported post-Ucrit Cortisol and glucose levels in Atlantic salmon to be in the range of approximately 130-180 ng mL" 1 and 5-7 mmol L " 1 , respectively. Ytrestoyl et. al. (2001) noted post-UCrit Cortisol and glucose concentrations in Atlantic salmon to range from about 140-225 ng mL" 1 and 5-8 mmol L " 1 , respectively. The disparities in post-Ucrit Cortisol and glucose levels between the present study and literature values may be the result of the wi ld and restless nature 87 o f the fish used in this study, as the salmon used by Wagner et al. (2003a, b) and Ytrestoyl et. al. (2001) were hatchery-reared. Furthermore, the fish in the present experiments were fasted, which may have caused the severe post-Ucrit hyperglycaemia. However, starvation may not be the only explanation for the high post-Ucrit glucose values reported here because the salmon used by Ytrestoyl et. al. (2001) had not been fed for at least several weeks before being used in experimentation. Nonetheless, the post-Ucrit Cortisol and glucose levels reported by Wagner et al. (2003a, b) were the most divergent from those of the present study, and his fish had been fed daily up to 24 hours pre-experimentation. Exercising, fasted fish preserve glycogen and access a greater proportion of glucose via gluconeogenesis pathways than fed fish (Vijayan et al. 1993; French et al. 1981, 1983). Therefore, perhaps this difference in glucose-accessing pathways is linked to the post-Ucrit hyperglycaemia observed in the fasted fish used in this study. Osmoregulatory Changes In the present study, the post-experimental osmoregulatory profiles of the salmon were slightly different from pre-surgery levels, although these changes were largely consistent between the two experiments. The blood ionic profile of Atlantic salmon following Experiments 1 and 2 were also comparable over all treatment groups, which coincides with reports that plasma osmolality in Atlantic salmon is not significantly altered by intense exercise (Byrne et al. 1972; Ytrestoyl, et al. 2001). Furthermore, there were no major changes in the osmoregulatory profile of the salmon following exposure to any concentration of T M P effluent. Effluent exposure had no effect on N a + and CI" levels. Potassium concentrations were highly variable in all treatment groups following both experiments. Correspondingly, levels o f this ion are erratic in pike (Esox leucius) following handling stress (Soivio and Oikari , 1976). There is no literature to support the present findings that whole T M P effluent does not elicit a strong change in the hydromineral profile o f salmon. However, resin acids do not cause significant osmoregulatory 88 stress in salmonids (Nikinmaa and Oikari , 1982; Oikari et al. 1983). Kruzynski (1979) provided the most convincing evidence that resin acids cause ionic imbalances, as all plasma electrolytes except N a + changed significantly in sockeye salmon following 5-day exposure to 0.65 mg L" 1 D H A . Comparison of Routine Q to Literature Values Literature values of the routine Q in Atlantic salmon average 11-18 m L min" 1 kg"1 at 9-10 °C (Brodeur et al. 1999, 2001a) and reach 47 m L min" 1 kg"1 at 16.5-17.5 °C (Brodeur et al. 2001b). In rainbow trout, routine Q normally ranges from, on average, 15-46 m L min" 1 kg" 1 between temperatures of 4-16 °C (see review in Farrell and Jones, 1992; Thorarensen et al. 1996b; Brodeur et al. 2001c). Randall (1970) summarized that, in teleosts, routine Q largely falls within the range of 15-30 m L min" 1 kg" 1. Given these literature values, the pre-exposure, routine Q values observed in the present study (c. 16-24 m L min" 1 kg"1) during routine (0.2 b l s"1) and low-level activity (0.6 b l s"1) are at the low end of the expected range of Q values given the experimental temperatures employed (c. 17-19 °C). However, keeping in mind that fish with a routine / H in excess of 60 beats min" 1 were treated as outliers and eliminated from the experiment, these routine Q values are more logical. Therefore, the routine Q values reported for this study are deemed to be within a normal range and observed changes in Q caused by , exposure to T M P effluent can be discussed with confidence. Effects of TMP Effluent on Q During Routine and Low-Level Activity During routine or low-level activity, acute exposure to at least 12% T M P effluent caused a 7-10% increase in Q relative to the control and 6% exposure groups. This was largely 89 achieved through an elevation in V s and a reduction in / H . Cardiac output is a measure of blood flow through the ventral aorta, the only artery leading away from the heart. Elevations in Q reflect increases in oxygen supply to tissues in need given all other things are equal. For example, Q increases to support the elevation in gut blood flow that occurs postprandially (Axelsson et al. 2000; Farrell et al. 2001). Additional energetic resources are required in the gut after feeding to aid in digestion. It is also a well-known fact that Q increases during exercise to help fuel the energetic costs of swimming in the red muscle (Randall, 1968; Axelsson, 1988; Axelsson and Fritsche, 1991; Farrell and Jones, 1992; Ko lok et al.1993; Wagner et al. 2003a, b). The fish in the present study had not fed for many weeks, and were operating at a routine, or very low, state of metabolism when the increase in Q was observed. Therefore, what can explain the relative increase in Q in fish exposed to at least 12% T M P effluent? One possibility is that the changes in behaviour (increased and restless activity, irregular pectoral fin movements) that are often displayed by fish acutely exposed to toxicants caused an increase in the blood supply to some muscle tissues (Jones, 1947; Calow, 1991; Laitinen and Valtonene, 1995; Brodeur et al. 2001a). However, given that the fish were somewhat restrained in the respirometer and that no restless behaviour was observed during effluent exposure, hyperactivity is probably not the cause of the observed increase in Q . Another potential explanation is that Q increased to modulate respiratory or metabolic stress. For example, mucus secretion by the gills could have caused a reduction in oxygen uptake from the water. In an attempt to compensate, Q increased to circulate any available oxygen to the tissues at a faster rate. This leads to another potential explanation for the observed increase in Q , as there are energy demands involved in the activation of the internal defence systems such as mucus secretion (Calow, 1991). 90 The final explanation for the observed increase in Q during routine and low-level activity can be better understood i f the Fick equation, VO2 - Q • EO2, is reintroduced. According to this principle, Q and EO2 are the only two variables that can be altered to satisfy increased metabolic demands and thus elevate VO2. Although the Fick equation overlooks some factors that govern metabolic processes (discussed in Chapter 1), these inherent inaccuracies w i l l be disregarded at this time and we w i l l focus on the equation itself. A s such, i f it is assumed that £02 remained stable or systematically changed with Q during the present experiments, it can be inferred that the.increase in Q observed in fish exposed to at least 12% effluent reflected an increase in metabolic rate. This is reasonable, since upholding homeostasis in the presence of stressors requires that compensatory processes and defence mechanisms (ie: mucus secretion) are initiated and maintained. Taken together, these processes are energy demanding and have generally been observed to cause metabolic loading (Seyle 1973; Calow, 1991). However, it is impossible to substantiate the claim that metabolic rate increased, or to determine i f there were changes in EO2 that would negate or amplify this suggested increase in metabolic rate, as neither EO2 nor VO2 were measured in this study. Consequently, we cannot be certain that the observed 7-10% increase in Q is evidence that acute exposure to T M P effluent caused metabolic loading in adult Atlantic salmon. Respiratory and metabolic stress has been observed in fish exposed to various types of toxicants. In salmonids exposed to B K M E overnight, ventilation volume doubles and oxygen saturation of arterial blood decreases by 20%. These responses could be indicative of an impairment of gas exchange at the gills due to mucus secretion, altered oxygen-carrying capacity of the blood, or reduced circulation (Davis, 1973). Numerous studies have also reported that routine metabolic rate (measured using respirometry) increases as a result o f toxicant exposure. 91 Some of these toxicants include B K M E (Davis, 1973), D H A (Jain et al. 1998), T C G (Brodeur et al. 2001b), copper (O'Hara, 1971), selenium (Lemly, 1993), P C P (Holmberg and Saunders, 1979; Farrell et al. 1998), and low p H (Butler et al. 1992). It is important to note that B K M E is whole effluent from bleached kraft mills, and D H A , T C G , and copper are common constituents of chemical and mechanical pulp mi l l effluents. In fact, copper and D H A were regularly present in the effluent used in this study (Table 2.1; Appendix I). However, in bluegill sunfish (Lepotnis macrochirus), copper concentrations <0.5 mg L" 1 elicit an increase in routine VO2 o f less than 5% during 18 hours of exposure (O'Hara, 1971). Copper levels in the T M P effluent used in this study were generally much lower than this (2-8 /xg L " 1 ; Appendix I). Concentrations of D H A between 0.12-0.88 mg L" 1 cause a 48% increase in routine VO2 in sockeye salmon following 8-14 hours of exposure (Jain et al. 1996). These concentrations are similar to the D H A levels that were commonly present in the effluent used in this study (0.11-0.85 mg L " 1 ; Appendix I). Although these are pertinent comparisons, it is evident from both Table 2.1 and Appendix I that T M P effluent is literally a cocktail o f cations, anions, metals, and wood extractives. Since whole effluent was used in the present experiments, it is very difficult to pinpoint the compounds that caused the observed physiological effects, particularly since various constituents may have acted synergistically. The increase in Q during routine and low-level activity indicates that respiratory or metabolic processes were in some way affected by the T M P effluent. For short periods, increasing Q is probably not too physiologically taxing for a fish, as Q regularly increases 2-3 fold during exercise and postprandially. However, i f Q remains heightened for an extended period of time, fish w i l l have a less Q s c o p e to escape from predators or deal with other stressors, which would negatively impact overall fitness (Priede, 1977). Furthermore, i f Q is elevated for a prolonged period energy reserves in these migrating, non-feeding fish would be taxed and fish 92 may more easily succumb to the toxic effects imposed by the effluent. This would be classified by Seyle (1973) as the third stage of stress - exhaustion. With respect to the scenario in the Exploits River, many of the adult salmon returning to spawn either migrate upstream past the effluent outfall, or enter tributaries of the Exploits River downstream of the effluent outfall. Therefore, it is hypothesized these fish are not exposed to T M P effluent for prolonged periods and i f Q does increase, this response is l ikely transient. Moreover, effluent concentrations in the Exploits River rarely exceed 1%. A s the results of the present study indicate that concentrations of T M P effluent of 12% and higher cause Q to increase during routine and low-level activity, it is unlikely that salmon in the Exploits River experience elevations in Q . Achievement of Maximum Swimming Performance It is possible that the adult Atlantic salmon used in both 2001 and 2002 experiments were unable to achieve maximum swimming velocities during experimentation due to their large size (fork length = 54-56 cm) relative to the swimming chamber (length = 80 cm; diameter = 24 cm). It is a concern that fish were unable to make full tail beats and that burst and coast swimming behaviours were inhibited in a swimming chamber of this size. Critical swimming speed is not only a useful endpoint by which to compare between-treatment effects, it also allows fish to be brought to a similar state of active metabolism so that physiological data can be compared between subjects that are in a similar energetic state. Indeed, when XJcrit is achieved, fish are swimming at their maximum prolonged swimming capability (Brett, 1964). When a fish is at or approaching U c r i t , maximum Q , / H , V s , and VO2 are also achieved (Kiceniuk and Jones, 1977; Kolok et al. 1993; Thorarensen et al. 1996b). Furthermore, Lee (2002) illustrated that at approximately 60% of U c r j t , sockeye and coho salmon (Oncorhynchus kisutch) begin to fuel swimming activity by anaerobiosis. Therefore, fish swum to