UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Competition in cultured and wild salmonid juveniles with emphasis on competitive interactions between… Blann, Camela A. 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_2003-0516.pdf [ 6.16MB ]
Metadata
JSON: 831-1.0091402.json
JSON-LD: 831-1.0091402-ld.json
RDF/XML (Pretty): 831-1.0091402-rdf.xml
RDF/JSON: 831-1.0091402-rdf.json
Turtle: 831-1.0091402-turtle.txt
N-Triples: 831-1.0091402-rdf-ntriples.txt
Original Record: 831-1.0091402-source.json
Full Text
831-1.0091402-fulltext.txt
Citation
831-1.0091402.ris

Full Text

COMPETITION IN CULTURED AND WILD SALMONTD JUVENILES WITH EMPHASIS ON COMPETITIVE INTERACTIONS BETWEEN FARMED ATLANTIC SALMON {Salmo salar) AND WILD COHO SALMON (Oncorhynchus kisutch) OR COASTAL CUTTHROAT TROUT (O. clarki clarki) by C A M E L A A. B L A N N B . S c , Simon Fraser University, 1993 A T H E S I S S U B M I T T E D I N P A R T I A L F U L F I L M E N T OF T H E R E Q U I R E M E N T S F O R T H E D E G R E E OF M A S T E R OF S C I E N C E in T H E F A C U L T Y OF G R A D U A T E STUDIES (Department o f Zoology)  We accept this thesis as conforming to the required standard  T H E U N I V E R S I T Y OF BRITISH C O L U M B I A October 2003 © Camela A . Blann, 2003  In presenting this thesis i n partial fulfilment o f the requirements for an advanced degree at the University o f British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying o f this thesis for scholarly purposes may be granted by the head o f my department or by his or her representatives. It is understood that copying or publication o f this thesis for financial gain shall not be allowed without my written permission.  Department o f Zoology The University o f British Columbia Vancouver, Canada  Date  ABSTRACT The purpose o f my research was to investigate the relative competitive ability o f cultured and w i l d salmon to provide insight into the potential effects o f introduction o f cultured salmon on Pacific salmon species. Aquarium trials involving equal contests (i.e. size matched, simultaneously introduced individuals) indicate that w i l d coho salmon (Oncorhynchus  kisutch)  populations (from Salmon River and Street Creek) were competitively matched i n contests with farmed coho salmon (originally from Kitimat River). In equal contests between farmed Atlantic salmon (Salmo salar) ( M o w i strain) and these w i l d coho salmon populations or coastal cutthroat trout (O. clarki clarki), Atlantic salmon were subordinate in all cases. W h e n Atlantic salmon were given a residence advantage, however, they were competitively equal to both w i l d coho salmon populations, but remained subordinate to coastal cutthroat trout. Contests i n which Atlantic salmon were given a 10-30% length advantage indicate that Atlantic salmon juveniles competitively match Street Creek coho salmon, but remained subordinate to Salmon River coho salmon.  Behaviour and growth rates o f w i l d or hatchery coho salmon and farmed Atlantic salmon competing for food and space i n artificial stream channels were also investigated. Individually marked fry were introduced to stream channel sections consisting o f both riffle and pool habitat and behavioural interactions and distribution were recorded daily. Length and weight were measured prior to introduction and at the end o f the 12 or 15 day experiments. In equal contests, both coho and Atlantic salmon grew significantly more i n the presence o f the other species than when alone. It appears that coho salmon obtain additional food ration by out-competing Atlantic salmon, whereas Atlantic salmon are stimulated to feed more i n the presence o f coho salmon competitors.  M y results suggest that underyearling w i l d coho salmon and cutthroat trout outcompete farmed Atlantic salmon o f a similar size, however, size and residence advantage improves the competitive ability o f farmed Atlantic salmon. The greater competitive ability o f Pacific salmon species may be, i n part, responsible for the failed attempts to introduce Atlantic salmon to the Pacific coast i n the past. However, as farmed salmon are competitively equal i n some instances,  farmed salmon could cause adverse effects on Pacific salmonid populations, particularly if farmed fish are introduced in large numbers relative to wild populations.  TABLE OF CONTENTS ABSTRACT  ii  TABLE OF CONTENTS  iv  LIST OF TABLES  vi  LIST OF FIGURES  vii  ACKNOWLEDGEMENTS  viii  CHAPTER I - INTRODUCTION  1  CHAPTER n - COMPETITION BETWEEN CULTURED AND WILD SALMONIDS  5  2.1 Intraspecific Competition between Native Salmonids  6  2.1.1. Intraspecific Competiton between Hatchery and Wild Salmonids 2.1.2 Intraspecific Competition between Farm and Wild Salmonids  7 14  2.2 Interspecific Competition between Native and non-Native Salmonids  19  2.3 Research Addressed  22  CHAPTER ffl - RELATIVE COMPETITIVE ABILITY OF CULTURED AND WILD SALMONIDS  24  3.1 Introduction  24  3.2 Materials and Methods  26  3.2.1 Source and Rearing Conditions for Juvenile Salmonids 3.2.2 Experimental Setup and Data Collection 3.2.3 Data Analysis 3.3 Results 3.3.1 Relative Competitive Ability 3.1.2 Behavioural Interactions 3.1.3 Growth 3.4 Discussion 3.4.1 Relative Competitive Ability 3.4.2 Growth '. 3.4.3.Behaviour 3.4.4.Implications  26 28 32 34 34 35 43 46 ,.... 46 48 48 50  CHAPTER IV - COMPETITIVE INTERACTIONS BETWEEN FARMED ATLANTIC SALMON AND NATIVE COHO SALMON IN SEM-NATURAL STREAM CHANNELS 54 4.1 Introduction  54  4.2 Methods  56  4.2.1 Source and Rearing Conditions for Juvenile Salmonids 4.2.2 Experimental Setup and Procedure 4.2.3 Data analysis 4.3 Results 4.3.1 Growth  56 56 59 60 60 iv  4.3.2 Behavioural Interactions 4.3.3 Distribution and Activity 4.4 Discussion 4.4.1 Growth 4.4.2 Behavioral Interactions 4.4.3 Distribution and Activity 4.4.4 Implications  64 67 70 70 72 73 74  CHAPTER V - GENERAL DISCUSSION AND CONCLUSIONS  79  REFERENCES  87  v  LIST OF TABLES Table 1 - Source Information for Juvenile Salmonids 27 Table 2 - Summary of Aquarium Experiment Conditions, Date and Fish Size 30 Table 3 - Summary of Aquarium Experiment Results on Relative Competitive Ability 35 Table 4 - Mean Number Behavioural Interactions per Observation Period for each Population (averaged over the course of each experiment) ^ 36 Table 5 - Intraspecific Coho Salmon Contests - Mean Daily Length and Weight Change 44 Table 6 - Interspecific Atlantic Salmon and Coho Salmon or Coastal Cutthroat Trout Contests - Mean Daily Length and Weight Change 45 Table 7 - Start Size of Juvenile Salmon, Experimental Dates and Mean Water Temperature 57 Table 8 - ANOVA Results for Growth - Set 1 and 2, Farm Atlantic salmon versus Wild coho salmon, simultaneous introduction 61 Table 9 - Mean growth of Atlantic and Coho salmon by Assemblage - Set 1 and 2, Farm Atlantic salmon versus Wild coho salmon, simultaneous introduction 62 Table 10 - Mean growth of Atlantic and Coho salmon by Assemblage - Set 3, Farm Atlantic salmon versus Wild coho salmon, Atlantic salmon residence advantage 63 Table 11 - Mean growth of Atlantic and Coho salmon by Assemblage - Set 3, Farm Atlantic salmon versus Hatchery coho salmon, Atlantic salmon residence advantage 64 Table 12 - ANOVA Results for Behavioural Interactions - Set 1 and 2, Farm Atlantic salmon versus Wild coho salmon, simultaneous introduction 65 Table 13 - Mean number of Behavioural Interactions per Observation Period - Set 1 and 2, Farm Atlantic salmon versus Wild coho salmon, simultaneous introduction 65 Table 14 - Mean number of Behavioural Interactions per Observation Period - Set 3, Farm Atlantic salmon versus Wild coho salmon, Atlantic salmon residence advantage 66 Table 15 - Mean number of Behavioural Interactions per Observation Period - Set 3, Farm Atlantic salmon versus Hatchery coho salmon, Atlantic salmon residence advantage 67 Table 16 - ANOVA Results for Distribution - Set 1 and 2, Farm Atlantic salmon versus Wild coho salmon, simultaneous introduction 67 Table 17 - Distribution of Atlantic and Coho Salmon - Set 1 and 2, Farm Atlantic salmon versus Wild coho salmon, simultaneous introduction 68 Table 18 - Atlantic Salmon Activity - Set 1 and 2, Farm Atlantic salmon versus Wild coho salmon, simultaneous introduction 68 Table 19 - Distribution of Atlantic and Coho Salmon - Set 3, Farm Atlantic salmon versus Wild or Hatchery coho salmon, Atlantic Salmon residence advantage 69 Table 20 - Atlantic Salmon Activity - Set 3, Farm Atlantic salmon versus Wild or Hatchery coho salmon, Atlantic Salmon residence advantage 69  LIST OF FIGURES Figure 1 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 1, Salmon R. (SR) / Street Cr. (SC) coho salmon 38 Figure 2 - Mean number of behavioural interactions by observation period - Aquarium Exp. #12, Street Cr. (SC) / Kitimat (farm) coho (KT) salmon 38 Figure 3 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 6, Salmon R. (SR) / Kitimat (farm) coho (KT) salmon (1 exp.) 38 Figure 4 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 10, Salmon R. (SR) / Kitimat (farm) coho (KT) salmon (2 exp.) 39 Figure 5 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 7, Street Cr. (SC) / Chilliwack Hatchery (CH) coho salmon (1 exp.) 39 Figure 6 - Mean number of behavioural interactions by observation period - Aquarium Exp. #11, Street Cr. (SC) / Chilliwack Hatchery (CH) coho salmon (2 exp.) 39 Figure 7 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 2, Salmon R. (SR) coho / Atlantic (farm) (AT) salmon 41 Figure 8 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 3, Salmon R. (SR) coho / Atlantic (farm) (AT) salmon prior residence 41 Figure 9 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 4, Street Cr. (SC) coho / Atlantic (farm) (AT) salmon 41 Figure 10 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 5, Street Cr. (SC) coho / Atlantic (farm) (AT) salmon prior residence 42 Figure 11 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 8, Coastal Cutthroat trout (CT) / Atlantic (farm) (AT) salmon 42 Figure 12 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 9, Coastal Cutthroat trout (CT) / Atlantic (farm) (AT) salmon prior residence 42 Figure 13 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 13, Salmon R. (SR) coho / Atlantic (farm) (AT) salmon size advantage 43 Figure 14 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 14, Street Cr. (SC) coho / Atlantic (farm) (AT) salmon size advantage 43 Figure 15 - Mean Length Change by Treatment, Set 1 and 2, Farm Atlantic salmon versus Wild coho salmon, simultaneous introduction 62 st  nd  st  nd  ACKNOWLEDGEMENTS I would like to thank Yuho Okada, Rowenna Flynn, Tyese Patton and Shannon MacLachlan for their assistance in conducting the experiments as well as the staff of the Department of Fisheries and Oceans at the Cultus Lake Salmon Research Laboratory for the use of the facility. Cultured juvenile salmon were provided by local aquaculture facilites; Target Marine and the Chilliwack Hatchery. Comments on experimental design were provided by an external review committee consisting of Helen Kerr of the Department of Fisheries and Oceans, Justin Henry of Target Marine, and John Nightingale of the Vancouver Aquarium. Dr. Valerie Lemay, Dr. Scott Hinch, Dr. Eric Taylor and my supervisor, Dr. Michael Healey provided useful comments on this thesis. This project has been funded through Aquanet and the Network of Centres of Excellence.  