U c r i t under similar conditions 93 should have comparable blood lactate levels. A s such, we are able to elucidate i f the salmon in the present study were able to achieve U c r i t in the respirometer by comparing the applicable physiological variables listed above to reported literature values (Table 2.4). A s Q m a x is the product o f the relative contributions of /Hmax and Vs m ax , Q max values w i l l be the only cardiac variable reported in Table 2.4. A s outlined in Table 2.4, U c r j t values observed in the present study are very similar to those reported by Farrell et al. (1996) and Jain et al. (1996) for sockeye salmon of similar length and swimming at like water temperatures. Comparisons of Q max values for similar species, U c rjt values, and water temperatures are limited. However, according to the literature, the value reported for the present study is very close to what would be expected for salmonid species swimming at U c rit- A s mentioned earlier, post-U c rj t blood lactate levels fall within the range of values reported in other salmonids. These comparisons indicate that salmon in the present study were able to reach U c rit- Nonetheless, it should be reiterated that the physiological variables listed in Table 2.4 can vary with fish size, stock, species, and experimental protocol. To provide additional evidence that fish in the present study were able to swim to their maximum capability, it was also determined i f there was a systematic effect of fish length on U c rit- The fork lengths of fish swam to U c r i t were between 46.0-60.9 cm in 2001 and 52.8-61.0 cm in 2002. It was found that fork length did not elicit a statistical covariate effect on U c r i t i n 2002 (p = 0.603 ). In 2001 fork length acted as a significant covariate (p = 0.011), however the interaction was.such that shorter fish actually had a slightly lower U c r i t than longer fish. Based on the above evidence, it can be concluded that the salmon in the present study achieved, or came very close to achieving, maximum swimming performance. Table 2.4. Comparison o f critical swimming speed (U c rit), maximum cardiac output (Q m a x ) during exercise, and post-U c rit blood lactate levels in various fish species. A l l entries are mean values unless indicated below. Species Temp. (°Q Length2 (cm) Ucrit (bl s"1) Qmax (mL min"1 kg"1) Lactate (mmol L"1) Reference Atlantic 15 53 1.9 — — Present study salmon 18 56 1.6 48.2 4.7 Present study 9 .38 2.4-2.6 60.7-69.4 3-4.5 Wagner et al. 2003b 10 38 2.6 44 5.5 Wagner et al. 2003a 10 40-56c 3.1 Ytrestoyl et al. 2001 12 55-60b 1.8 Booth etal. 1997 18 55-60b 2.2 Booth etal. 1997 Sockeye 12 64 1.4 Lee, 2002 salmon 16 63 1.8 Lee, 2002 18 64 2.1 Lee, 2002 19-20 50-59 1.4 3.5 Farrell et al. 1998 19-21 59 1.6 3 Jain etal. 1998 - Coho 8 58 1.7 Lee, 2002 salmon Rainbow 7 29 2.5 Daxboeck, 1982 trout 9-11 40-53 b ~2 52.6 Kiceniuk and Jones, 1977 10 32-42 b 2.4 48.7 Thorarensen et al. 1996b Largescale 5 40 0.8-1.0 18 Koloketal . 1993 suckerd 10 40 1.3-1.4 37 Koloketal . 1993 16 40 1.3-1.4 43 Koloketal . 1993 a fork length b fork or total length not specified c total length Catostomus macrocheilus 95 Effects of TMP Effluent on Ucrit, Q max, Q scope, and Post- Ucru Lactate Levels Although Q increased during routine and low-level activity in salmon exposed to the highest effluent concentrations, such effects were masked at progressively higher swimming speeds during the U c r i t challenge. This may suggest that fish could defer metabolic costs when swimming at higher velocities, but is more likely a factor of the low power of the between treatment statistical tests performed at each consecutive swimming speed. It would be interesting to further investigate the how T M P effluent or other toxicants impact Q during various intensities of exercise, as this has largely been overlooked by scientists who commonly assess the energetic state of a fish swimming at, or very near, U c r i t- Nonetheless, interesting comparisons can be made between treatment groups when fish are performing at their maximum capability. There were no relative changes in U c r i t , Q scope, or Q max at effluent concentrations between 0-12%. Despite the low power of the statistical tests, the mean values of U c r i t , Q scope, and Q m a x between treatment groups were numerically similar. It is possible that i f the fish were not able to achieve U c r j t due to their large size relative to the respirometer, changes in Q s c o p e and Q m a x may not have been detected at these low effluent concentrations and this could explain why metabolic limiting was only apparent at the highest effluent concentration. In the 25% exposure group, Q scope was distinctly reduced compared to all other treatment groups. Also , Q m a x in the 25% exposure class was noticeably depressed relative to fish exposed to lower effluent concentrations. Although neither of these changes were statistically significant, p-values approaching the 5% level of significance (0.079 and 0.069 for Q s c o p e and Q m a x , respectively) and the moderate power of the statistical tests (55 and 57% for Q s c o p e and Q m a x , respectively) demonstrate that a type II error may have occurred. It is logical that relative Q S Co Pe and Q m a x 96 were depressed in the 25% exposure group, because the mean U c r i t o f fish exposed to 25% effluent following 6-hours was reduced compared to fish exposed to lower effluent concentrations. However, it is possible that a limitation on Q m a x reduced U c r i t rather than vice versa. The relative depression in U c r j t in this treatment group was not statistically significant, although the fact that the p-value was very close to the 5% level of significance (p = 0.067) and the statistical power of the test was low (58%) indicate that concentrations of 25% T M P effluent could have some impact on U c r i t . Further proof that T M P effluent impairs swimming performance is that following 12-hour exposure to T M P effluent, U c r i t in the 25% exposure group was significantly depressed. Indeed, these results strongly suggest that fish exposed to 25% effluent were energetically taxed. Given that routine Q was elevated in the 25% exposure group, which implies that metabolic loading may have been occurring, it is reasonable that fish in this exposure class had a reduced U C rit , Q scope and Q m a x relative to fish exposed to lower effluent concentrations. Fish exposed to 25% effluent likely reduced the metabolic costs incurred during swimming to allow for increased maintenance and repair costs associated with effluent exposure. If metabolism is loaded prior to the initiation of a swimming challenge, the additional energetic costs of active metabolism often cannot be met by the already challenged aerobic supply system. To further support for this notion, post-U c rit plasma lactate levels increased in direct proportion with effluent concentration. This suggests that as effluent levels rose, more anaerobic effort was required to meet the costs incurred during active metabolism. It is particularly interesting that, relative to the other treatment groups, average plasma lactate level was highest and U c r i t was lowest in the 25% exposure class. A s there is an indication that salmon exposed to T M P effluent are metabolically taxed during exhaustive exercise, further exploration o f the metabolic demands placed on these fish during intense levels of activity would have great ecological relevance. 97 Atlantic salmon migrating up the Exploits River must surpass waterfalls, fishways, and navigate through numerous turbulent areas of the river. Similar findings of metabolic loading during routine activity and apparent trade-offs in the allocation of metabolic resources during exercise have been reported in fish pre-exposed to P C P and D H A . Farrell et al. (1998) pre-exposed sockeye salmon to P C P and challenged them to perform three consecutive U c r j t trials interspersed with a 45-minute recovery period. He found that recovery ratios were less than unity, and that relative to control fish, VO2 was elevated during routine activity and recovery, plasma lactate levels were higher immediately post-U c r i t and during recovery, and ventilation rate during recovery was elevated (Farrell et al. 1998). Jain et al. (1998) employed similar methodology, however in this case sockeye salmon were pre-exposed to D H A and challenged to perform two consecutive U c r i t challenges separated by a 40-minute recovery period. In these fish, routine VO2 was higher than that of the control group, the recovery ratio was less than unity, and at U c r i t and during recovery plasma lactate levels were always higher than routine levels, whereas those of control fish remained unchanged throughout testing. In other studies, where swimming performance was the main energetic parameter examined, varying results have been reported. Chronic B K M E exposure and 24-hour exposure to 0.2 mg L" 1 T C G cause significant reductions in U c r i t relative to controls (Howard, 1975; Johansen et al. 1994). Critical swimming speed is unaffected following 90-day exposure to B K M E (McLeay and Brown, 1979), 24-hour exposure to 0.3 and 0.4 mg L" 1 T C G (Johansen et al. 1994), 25 day exposure to T C G (Johansen et al. 1994), and 24-hour exposure to mono- or di-chlorinated D H A (Kennedy et al. 1995). Furthermore, exposure to conifer pulp causes a depression in swimming endurance but not in maximum swimming speeds compared to control fish (MacLeod and Smith, 1966). Very little literature is available pertaining to the effects of mechanical pulping effluents on lactate levels, although Tana (1988) recorded that there were 98 initial, but transient, increases in lactate levels in rainbow trout exposed to 5 or 50 ug L" 1 D H A for 80 days. Explanations for Lack of Effects and Caveats A s indicated by the hierarchy of the nomenclature, primary, secondary, and tertiary stress responses are initiated in a sequential order. It is interesting to ponder why a significant decrease in Ucnt and increase in Q , tertiary stress responses, were observed in this study, while none of the primary and secondary stress responses examined changed substantially. For many comparisons in the present study, particularly of the hematological data, the power of the statistical tests employed was very low. When variability was small, sample sizes were increased through repeated measures or pooling data over time, or when there was a large effect (difference between the null and alternative hypotheses), statistical power was usually > 80%. A s such, a recommendation for future work is to increase sample sizes in each treatment group. Cannulating fish so that repeated blood samples can be taken may also prove to be useful in increasing statistical power. Furthermore, variability in the hematological data would have been reduced i f blood samples were taken from unanaesthetized, cannulated fish. Other suggestions for future studies are to hold each experimental fish for the same length o f time in the laboratory before subjecting it to experimentation to ensure that fish have more equally been able to recover from capture and transport stress. Using a swimming chamber that is long enough to allow for burst and coast swimming and wide enough to enable full-amplitude stride lengths is also recommended. Nonetheless, there may be other factors that contributed to the lack o f adverse physiological effects. For example, adult fish were chosen as the experimental subjects, and this is not always, but usually, the most resilient life stage of fish. Secondly, effluent used in this study had undergone primary and secondary treatment. Secondary treatment of T M P effluents is 99 effective in largely reducing or eliminating acute lethal and chronic sublethal toxicities, biochemical responses (ie: M F O induction), wood extractive levels, B O D , and C O D (Gibbons et al. 1992; Kovacs et al. 1995; Martel et al. 1997; Martel and Kovacs, 1997; Appendix I). Although secondary treatment provides the greatest reduction in effluent toxicity, there are other contributing factors that dictate an effluent's potency. For example, fir species exclusively contain two naturally occurring, highly toxic extractives, juvabione and dehydrojuvabione (Leach and Thakore, 1976; Wong et al. 1978; O'Connor et al. 1992; Martel et al. 1997). Moreover, mechanical pulping processes that produce high pulp yields (yield range 55-95%) generate effluents with lower acute toxicity and B O D than effluents from low pulp-yielding processes (Kovacs and Voss, 1992). Comparison of2001 and 2002 Ucril Values for Control Fish It is difficult to compare U c r i t values between the two years, as many factors may have confounded the results. It is possible that U c r j t was lower, although not statistically, at 18 °C (c. 1.55 bl s"1) than at 15 °C (c. 1.90 b l s"1) because 18 °C is above the optimal temperature that permits the most favourable functioning of internal biochemical, cellular, and physiological mechanisms that allow for maximal swimming performance to be achieved (Lee, 2002). Certainly, various sources document that temperature extremes ultimately inhibit U c r i t performance, but somewhere in the middle is an optimal temperature at which peak performance occurs. Brett has extensively researched the temperature optima for swimming in salmon, and has deduced that sockeye salmon achieve maximal U c r i t at temperatures near 15 °C, while coho salmon swim best at temperatures approaching 20 °C (Brett et al. 1958; Brett, 1964; Brett and Glass, 1973). The temperature optima for swimming in various other fish species are 18 °C in rainbow trout (Keen and Farrell, 1994), 15 °C in brown trout (Salmo trutta; Butler et al. 1992), 12 °C in lake whitefish (Coregonus clupeaformis; Bernatchez and Dodson, 1985), and 11 °C for purely aerobic swimming in rainbow trout (Taylor et al. 1996). In light of this review, it is possible that 15 °C is close to the optimal temperature for peak swimming performance in Atlantic salmon. However, in direct contrast to these results, Booth et al. (1997) found that the mean (± S E M ) TJcrit in pre-spawning adult Atlantic salmon from the Exploits River, N L , was 1.76 ± 0.06 bl s"1 at a water temperature of 12 °C, and was 2.16 ± 0.18 b l s"1 at 18 °C (mean ± S E M ) . Such fish were equipped with an E M G radio tag, yet the presence of the tag had no effect on swimming performance. One possible explanation for the inverse relationship between temperature and U c r i t in the current study compared to that of Booth et al. (1997) is that the influences of temperature were overridden by the effects that the different types of flowprobes employed in 2001 and 2002 had on swimming performance. Transonic flowprobes were employed in 2002, and they have thicker and bulkier lead wires than Doppler flowprobes (used in 2001). Researchers have suggested that Transonic flowprobes are not as wel l suited for measuring blood flow in exercising fish, as more swimming effort may be required to overcome the drag from the lead wire (Thorarensen et al. 1993). This could be the reason why U c r i t in the present study was substantially lower than that found by Booth et al. (1997) when fish were swam at 18 °C, and could also help to explain why the mean U c r j t o f control fish tested in 2001 was higher than in 2002. There are other disparities in the treatment of the fish between the two years that may have confounded the observed results. There was great inequality in length of time a fish was held in the laboratory before being used in experimentation in each year (41-100 days in 2001; 2-5 days in 2002). Furthermore, in 2001 salmon were held in a large tank with no internal current. Based on these facts, one would expect that U c r i t of 2001 fish would be lower than that of 2002 fish, as their migratory instinct may have been lost after being held in the laboratory in still water for such a long time. Additionally, fish in 2001 had varying degrees of Saprolegnia spp. infections prior to experimentation and had successfully been treated in numerous anaesthetic baths of Prefuran and Malachite Green. Prefuran is a broad spectrum antibiotic and Malachite Green is a powerful anti-parasitic agent that acts as a respiratory inhibitor (FishDoc, 2003). It was suspected that fish in 2001 would have lower U c r j t than fish in 2002 because they had all been treated with strong antibiotics and some had previously had serious fungal infections. However, U c r i t in 2001 fish was higher than in 2002, therefore water temperature or the type of flowprobe employed may have had more of an influence on U c r i t than the holding duration and conditions and the antibiotic treatment of the fish. Overall, given that there were numerous disparities between the experimental conditions and the treatment of the fish in 2001 and 2002, we cannot say with certainty why U c r i t for control fish in 2001 experiments was subtly higher than in 2002. Effect of Water Temperature on Routine and Maximum fu It was observed in this experiment that both routine and maximum / H ' s are positively correlated with water temperature. It has been reported that in lower vertebrates, with the exception of tuna fish, there is a wide range of normal / H ' S that extends to 120 beats min" 1 (Farrell, 1991). The highest / H ' S are commonly displayed at the wannest water temperatures (reviewed in Farrell and Jones, 1992). For example, in intact rainbow trout, resting / H ' S are approximately 38, 46, and 91 beats min" 1 at water temperatures of 4, 11, and 18 °C, respectively (Taylor et al. 1996). In perfused in situ rainbow trout hearts, / H ' S range from approximately 69-81 beats min" 1 at 15 °C, 78-88 beats min" 1 at 18 °C, and 81-101 beats min" 1 at 22 °C, depending on the concentration of adrenaline administered in the perfusate (Farrell et al. 1996). These literature values, as well as the present data, clearly illustrate that the fu o f fish varies directly 102 with water temperature. Furthermore, these data and Figure 2.19 illustrate that fu is a weak means of indirectly estimating FO2. Routine fu is not only highly variable at any given water temperature, it is positively correlated with temperature. Undoubtedly, any evaluation of VO2 using / H telemetry in free-swimming fish while they are patrolling water of different temperatures would undoubtedly be wrought with error. Large Range in Routine fn Values and Effects of High fn on Q The pre-exposure, routine fu of experimental fish fell within an extremely wide range (c. 30 to 90 beats min" 1). Furthermore, for reasons that were not observed and are difficult to explain, some of the highest routine / H values approached expected maximum / H values, which is not consistent with what is known for exercising salmonids. A n examination of the literature indicates that the range in routine fu values at similarly high temperatures is not consistent between studies. For example, routine fu is 78 ± 1.0 beats min" 1 in in situ perfused rainbow trout hearts bathed in 18 °C solution (Keen and Farrell, 1994; mean ± S E M ) . The routine fu o f intact rainbow trout is 91 ± 4.8 beats min" 1 at 18 °C (Taylor et al. 1996) and is 74.8 ± 13.7 beats min" 1 at temperatures between 16.5-17.5 °C (Brodeur et al. 2001b; mean ± S E M ) . In Atlantic salmon grilse angled from the Exploits River system, routine fu ranges from 55-89 beats min" 1 (66.9 ± 0.5 beats min" 1; mean ± S E M ) at 17 °C and 53-87 beats min" 1 (72.3 ± 0.4 beats min" 1; mean ± S E M ) at 20 °C (Anderson et al. 1998). Although the tremendous range in routine fu values recorded in the present study is unusual, the data of Brodeur et al. (2001b) and Anderson et al. (1998) also illustrate that routine fu can diverge greatly from the mean and fall within a wide range. One cannot be certain why some studies report that fu is regulated within a narrow range and others do not. It should be noted that in this study, raising the dissolved oxygen levels, increasing the rate of water replenishment, removing suspect visual or audible disturbances, and 103 employing longer post-surgery recovery times had little or no impact on changing the routine fu of the fish. It was initially suspected that at warm water temperatures there would be greater variance around the mean routine fu and that this would explain the large range in routine fu that was observed in the 2002 experiments. However, the S E M of the routine / H ' S in 2001 and 2002 studies are similar (homogeneity of variance p = 0.205), although these error bars obviously span different values. Similar deviations around mean / H values at low and high water temperatures have been observed elsewhere. In in situ rainbow trout heart preparations fu is approximately 52 ± 1.4 beats min" 1 at 8 °C and 78 ± 1.0 beats min" 1 at 18 °C (mean ± S E M ; Keen and Farrell, 1994). In the study by Taylor et al. (1996), the S E M of the resting fu o f rainbow trout at 4, 11, and 18 °C are also comparable. Therefore, it can be concluded that there is similar variability in fu at warm and cool water temperatures. Routine / H and Q were positively and significantly correlated, however when fu and Q were at a maximum, this relationship broke down. This indicates that routine and maximum Q may be affected differently by fu, and as a result, temperature. Farrell and Jones (1992) also raised this point in their review of the fish heart. It has been observed that elevated water temperatures can increase routine Q through chronotropy, yet V s can decrease concurrently, thus Q can also remain stable (Fig 2.4; Butler and Taylor 1975; Farrell et al. 1996). Maintaining a relatively constant Q highlights the importance of a constant blood flow to the body of the fish. However, in the present experiments it was also observed that there was a significant positive relationship between routine Q and / H . It was expected that i f T M P effluent influenced Q , that it would be in a very subtle fashion. In order to obtain precise Q data with low variability and because some routine / H ' S approached expected maximum fu values, fish with a routine fu in .104 excess of 60 beats min" 1 were treated as outliers and eliminated from experimentation and analysis. However, this practice resulted in low sample sizes, as 33% of the attempted trials were prematurely terminated. Employing greater sample sizes would have allowed for more robust statistical tests, however i t was unfeasible given the time restraints of the study. These time restraints were dictated by the finite migratory period of the salmon, an unscheduled mi l l shutdown in mid-August, and ambient water temperatures approaching 25 °C in mid-summer. It would be interesting to compare the statistically resolvable changes in Q for data sets where fish having a routine / H above 60 beats min" 1 were and were not included in analysis. Unfortunately, in this study all such trials were prematurely terminated, thus no comparison can be made. Nevertheless, Q has been measured in rainbow trout equipped with Doppler flowprobes at similarly warm water temperatures (16.5 -17.5 °C; Brodeur et al. 2001b). In this study, routine / H was 74.8 ± 13.7 beats min" 1 (mean ± S E M ) and data from all fish were analyzed (n = 8 per treatment group). N o significant changes in routine Q after 24-hour exposure to either P C P or T C G at their respective 25, 50, or 100% 96-hour LC50 concentrations were observed. One would expect an increase in Q following P C P exposure. It is a know uncoupler o f oxidative phosphorylation and has been reported to load the routine metabolic rate o f other fish species (Holmberg and Saunders, 1979; Farrell et al. 1998). Perhaps i f data from fish with the highest routine / V s were eliminated from analysis, significant differences in Q between treatment groups would have been observed. However, the inherent inaccuracies associated with Doppler flowprobes (Farrell and Jones, 1992; Thorarensen et al. 1996b), or non-robust statistical analyses may have also contributed to the irresolvable changes in Q . 105 Sudden and Extreme Increases in Routine fn In some fish in the present experiments, routine fu changed suddenly and extremely. Elevations in fu are mediated by increased adrenergic stimulation of the heart, while decreases are modulated by valgal release of acetylcholinesterase (Farrell and Jones, 1992). Increased activity and exposure to stressors such as predators (Johnsson et al. 2001; Cooke et al. 2003) and toxicants (Brodeur et a l l 999 ) can result in extreme and often rapid elevations in fu- However, upon the recommencement of normal conditions, fu generally falls to routine levels. When abrupt and extreme increases in fu were observed in the present study, neither the activity of the salmon nor the experimental or environmental conditions changed appreciably. In fact, care was taken to ensure that the experimental and environmental conditions remained constant throughout each trial. Furthermore, the sudden elevations in fu that were observed were not transient. Perhaps these extreme and sudden increases in / H can be attributed to the wi ld nature of the salmon and the cumulative stress placed on the fish by the high water temperatures, confinement to the respirometer, or the changing position of the flowprobe around the ventral aorta. If the flowprobe twisted while affixed around the ventral aorta, blood flow could have been impaired such that fu increased to compensate. Measures were taken to ensure that the flowprobe was affixed properly and securely around the blood vessel. The flow profile was checked for all fish while they were still on the surgery table and the sutures, which held the lead wire to the fish's body and ultimately kept the flowprobe in position, were examined for fastness following each trial. I f any sutures had come loose, data for that fish were either discarded or carefully scrutinized before use. However, this procedure did not completely eliminate the possibility that the probe could have twisted slightly during experimentation. 