viii  CHAPTER I - INTRODUCTION Aquaculture, i n various forms, has been utilized for thousands o f years to improve seasonal food availability and increase production o f desirable products. Favourable attributes o f plants or animals can be enhanced through artificial selection, to develop highly productive, fast growing, domesticated strains. Salmonids (i.e. salmon, trout and char), the subject o f this thesis, have been cultured extensively since the late 19 century (Lichatowich 1999). Hatchery production o f th  juvenile salmonids for release into the natural environment has been utilized i n attempts to increase the production o f adult fish. M o r e recently, farming o f salmonids from egg to adult has been undertaken as a commercial enterprise to extend their local availability beyond their natural range or to increase their availability i n areas where w i l d stocks have declined and to meet offseason market demand.  1  Cultured fish have been observed to differ from their w i l d progenitors i n a variety o f ways. M a n y studies have noted morphological and behavioural differences i n traits such as aggression and predator avoidance (Taylor 1986; E i n u m and Fleming 2001). In addition, some studies suggest changes i n feeding behaviour, habitat use (Einum and Fleming 2001) and spawning behaviour (Fleming et al. 2000). These differences have been shown to have both an environmental (i.e. related to rearing environment) and genetic basis. Such differences can be accentuated depending on the extent o f culture o f the population i n question. Extent o f culture relates to the percentage o f the organism's lifespan during which it is reared i n an artificial environment and the number o f generations the cultured strain has been isolated from the w i l d founder population(s). Hatchery and sea-ranched fish generally have a relatively l o w extent o f 2  culture whereas farmed fish generally have a higher extent. Currently, on the Pacific coast o f North America, hatchery and sea-ranched salmon are typically derived from naturally spawning broodstock. These fish are artificially reared during their juvenile life-stage and released into freshwater or into marine waters to complete their life cycle under natural conditions. Hatchery and sea-ranched fish are released i n order to supplement naturally spawning populations or  Spawning channels which can be considered a type of culture activity was considered beyond the scope of this thesis. Hatchery salmonids are typically released to freshwater or sometimes marine waters to mature. Releases are undertaken to increase adult production forfisheriesand enhancement purposes. Sea-ranched salmonids are released to marine waters to mature forfisheries(Isaksson et al. 1997). 1  2  1  enhance local fisheries.  In contrast, farmed fish are typically derived from broodstock selected  from cultured populations and are artificially reared to adults at which time they are harvested. Broodstock fish are often subjected to artificial selection over multiple generations i n order to enhance favourable characteristics such as survival and growth under farm rearing conditions.  Cultured fish frequently show evidence o f morphological and behavioural change due to adaptation to the artificial rearing environment, natural selection for favourable traits i n the artificial rearing environment, and artificial selection due to broodstock selection. These changes are most pronounced i n fanned fish due to their greater extent o f culture. Hatchery and sea-ranched salmonids spend much o f their lives i n the natural environment and millions o f f i s h are released annually so these cultured fish interact with w i l d salmonids frequently and continually. Whereas hatchery salmonids are intentionally released, farmed salmonids may be introduced into the w i l d through accidental releases from juvenile rearing facilities, during transfers between holding sites and from compromised net cages . Although rates o f escape 3  from w e l l run aquaculture operations are small (1-2%) (Alverson and Ruggerone 1997), as aquaculture production increases, the number o f escapees may be considerable and rival or even outnumber w i l d fish in the natural environment (Gausen and M o e n 1991; Saegrov et al. 1997) with poorly understood consequences.  Due to the morphological and behavioural differences between cultured and w i l d fish, there are potential ecological and genetic consequences from the introduction o f cultured fish to the w i l d populations. Potential ecological effects include predation, competition and disease transfer. Potential genetic effects include introduction o f maladapted alleles into w i l d populations and reduced genetic diversity. This thesis explores aspects o f competition for food and space between cultured and w i l d salmonids.  Biological competition, as defined by B i r c h (1957), occurs when a number o f animals (of the same or different species) use common resources, the supply o f which is short; (exploitation Net pens and net cages are structures in which organisms are held within an enclosed space while maintaining a free exchange of water. Though the terms are often used interchangeably, they are distinct in that a cage is totally enclosed on all sides, or on all sides except the top, whereas a pen is enclosed on the bottom by a lake, river or seabed (Beveridge 1984 cited in Alverson and Ruggerone 1997). 3  2  competition) or i f the resources are not i n short supply, competition occurs when the animals seeking that resource nevertheless harm one another i n the process (interference competition). Competition for various resources, including food, space and mates occurs throughout the life o f an organism. A s described by Hearn (1987), during their stream resident stage, juvenile salmonids compete for space rather than directly for food, cover or other resources (Chapman 1966). Individuals compete for favourable stream positions based on their value as feeding sites (Fausch and White 1981; Bachman 1984; Fausch 1984). Competition i n salmonids and the rationale for this work in the context o f previous work on cultured salmonids is described further in Chapter 2.  Both native and non-native salmonid species are currently being utilized i n finfish aquaculture on the Pacific coast o f North America. Hatchery and sea-ranching programs produce a l l species o f native Oncorhynchus for release (Ruggerone et al. 1995 cited i n Alverson and Ruggerone 1997; Gustafson etal. 1997). Atlantic salmon, (Salmo salar) an introduced species, is currently 4  being farmed i n open net-cages i n Washington state and i n British Columbia. In addition, two native species, chinook (Oncorhynchus  tshawytscha) and coho salmon (O. kisutch) are being  farmed i n open net-cages i n British Columbia, but to a lesser extent than Atlantic salmon (i.e. native species comprise approximately 20 percent o f farmed salmon i n B . C . ) (B.C. Salmon Farmers Assoc. 2003).  The potential ecological and genetic consequences o f salmonid aquaculture began to be seriously questioned by the 1930s (Lichatowich 1999). Recently, this issue has received increased attention on the Pacific coast o f North America due to reduced population sizes o f many salmonid species over large portions o f their ranges, the limited success o f hatchery In this thesis, the term "introduced" is defined as follows: "A plant or animal movedfromone place to another by man (i.e. an individual, group or population of organisms that occur in a particular locale due to man's actions" (Shafland and Lewis 1984) as per the American Fisheries Society's "Recommended Standardized Terminology." The terms exotic and transplanted will not be used as they refer to the organisms country of origin; [Exotic is defined as " An organism introduced from a foreign country (i.e., one whose entire native range is outside the country where found) and Transplanted is defined as " An organism moved outside its native range but within a country where it occurs naturally (i.e.,one whose native range includes at least a portion of the country where found] (Shafland and Lewis 1984). In this thesis, the term native is defined as occurring naturally within an area. I will differentiate between an introduced species which is non-native (e.g. Atlantic salmon in B.C.) and an introduced population which may be native (e.g. hatchery coho salmon stocked to the systemfromwhich the founder population was derived) or 4  3  programs i n increasing production o f native species and the development o f a fish farming industry based predominantly on an introduced species. Atlantic salmon that have escaped from fish farming operations i n Washington and B . C . are reported to have produced offspring i n several British Columbia rivers (Volpe et al. 2000). This has raised public concern that Atlantic salmon may successfully colonize local rivers and expose native species to a competitor with which they have not coevolved. M a n y examples o f "biological invasions" that had severe effects on local flora and fauna are present i n the scientific literature and media (e.g. Drake et al. 1989; M i l l s et al. 1994; Lassuy 1995). Culture o f native species and potential intraspecific competition has also been a matter o f public concern, though perhaps to a lesser extent.  