106 Decreases in Routine fn During Exposure to Dark Coloured Solutions Bathing fish with a high routine fu in dark coloured effluent or food colouring caused / H to decrease, which was likely mediated by an increase in vagal tone (Farrell and Jones, 1992). The reduction in fu in response to darkly coloured solutions coincides with the rapid bradycardia that accompanies the fear response in fish during simulated bird attacks (Johnsson et al. 2001; Cooke et al. 2003). However, simulated avian attacks are difficult to compare to the results of the present study. In such experiments, a model predator is briefly introduced into the line of sight of the fish. This induces almost immediate, yet short-term (< 1 minute) bradycardia followed by tachycardia, which is characteristic of the flight or fight response (Johnsson et al. 2001; Cooke et al. 2003). In the present study, fish were exposed to the stimulus (dark coloured solution) for several hours (effluent) or minutes (food colouring), and bradycardia largely persisted over the entire exposure periods (Fig. 2.4, 2.21). It may be possible that the fish in the present study maintained a fear response to the dark coloured solutions for the entire exposure periods. However, a potentially superior explanation for the observed bradycardia is that the darkly coloured solutions had a calming effect on / H due to the reduction of visual stimuli. Indeed, visual stimuli such as the lights being switched on and off and the movement of people or objects near the housing chamber elicit a tachycardic response in the toad (Bufo marinus; Dumsday, 1990). Similarly, light stimuli cause fu to increase in rainbow trout (Takahito and Hideo, 1997) and light of increasing intensity causes graded tachycardia in goldfish (Carassius auratus; Rooney and Lamin, 1986). Although the entire circumference o f the respirometer except the posterior 30 cm was covered with black plastic sheeting, there may have been faint shadows that fell on the respirometer or small particulates in the water flowing within the respirometer that acted as visual stimuli to the fish. Furthermore, seeing that they were trapped within the respirometer may have acted as sufficient visual stimuli to induce confinement stress and cause tachycardia in some of the salmon used in the present study. Certainly, it is possible 107 that confinement within the respirometer was stressful to such large fish of a feral nature whose migratory and spawning instincts were at their peak. Therefore, it is reasonable to deduce that bradycardia was observed in fish with a high routine fu following exposure to dark coloured effluent or food colouring because visual cues, which otherwise could have caused tachycardia, were largely reduced or eliminated. Conclusions This thesis addressed many common themes in the field o f fish physiology. It also generated some new information. Data were provided, for the first time, regarding the routine hematological profile of anaesthetised adult Atlantic salmon from the Exploits River. Additionally, the influence of temperature on fu and consequently, Q , was discussed in detail. It was also demonstrated that bathing fish in a dark coloured solution induces bradycardia. Furthermore, evidence was provided to suggest that excluding fish with an excessively high routine fu from data analysis may aid in the resolution of treatment-related effects on Q . This study also adds to the small body of existing literature discussing the physiological responses of fish to whole T M P effluent. It was suggested that free-swimming Atlantic salmon in the Exploits River are unlikely to experience sublethal stress due to the presence of T M P effluent, because only relatively high effluent concentrations elicited statistically significant responses. Exposure to effluent concentrations of 12% or higher caused relative Q to increase during routine and low-level activity, and exposure to 25% effluent impacted both routine and active Q and swimming performance. Nonetheless, effluent discharge would have to increase and/or river flow would have to decrease substantially for riverine effluent concentrations to reach 12%. •108 This is the first document to report that Q is sensitive to toxicant exposure and to evaluate the use Q as a biological monitoring tool of pollution stress. If Q technology continues to be demonstrated as a useful tool by which to assess changes in water quality and the effects of xenobiotic exposure on fish health, it could have applications as a biological monitoring tool in E E M programs, impact assessments, and additional physiological and toxicological research. With this in mind, the final chapter w i l l revisit the key features that biological monitoring tools should possess and discuss the features of the present study that have aided in determining i f Q technology has potential for use as a biomonitoring tool. 109 Chapter 3: New Information Supporting the use of Cardiac Output as a Biological Monitoring Tool In this final chapter, the characteristics of suitable biological monitoring systems that were outlined in Chapter 1 are revisited to highlight how the present study has aided in understanding i f Q technology has utility in biomonitoring applications. Criterion 1. The response variable in the chosen fish species changes over a wide range of developing toxic conditions and is sensitive to the pollutant(s) being screened for. The present research only examined how Q in adult Atlantic salmon responded to one toxic condition; whole T M P effluent. The results of this study indicate that Q is sensitive enough to screen for concentrations of whole T M P effluent >12%. Criterion 2. Fish restraint and the direct attachment of devices to it are not recommended. Fish underwent brief (c. 18 min) surgery to affix the Q flowprobe around the ventral aorta and suture its lead wire to the body of the fish. Restraining fish in a cylindrically shaped respirometer (80 cm long, 24 cm diameter) did not hinder the results o f the present study. Occasionally, the lead wire from the flowprobe became tangled around the fish. B y slackening the lead wire and stimulating the fish to move within the swimming chamber using changes in water velocity the situation was easily rectified with minimal disruption to the fish. Unless a fish was extremely restless within the respirometer, flowprobes remained secure around the ventral aorta and sutures held fast. 110 Criterion 3. The operation system should be continuous, automatic, cost effective, and require little maintenance. The equipment from Transonic Systems Inc. was more reliable and straightforward to use than the Doppler system. The signal strength of individual Doppler flow probes diminished over time, therefore many flow probes needed to be replaced during experimentation. Furthermore, Doppler flowprobes had to be calibrated in situ after each use. Absolute Q data were not collected in real-time but were calculated post-experimentation following flowprobe calibration. Consequently, Doppler flowprobes would not be well suited for use in biomonitoring. The major downfall to using the highest quality flowprobes and the flowmeter manufactured by Transonic Systems Inc. is that they are 20-times the cost of the Doppler probes and meter system. A s explorations into the utility of Q as a biomonitoring tool are in their infancy, how cost effective using this technology in full-scale biomonitoring systems is unknown. Furthermore, i f a Q tag reaches the market it w i l l l ikely be even more costly than the current system. Cardiac output data was recorded for long periods (up to 14 hours) using both the Doppler and Transonic system's interfaced with the same computer, software, and D A Q card. Therefore, in biomonitoring applications, Q data would need downloading only once or twice each day. The initial surgical attachment of the flowprobes would l ikely be the most time-consuming process. In this study, each surgery was approximately 18 minutes long. Fish recovered from surgery and were able to maintain an upright position within the respirometer approximately 10 minutes following surgery. However, post-surgery recovery times were observed to increase with water temperature, particularly when temperatures were >20 °C. A flowprobe affixed properly around the ventral aorta should remain in position for many months. However, Doppler flowprobes were observed to shift position and subsequently restrict flow through the ventral aorta more often than Transonic flowprobes. When using Q technology for i l l biomonitoring, a flowprobe that allows for vessel growth over time needs to be employed. Unlike Doppler flowprobes, Transonic flowprobes are designed for chronic applications and can be loosely fit around the ventral aorta when they are initially attached. Criterion 4. Simple and appropriate methods of data analysis and results interpretation should be available. The methods used to analyze data in this study may be appropriate for detecting subtle pollution-induced effects on Q . For example, treating fish with a routine fu above 60 beats min" 1 as outliers may have aided in resolving the between treatment effects of T M P effluent on Q . Furthermore, increases in Q may have been identified because the pre-exposure, routine Q value for each fish was applied as a covariate during data analysis. Additionally, since Q did not change within each treatment group during the exposure period, Q data for each treatment group were pooled. The analysis of pooled exposure data allowed for statistically significant increases in Q to be detected in fish exposed to at least 12% T M P effluent relative to the control and 6% exposure groups. With the exception of pooling cardiac data obtained during an exposure period, none of the above methods of data analysis have been used by Brodeur et al. (1999, 2001a, 2001b). It is interesting that they have never reported that Q is sensitive indicator of sublethal toxicant levels. However, this group always employed Doppler flowprobes, which, as previously discussed, are inherently less accurate and precise than Transonic flowprobes (Farrell and Jones, 1992; Thorarensen et al. 1996b). 112 Criterion 5 . The response variable must be continuously expressed and change rapidly and reliably in response to low toxicant concentrations. Cardiac output during routine and low-level activity increased in adult salmon exposed to 12%, 25%, or an increasing concentration gradient (0-25%) of T M P effluent relative to the control and 6% groups. Although routine Q did not respond to toxicant concentrations as low as 6%, it consistently increased when fish were exposed to effluent concentrations of 12% and greater. This variable responded reasonably quickly to T M P effluent, as significant increases in Q were detected during the 6-hour exposure period. Criterion 6. The normal range of the response variable under common conditions (diurnal fluctuations, temperature, changes in hardness, oxygen) should be determined prior to its use in biological monitoring to reduce the occurrence of 'false alarms'. The present work has yielded information about the normal range of Q during routine and prolonged exercise in fasting, wi ld , adult Atlantic salmon at water temperatures of approximately 18 °C. A s indicated above, this study has generated new information supporting the use of Q as a biomonitoring tool of aquatic pollution and has potentially aided in the development of a standardized biomonitoring protocol when Q is used as the response variable. Methods of data analysis that have not been employed before and high quality Q flowprobes from Transonic Systems Inc. l ikely contributed to the detection of toxicant-induced elevations in Q . Cardiac output may also be more sensitive to T M P effluent, a mixture of various toxic compounds, than to xenobiotics that have been the subject o f other experiments (Brodeur et al. 1999, 2001a, b). 