In this study I investigated competitive interactions between cultured and w i l d salmonids during their juvenile freshwater phase with a view to assessing the risk to w i l d populations from cultured salmonids. Particular emphasis was given to competitive interactions between farmed Atlantic salmon and w i l d coho salmon or coastal cutthroat trout (O. clarki clarki) to assess the potential effects o f feral Altantic salmon on native Pacific coast species.  Chapter 2 provides additional background information on salmonid competition, summarizes published research specifically on cultured salmonids, and describes the rationale for research addressed i n this work. Chapter 3 describes aquarium experiments that were designed to assess the relative competitive ability o f cultured versus w i l d salmonids, competing for food in a simple environment. Chapter 4 describes stream channel experiments that were designed to investigate behaviour and growth rates i n groups o f salmonids competing for food and space i n a seminatural setting. Chapters 3 and 4 are being submitted for publication and so include some repetition o f introductory information and discussion. Chapter 5 synthesizes the results and provides a discussion and conclusions i n terms o f risks to w i l d salmon from salmonid culture.  non-native (e.g. hatchery coho salmon stocked to a system other than thatfromwhich the founder population was derived). 4  CHAPTER H - COMPETITION BETWEEN CULTURED AND WILD SALMONIDS The various models of competition, particularly the Lotka-Volterra model (Lotka 1925; Volterra 1926; Krebs 2001), the competitive exclusion principal (Gause 1934; Hardin 1960) and the theory of the evolution of the niche (MacArthur 1972), have been useful in developing our understanding of competition but they are of limited use in predicting the outcome of competitive interactions in nature. In salmonids, competitive outcomes may be affected by density or environmental conditions, as well as age or size structure of populations (Law and Watkinson 1989). Such effects have been demonstrated in empirical studies of salmonids. For example, Cunjak and Green (1986) showed that the relative competitive ability of brook trout (Salvelinus fontinalis) and rainbow trout (O. mykiss) is affected by temperature. Differential effects on various age classes of rainbow trout after introduction of the redside shiner (Richardsonius balteatus) in Paul Lake, British Columbia have also been observed (Crossman and Larkin 1959 cited in Werner 1986). The shiner had a negative effect on the growth rate of young trout due to competition for food, whereas larger trout preyed on the shiners and their growth increased. In addition, studies on salmonids are typically limited to interaction effects between two species and on small scale studies both spatially and temporally (Fausch 1998).  Fausch (1998) pointed out that interspecific contests between stream salmonids for favourable feeding locations satisfy many criteria needed to demonstrate existence of competition; namely, salmonid species overlap in resource use, intraspecific competition occurs and resources are limiting (Weins 1989). Stream dwelling salmonids have been observed to exhibit resource partitioning along several niche axes including food, habitat and time (Werner 1986) but it is generally thought that they compete for space rather than food directly (Hearn 1987), i.e. interference rather than exploitation competition is most important. It is in this context, i.e. with indications that competition is occurring, yet limitations in our ability to assess competitive interactions in nature, that we approach the problem of determining how the culture of fish may affect competitive interactions and population or community structure.  The culture of fish has implications for ecology, in part, due to effects on competitive interactions. First, the culture of fishes results in morphological and behavioural changes that 5  may affect species specific responses and, therefore, the outcome o f both intraspecific and interspecific interactions. Second, culture o f fishes may cause additional interspecific competition due to introduction o f a non-native species. This potentially results i n a novel type o f interspecific competition wherein species that are not coevolved are exposed to one another. Considerable niche overlap may occur as there has been no opportunity for niche differentiation over evolutionary time. In addition, appropriate defences may not have been evolved by the native species to deal with a novel competitor and as such there is a potential for dramatic effects on native fauna. There have been many examples i n which the introduction o f non-native organisms resulted i n displacement o f native species (e.g., Larson and Moore 1985; M o y l e and Vondracek 1985; Herbold and M o y l e 1986).  The purpose o f my study is to investigate competitive interactions between cultured and w i l d salmonids during their juvenile freshwater phase with a view to assessing the risk to native w i l d populations from cultured salmonids. The following is a summary o f the research that has been conducted on intraspecific (Section 2.1) and interspecific (Section 2.2) competition between w i l d and cultured salmonids. The rationale for research addressed i n this thesis is then provided (Section 2.3).  2.1 Intraspecific Competition between Native Salmonids  Investigations o f effects o f intraspecific competition between w i l d and cultured native salmonids has generally involved comparing cultured fish to w i l d stocks o f the same species and ideally from the same population. These studies can be subdivided into two types based on the extent o f culture. Studies on fish with a l o w extent o f culture typically compare hatchery or sea-ranched fish (collectively referred to as "hatchery" i n the following section) with w i l d populations (Section 2.1.1). Studies on fish with a high extent o f culture typically compare farmed fish that have been i n culture for many generations with w i l d populations (Section 2.1.2).  6  2.1.1. Intraspecific Competiton between Hatchery and Wild Salmonids  Hatchery production o f salmonids has been utilized i n an attempt to increase production through improved returns o f adult fish. Use o f hatchery programs to achieve this goal has been widespread and intense. Concerns began to be raised regarding potential ecological and genetic consequences o f hatchery programs (Winton and Hilborn 1994; Lichatowich 1999) due to theoretical or observed differences between cultured and w i l d fish. Initially there was skepticism regarding differences between hatchery and w i l d salmonids, particularly anadromous salmon, due to the limited time they were reared i n the artificial culture environment, followed by natural rearing to maturity (Reisenbichler and R u b i n 1999). Changes due to natural selection in the culture environment were considered unlikely because mortality i n the hatcheries is usually very l o w providing little opportunity for natural selection to occur.  Recently, several potential causes o f differences between hatchery and w i l d fish have begun to receive increased attention. Particularly i n the early years o f supplementation, there was limited appreciation o f population differences and juveniles were transplanted beyond their population's natural range. Even recent hatchery operations have unintentionally collected broodstock from outside the target population (Waples et al. 2001) causing additional mixing o f populations with attendant potential adverse effects such as outbreeding depression. Recent work has also shown that hatchery fish may have behavioural and physiological differences from w i l d conspecifics due to the culture environment and management practices. Cultured salmonids can potentially differ, not only due to natural selection for traits i n the artificial rearing environment, but also due to physical adaptation to the rearing environment and artificial selection due to broodstock selection practices even i f broodstock are selected from the receiving population.  M a n y studies have documented a variety o f traits for which cultured fish differ from their w i l d counterparts. A review by E i n u m and Fleming (2001) indicated that cultured fish exhibit increased aggression , reduced predator avoidance , altered patterns o f movement, feeding 5  5  5  Significant difference based on meta-analysis by Einum and Fleming (2001). 7  behaviour, habitat use, morphology, age and size-at-maturity, spawning behaviour and disease resistance (see also Hindar et al. 1991, Fleming and Petersson 2001).  The above studies summarized research comparing hatchery populations with w i l d populations from which they were derived as well as those from different source populations. In some instances the differences noted could, therefore, be due to a population effect rather than culture effect (e.g. Rosenau and M c P h a i l 1987), particularly i n instances when effects are not significant by meta-analysis and when an introduced population may be at a disadvantage (such as homing and therefore reproductive success). For example, increased aggression is likely due to culture as it is significant by meta-analysis. A habitat use difference between w i l d and hatchery fish however, was not and the analysis did not include any hatchery populations derived from the w i l d population with which they were being compared. In this instance, the observed difference could be due to a population rather than culture effect. In addition, some studies cited involved multi-generation hatchery populations, which could have differentiated more than hatchery fish derived from w i l d broodstock (e.g. Swain et al. 1991). Under current hatchery practice, broodstock is typically taken from the w i l d spawning population (Bonnell 1999). Broodstock collection and hatchery culture practices can affect the quality o f hatchery juveniles and this was not considered i n the review by E i n u m and Fleming (2001)(see below).  