113 Nonetheless, this is the first report to indicate that Q was sufficiently sensitive to detect sublethal stress in fish resulting from toxicant exposure. Since VO2 and EO2 were not measured in the present study, very little can be said regarding the potential for Q technology to accurately estimate metabolic rate in adult salmon. Nonetheless, changes in Q still shed light on the energetic challenges faced by fish in the presence of toxicants, not to mention elucidate changes in water quality. It is recommended that research concerning the response of Q to xenobiotics and its potential as a biomonitoring tool continue, keeping in mind the new information provided by the present study as well as the above criteria. Moreover, investigations pertaining to the reliability of Q as a predictor of metabolic rate should also persist, as Q could be a suitable estimate of VO2 in other fish species than cod. To expand on such databases of knowledge w i l l take years of research and countless experiments. However, i f Q proves to be sensitive to a wide range of xenobiotics and an accurate estimate of metabolic rate in numerous fish species, a tool that has the ability to concurrently monitor water quality and fish health w i l l have widespread uses in today's anthropogenically challenged world. "114 Bibliography Adams, S . M . and Breck, J.E. 1990. Bioenegetics. In, Methods for Fish Biology. (Eds.) C . B . Shreck and P .B . Moyle . American Fisheries Society, Bethsda, M D , U S A . Pp. 389-415. Alsop, D . H . , McGreer, J.C., McDonald, D . G . and Wood, C M . 1999. Costs of chronic waterborne zinc exposure and the consequences o f zinc acclimation on the gill/zinc interactions of rainbow trout in hard and soft water. Environmental Toxicology and Chemistry 18: 1014-1025. Anderson, W . G . , Booth, R., Beddow, T .A . , McKin l ey , R.S. , Finstad, B . , 0kland, F. and Scruton, D . 1998. Remote monitoring of heart rate as a measure o f recovery in angles Atlantic salmon, Salmo salar, L . Hydrobiologia 371/2: 233-240. Armstrong, J.D. 1986. Heart rate as an indicator of activity, metabolic rate, food intake and digestion in pike, Esox lucius. Journal of Fish Biology 29 Supp.A: 207-221. Armstrong, J.D. 1998. Relationships between heart rate and metabolic rate of pike: Integration of existing data. Journal of Fish Biology 52: 362-368. Axelsson, M . 1988. The importance of nervous and humoral mechanisms in the control of cardiac performance in the Atlantic cod, Gadus morhua, at rest and during non-exhaustive swimming. Journal of Experimental Biology 137: 287-303. Axelsson, M . and Fritsche, R. 1991. Effects of exercise and feeding on the gastrointestinal blood flow in the Atlantic cod, Gadus morhua. Journal of Experimental Biology 158: 81-198. Axelsson, M . , Thorarensen, H . , Nilsson, S. and Farrell, A . P . 2000. Gastrointestinal blood flow in the red Irish lord, HemUepidotus hemUepidotus: Long-term effects of feeding and adrenergic control. Journal of Comparative Physiology 170B: 145-152. B . A . R . Environmental. 1996. Environmental Effects Monitoring. Abitibi-Price, Inc., Grand Falls Division, N L . 83 p., 4 Appendices. Bamett, V . and Lewis, T. 1978. Outliers in Statistical Data. John Wi ley and Sons, Chichester, U K . Pp. 1-365. Barron, M . G . , Tarr, B . D . and Hayton, W . L . 1987. Temperature-dependence of cardiac output and regional blood flow in rainbow trout, Salmo gairdneri Richardson. Journal of Fish Biology 31: 735-744. Beamish, F .W. 1978. Swimming capacity. In, Fish Physiology VII . (Eds.) W.S . Hoar and D.J . Randall. Academic Press, N Y , U S A . Pp. 101-187. Bernatchez, L . and Dodson, J.J. 1985. Influence of temperature and current speed on the swimming capacity of lake whitefish (Coregonus clupeaformis) and ciso (C. artedii). Canadian Journal of Fisheries and Aquatic Sciences 42: 1522-1529. Biological Monitoring Inc. 2003. BioSensor® : Automated and Continuous Water Quality Monitoring. Accessed from the World Wide Web on A p r i l 14, 2003. http://www.biomon.com/biosenso.html. Black, E . C . 1955. Blood levels of hemoglobin and lactic acid in some freshwater fishes following exercise. Journal of the Fisheries Research Board o f Canada 12: 917-929. Blazka, P., Bolf, M . and Cepela, M . 1960. A new type o f respirometer for the determination of the metabolism of fish in an active state. Physiologia Bohemoslovenica 9:553-558. Boisclair, D . and Tang, M . 1993. Empirical analysis for the influence of swimming pattern on the net cost of swimming in fishes. Journal of Fish Biology 42: 169-183. Booth, R . K . , McKin l ey , R.S. , Okland, F. and Sisak, M . M . 1997. In situ measurement of swimming performance in wi ld Atlantic salmon (Salmo salar), using radio transmitted electromyogram ( E M G ) signals. Aquatic L iv ing Resources 10: 213-219. Bourgeois, C .E . , Murray, J. and Mercer, V . 1994. Status of the Exploits River stock of the Atlantic salmon (Salmo salar L.) in 1993. D F O Atlantic Fisheries. Research Document 94/73. 13 p. Brauner, C J . , Randall, D.J . , Neuman, J.F. and Thurston, R . V . 1994. The effect of exposure to 1,2,4,5-tetrachlorobenzene and the relationship between toxicant and oxygen uptake in rainbow trout (Oncorynchus mykiss) during exercise. Environmental Toxicology and Chemistry 13: 1813-1820. Brett, J.R., Hollands, M . and Alderice, D .F . 1958. The effect of temperature on the cruising speed of young sockeye and coho salmon. Journal o f the Fisheries Research Board of Canada 15: 587-605. Brett, J.R. 1964. The respiratory metabolism and swimming performance of young sockeye salmon. Journal of the Fisheries Research Board o f Canada 21: 1183-1226. Brett, J.R. 1965. The relation of size to rate of oxygen consumption and sustained swimming speed of sockeye salmon (Oncorhynchus nerka). Journal of the Fisheries Research Board of Canada 22: 1491-1501. Brett, J.R. and Glass, N . R . 1973. Metabolic rates and critical speeds o f Sockeye salmon (Oncorhynchus nerka) in relation to size and temperature. Journal of the Fisheries Research Board of Canada 21: 1183-1226. Briggs, C.T. and Post, J.R. 1997. Field metabolic rates of rainbow trout estimated using electromyogram telemetry. Journal of Fish Biology 51: 807-823. Brodeur, J.C., Ytestoyl, T., Finstad, B . and McKin ley , R.S . 1999. Increase o f heart rate without elevation of cardiac output in Atlantic salmon (Salmo salar) exposed to acidic water and aluminium. Canadian Journal of Fisheries and Aquatic Sciences 56:184-190. Brodeur, J .C., 0kland, F., Finstad, B . , Dixon, D . G . and M c K i n l e y , R .S . 2001a. Effects of subchronic exposure to aluminium in acidic water on bioenergetics of Atlantic salmon (Salmo salar). Ecotoxicology and Environmental Safety 49: 226-234. Brodeur, J.C., Dixon, D . G . and McKin l ey , R.S. 2001b. Inhibition of oxygen consumption by pentachlorphenol and tetrachloroguaiacol in rainbow trout (Oncorhynchus mykiss). Aquatic Toxicology 54: 143-148. Brodeur, J .C., Dixon, D . G . and McKin l ey , R.S . 2001c. Assessment of cardiac output as a predictor o f metabolic rate in rainbow trout. Journal of Fish Biology 58: 439-452. Brownlee, B . , Fox, M . E . , Strachan, W . M . and Joshi, S.R. 1977. Distribution of dehydroabietic acid in sediments adjacent to a kraft pulp and paper mi l l . Journal o f the Fisheries Research Board of Canada 34: 838-843. Butler, P.J. and Taylor, E . W . 1975. The effect of progressive hypoxia on respiration in the dogfish (Scyliorhinus canicula) at different seasonal temperatures. Journal of Experimental Biology 63: 117-130. Butler, P.J., Day, N . , and Namba, K . 1992. Interactive effects o f seasonal temperature and low p H on resting oxygen uptake and swimming performance of adult brown trout Salmo trutta. Journal of Experimental Biology 165: 195-212. Byrne, J . M . , Beamish, F . W . and Saunders, R . L . 1972. Influence o f salinity, temperature, and exercise on plasma osmolality and ionic concentrations in Atlantic salmon (Salmo salar). Journal of the Fisheries Research Board of Canada 29: 1217-1220. Calow, P. 1991. Physiological costs of combating chemical toxicants: Ecological implications. Comparative Biochemistry and Physiology 100C: 3-6. Cairns, J. and van der Schalie, W . H . 1980. Biological monitoring Part I: Early warning systems. Water Research 14: 1179-1196. Canadian Pulp and Paper Association. 2001. Statistical Bulletin. Canadian Pulp and Paper Association, Montreal, PQ , Canada. Canadian Pulp and Paper Association. 2002. Statistical Bulletin. Canadian Pulp and Paper Association, Montreal, PQ , Canada. Cech, J.J. 1990. Respirometry. In, Methods for Fish Biology. (Eds.) C . B . Shreck and P .B. Moyle . American Fisheries Society, Bethsda, M D , U S A . Pp. 335-362. Claireaux, G . , Webber, D . M . , Kerr, S.R. and Boutilier, R . G . 1995a. Physiology and behaviour of free-swimming Atlantic cod (Gadus morhua) facing fluctuating temperature conditions. Journal of Experimental Biology 198: 49-60. Claireaux, G . , Webber, D . M . , Kerr, S.R. and Boutilier, R . G . 1995b. Physiology and behaviour of free-swimming Atlantic cod (Gadus morhua) facing fluctuating salinity and oxygenation conditions. Journal of Experimental Biology 198: 61-69. Cooke, S.J., Schreer, J.F. and McKin l ey , R.S . 1999. Evaluating catch-and-release angling practices from the fish's perspective. (Ed.) T . V . Pitcher. Fisheries Centre Research Reports, V o l . 7, No. 2. Fisheries Centre, University of British Columbia, Vancouver, B C , Canada. Cooke, S.J., Steinmetz, J., Degner, J.F., Grant, E . C . and Phil l ip, D.P . 2003. Metabolic fright responses of different-sized largemouth bass (Micropterus salmoides) to two avian predators show variations in nonlethal energetic costs. Canadian Journal of Zoology 81: F699-709. Crocker, C .E . , Farrell, A . P . , Gamperl, A . K . and Cech, J.J. 2000. Cardiorespiratory responses of white sturgeon to environmental hypercapnia. American Journal of Physiology 279:R617-R628. Davie, P.S. and Forster, M . E . 1980. Cardiovascular responses to swimming in eels. Comparative Biochemistry and Physiology 67A: 367-373. Davis, J .C. 1973. Sublethal effects of bleached kraft pulp m i l l effluent on respiration and circulation in sockeye salmon (Oncorhynchus nerka). Journal of the Fisheries Research Board o f Canada 30: 369-377. Daxboeck, C. 1982. Effect of coronary artery ablation on exercise performance in Salmo gairdneri. Canadian Journal of Zoology 60: 375-381. Demers, E . M c K i n l e y , R.S. , Weatherly, A . H . and McQueen, D.J . 1996. Act ivi ty of largemouth 119 and smallmouth bass determined using electromyogram telemetry. Transactions of the American Fisheries Society 125: 434-439. Diamond, J., Collins, M . , and Gruber, D . 1988. A n overview of automated biomonitoring - past developments and future needs. In, Automated Biomonitoring: L iv ing Sensors as environmental Monitors. (Eds.) D.S. Gruber and J.M.-Diamond. El l i s Horwood, N Y , U S A . Pp. 23-39. Discovery Channel. 2003. Great Canadian Rivers Series: The Exploits River. Discovery Channel, January 2003. Duan, C. and Plisetskaya, E . M . 1993. Nutritional regulation of insulin-like growth factor-I m R N A expression in salmon tissues. Journal of Endocrinology 139: 243-252. Dumsday, B . 1990. Resting heart rate o f the toad Bufo marinus: A long-term study of individual differences and environmental influences. Physiological Zoology 63: 420-431. Environment Canada. 1992. Pulp and Paper Effluent Regulation SOR/92-269. Accessed from the World Wide Web on September 10, 2002. www3.ec.gc.ca/EnviroRegs/eng/SearchDetail.cfm?intCategory==ll&intact=1017. Farrell, A . P . , MacLeod, K . R . , Driedzic, W . R . and Wood, S. 1983. Cardiac performance during hypercapnic acidosis in the in situ perfused fish heart. Journal of Experimental Biology 107: 415-429. Farrell, A . P . , Small, S. and Graham, M . S . 1989. Effect of heart rate and hypoxia on the performance of a perfused trout heart. Canadian Journal of Zoology 67: 274-280. Farrell, A . P . 1991. From hagfish to tuna: A perspective on cardiac function in fish. Physiological Zoology 64: 1137-1164. Farrell, A . P . and Jones, D .R. 1992. The Heart. In, Fish Physiology, Volume X I I A . (Eds.) W.S . Hoar and D.J . Randall. Academic Press Inc., N Y , U S A . Pp 1-88. Farrell, A . P . , Gamperl, A . K . , Hicks, J . M . , Shiels, H . A . and Jain, K . E . 1996. Maximum cardiac 120 performance o f rainbow trout (Oncorhynchus mykiss) at temperatures approaching their upper lethal limit. Journal of Experimental Biology 199: 663-672. Farrell, A . P . , Gamperl, A . K . and Birtwell , I .K. 1998. Prolonged swimming, recovery and repeat swimming performance of mature sockeye salmon Oncorhynchus nerka exposed to moderate hypoxia and pentachlorophenol. Journal of Experimental Biology 201:2183-2193. Farrell, A . P . , Thorarensen, H . , Axelsson, M . , Crocker, C .E . , Gamperl, A . K . and Cech, J.J. 2001. Gut blood flow in fish during exercise and severe hypercapnia. Comparative Biochemistry and Physiology 128A: 551-563. Farrell, A . P . , Lee, C . G . , Tierney, K . , Hodaly, A . , Clutterham, S., Healey, M . , Hinch, S. and Lotto, A . 2003. Field-based measurements of oxygen uptake and swimming performance with adult Pacific salmon using a mobile respirometer swim tunnel. Journal of Fish Biology 62: 64-84. FishDoc: The Home of Fish Health. 2003. Malachite Green and Formalin: A Good General-Purpose Anti-Parasite Treatment. Accessed from the Wor ld Wide Web on Apr i l 24, 2003. http://www.fishdoc.co.uk/treatments/malachite.htm. Food and Agriculture Organization of the United Nations. 1999. Pulp and paper capacities survey 1998-2003. 160 p., 5 Appendices. French, C.J . , Mommsen, T.P. and Hochachka, P .W. 1981. Amino acid utilization in isolated hepatocytes from rainbow trout. European Journal of Biochemistry 113:311-317. French, C.J . , Hochachka, P .W. and Mommsen, T.P. 1983. Metabolic organization of liver during spawning migration of sockeye salmon. American Journal of Physiology 245: R827-R830. 121 Fry, F .E . 1971. The Effect of Environmental Factors on the Physiology of Fish. In, Fish Physiology, V o l . V I : Environmental Relations and Behaviour. (Eds.) W.S . Hoar and D J . Randall. Academic Press, London, U K . Pp. 1-98. Gallaugher, P. and Farrell, A . P . 1998. Hematocrit and Blood OxygenTCarrying Capacity. In, Fish Respiration, Fish Physiology Series V X V I I . (Eds.) S.F. Perry and B . Tufts. Academic Press, San Diego, C A , U S A . Pp. 185-219. Gibbons, J.S., Kovacs, T .G . , Voss, R . H . , O'Connor, B . L and Diorica, J .G. 1992. Appraisal of the effect of biological treatment on the environmental quality of high-yield mechanical pulping effluents; Water Research 26: 1425-1432. Grottum, J .A. and Sigholt, T. 1998. A model for oxygen consumption of Atlantic salmon (Salmo salar) based on measurements of individual fish in a tunnel respirometer. Aquaculture Engineering 17: 241 -251. Gruber, D . 1988. A historical perspective. In, Automated Biomonitoring: L iv ing Sensors as Environmental Monitors. (Eds.) D.S. Gruber and J . M . Diamond. El l i s Horwood, N Y , U S A . Pp. 15-22. Hinch, S.G., Diewert, R . E . , Lissimore, T.J., Prince, A . M . , Healy, M . C . and Henderson, M . A . 1996. Use of electromyogram telemetry to assess difficult passage areas for river-migrating adult sockeye salmon. Transactions of the American Fisheries Society 125: 253-260. Hinch, S. 2002. Salmon at their southern edge: Current challenges to survival and prognosis for long-term sustainability. Jubilee Lecture Series, Faculty o f Forestry, University of British Columbia, Vancouver, B C , Canada. 18 p. Holmberg, B . and Saunders, R . L . 1979. The effects of pentachlorophenol on swimming performance and oxygen consumption in the American eel (Anguilla rostrata) Rapports et Proces-Verbaux des Reunions Conseil International pour 1' Exploration de la Mer 174: 144-149. Howard, T .E. 1975. Swimming performance o f juvenile coho salmon (Oncorhynchus kisutch) exposed to bleached kraft pulpmill effluent. Journal of the Fisheries Research Board of Canada 32: 789-793. Iwama, G . K . , Greer, G . L . and Larkin, P . A . 1976. Changes in some hematological characteristics of coho salmon (Oncorhynchus kisutch) in response to acute exposure to dehydroabietic acid ( D H A A ) at different exercise levels. Journal of the Fisheries Research Board of Canada 33: 285-289. Jain, K . E . , Birtwell , I .K. and Farrell, A . P . 1998. Repeat swimming performance of mature sockeye salmon following a brief recovery period: A proposed measure of fish health and water quality. Canadian Journal of Zoology 76: 1488-1496. Janz, D . M . , Farrell, A . P . , Morgan, J.D. and Vigers, G . A . 1991. Acute physiological stress response of juvenile coho salmo (Oncorhynchus kisutch) to sublethal concentrations of Garlon 4, Garlon 3 A , and Vis ion herbicides. Environmental Toxicology and Chemistry 10: 81-90. Johansen, J .A. , Kennedy, C.J . , Sweeting, R . M . , Farrell, A . P . and M c K e o w n , B . A . 1994. Sublethal effects of tetrachloroguaiacol on juvenile rainbow trout, Oncorhynchus mykiss, following acute and chronic exposure. Canadian Journal of Fisheries and Aquatic Sciences 51: 1967-1974. Johnsen, K . , Mattsson, K . , Tana, J., Stuthridge, T.R., Hemming, J. and Lehtinen, K . J . 1995. Uptake and elimination of resin acids and physiological responses in rainbow trout exposed to total mi l l effluent from an integrated newsprint mi l l . Environmental Toxicology and Chemistry 14: 1561-1568. Johnsen, K . , Tana, J., Lehtinen, K . , Stuthridge, T., Mattsson, K . , Hemming, J. and Carlberg, G .E . 1998. Experimental field exposure of brown trout to river water receiving effluent from 123 an integrated newsprint mi l l . Ecotoxicology and Environmental Safety 40: 184-193. Johnstone, A . D . , Lucas, M . C . , Boylan, P. and Carter, T. 1992. Telemetry of tail beat frequency in Atlantic salmon, Salmo salar L . , during spawning . In, Wildl i fe telemetry: Remote monitoring and tracking of animals. (Eds.) I . M . Priede and S . M . Swift. El l i s Horwood, N Y , U S A . Pp. 456-465 Jones, J.R. 1947. The oxygen consumption of Gasterosteus aculeatus L . in toxic solutions. Journal of Experimental Biology 23: 298-311. Jones, D .R. and Randall, D.J . 1978. The respiratory and Circulatory Systems During Exercise. In, Fish Physiology, V o l . VII . (Eds.) W . S. Hoar and D.J . Randall. Academic Press, N Y , U S A . Pp. 425-501. Jonssen, N . , Jonssen, B . , and Hansen, L .P . 1991. Energetic costs of spawning in male and female Atlantic salmon (Salmo salar). Journal of Fish Biology 39: 739-744. Johnsson, J.I, Hojesjo, J. and Fleming, L A . 2001. Behavioural and heart rate responses to predation risk in w i ld and domesticated Atlantic salmon. Canadian Journal of Fisheries and Aquatic Sciences 58: 788-794. Kaseloo, P .A . , Weatherly, A . H . , Lotimer, J. and Farina, M . D . 1992. A biotelemetry system recording fish activity. Journal of Fish Biology 40: 165-179. Keen, J.E. and Farrel, A . P . 1994. Maximum prolonged swimming speed and maximum cardiac performance of rainbow trout, Oncorhynchus mykiss, acclimated to two different water temperatures. Comparative Biochemistry and Physiology 108A: 287-295. Kennedy, C. J., Sweeting, R . M . , Johansen, J .A. , Farrell, A . P . and M c K e o w n , B . A . 1995. Acute effects of chlorinated resin acid exposure on juvenile rainbow trout, Oncorhynchus mykiss. Environmental Toxicology and Chemistry 14: 977-982. Kicenuik, J .W. and Jones, D .R. 1977. The oxygen transport system in trout (Salmo gairdneri) during sustained exercise. Journal of Experimental Biology 69: 247-260. Klaverkamp, J.F. 1982. The physiological pharmacology and toxicology o f fishcardiovascular systems. In, Aquatic Toxicology, V o l . 1. (Ed.) L J . Weber. Chapter Raven Press, N Y , U S A . Pp. 1-54. Kloepper-Sams, P.J. and Benton, E . 1994. Exposure of fish to biologically treated bleached kraft effluent. 2. Induction of hepatic cytochrome P4501A in Rocky Mountain whitefish (Prosopium williansoni) and other species. Environmental Toxicology and Chemistry 13: 1483-1496. Kolok, A . S . , Spooner, R . M . and Farrell, A . P . 1993. The effect o f exercise on the cardiac output and blood flow distribution of the largescale sucker Catostomus macrocheilus. Journal of Experimental Biology 183: 301-321. Kovacs, T . G . and Voss, R . H . 1992. Biological and chemical characterization of newsprint/ specialty m i l l effluents. Water Research 26: 771-780. Kovacs, T .G . , Gibbons, J.S., Martel, P .H . , O'Connor, B . L and Voss, R . H . 1995. Effects of a secondary-treated thermomechanical pulp m i l l effluent on aquatic organisms as assessed by short- and long-term laboratory tests. Journal of Toxicology and Environmental Health 44: 485-502. Kruzynski , G . H . 1979. Some effects of dehydroabietic acid ( D H A ) on hydromineral balance and other physiological parameters in juvenile sockeye salmon Oncorhynchus nerka. PhD. Thesis, University of British Columbia, Vancouver, B C , Canada. Laitinen, M . and Valtonen, T. 1995. Cardiovascular, ventilatory and hematological responses of brown trout (Salmo trutta L.) to the combined effects of acidity and aluminum in humic water at winter temperatures. Aquatic Toxicology 31: 99-112. Larsson, A . , Andersson, T., Forlin, L . and Hardig, J. 1988. Physiological disturbances in fish exposed to bleached kraft mi l l effluents. Water Science and Technology 20: 67-76. Leach, J . M . and Thakore, A . N . 1975. Isolation and identification of constituents toxic to juvenile rainbow trout (Salmo gairdneri) in caustic extraction effluents from kraft pulpmill bleach plants. Journal of the Fisheries Research Board of Canada 32: 1249-1257. Leach, J . M . and Thakore, A . N . 1976. Compounds toxic in mechanical pulping effluents. Tappi 59: 129-132. Lee, C G . 2002. Swimming performance, metabolic rate and recovery of adult Pacific salmon (Oncorhynchus sp.). M S c . Thesis, Simon Fraser University, Burnaby, B C , Canada. Lemly, A . D . 1993. Metabolic stress during winter increases the toxicity of selenium to fish. Aquatic Toxicology 27: 133-158. Lowry, O .H. , Rosebrough, N . J . , Fair, A . L . and Randall, R .J . 1951. Protein measurement with the folin phenol reagent. Journal of Biological Chemistry 193: 265-275. Lucas, M . C Priede, I.G., Armstrong, J.D. Gindy, A . N . and De Vera, L . 1991. Direct measurements of metabolism, activity and feeding behaviour o f pike, Esox lucius L . , in the wi ld , by the use of heart rate telemetry. Journal of Fish Biology 39: 325-345. Lucas, M . C , Johnstone, A . D . and Priede, I.G. 1993. Use of physiological telemetry as a method of estimating metabolism of fish in the natural environment. Transactions of the American Fisheries Society 122: 822-833. Lucas, M . C . 1994. Heart rate as an indicator of metabolic rate and activity in adult Atlantic slamon, Salmo salar. Journal of Fish Biology 44: 889-903. MacLeod, J .C. and Smith L . L . 1966. Effect of pulpwood fibre on the oxygen consumption and swimming endurance of the fathead minow, Pimephales promelas. Transactions of the American Fisheries Society 95: 71-84. Martel, P .H . , Kovacs, T .G . , O'Connor, B . L and Voss, R . H . 1994. A survey of pulp and paper m i l l effluents for their potential to induce mixed function oxidase enzyme activity in fish. Water Research 28: 1835-1844. 126 Martel, P . H . and Kovacs, T. 1997. A comparison of the potential of primary and secondary treated pulp and paper m i l l effluents to induce mixed function oxygenase (MFO) activity in fish. Water Research 31: 1482-1488. Martel, P .H . , Kovaks, T .G . , O'Connor, B. I . and Voss, R . H . 1997. Source and identity of compounds in a thermomechanical pulp m i l l effluent inducing hepatic mixed-function oxygenase activity in fish. Environmental Toxicology and Chemistry 16: 2375-2383. Mazeaud, M . M . , Mazeaud, F. and Donaldson, E . M . 1977. Primary and secondary effects of stress in fish: Some new data with a general review. Transactions of the American Fisheries Society 106: 201-212. McKin l ey , R.S . and Power, G . 1992. Measurement of activity and oxygen consumption for adult lake sturgeon (Acipenser fulvescens) in the wi ld using radio-transmitted E M G signals. In, Wildl ife Telemetry: Remote monitoring and tracking o f animals. (Eds.) I .M. Priede and S . M . Swift. Ellis-Horwood, N Y , U S A . Pp. 307-318. McLeay, D.J . and Brown, D . A . 1975. Effects of acute exposure to bleached kraft pulpmill effluent on carbohydrate metabolism of juvenile coho salmon (Oncorhynchus kisutch) during rest and exercise. Journal of the Fisheries Research Board of Canada 32: 753-760. McLeay, D .J . and Brown, D . A . 1979. Stress and chronic effects o f untreated and treated bleached kraft pulpmill effluent on the biochemistry and stamina of juvenile coho salmon (Oncorhynchus kisutch ). Journal of the Fisheries Research Board of Canada 36: 1049-1059. Mi l l igan , C L . 1996. Metabolic recovery from exhaustive exercise in rainbow trout. Comparative Biochemistry and Physiology 113A: 51-60. Neumann, P. Holeton, G.F. and Heisler, N . 1983. Cardiac output and regional blood flow in gills and muscles after exhaustive exercise in rainbow trout (Salmo gairdneri). Journal of Experimental Biology 105: 1-14. Nikinmaa, M . and Oikari , A . 1982. Physiological changes in trout (Salmo gairdneri) during short-term exposure to resin acids and during recovery. Toxicological Letters 14: 103-110. O'Connor, B . L , Kovacs, T .G . and Voss, R . H . 1992. The effect o f wood species composition on the toxicity of simulated mechanical pulping effluents. Environmental Toxicology and Chemistry 11: 1259-1270. O'Hara, J. 1971. Alterations in oxygen consumption by bluegills exposed to sublethal treatment with copper. Water Research 5: 321-327. Oikari, A . , Lonn, M . , Castren, T., Nakari, B . , Snickars-Nikinmaa, H . and Virtanen, E . 1983. Toxicological effects of dehydroabietic acid ( D H A A ) on the trout, Salmo gairdneri Richardson, in fresh water. Water Research 17: 81-89. O 'Rei l ly , F . L . 1959. A preliminary report on the Exploits River. Department of Fisheries of Canada. St. John's, N L , Canada. 