If large numbers o f hatchery salmonids are released into a w i l d population then any differences in the hatchery fish, whether genetic or phenotypic, can affect the w i l d populations. The effects can be genetic (from selection, drift or inbreeding) or ecological (from competition, predation or disease transfer). The following review specifically describes competitive interactions between hatchery and w i l d salmonids by life history stage and observed population response to introduction o f hatchery salmonids.  Juveniles:  Juvenile salmonids have been intentionally introduced extensively and i n large  numbers (Isaksson 1988, Hilborn and Winton 1993). There is potential for increased frequency o f competitive interactions and adverse effects on w i l d juveniles, simply due to the increased density o f animals (Petrosky and Bjornn 1998). In addition, the relative competitive ability o f hatchery to w i l d salmonids must be considered. Stream dwelling juvenile salmonids compete  for space (Chapman 1966; Hearn 1987). Dominant individuals have more profitable feeding territories and grow faster (Fausch 1984; Nielsen 1992). If hatcheryfishhave a higher relative competitive ability than wild salmonids there is potential for displacement of wild fish (Nickelson et al. 1986; McMichael et al. 1999) potentially leading to reduced growth or increased mortality.  The relative competitive ability of juvenile fish is related to multiple factors including metabolic rate, aggression, prior experience, prior territory residence and size (Grant 1990; Berejikian et al. 1996; Rhodes and Quinn 1998; Cutts et al. 1999; Johnsson et al. 1999). Hatchery salmonids have often been found to exhibit higher levels of aggression (e.g. Berejikian et al. 1996; Rhodes and Quinn 1998), are often stocked at larger sizes relative to wild conspecifics of the same age and may exhibit higher growth rates (see studies cited in Einum and Fleming 2001). As these factors are associated with dominance (Fausch 1984; Holtby et al. 1993 but also see studies cited in Jonsson 1997), hatcheryfishmay have a competitive advantage in this regard. Conversely, wild salmonids may have prior residence, which could confer an advantage.  Studies directly investigating relative competitive ability of size matched hatchery and wild salmonids indicate single generation hatchery salmonids are dominant (Rhodes and Quinn 6  1998; Berejikian et al. 1999). Individualsfroma multi-generation hatchery population, however, were subordinate (Berejikian et al. 1996). In experiments in which hatchery fish had a size advantage, hatchery salmonids were dominant to their wild conspecifics, often despite a residence advantage of the wildfish(Berejikian et al. 1996; Rhodes and Quinn 1998; McMichael et al. 1999).  Greater growth of hatcheryfishrelative to wildfishmay indicate superior competitive ability. In addition, large size is typically correlated with improved competitive ability, reduced predation risk and, therefore, improved survival and reproductive success (Werner 1986, but also see Healey 1982). Einum and Fleming (2001) found inconsistent results in the literature. In four studies wildfishoutgrew hatchery salmonids (also Rhodes and Quinn 1999), in two the opposite Single generation hatchery salmonids are produced from broodstock collected in the wild. Multi-generation hatchery salmonids are produced from broodstock returning to the hatchery. 6  9  was found, and i n one no difference was observed (also Berejikian et al. 1999). In addition, Reisenbichler and Mclntyre (1977) determined that hatchery/wild hybrids had greater growth than pure hatchery or w i l d fish i n stream enclosures. E i n u m and Fleming (2001) found no statistical difference i n growth based on their meta-analysis. In addition, three o f four studies showing inferior performance by hatchery fish were studies i n which the hatchery and w i l d fish were from different source populations and the results could, therefore, be due to population as well as culture effects.  In terms o f survival i n the wild, hatchery fish typically have lower survival rates than w i l d fish (Leider et al. 1990; statistically significant by meta-analysis o f E i n u m and Fleming 2001, but also see Berejikian et al. 1999; Rhodes and Quinn 1999). This may be due, i n part, to competitive inferiority, however, several authors suggest it could be due to higher relative predation rates as hatchery fish consistently have exhibited reduced predator avoidance behaviour (Einum and Fleming 2001), and their increased aggressiveness could increase susceptibility to predation i n the w i l d (Rosenau and M c P h a i l 1987) or unnecessary energy expenditure (Bachman 1984, M c M i c h a e l et al. 1999). Again, juvenile survival is affected by hatchery culture practices and this was not considered i n the review by E i n u m and Fleming (2001) (see below).  Spawning adults: Salmonids reproduce i n freshwater and compete for space i n which to spawn as well as for mates. Intrasexual competition occurs between males for access to females and between females for spawning sites. Females appear to express mate choice through delays i n breeding and aggression though this expression is constrained because dominant males can monopolize access to females (Fleming and Petersson 2001). M a n y additional factors including morphological traits, age and size at maturity, mating tactics and timing affect their ability to obtain a mate.  Investigations o f the relative competitive ability o f hatchery and w i l d fish under semi-natural conditions, have generally shown that hatchery fish, particularly males, are competitively inferior to w i l d salmonids. Work on coho salmon has shown that hatchery males are less aggressive, attacked more by females, are most often dominated i n spawning events, and have 10  lower breeding success than w i l d males (Fleming and Gross 1992,1993; Berejikian et al. 1997; reviewed i n Fleming and Petersson 2001). Hatchery females establish nesting territories and breed later, construct fewer nests, retain more eggs, lose more eggs to nest destruction by other females and have lower breeding success than w i l d females (Fleming and Gross 1993; Berejikian et al. 1997; reviewed i n Fleming and Petersson 2001). Fleming (1994) suggested that selection against hatchery coho salmon was, i n part, due to morphological differences as well as inferior competitive ability due to behavioural differences. Thompson et al. (1998) determined, through the use o f genetic markers, that sea ranched Atlantic salmon had a significantly greater percentage eggs fertilized by secondary males (i.e. subordinate males or precocious male parr) than w i l d salmon again suggesting behavioural differences i n the cultured fish that effect their spawning success. A field study by Jonsson et al. (1990) indicated that hatchery Atlantic salmon ascended their natal river later, had more injuries and a greater proportion left the river unspawned than w i l d fish.  Chilcote et al. (1986) compared the relative reproductive success o f naturally spawning hatchery and w i l d steelhead trout (O. mykiss) and found that hatchery fish had a reproductive success to the smolt stage o f 28% that o f w i l d fish. However the majority o f smolts were offspring o f hatchery spawners as hatchery spawners outnumbered w i l d spawners by at least 4.5 to 1. In an extension o f the above study, Leider et al. (1990) showed that reproductive success o f hatchery steelhead decreased from approximately 7 5 % at the subyearling stage to approximately 12 % percent at the adult stage, so ultimately the percentage o f offspring produced from hatchery steelhead was 4 2 % despite a 4.5 fold majority i n the previous spawning population. Several assumptions, however, are implicit i n these calculations (e.g. random mate selection, no interbreeding among types, Leider et al. 1990), and additional research is required to determine i f the assumptions are realistic. Lower reproductive success o f hatchery salmonids has also been inferred by Nickelson et al. (1986). M c L e a n et al. (2003) also found that w i l d steelhead trout (O. mykiss) females had far higher reproductive success than hatchery steelhead trout, producing nine and 42 times as many adult offspring as the hatchery females per capita, over two years, respectively. These hatchery steelhead trout were o f non-native origin and were artificially selected for early spawning which they exhibited relative to w i l d trout.  n  Population Response: Assessments o f population response to hatchery supplementation programs do not separate the role o f competition from factors such as predation and environmental effects. Population response however, is the criterion on which the programs should ultimately be tested. The relative importance o f competition between hatchery and w i l d salmon and long-term population level outcomes w i l l depend upon the number o f hatchery juveniles introduced, their relative competitive ability, survival and reproductive success as well as the relative competitive ability and fitness o f their offspring. Population level outcomes w i l l also depend on the size o f w i l d populations and the relative rate o f purging o f maladaptive traits through natural selection versus swamping from annual introduction o f large numbers o f hatchery fish.  