67 p. Perry, S.F. and McKendry, J .E. 2001. The relative roles of external and internal CO2 versus H + in eliciting the cardiorespiratory responses of Salmo salar and Squalus acanthias to hypercarbia. Journal of Experimental Biology 204: 3963-3971. Porter, E . L . , Payne, J.F., Kiceniuk, J., Fancey, L . and Melv in , W . 1989. Assessment of the potential for mixed-function oxygenase enzyme induction in the extrahepatic tissues of cunners (Tautogolabrus adspersus) during reproduction. Marine Environmental Research 28: 117-121. Portt and Associates. 1999. Pre-design and Study Design - Cycle 2 E E M . Abitibi-Consolidated Inc., Grand Falls Division, N L , Canada. 78 p. Portt and Associates. 2000. Cycle 2 Environmental Effects Monitoring - Interpretive Report Abi t ib i Consolidated Inc., Grand Falls Division, N L , Canada. 36 p. Priede, I.G. 1977. Natural selection for energetic efficiency and the relationship between 128 activity levels and mortality. Nature 267: 610-611. Priede, I.G. and Tytler, P. 1977. Heart rate as a measure of metabolic rate in teleost fishes: Salmo gairdneri, Salmo trutta, and Gadus morhua. Journal of Fish Biology 10: 231-242. Randall, D.J . 1968. Functional morphology of the heart in fishes. American Zoologist 8: 179-189. Randall, D.J . 1970. The Circulatory System. In, Fish Physiology, V o l . I V . (Eds.) W . S. Hoar and D.J . Randall. Academic Press, N Y , U S A . Pp. 133-168. Randall, R . G . , O'Connel l , M . F . and Chadwick, E . M . 1989. Fish production in two large Atlantic coast rivers: Miramichi and Exploits. In, Proceedings of the International Large River Symposium. (Ed.) D.P. Dodge. Canadian Special Publications in Fisheries and Aquatic Sciences 106: 292-308. Rogers, S.C. and Weatherly, A . H . 1983. The use of opercular muscle electromyograms as an indicator of the metabolic costs of fish activity in rainbow trout, Salmo gairdneri Richardson, as determined by radiotelemetry. Journal of Fish Biology 23: 535-547. Rooney, D.J . and Laming, P.R. 1986. Cardiac and ventilatory arousal responses and their habituation in goldfish: Effects of the intensity of the eliciting stimulus. Physiology and Behaviour 37: 11-14. Ross, L . G . , Watts, W . and Young, A . H . 1981. A n ultrasonic biotelemetry system for continuous monitoring of tail-beat rate from free-swimming fish. Journal of Fish Biology 18: 479-490. Sandnes, K . , L ie , O. and Waagbo, R. 1988. Normal ranges of some blood chemistry parameters in adult farmed Atlantic salmon, Salmo salar. Journal of Fish Biology 32: 129-136. Scharold, J., La i , N . C . , Lowel l , W . R . and Graham, J .B. 1989. Metabolic rate, heart rate and tailbeat frequency during sustained swimming in leopard shark, Triakis semifasciata. 129 Experimental Biology 48: 223-230. Scharold, J. and Gruber, S.H. 1991. Telemetered heart rate as a measure of metabolic rate in the lemon shark, Negaprion brevirostris. Copeia 1991: 942-953. Seyle, H . 1973. The evolution of the stress concept. American Scientist 61: 692-699. Sheridan, M . A . and Mommsen, T.P. 1991. Effects of nutritional state on in vivo l ipid and carbohydrate metabolism of coho salmon, Oncorhynchus kisutch. General and Comparative Endocrinology 81: 473-483. Sinnott, A . and Mackey, J. 1997. The effect of sulphite-treated thermomechanical pulp (TMP) m i l l effluent on the mixed-function oxygenase (MFO) system in Atlantic salmon {Salmo salar) liver. Biochemistry 4211 Research Paper, Department of Biochemistry, Memorial University, N L , Canada. Smit, FL, Amelink-Koutstall, J . M . , Vijverberg, J. and Vaupel-Klein, J .C. 1971. Oxygen consumption and efficiency of swimming goldfish. Comparative Biochemistry and Physiology 39A: 1-28. Soivio, A . and Oikari , A . 1976. Hematological effects of stress on a teleost, Esox leucius L . Journal o f Fish Biology 8: 397-411. Stasko, A . B . and Horrall , R . M . 1976. Method of counting tailbeats of free-swimming fish by ultrasonic telemetry techniques. Journal of the Fisheries Research Board o f Canada 33: 2596-2598. Stecyk, J .A. and Farrell, A . P . 2002. Cardiorespiratory responses o f the common carp (Cyprinus carpio) to severe hypoxia at three acclimation temperatures. Journal of Experimental Biology 205: 759-768. Stevens, E . D . and Randall, D.J . 1967. Changes in gas concentrations in blood and water during moderate swimming activity in rainbow trout. Journal of Experimental Biology 46: 329-337. 130 Sureau, D . and Lagardere, J.P. 1991. Coupling of heart rate and locomotor activity in sole Solea solea (L.), and bass, Dicentrarchus labrax (L.), in their natural environment by using ultrasonic telemetry. Journal of Fish Biology 38: 399-405. Takahito, K . and Hideo, S. 1997. Heart rate change of rainbow trout under long period illumination and sound stimuli. Nippon Suisan Gakkaishi-Bulletin of the Japanese Society of Scientific Fisheries 63: 905-911. Takeda, T. 1993. Effects of exercise-stress on ventilation, cardiac output and blood respiratory parameters in the carp, Cyprinus carpio. Comparative Biochemistry and Physiology 106A: 277-283. Tana, J. 1988. Sublethal effects of chlorinated phenols and resin acids on rainbow trout (Salmo gairdneri). Water Science and Technology 20: 85-88. Taylor, S.E., Egginton, S. and Taylor, E . W . 1996. Seasonal temperature acclimatisation of rainbow trout: Cardiovascular and morphometric influences on maximal sustainable exercise level. Journal of Experimental Biology 199: 835-845. Thorarensen, H . , Gallaugher, P.E. , Kiessling, A . K . and Farrell, A . P . 1993. Intestinal blood flow in swimming chinook salmon Oncorhynchus tshawytscha and the effects of heamatocrit on blood flow distribution. Journal of Experimental Biology 179: 115-129. Thorarensen, H . , Gallaugher, P .E . and Farrell, A . P . 1996a. The limitations of heart rate as a predictor of metabolic rate in fish. Journal of Fish Biology 49:226-236. Thorarensen, H . , Gallaugher, P. and Farrell, A . P . 1996b. Cardiac output in swimming rainbow trout, Oncorhynchus mykiss, acclimated to seawater. Physiological Zoology 69:139-153. Vavasour, G.R. and Blair , A . A . 1955. Pollution survey of Humber A r m and Exploits River. Journal of the Fisheries Research Board 12: 682-697. Vijayan, M . M . , Ballantyne, J.S. and Letherland, J.F. 1991. Cortisol-induced changes in some aspects of the intermediary metabolism of Salvelinus fontinalis. General and Comparative Endocrinology 82: 476-486. Vijayan, M . M . and Leatherland, J.F. 1992. In vivo effects of the steroid analogue RU486 on some aspects of intermediary and thyroid metabolism of brook charr, Salvelinus fontinalis. Journal of Experimental Zoology 263: 265-271. Vijayan, M . M . , Maule, A . G . , Schreck, C . B . and Moon , T .W. 1993. Hormonal control of hepatic glycogen metabolism in food-deprived, continuously swimming coho salmon (Oncorhynchus kisutch). Canadian Journal of Fisheries and Aquatic Sciences 50: 1676-1682. Wagner, G . N . , McKin l ey , R.S. , Bjorn, P. A . and Finstad, B . 2003a. Physiological impact of sea lice on swimming performance of Atlantic salmon. Journal of Fish Biology, in press. Wagner, G . N . , McKin l ey , R.S. , Bjorn, P . A . and Finstad, B . 2003b. Short-term freshwater exposure benefits sea lice-infected Atlantic salmon. Journal of Fish Biology, in review. Weatherly, A . H . , Rogers, S.C., Pinock, D . G . and Patch, J.R. 1982. Oxygen consumption of active trout, Salmo gairdneri Richardson, derived from electromyograms obtained by radiotelemetry. Journal of Fish Biology 20: 479-489. Webb, P .W. 1971. The swimming energetics of trout II: Oxygen consumption and swimming efficiency. Journal of Experimental Biology 55: 489-520. Webber, D . M . , Boutilier, R . G . and Kerry, S.R. 1998. Cardiac output as a predictor of metabolic rate in cod, Gadus morhua. Journal of Experimental Biology 201: 2779-2789. Wedemeyer, G . A . , Barton, B . A . and McLeay , D.J . 1990. Stress and acclimation. In, Methods for Fish Biology. (Eds.) C . B . Schreck and P .B . Moyle . American Fisheries Society, Bethsda, M D , U S A . Pp. 451-488. Wendelaar Bonga, S.E. 1997. The stress response in fish. The American Physiological Society, July 1997. 77:591-525. 132 Widdows, J. and Donkin, P. 1991. Role of physiological energetics in ecotoxicology. Comparative Biochemistry and Physiology 100C: 69-75. Wong, A . , Breck, D . and Costantino, J. 1978. T M P effluents, a review. Tappi Journal 61: 19-22. Yang, R. and Randall, D .J . 1997. Oxygen consumption of juvenile rainbow trout (Oncorhynchus mykiss) exposed to sublethal concentrations of 1,2,4,5-tetrachlorobenzene and tetrachloroguaiacol. Bulletin of Environmental Contamination and Toxicology 59: 479-485. Young, A . H . , Tytler, P., Holliday, F . G . and MacFarlane, A . 1972. A small sonic tag for measurement of locomotor activity in fish. Journal of Fish Biology 4: 57-65. Ytrestoyl, T., Finstad, B . , and McKin ley , R.S. 2001. Swimming performance and blood chemistry in Atlantic salmon (Salmo salar) spawners exposed to acid river water with elevated aluminium concentrations. Journal o f Fish Bio logy 58: 1025-1038. 133 Appendix I: Effluent Characteristics for Various Sample Locations and Dates Effluent Characteristics - Abitibi Consolidated, Grand Falls-Windsor, NL 13-Dec-01 Variable Units WQG" MC b wc c Cell l d Cell 2d Cell 3d ASBe Nitrate + Nitrite mg/1 NE ND ND 0.1 0.1 0.1 ND Nitrate mg/1 Narrativer ND ND 0.1 0.1 0.1 ND Nitrite mg/1 0.06 ND 0.0 0.0 ND 0.0 ND Ammonia mg/1 Narrativer ND ND ND 0.1 ND ND Colour TCU Narrativer 68.0 110.0 120.0 100.0 130.0 100.0 Sodium . mg/1 NE 77.4 89.0 94.9 94.3 86.4 88.5 Potassium mg/1 NE 14.4 14.1 14.7 14.3 14.4 14.1 Calcium mg/1 NE 16.2 13.9 15.3 16.1 14.5 14.7 Magnesium mg/1 NE 3.1 2.9 3.1 3.2 3.0 3.0 Alkalinity mg/1 NE 24.0 31.0 38.0 34.0 37.0 32.0 Sulfate mg/1 NE 158.0 161.0 187.0 185.0 176.0 167.0 Chloride mg/1 NE 11.5 9.8 10.5 9.7 9.3 9.5 Reactive silica mg/1 NE 2.8 2.6 2.8 2.8 2.8 2.7 Orthophospate mg/1 NE 0.1 1.4 1.3 1.4 1.4 1.4 Phosphorus mg/1 NE 0.4 2.0 2.3 2.4 2.0 2.2 DOC mg/1 NE 563.0 272.0 283.0 282.0 265.0 241.0 Turbidity NTU Narrative' 149.0 120.0 176.0 253.0 116.0 127.0 Conductivity uS/cm NE 481.0 490.0 505.0 496.0 493.0 494.0 pH 6.5-9.0 4.6 4.8 4.9 4.8 4.9 4.7 Hardness (CaC03) mg/1 NE 53.2 46.6 51.0 53.4 48.6 49.1 Bicarbonate (CaCOs) mg/1 NE 24.0 31.0 38.0 34.0 37.0 32.0 Carbonate (CaC03) mg/1 NE ND ND ND ND ND ND TDS mg/1 NE 2600.0 700.0 550.0 510.0 460.0 480.0 Phenolics mg/1 0.001 Sulphide mg/1 0.05 METALS Aluminum ug/1 5.0-100 90.0 120.0 130.0 170.0 120.0 140.0 Antimony ug/1 NE ND ND ND ND ND ND Arsenic ug/1 5.0 ND ND ND ND ND ND Barium ug/1 NE 170.0 120.0 130.0 . 150.0 130.0. 120.0 Beryllium ug/1 NE ND ND ND ND ND ND Bismuth ug/1 NE ND ND ND ND ND ND Boron ug/1 NE 27.0 30.0 30.0 31.0 30.0 29.0 Cadmium ug/1 0.017 ND ND ND ND ND ND Chromium ug/1 NE ND ND ND 2.0 ND ND Cobalt ug/1 NE ND ' ND ND ND ND ND Copper ug/1 2.0-4.0 4.0 2.0 2.0 3.0 2.0 3.0 Iron ug/1 300 250.0 730.0 710.0 980.0 740.0 620.0 Lead ug/1 1.0-7.0 1.0 ND 1.0 ND ND ND Manganese ug/1 NE 2800.0 2500.0 2800.0 2900.0 2600.0 2400.0 Molybdenum ug/1 73 ND ND ND ND ND ND Nickel ug/1 25-150 3.0 ND ND 2.0 ND ND Selenium ug/1 1.0 ND ND ND ND ND ND Silver ug/1 0.1 ND ND ND ND ND ND 134 Variable Units WQG" MC wc Cell l b Cell2b Cell 3b ASBC Strontium ug/1 NE 46.0 42.0 46.0 48.0 43.0 43.0 Thallium ug/1 0.8 0.1 ND ND ND ND ND Tin ug/1 NE ND ND ND ND ND ND Titanium ug/1 NE 3.0 3.0 3.0 3.0 3.0 4.0 Uranium ug/1 NE ND 0.4 0.6 0.8 0.5 0.7 Vanadium ug/1 NE : ND ND ND 2.0 ND 2.0 Zinc ug/1 30 100.0 16.0 30.0 32.0 15.0 36.0 RESIN & FATTY ACIDS MDLS Abietic acid mg/1 0.005 1.092 0.640 7.317 0.791 . 0.470 0.689 Arachidic acid mg/1 0.005 Behenic acid Chlorodehydroabietic acid mg/1 mg/1 0.005 0.005 0.017 ND ND ; ND ND ND Dehydroabietic acid mg/1 0.005 0.938 0.650 3.452 0.844 0.673 0.849 Dichlorodehydroabietic mg/1 0.005 ND ND ND ND ND ND 9,10-dichlorostearic acid mg/1 0.005 ND ND ND ND ND ND Isopimaric acid mg/1 0.005 0.192 0.109 1.428 0.163 0.110 0.123 Laurie acid mg/1 0.005 Levopimaric acid mg/1 0.005 Lignoceric acid mg/1 0.005 Linoleic acid mg/1 0.005 0.281 0.054 0.242 0.099 0.068 0.092 Linolenic acid mg/1 0.005 ND ND ND ND ND ND Myristic acid mg/1 0.005 Neoabietic acid mg/1 0.005 0.345 0.140 2.340 0.102 0.052 0.117 Oleic acid mg/1 0.005 0.150 0.033 0.090 0.050 0.035 0.031 Palmitric acid . mg/' 0.005 0.079 0.037 0.062 0.060 0.040 0.020 Palmitoleic mg/1 0.005 0.025 0.024 0.055 0.040 0.028 0.022 Palustric acid mg/1 0.005 0.718 0.185 4.002 0.233 0.121 0.179 Pimaric acid mg/1 0.005 0.079 0.032 0.545 0.055 0.034 0.039 Sandaraco Pimaric acid mg/1 0.005 0.195 0.058 1.095 0.096 0.056 0.065 Stearic acid mg/1 0.005 TOTAL RFA mg/1 4.111 1.962 20.628 2.533 1.687 2.226 135 Effluent Characteristics - Abitibi-Consolidated, Grand Falls-Windsor, NL 20 -Dec -01 Variable Units WQG" MC b wcc Cell l d Cell 2d Cell 3d ASBe Nitrate + Nitrite mg/1 NE 0.1 0.1 ND 0.1 0.1 0.1 Nitrate mg/1 Narrativer 0.1 0.1 ND 0.1 0.1 0.1 Nitrite mg/1 0.C6 ND 0.0 0.0 0.0 0.0 0.0 Ammonia mg/1 Narrativef ND 0.2 ND 0.1 ND 0.1 Colour TCU Narrativef 84.0 160.0 160.0 200.0 180.0 170.0 Sodium mg/1 NE 119.0 12.6 116.0 103.0 112.0 110.0 Potassium mg/1 . NE 15.8 20.4 16.1 14.7 15.2 15.3 Calcium mg/1 NE 20.6 10.9 15.6 14.1 14.9 14.3 Magnesium mg/1 NE 4.0 2.8 3.2 2.9 3.1 3.0 Alkalinity mg/1 NE 12.0 12:0 33.0 36.0 34.0 36.0 Sulfate mg/1 NE 284.0 36.0 226.0 230.0 238.0 209.0 Chloride mg/1 NE 19.1 27.2 13.1 12.3 11.8 18.6 Reactive silica mg/1 NE 3.0 2.6 2.0 1.9 2.0 2.2 Orthophospate mg/1 NE 0.3 0.5 0.2 0.2 0.4 0.9 Phosphorus mg/1 NE 0.6 1.6 0.8 0.8 1.0 1.3 DOC mg/1 NE 545.0 453.0 337.0 321.0 15.8 275.0 Turbidity NTU Narrativer 107.0 635.0 100.0 78.5 80.5 82.5 Conductivity uS/cm NE 676.0 171.0 593.0 584.0 575.0 557.0 pH 6.5-9.0 4.8 4.5 5.6 5.7 5.5 5.5 Hardness (CaC03) mg/1 NE 67.9 38.7 52.1 47.1 50.0 48.1 Bicarbonate (CaCOs) mg/1 NE 12.0 12.0 33.0 36.0 34.0 36.0 Carbonate (CaC03) mg/1 NE ND ND ND ND ND ND TDS (calculated) mg/1 NE 1400.0 880.0 780.0 550.0 740.0 720.0 Phenolics mg/1 0.001 Sulphide mg/1 0.05 METALS Aluminum ug/1 5.0-100 120.0 1000.0 160.0 140.0 150.0 150.0 Antimony ug/1 NE ND ND ND ND ND ND Arsenic ug/1 5.0 ND ND ND ND ND Barium ug/1 NE 180.0 120.0 140.0 120.0 140.0 Beryllium ug/1 NE ND ND ND ND ND Bismuth ug/1 NE ND ND ND ND ND Boron ug/1 NE 32.0 31.0 32.0 29.0 33.0 Cadmium ug/1 0.017 ND ND ND ND ND Chromium ug/1 NE 3.0 3.0 ND ND ND Cobalt ug/1 NE ND 1.0 ND ND ND Copper ug/1 2.0-4.0 5.0 8.0 4.0 2.0 2.0 Iron ug/1 300 370.0 3900.0 660.0 520.0 650.0 Lead ug/1 1.0-7.0 0.8 2.5 1.4 0.6 0.7 Manganese ug/1 NE 3800.0 2300.0 2800.0 2500.0 2700.0 Molybdenum ug/1 73 ND ND ND ND ND Nickel . ug/1 25-150 7.0 4.0 2.0 ND ND Selenium ug/1 1.0 ND ND ND ND ND Silver ug/1 0.1 ND ND ND ND ND 136 Variable Units WQG" MC b wcc Ceil l d Cell 2d Cell 3d ASBe Strontium ug/1 NE 60.0 33.0 47.0 42.0 46.0 Thallium ug/1 0.8 0.2 0.4 ND ND ND Tin ug/1 NE ND ND ND ND ND Titanium ug/1 NE 5.0 2.0 4.0 4.0 4.0 Uranium ug/1 NE ND 0.1 0.4 0.4 0.4 Vanadium ug/1 NE ND 3.0 ND ND ND Zinc ug/1 30 180.0 130.0 28.0 ND 19.0 RESIN & FATTY ACIDS MDL 8 Abietic acid mg/1 0.005 0.08 0.04 0.28 0.25 Arachidic acid mg/1 0.005 1.60 9.91 Behenic acid Chlorodehydroabietic acid mg/1 mg/1 0.005 0.005 ND ND ND ND Dehydroabietic acid mg/1 0.005 0.04 0.03 0.07 0.05 0.19 0.11 Dichlorodehydroabietic mg/1 0.005 1.12 4.20 ND ND ND ND 9,10-dichlorostearic acid mg/1 0.005 ND ND ND ND ND ND Isopimaric acid mg/1 0.005 ND ND 0.18 0.05 0.18 0.10 Laurie acid mg/1 0.005 0.28 1.83 Levopimaric acid mg/I 0.005 Lignoceric acid mg/1 0.005 Linoleic acid mg/1 0.005 0.07 0.06 0.09 ND Linolenic acid mg/1 0.005 0.48 1.08 ND ND ND ND Myristic acid mg/1 0.005 ND 0.13 Neoabietic acid mg/1 0.005 0.06 0.02 0.17 0.22 Oleic acid mg/1 0.005 0.52 ND 0.04 0.03 0.04 ND Palmitric acid mg/1 0.005 0.19 0.75 0.05 0.03 0.07 0.01 Palmitoleic mg/1 0.005 0.09 0.23 0.05 0.03 0.10 0.04 Palustric acid mg/1 0.005 0.02 0.05 0.11 0.06 0.27 0.16 Pimaric acid mg/1 0.005 1.60 13.88 0.06 0.02 0.05 0.03 Sandaraco Pimaric acid mg/1 0.005 0.11 0.41 0.04 0.01 0.06 0.05 Stearic acid mg/1 0.005 0.24 1.16 TOTAL RFA mg/1 6.29 33.64 0.81 0.39 1.49 0.97 •137 Effluent Characteristics - Abitibi Consolidated, Grand Falls-Windsor, NL 15-Jan-02 Variable Units WQG" MC" wcc Cell l d Cell2d Cell 3d ASBe Nitrate + Nitrite mg/1 NE Nitrate mg/1 Narrative' Nitrite mg/1 0.06 Ammonia mg/1 Narrative' Colour TCU Narrative' Sodium mg/1 NE Potassium mg/1 NE Calcium mg/1 NE Magnesium mg/1 NE Alkalinity mg/1 NE Sulfate mg/1 NE Chloride mg/1 NE Reactive silica mg/1 NE Orthophospate mg/1 NE Phosphorus mg/1 NE DOC mg/1 NE Turbidity NTU Narrative' Conductivity uS/cm NE pH 6.5-9.0 Hardness (CaC03) mg/1 NE Bicarbonate (CaC03) mg/1 NE Carbonate (CaC03) mg/1 NE TDS (calculated) mg/1 NE Phenolics mg/1 0.001 Sulphide mg/1 0.05 METALS f Aluminum ug/1 5.0-100 120.0 680.0 680.0 250.0 50.0 90.0 Antimony ug/1 NE ND ND ND ND ND ND Arsenic ug/1 5.0 ND ND ND ND ND ND Barium ug/1 NE 110.0 120.0 160.0 100.0 40.0 73.0 Beryllium ug/1 NE ND ND ND ND ND ND Bismuth ug/1 NE ND ND ND ND ND ND Boron ug/1 NE 25.0 40.0 29.0 21.0 ND 16.0 Cadmium ug/1 0.017 ND ND 0.3 ND ND ND Chromium ug/1 NE 2.0 4.0 6.0 2.0 ND ND Cobalt ug/1 NE ND 1.0 1.0 ND ND ND Copper ug/1 2.0-4.0 7.0 8.0 14.0 7.0 4.0 2.0 Iron ug/1 300 240.0 4600.0 1600.0 660.0 50.0 300.0 Lead ug/1 1.0-7.0 1.6 2.8 3.5 5.9 1.8 0.5 Manganese ug/1 NE • 1800.0 2800.0 2500.0 2000.0 7.0 1700.0 Molybdenum ug/1 73 ND ND ND ND ND ND Nickel ug/1 25-150 4.0 4.0 6.0 4.0 ND 2.0 Selenium ug/1 1.0 ND ND ND ND ND ND Silver ug/1 0.1 ND ND ND ND ND ND 138 Variable Units WQG" MC b wcc Cell l d Cell 2d Cell 3d ASBe Strontium ug/1 NE 35.0 42.0 56.0 45.0 8.0 38.0 Thallium ug/1 0.8 0.1 0.4 0.2 0.1 ND ND Tin ug/1 NE ND ND ND ND ND ND Titanium ug/1 NE 4.0 2.0 12.0 4.0 ND 3.0 Uranium ug/1 NE 0.1 0.1 3.1 1.2 ND 0.4 Vanadium ug/1 NE 5.0 3.0 9.0 4.0 ND ND Zinc ug/1 30 160.0 230.0 330.0 90.0 20.0 32.0 RESIN & FATTY ACIDS MDL 8 Abietic acid mg/1 0.005 4.52 0.39 ND 0.15 Arachidic acid mg/1 0.005 2.05 10.60 Behenic acid Chlorodehydroabietic acid mg/1 mg/1 0.005 0.005 0.04 ND ND ND Dehydroabietic acid mg/1 0.005 ND 0.14 1.77 0.29 ND 0.12 Dichlorodehydroabietic mg/1 0.005 1.64 4.34 ND ND ND ND 9,10-dichlorostearic acid mg/1 0.005 0.04 ND ND ND ND ND Isopimaric acid mg/1 0.005 ND ND 0.37 0.03 ND 0.03 Laurie acid mg/1 0.005 0.34 1.88 Levopimaric acid mg/1 0.005 Lignoceric acid mg/1 0.005 Linoleic acid mg/1 0.005 0.13 0.02 ND 0.01 Linolenic acid mg/1 0.005 0.69 1.32 0.03 ND ND ND Myristic acid mg/1 0.005 0.01 0.14 Neoabietic acid mg/1 0.005 1.61 0.12 ND 0.06 Oleic acid mg/1 0.005 0.26 8.24 0.06 0.01 ND ND Palmitric acid mg/1 0.005 0.27 0.95 0.05 0.02 ND ND Palmitoleic mg/1 0.005 0.06 0.26 0.05 0.03 ND 0.04 Palustric acid mg/1 0.005 0.02 0.06 4.28 0.28 ND ND 0.10 Pimaric acid mg/1 0.005 2.38 18.00 0.32 0.03 0.02 Sandaraco Pimaric acid mg/1 0.005 0.17 0.50 0.77 0.05 ND 0.02 Stearic acid mg/1 ' 0.005 0.36 1.46 TOTAL RFA mg/1 8.29 47.88 14.01 1.27 0.00 0.54 i 139 Effluent Characteristics - Abitibi Consolidated, Grand Falls-Windsor, NL 15-May-02 Variable Units WQG" M C b Cell l d Cell 2 d Cell 3 d ASB e Nitrate + Nitrite mg/1 NE Nitrate mg/1 Narrative' Nitrite mg/1 0.06 Ammonia mg/1 Narrative' Colour TCU Narrative' Sodium mg/1 NE Potassium mg/1 NE Calcium mg/1 NE Magnesium mg/1 NE Alkalinity mg/1 NE Sulfate mg/1 NE Chloride mg/1 . NE Reactive silica mg/1 NE Orthophospate mg/1 NE Phosphorus mg/1 NE DOC mg/1 NE Turbidity NTU Narrative' Conductivity uS/cm NE PH 6.5-9.0 Hardness (CaC03) mg/1 NE Bicarbonate (CaC03) mg/1 NE Carbonate (CaC03) mg/1 NE TDS (calculated) mg/1 NE Phenolics mg/1 0.001 Sulphide mg/1 0.05 METALS Aluminum ug/1 5.0-100 140.0 2100.0 3300.0 2700.0 350.0 320.0 Antimony ug/1 NE ND ND ND ND ND ND Arsenic ug/1 5.0 ND ND ND ND ND ND Barium ug/1 NE 200.0 160.0 650.0 550.0 180.0 170.0 Beryllium ug/1 NE ND ND ND ND ND ND Bismuth ug/1 NE ND ND ND ND ND ND Boron ug/1 NE ND ND 97.0 85.0 ND ND Cadmium ug/1 0.017 ND ND ND ND ND ND Chromium ug/1 NE ND ND 28.0 22.0 ND ND Cobalt ug/1 NE ND ND ND ND ND ND Copper ug/1 2.0-4.0 ND ND 52.0 44.0 ND ND Iron ug/1 300 530.0 3800.0 6700.0 5500.0 870.0 800.0 Lead ug/1 1.0-7.0 ND ND 20.0 11.0 14.0 ND Manganese ug/1 . NE 4000.0 3400.0 8400.0 7200.0 3500.0 3500.0 Molybdenum ug/1 73 ND ND ND ND ND ND Nickel ug/1 25-150 ND ND ND ND ND ND Selenium ug/1 1.0 ND ND ND ND ND ND Silver ug/1 0.1 ND ND ND ND ND ND 140 Variable Units WQG° M C b wc c Cell l d Cell 2 d Cell 3 d ASBe Strontium ug/1 NE 70.0 62.0 140.0 120.0 67.0 67.0 Thallium ug/1 0.8 ND ND 1.6 1.2 ND 1.6 Tin ug/1 NE ND ND ND ND ND ND Titanium ug/1 NE ND 50.0 34.0 35.0 ND ND Uranium ug/1 NE ND 3.7 18.0 16.0 3.1 2.8 Vanadium ug/1 NE ND ND 47.0 36.0 ND ND Zinc ug/1 30 180.0 170.0 960.0 750.0 77.0 74.0 RESIN & FATTY ACIDS M D L e Abietic acid mg/1 0.005 0.58 0.32 0.04 0.04 Arachidic acid mg/1 0.005 ND 2.31 Behenic acid Chlorodehydroabietic acid mg/1 mg/1 0.005 0.005 0.01 ND ND 0.01 Dehydroabietic acid mg/1 0.005 0.06 0.01 0.38 ND ND ND Dichlorodehydroabietic mg/1 0.005 ND ND ND ND ND ND 9,10-dichlorostearic acid mg/1 0.005 ND ND ND ND ND ND Isopimaric acid mg/1 0.005 ND ND ND ND 0.05 0.05 Laurie acid mg/1 0.005 0.40 0.44 Levopimaric acid mg/1 0.005 Lignoceric acid mg/1 0.005 Linoleic acid mg/1 0.005 0.01 0.01 ND ND Linolenic acid mg/1 0.005 ND ND ND ND ND ND Myristic acid mg/1 0.005 ND ND Neoabietic acid mg/1 0.005 0.24 0.12 0.01 0.02 Oleic acid mg/1 0.005 0.73 1.22 ND ND ND ND Palmitric acid mg/1 0.005 ND ND 0.42 0.38 0.31 0.36 Palmitoleic mg/1 0.005 0.12 0.13 ND 0.02 ND ND Palustric acid mg/1 0.005 0.01 0.04 0.45 ND ND ND Pimaric acid mg/1 0.005 1.70 2.54 0.02 0.01 ND 0.01 Sandaraco Pimaric acid mg/1 0.005 0.20 0.18 0.05 ND ND ND Stearic acid mg/1 0.005 0.44 0.41 TOTAL RFA mg/1 3.65 7.28 2.16 0.86 0.41 0.48 141 Effluent Characteristics - Abitibi Consolidated, Grand Falls-Windsor, NL 26-Sep-02 Variable Units WQG" MC b wcc Cell l d Cell 2d Cell 3d ASBC Nitrate + Nitrite mg/1 NE 0.2 0.1 ND ND ND ND Nitrate - mg/1 Narrative' 0.2 0.1 ND ND ND ND Nitrite mg/1 0.06 ND ND ND ND ND ND Ammonia mg/1 Narrative' ND 0.3 ND ND ND ND Colour TCU Narrative' 120.0 210.0 19.0 200.0 180.0 190.0 Sodium mg/1 NE 91.3 15.5 88.8 90.1 81.8 82.9 Potassium mg/1 NE 17.2 25.9 17.9 20.0 13.6 14.2 Calcium mg/1 NE 15.1 14.4 21.6 29.9 10.5 11.0 Magnesium mg/1 NE 2.8 2.8 3.9 5.4 2.0 2.2 Alkalinity mg/1 NE 29.0 13.0 86.0 98.0 77.0 74.0 Sulfate mg/1 NE 200.0 44.0 150.0 130.0 160.0 160.0 Chloride mg/1 NE 11.0 22.0 7.0 8.0 7.0 7.0 Reactive silica mg/1 NE 3.2 3.6 2.2 2.3 1.8 1.8 Orthophospate mg/1 NE 0.3 1.9 0.1 1.0 0.6 0.5 phosphorus mg/1 NE 0.8 3.7 9.1 14.0 0.8 1.6 DOC mg/1 NE 492.0 482.0 49.0 46.3 41.4 37.2 Turbidity NTU Narrative' 223.0 27.6 685.0 >1000 18.2 57.7 Conductivity uS/cm NE 602.0 236.0 525.0 516.0 514.0 526.0 pH 6.5-9.0 4.7 4.3 6.7 6.7 7.4 7.4 Hardness (CaC03) mg/1 NE 49.2 .47.5 70.0 96.9 34.5 , 36.5 Bicarbonate (CaC03) mg/1 NE 29.0 13.0 86.0 98.0 77.0 74.0 Carbonate (CaCOj) mg/1 NE ND ND ND ND ND ND TDS (calculated) mg/1 NE 359.0 137.0 343.0 345.0 323.0 324.0 Phenolics mg/1 0.001 0.6 1.8 0.0 0.0 ND 0.0 sulphide mg/1 0.05 0.1 ND 0.2 4.8 ND 0.1 METALS Aluminum ug/1 5.0-100 90.0 1000.0 1500.0 2400.0 80.0 170.0 Antimony ug/1 NE ND ND ND ND ND ND Arsenic ug/1 5.0 ND ND 2.0 ND ND ND Barium ug/1 NE 170.0 190.0 330.0 460.0 88.0 99.0 Beryllium ug/1 NE ND ND ND ND ND ND Bismuth ug/1 NE ND ND ND ND ND ND Boron ug/1 NE 37.0 ND 55.0 74.0 20.0 18.0 Cadmium ug/1 0.017 0.3 ND 0.9 ND ND ND Chromium ug/1 NE 2.0 ND 12.0 ND ND 2.0 Cobalt ug/1 NE ND ND 2.0 ND ND ND Copper ug/1 2.0-4.0 7.0 ND 26.0 43.0 2.0 8.0 Iron ug/1 300 470.0 5200.0 3200.0 5400.0 300.0 530.0 Lead ug/1 1.0-7.0 1.1 ND 5.4 9.2 1.7 0.7 Manganese ug/1 NE 2700.0 3000.0 4500.0 6500.0 1800.0 2000.0 Molybdenum ug/1 73 ND ND ND ND ND ND Nickel ug/1 25-150 2.0 ND 10.0 ND ND ND Selenium ug/1 1.0 ND ND ND ND ND ND Silver ug/1 0.1 0.5 ND 1.0 ND ND ND 142 Variable Units WQG" M C b w e Cell l d Cell 2 d Cell 3 d ASB e Strontium ug/1 NE 47.0 ND 72.0 100.0 36.0 38.0 Thallium ug/1 0.8 0.1 ND 0.6 ND ND ND Tin ug/1 NE ND ND ND ND ND ND Titanium ug/1 NE 4.0 ND 14.0 ND 3.0 5.0 Uranium ug/1 NE ND ND 6.5 8.3 0.4 1.1 Vanadium ug/1 NE ND ND 21.0 30.0 2.0 3.0 Zinc ug/1 30 110.0 170.0 400.0 640.0 15.0 40.0 RESIN & FATTY ACIDS M D L g Abietic acid mg/1 0.005 ND ND ND ND Arachidic acid mg/1 0.005 2.3 12.0 Behenic acid Chlorodehydroabietic acid mg/1 mg/1 0.005 0.005 ND ND ND ND ND ND ND ND Dehydroabietic acid mg/1 0.005 0.0 0.1 ND ND ND ND Dichlorodehydroabietic mg/1 0.005 ND ND ND ND ND ND 9,10-dichlorostearic acid mg/1 0.005 ND ND ND ND ND ND Isopimaric acid mg/1 0.005 ND ND ND ND ND ND Laurie acid mg/1 0.005 0.2 0.8 Levopimaric acid mg/1 0.005 ND ND ND ND Lignoceric acid mg/1 0.005 2.3 16.4 Linoleic acid mg/1 0.005 ND ND ND ND Linolenic acid mg/1 0.005 0.5 0.9 ND ND ND ND Myristic acid mg/1 0.005 ND 0.1 Neoabietic acid mg/1 0.005 ND ND ND ND Oleic acid mg/1 0.005 13.1 143.2 ND ND ND ND Palmitric acid mg/1 0.005 0.3 0.5 0.0 0.0 0.0 0.0 Palmitoleic mg/1 0.005 0.1 0.4 0.0 0.0 0.0 0.0 Palustric acid mg/1 0.005 0.0 0.0 Pimaric acid mg/1 0.005 ND ND ND ND Sandaraco Pimaric acid mg/1 0.005 0.1 0.2 ND ND ND ND Stearic acid mg/1 0.005 0.3 0.9 TOTAL RFA mg/1 19.3 175.6 0.0 0.0 0.0 0.0 a WQG; Canadian Water Quality Guidelines for the protection of freshwater aquatic life b M C ; mill clarifier c WC; woodroom clarifier d cells 1, 2, and 3 are within the aerated stabilization basin. Treatment progresses as effluent moves through subsequent cells e A S B ; aerated stabilization basin. Secondary effluent treatment occurs here. Sample taken from effluent that has completed both primary and secondary treatment processes and is ready to be discharged into the river f see narrative at http://www.ec.gc.ca/ceqg-rcqe/English/download/default.cfm 8 M D L ; maximum detection limit of the analytical instrument 'ND; not detected. NE; no guideline established. TDS; total dissolved solids DOC; dissolved organic carbon a blank space indicates that the test was not performed 143 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0091388/manifest

Comment

Related Items