Several studies have demonstrated l o w reproductive success o f released hatchery salmonids (Chilcote et al. 1986; Leider et al. 1990; M c L e a n et al. 2003). There is high variability i n level of introgression o f hatchery alleles into w i l d populations. Introgression ranging from 0 to over 75% has been observed, with resistance to introgression possibly greater i n anadromous than resident populations (Hindar et al. 1991; Fleming and Petersson 2001; Utter 2001). In some instances l o w levels o f introgression are observed despite large scale introductions (Isaksson 1988; Hilborn and Winton 1993).  Overviews o f the effectiveness o f hatchery programs i n several jurisdictions indicate that longterm hatchery programs are typically evaluated i n terms of juvenile survival to time o f release, contribution to the commercial catch and adult returns o f fish o f hatchery origin (Winton and Hilborn 1994; Finstad and Jonsson 2001; Fjellheim and Johnsen 2001; Waples et al. 2001). Often corresponding data are not collected for w i l d populations (Hilborn and Winton 1993) or evaluation studies have limitations that do not allow definitive answers regarding changes i n abundance. A demographic boost is necessary but not sufficient for improved natural production ( W a p l e s e t a l . 2001).  In order for supplementation programs to be effective and provide a long-term benefit hatchery fish must produce viable progeny that contribute to natural productivity without adversely affecting production from w i l d fish and, together with w i l d fish, form an integrated self-  12  sustaining natural population (Waples et al. 2001). Evaluations of supplementation programs, in which an increase in natural spawners or a net gain in productivity has been measured are few in number. Results of such evaluations often indicate that the desired increase in productivity is not being obtained (Nickelson et al. 1986; Winton and Hilbora 1994; Unwin and Glova 1997; Fleming and Petersson 2001; V0llestad and Hesthagen 2001; Waples et. al. 2001).  Furthermore, published reviews of the literature on supplementation often do not differentiate between multigeneration or single generation hatchery stocks or the quality of hatchery salmonids introduced. Several factors have been determined to affect the survival of hatchery salmonids such as rearing techniques (Berejikian 1999), growth rate (Beckman 1999), time and site of release (studies cited in Jonsson 1997; Youngson and Verspoor 1998), age and size of the released fish, water quality, sexual maturity and conditioning for survival in the wild ( reviewed in Finstand and Jonsson 2001; V0llestad and Hesthagen 2001). In addition, time and site of release affect the homing of the adultfish(Jonsson 1997; Finstad and Jonsson 2001). Stocking density is also important as large scale introductions may cause density dependent effects (Finstad and Jonsson 2001; Vollestad and Hesthagen 2001). Attention to these factors will likely improve the effectiveness of supplementation programs (Campton 1995; Einum and Fleming 2001).  Waples et al. (2001) reviewed hatchery programs in which single generation hatchery broodstock (i.e. adults for broodstock collected from the wild and juveniles released into the wild) derived primarily from a local, native population were used. Waples et al. (2001) determined that there were no controlled studies with multiple replicates with pre- and post-treatment data that were complete. Existing controlled studies had datafrompre-treatment and while stocking was ongoing. Paired supplemented and control populations had a similar abundance trend infiveof eight cases, the supplemented population outperformed the control population in two cases, while the reverse was true for one case. Waples et al. (2001) concluded that the premise that supplementation can be used to provide a net long-term benefit to natural populations remains an untested hypothesis.  13  The goal o f conventional hatcheries, i.e. to increase fish availability for harvest should be differentiated from the goal o f supplementation hatcheries, i.e. to increase production over the long-term, as they pose different risks and benefits (Fleming and Petersson 2001; Waples et al. 2001). The effectiveness o f hatchery programs, their risk (Hindar et al. 1991; Busack and Currens 1995) and their cost (Hilborn and Winton 1993; Winton and Hilborn 1994; Naylor et al. 1998) must be weighed against alternative methods o f achieving improved natural production such as habitat improvements and increased escapement (Hilborn and Winton 1993; VMlestad andHesthagen 2001).  2.1.2 Intraspecific Competition between Farm and Wild Salmonids  The differences between cultured and w i l d fish can be particularly extreme i n farm salmonids as they are reared i n an artificial environment for their entire lifespan and broodstock is artificially selected over multiple generations for specific characteristics that are favourable i n the farm rearing environment (Fleming et al. 1994). For example, rapid growth i n the farm rearing environment is typically artificially selected for and farm fish often outgrow w i l d conspecifics (Hershberger et al. 1990, see references below). In addition, farmed salmon are often derived from several populations that may be non-native (Gjedrem et al. 1991).  The following review o f intraspecific competition between farm and w i l d salmonids is based on Atlantic salmon as many studies have been conducted on this species, and information on other species was not encountered during the literature review. A s with hatchery salmonids, relative 1 ft to w i l d salmonids, farm fish exhibit increased aggression (Einum and Fleming 1997 ' ; Fleming and E i n u m 1997 ' ), reduced predator avoidance (Einum and Fleming 1997 ; Fleming and E i n u m 7 9  8  1997 ), altered patterns o f movement (Carr et al. 1997 ; see references i n Jonsson 1997), 9  8  morphology (Fleming et al. 1994; Fleming and E i n u m 1997 ), male parr maturity (Fleming and 9  E i n u m 1997 ), and spawning behaviour (Gausen and M o e n 1991; Webb et al. 1991, 1993; 9  0 k l a n d et al. 1995; Heggberget et al. 1996; Sasgrov et al. 1997 ). 8  Comparison of intrapopulation aggression. Comparison between farm salmon and wild populations other than the primary founder populations as well as hybrids. Comparison between farm salmon and primary founder population.  7  8  9  14  Juveniles: As discussed above, stream dwelling juvenile salmonids compete for space with dominant individuals obtaining more profitable feeding territories and achieving higher growth. The relative competitive ability of farm and wild salmonids may therefore affectfitness.As with hatchery salmonids, farmfishoften exhibit higher levels of aggression (Einum and Fleming 1997; Fleming and Einum 1997), which may be associated with dominance, and farm fish potentially have a competitive advantage in this regard. A residence and/or size advantage is possible for either group as conflicting results have been obtained in various experiments on different populations measuring time of spawning or relative growth of farm and wild juveniles. In some instances farm Atlantic salmon spawn earlier (Lura and Saegrov 1991; Saegrov et al. 1997; Fleming et al. 2000) while in others they spawn after wild conspecifics (Webb et al. 1991). Earlier emergence of farm juveniles may therefore occur (Lura and Saegrov 1991) resulting in reduced survival if unfavourable environmental conditions are encountered or, alternatively, a residence and size advantage.  Experiments specifically investigating relative competitive ability of size matched farm versus wild juvenile salmonids has also yielded variable results. Fleming and Einum (1997) investigated the competitive ability of farm Atlantic salmon relative to their principal founder population and found that there was no difference in a tank environment but wild salmon dominated in a stream-like environment. In a comparison of the same farm strain with two other wild populations (River Imsa and River Lone) and hybrids of the farm strain and wild populations, variable results were obtained in a stream-like environment. Farmfishdominated wild Imsa juveniles but were competitively equal to the hybrid juveniles (as were the hybrid and wild Imsa juveniles). The hybrids between the farm strain and the Lone juveniles dominated both populations they originatedfrom,a result that may stemfromhybrid vigor. The farm and wild Lone juveniles were competitively equal (Einum and Fleming 1997). Resultsfroman experiment in an Irish river also suggested that farm salmon dominate wild juveniles, as the latter were displaced downstream (McGinnity et al. 1997). Fleming et al. (2000) also reported differential distribution of smaller wild and farm or hybrid progeny in a Norwegian River despite similar nest distributions of spawners, with wild progeny located further upstream suggesting segregation may be occurring due to competition. 15  Several studies have compared growth o f farm Atlantic salmon relative to w i l d conspecifics or hybrids between farm and w i l d Atlantic salmon. Greater growth o f farm fish can be due to superior competitive ability, but can also be explained by their greater growth efficiency. Regardless o f the cause, greater growth can result i n a competitive edge as greater size is associated with dominance and improved odds o f survival. In comparing farm salmon juveniles and the primary founder population from which they were derived, Fleming and E i n u m (1997) found that i n the hatchery environment, farm salmon outgrew w i l d salmon i n pure groups, but not i n competition. Competition resulted i n a decline i n the growth performance o f farm salmon, but did not significantly affect the w i l d salmon. In contrast, i n a stream-like environment w i l d juveniles outgrew the farmed juveniles across the experiments consistent with their observed dominance discussed above. E i n u m and Fleming (1997) also compared the growth o f the same farm strain with River Imsa juveniles and hybrids i n the hatchery and i n the wild. They determined that i n the hatchery growth performance was significantly higher for the farm juveniles than the w i l d or hybrid fish. In the w i l d , however, both farm and hybrid fish grew significantly faster than w i l d fish.  Performance o f farm (a Norwegian strain), w i l d and hybrid Atlantic salmon progeny i n an Irish river was investigated by M c G i n n i t y et al. (1997), by planting equivalent numbers o f eyed eggs over two consecutive years. Although survival (particularly early survival potentially from small egg size or predation) and condition were poorer i n farm progeny, they grew fastest and as mentioned above, competitively displaced the smaller w i l d fish to downstream river sections. Performance o f hybrids was generally intermediate between farm and w i l d juveniles. Fleming et al. (2000) obtained similar results i n a Norwegian river experiment, with progeny o f native spawners being smaller (due to later female spawning dates and lower offspring growth rates relative to farm juveniles) and differentially distributed as mentioned above. Fleming et al. (2000) determined that there was no difference i n survival o f farm, w i l d and hybrid progeny between their seaward migration and maturity.  It should be noted that many o f the studies cited above do not contrast farm fish with their founder population. The comparisons are still relevant, however, as Norwegian strains are being 16  farmed globally and are competing against various native populations i n nature. Nevertheless, it should be kept i n mind that the dominance and greater growth o f the farm fish relative to nonfounder populations may be due i n part to culture effects, but can also be explained by competitive superiority or growth performance o f founding populations o f the farm strains.  Spawning Adults: Fleming et al. (1996) investigated the relative competitive ability o f Atlantic salmon spawners i n experimental arenas. Farm Atlantic salmon were compared with a w i l d population from the River Imsa (i.e. not a principal founder population o f the farm population). Fleming et al. (1996) found that relative to w i l d salmon, farm females i n mixed groups, exhibited similar levels o f aggressive and submissive behaviours. Farm females were courted less frequently, constructed fewer nests, retained greater weight o f eggs unspawned, and had greater egg mortality than w i l d females. Consequently, i n mixed groups, farm female salmon had significantly lower reproductive success; approximately one third that o f w i l d females. In the absence o f intergroup competition farm female reproductive success was significantly reduced relative to mixed groups as eggs were typically unfertilized when only farm males were present. Relative to w i l d males, farm males were less aggressive, and courted and spawned less frequently. Farm males achieved only one to three percent o f the reproductive success o f the w i l d males. Intergroup competition did not reduce the reproductive ability or success o f either the farmed or w i l d males, indicating reproductive inferiority o f the farm males even i n the absence o f competition from w i l d fish.  In a larger scale study, Fleming et al. (2000) determined that farm females and males were competitively and reproductively inferior to the River Imsa spawners, attaining less than onethird their breeding success. A s before, farm males were reproductively inferior to farm females and the majority o f farm genetic contribution to progeny was through hybridization o f farm females with w i l d males. Progeny were followed to maturity and the lifetime reproductive success o f the farm salmon was 16% that o f the w i l d fish.  In contrast to Fleming et al. (1996), Lura (1995, cited i n Saegrov et al. 1997) found that spawning success o f farm Atlantic salmon females was affected by w i l d spawner density. Lura (1995) observed that at comparatively high densities o f w i l d salmon, farmed females have lower 17  spawning success however, at l o w densities o f w i l d spawners, the spawning success o f farmed females appeared to be similar to that o f w i l d fish.  In addition, a recent study by Garant et al. (2003), showed that precocious male parr may hasten introgession o f farm alleles into w i l d populations o f Atlantic salmon. In spawning arenas, farm parr had significantly higher fertilization success than w i l d parr o f the primary founder population o f the farm strain. The relative reproductive success o f the farm parr was almost twice that o f a hybrid strain and four times that o f the w i l d parr. A s the parr were raised in a common environment, the observed differences were inferred to be genetic i n origin. These differences may relate to behaviour or physiological traits, such as increased aggression and the risk prone nature o f farm salmonids which may influence dominance and therefore reproductive success.  Population Response: A s with hatchery salmonids, the relative importance o f competition between farm and w i l d salmon and long-term population level outcomes w i l l depend upon the number o f feral spawning adults relative to w i l d adults, their relative competitive ability and reproductive success as well as the relative competitive ability and fitness o f their offspring.  In some locations i n eastern Canada and Europe farming o f Atlantic salmon is particularly intense. Farm salmon that have escaped into the w i l d spawn successfully i n rivers (Lura and Saegrov 1991; Webb et al. 1993) and i n some rivers farm spawners outnumber w i l d spawners (Gausen and M o e n 1991; see references i n Heggberget et al. 1993; Sa?grov et al. 1997). Farm fish have been confirmed to be contributing progeny to juvenile salmon populations (Crozier 1993), often through hybridization with w i l d salmon (Crozier 1993; Fleming et al. 1996). The level o f introgression o f alleles from farmed fish is variable (Heggberget et al. 1993); however, in some systems a substantial percentage o f the genetic makeup o f the population is derived from farm fish (e.g. Sasgrov et al. 1997).  18  2.2 Interspecific Competition between Native and non-Native Salmonids Culture offishesmay cause additional interspecific competition due to introduction of a nonnative species. Such an introduction is of particular concern because many past cases have resulted in displacement of native species (e.g. Larson and Moore 1985; Moyle and Vondracek 1985; Herbold and Moyle 1986). There is potential for adverse effects as native species are not coevolved with the competitor. Niche overlap may occur and native species may lack appropriate defenses to deal with the novel competitor. Pacific salmonids have been introduced in many areas beyond their native range including the east coast of North America, Europe and New Zealand. The widespread introduction of hatchery salmonids has been motivated by a variety of factors including the relative hardiness of some species, fast growth and desirability as sportfish.More recently, Pacific salmonids are also being farmed beyond their native range. In Chile and on the east coast of North America for example, coho salmon and rainbow trout are farmed. Nevertheless, Atlantic salmon remain by far the most common farmed salmon. They have been farmed in Europe and on the east coast of North America for several decades but now additional farming of this species is being done in Chile and on the Pacific coast of North America. As the emphasis of my thesis is competitive interactions between farmed Atlantic salmon and wild coho salmon or coastal cutthroat trout, the following review of interspecific interactions between native and non-native species will focus on competition between Atlantic salmon and Pacific salmon species. Only a single study has been done with farmed Atlantic salmon, most investigations focusing on wild or hatchery Atlantic salmon and Pacific salmonid competition. The following review is based on competition between hatchery or wildfishand the single study on farmed Atlantic salmon has been incorporated.  Juveniles:  Several studies have investigated rainbow/steelhead competition with Atlantic  salmon as the juveniles of both species utilize similar ecological niches (i.e. they often are found inrifflehabitat and feed largely on invertebrate drift (Wankowski and Thorpe 1979; Kwain 1983). Three of four studies on juvenile competition were conducted due to concern related to  the introduction o f Pacific salmonids into the historical range o f Atlantic salmon and the potential difficulty o f re-establishing Atlantic salmon, i n the presence o f these introduced species. Based on a series o f experiments, Gibson (1981) concluded that steelhead are more aggressive and able to displace Atlantic salmon o f similar or slightly larger size from preferred locations. These results should be interpreted with caution, however, because the experiments were unreplicated and individual fish were often used i n multiple trials.  Hearn and Kynard (1986) also found that rainbow trout were more aggressive than Atlantic salmon. In artificial stream channels, Atlantic salmon utilized riffle habitat more frequently i n the presence o f rainbow trout, whereas rainbow trout distribution was not affected by salmon. In a field study, stocking o f Atlantic salmon fry did not affect the rate o f movement o f rainbow trout fry. The shift i n distribution o f Atlantic salmon i n the presence o f rainbow trout appears to be i n response to aggressive interactions with rainbow trout. Hearn and Kynard (1986) concluded that interspecific interactions may cause reductions i n Atlantic salmon production, although total displacement from stream sections appears unlikely.  Jones and Stanfield (1993) found that the growth and survival o f hatchery-reared Atlantic salmon juveniles i n stream enclosures was significantly reduced i n the presence o f underyearling coho salmon, yearling rainbow trout and brown trout (Salmo trutta) relative to sections where these fish had been removed. This effect is potentially attributable to either competition and/or predation. Though the study met its stated objective, the hatchery origin o f the newly introduced Atlantic salmon and density differences (i.e. higher density i n control sections) could explain or influence the results. These factors become important when the intention is to assess the relative competitive ability o f the two species and potential outcomes o f competition.  Volpe et al. (2001) investigated competition between steelhead trout and farmed Atlantic salmon due to concerns relating to introduction o f farmed Atlantic salmon on the Pacific coast o f North America. Steelhead trout and Atlantic salmon performance was investigated i n laboratory experiments by comparing performance o f the two species i n competition relative to intraspecific competition. Again, steelhead trout were found to be more aggressive than Atlantic salmon. Growth comparisons were made without statistical analysis and indicated that fish o f 20  either species with a residence advantage had greater growth than challengers. When growth o f each species i n interspecific competition was assessed relative to intraspecific competition however, the results.for steelhead growth generally indicated that steelhead trout outcompete Atlantic salmon. W h e n growth o f Atlantic salmon was assessed i n a similar manner, conflicting results are obtained, with Atlantics dominating i n some instances and steelhead trout i n others.  Several studies have investigated the relative competitive ability o f Atlantic salmon and coho salmon. The potential exists for food and space competition between juveniles o f these species as both feed largely on invertebrate drift (i.e. Atlantic salmon forage both on surface and suspended drift, Wankowski and Thorpe 1979; coho salmon primarily forage on surface drift, G l o v a 1984; Sandercock 1991). Coho salmon typically utilize pool habitat (Hartman 1965; G l o v a 1984) but are also found i n run and riffle habitat (Glova 1984). Atlantic salmon, while most often found i n riffle habitat (Heggenes et al. 1999), also utilize runs and pools (Rimmer et al. 1983; Hearn and Kynard 1986). Fausch (1988) indicated that Atlantic salmon are more abundant i n riffles during summer, especially i n the presence o f pool dwelling competitors (Gibson 1966,1978; Kennedy and Strange 1986). Fausch (1988), however, noted that Atlantic salmon may also shift to pools during winter, as they grow larger or when competitors are absent (see also Gibson 1993; Heggenes et al. 1999; Fjellheim and Johnson 2001).  A l l work to date related to coho salmon and Atlantic salmon interactions has been undertaken due to concern regarding coho salmon introduction i n areas where Atlantic salmon are native. Hearn (1978 cited i n Gibson 1981) found that coho salmon dominated juvenile Atlantic salmon o f a similar size i n a pool environment. Gibson (1981), however, concluded that coho salmon were less aggressive than Atlantic salmon but could displace Atlantic salmon smaller than themselves. A s discussed above, Gibson's (1981) conclusions must be interpreted with caution.  Jones and Stanfield (1993) results, summarized above, indicate that Atlantic salmon growth and survival could have potentially been reduced by competition with underyearling coho salmon. Alternatively, these results could be explained or affected by the presence o f other species (rainbow and brown trout) or effects such as predation, hatchery origin o f Atlantic salmon or density differences. 21  Beall et al. (1989) conducted a series o f experiments to investigate the effect o f the presence o f w i l d coho salmon at various juvenile stages on emerging w i l d Atlantic salmon fry i n seminatural stream enclosures. A n unreplicated experiment indicated that emerging coho salmon quickly developed a size advantage and appeared to have a negative effect on Atlantic salmon growth. Coho salmon juveniles and yearlings also adversely affected Atlantic salmon growth and survival and this effect was not due to predation. Increased or hastened Atlantic salmon emigration was also noted.  Spawning Adults:  T o my knowledge, there are no investigations o f competition between adult  Atlantic salmon and Pacific salmon species.  2.3 Research Addressed  The literature reviewed generally indicates that Atlantic salmon juveniles are competitively inferior to size-matched or larger rainbow/steelhead trout and coho salmon. Most o f the studies, however, were designed to test potential adverse impacts o f Pacific salmonids on Atlantic salmon. O w i n g to the large scale farm production o f Atlantic salmon on the Pacific coast, and the documentation o f offspring o f feral Atlantic salmon, additional work is critical to test scenarios that are relevant to the affect o f Atlantic salmon on Pacific salmonid species. In addition, only a single study utilized farmed Atlantic salmon, which have been shown to differ from w i l d or hatchery populations. A recent review o f issues related to w i l d salmon and aquaculture i n British Columbia identified considerable gaps i n our knowledge i n regards to interactions between farm and w i l d salmon and recommends a wide-ranging research and monitoring program o f these interactions (Gardner and Peterson 2003, P F R C C 2003).  M y thesis describes research undertaken to assess the relative competitive ability o f cultured and w i l d salmonids. The experiments primarily tested the relative competitive ability o f sizematched farm Atlantic salmon and w i l d coho salmon or w i l d coastal cutthroat trout, competing one-on-one for food. Factors observed to affect relative competitive ability, such as prior residence and size advantage were also investigated. Additional experiments were undertaken to 22  assess the relative competitive ability o f farm coho salmon or hatchery coho salmon and w i l d conspecifics. Results o f these experiments provide a context for more clearly formulated hypotheses about competition between these species i n the wild. Artificial stream channel experiments were also undertaken to investigate behaviour and growth rates i n groups o f farm Atlantic salmon competing with coho salmon for food and space i n a more natural setting.  23  CHAPTER HI - RELATIVE COMPETITIVE ABILITY OF CULTURED AND WILD SALMONIDS 3.1 Introduction  Currently on the Pacific coast o f North America, there are extensive salmonid culture programs which exist i n a variety o f forms including spawning channels and hatchery production o f juveniles for supplementation o f w i l d populations as w e l l as commercial farming enterprises. Concerns have arisen regarding the overall success o f supplementation programs (Hilborn and Winton 1993; Waples 2001) and potential negative effects on w i l d populations through a variety o f genetic and ecological processes (e.g. Hindar et al. 1991; Busack and Currens 1995; E i n u m and Fleming 2001).  The purpose o f my research was to investigate the relative competitive ability o f cultured and w i l d salmon to provide insight into the potential effects o f introduction o f cultured salmon on w i l d fish on the Pacific coast o f North America. M y research has focused on competitive interactions between juvenile farmed coho or Atlantic salmon and w i l d coho salmon or coastal cutthroat trout. M y working hypothesis is that cultured fish are competitively inferior to w i l d fish o f the same size, but are competitively superior i n the w i l d because they are generally larger for a given age.  Cultured fish have been observed to differ from w i l d populations from which they were derived i n a variety o f ways. M a n y studies have noted morphological and behavioural differences i n traits such as aggression and predator avoidance (Taylor 1986; E i n u m and Fleming 2001). In addition, some studies suggest changes i n feeding behaviour, habitat use (Einum and Fleming 2001) and spawning behaviour (Fleming et al. 2000).  Spawning channels and hatchery supplementation programs on the Pacific coast result i n the introduction o f large numbers o f cultured native salmonid species that interact with w i l d salmonids. Comparison o f cultured and w i l d conspecifics provides insight into differences o f cultured salmonids that may affect these interactions. Such differences are potentially caused by physical adaptation to the artificial rearing environment, natural selection for favourable traits i n  the culture environment, and artificial selection due to broodstock selection. They may affect the relative competitive ability o f the cultured fish and therefore impact w i l d populations.  Farmed fish experience longer and more intensive culture than juveniles derived from spawning channels and hatchery reared fish. Farmed fish are artificially reared to adults for harvest but may be introduced into the w i l d through accidental releases from juvenile rearing facilities, during transfers and from compromised net pens. Changes from culture are particularly extreme in farmed fish as they are reared i n an artificial environment for their entire lifespan and broodstock is artificially selected over multiple generations for specific characteristics that are favourable i n the farm rearing environment and for production efficiency.  Currently, Atlantic salmon (a non-native species) comprises about 80% o f production from netpen farming on the Pacific coast o f North America with chinook and coho salmon, two native species, making up the rest ( B C Salmon Farmers Assoc. 2003). Farming o f native species may result i n the introduction o f farmed fish that w i l l interact with w i l d fish, particularly native conspecifics. Farming o f non-native species may potentially result i n novel interspecific interactions as species that have not coevolved are exposed to one another. There are potential negative effects to native species due to niche overlap or lack o f appropriate defence mechanisms for the novel competitor. There have been many examples o f introduction o f nonnative organisms that have resulted i n displacement o f native species (Larson and Moore 1985; Lassuy 1995).  Research on competitive interactions between Atlantic salmon and Pacific salmonids has been limited to work on rainbow/steelhead trout and coho salmon. Generally these studies indicate that Atlantic salmon juveniles are competitively inferior to size-matched or larger rainbow/steelhead trout and coho salmon (e.g. Beall et al. 1989; V o l p e et al. 2001). Most o f the studies o f competition between Pacific salmonids and Atlantic salmon, however, were conducted on the Atlantic coast o f North America and Europe to test the possible effects o f introduced Pacific salmonids on native Atlantic salmon using w i l d or hatchery populations. The exception is Volpe et al.'s (2001) study, which examined competition between farmed Atlantic salmon and steelhead trout on the Pacific coast. Additional research is therefore required to assess the  potential effects o f introduction o f farmed Atlantic salmon to coho salmon and cutthroat trout populations on the Pacific coast o f North America, particularly, as farmed Atlantic salmon have been shown to differ from w i l d and hatchery Atlantic salmon (Fleming et al. 1994,1996; E i n u m and Fleming 1997; Fleming and E i n u m 1997). N o research has been conducted to my knowledge, on competitive interactions between farmed and w i l d coho salmon. Research on interactions between farmed and w i l d native salmonids is available for Atlantic salmon however, and indicates that farmed salmon may have negative effects on the fitness o f w i l d conspecific populations (Heggberget et al. 1993; Fleming et al. 2000) and as such there is the potential for similar effects i n coho salmon.  This chapter describes a series o f aquarium experiments i n which I investigated the relative competitive ability o f individuals competing one-on-one for food. T w o kinds o f contests were completed: Equal contests i n which fry o f equal size were simultaneously introduced and unequal contests i n which fry from one population were given 3.5 days prior residence or a size advantage. Prior residence and size advantage have often been found to be associated with superior relative competitive ability i n salmonids (Berejikian et al. 1996; Rhodes and Quinn 1998; Cutts et al. 1999; Johnsson et al. 1999). Results o f these experiments provide a context for more clearly formulated hypotheses about competition between these species i n the wild.  3.2 Materials and Methods 3.2.1 Source and Rearing Conditions for Juvenile Salmonids  Underyearling farm Atlantic salmon and coho salmon, hatchery coho salmon and w i l d coho salmon and coastal cutthroat trout were obtained from a number o f sources (Table 1). N o stocking o f the species o f interest has been conducted i n the Salmon River, Street, L o o n or Blaney creeks with the exception o f approximately 150,000 unfed coho salmon fry that were transplanted into the Salmon River between 1980 and 1988 ( B C Fisheries 2002; R. Cooke, Fisheries and Oceans, pers. comm. 2003; S. Hinch, U B C Forest Sciences, pers. comm. 2003; B . Stanton, Fisheries and Oceans, pers. comm. 2003), so I inferred that the coho salmon and coastal cutthroat trout collected from these locations were essentially wild. The Chilliwack hatchery 26  coho salmon were derived from broodstock originally collected from Salwein Creek (B. Stanton, Fisheries and Oceans, pers. comm. 2003). The farm Atlantic salmon were o f the M o w i strain and the farm coho salmon were derived from broodstock from the Kitimat River on the central British Columbia coast (J. Henry, Target Marine Products, pers. comm. 2002).  Table 1 - Source Information for Juvenile Salmonids Species  Type  Coho salmon  Hatchery, originally Salwein Creek stock Wild  Source  Method of Capture  Date of Arrival  Chilliwack Hatchery, May 2, 2001 Dip net Chilliwack, B.C. Coho salmon Salmon River, May 3, 2001 Pole Seine Langley, B . C . Coho salmon Wild Street Creek, Pole Seine May 6, 2001 Chilliwack, B . C . Coho salmon Farmed - originally May 10, 2001 Target Marine, Dip net Kitimat River stock Sechelt, B.C. Atlantic Farmed - Mowi May 10, 2001 Target Marine, Dip net salmon Strain Sechelt, B.C. Coastal Wild Loon Creek and Pole Seine Aug. 15, 2001 Blaney Creek, Maple cutthroat trout Ridge, B . C . salmon obtained from the Salmon River (a tributary of the Fraser River) near its confluence with Coglan Creek salmon obtained from Street Creek (a tributary of Hopedale Channel and the ChilliwackA'edder River) trout obtained from Loon Creek and Blaney Creek, tributaries of the North Alouette River) 1  2  3  2  3  A l l fish were immediately transported to the Department o f Fisheries and Oceans' Cultus Lake facility. E a c h population was held i n 4 m by 0.4 m troughs from the date o f capture (Table 1) until they were used in experiments, a minimum o f three weeks. A l l fish were fed commercial starter or #2 feed and typically maintained on a ration o f 1-2 percent body mass/day with the exception o f Atlantic salmon which were fed to satiation. In some instances ration was temporarily modified (up to 4%) for Pacific salmonid populations to attain rearing groups o f similar mean body size. Density and ration were equalized for specific populations o f Pacific salmonids at least one week prior to experiments. Coho salmon and coastal cutthroat trout were maintained at 7.5 to 11.5°C. Atlantic salmon were maintained at 10 to 14°C to encourage growth i n order to size match the populations. Fish were acclimated to experimental temperatures as required, by adjusting temperature by a maximum o f 1°C per hour. A l l experiments were completed by November 5,2001.  27  3.2.2 Experimental Setup and Data Collection  Contests between fry from different populations were conducted i n 40 L aquaria which were isolated with opaque plastic screens on four sides and from above with the exception o f a small (approximately 3 c m by 10 cm) screened viewing port. A layer o f small gravel covered the bottom o f each aquarium to a depth o f approximately 2 to 4 cm. Water was circulated through each aquarium at approximately 1 to 2 L/min. Water temperature was maintained at 11 °C for a l l experiments with the exception o f the contest between the two w i l d coho salmon populations (Table 2) which was conducted at 8°C. Three kinds o f experiments were conducted: 1) Equal experiments i n which size matched fry o f two different types were simultaneously introduced into the aquaria where they competed for access to food; 2) residence advantage experiments i n which the subordinate member o f a pair (as determined i n the equal experiments) was given a 3.5 day residency advantage i n the aquarium before its competitor was introduced; and 3) size advantage experiments i n which the subordinate member o f a pair was given a 2-20 m m advantage i n fork length but introduced simultaneously with its competitor into the aquarium. E a c h experiment consisted o f approximately 20 replicate aquaria. Size advantage experiments include additional replicates from a pilot experiment.  Populations to be used i n an experimental trial were not fed the day that they were introduced to the aquaria. Fish were collected from holding troughs and anaesthetized using clove o i l (40 ppm based on Taylor and Roberts (1999); 1 part clove o i l dissolved i n 10 parts ethanol as per Keene et al. (1998)), measured and weighed. In addition, to identify fish during intraspecific contests, caudal fins were clipped so that half o f the individuals from each population received clips which were randomly assigned to each replicate. Fry were placed separately i n 10 L holding containers to recover from anaesthetic prior to introduction. For equal contests, competing fry were matched for size (fork length differed by no more than 1 m m and weight by no more than 16%. In more than 90% o f cases fry differed i n weight by no more than 10%). Because Atlantic salmon tended to grow more slowly than Pacific salmon, size matching sometimes required that the smallest Atlantic salmon and the largest coho salmon be avoided (see Discussion). For some comparisons, (Salmon River and Kitimat coho salmon; Street Creek and Chilliwack Hatchery coho salmon) size matching i n the first set o f experiments could only be accomplished by 28  selecting the larger fish from one group and the smaller from the other. For these comparisons, a second set o f experiments was conducted after the experimental groups were better matched for size through control o f ration.  For residence advantage trials, competitors were size matched as above. In the Street Creek coho/Atlantic salmon comparison, however, 7 coho salmon exceeded Atlantic salmon i n weight by 11 to 29% although they met the 1 m m restriction for difference i n length. In the coastal cutthroat trout / Atlantic salmon comparison, 4 trout exceeded salmon fork length by 1.5 to 2 mm. In addition, 7 pairs that were matched for length differed i n weight from 11 to 20% (2 coastal cutthroat trout and 5 Atlantic salmon exceeding the weight matching criterion).  These  differences did not appear to effect dominance outcomes, i.e. fish with a weight disadvantage were often dominant and similar dominance results were obtained compared to fish matched within +/-10%. For size advantage experiments, Atlantic salmon were given a size advantage o f 2 to 20 m m (i.e. 3 to 38% length advantage) over a coho salmon competitor. In each experiment the size advantage given increased i n approximately 2 m m increments and each size had two replicates (i.e. 2 X 2mm, 2 X 4mm,...2 X 20 mm).  In equal contests and size advantage contests, each replicate consisted o f placing a pair o f fry into an aquarium simultaneously and allowing them to acclimate overnight for approximately 18 hours. For residence advantage experiments, Atlantic salmon were introduced i n the late afternoon, acclimated overnight and then fed at a ration o f approximately 2 % biomass/day over three subsequent days. Competitors were introduced on the morning o f the 5th day, resulting i n a residency advantage o f approximately 3.5 days. O n subsequent days food was introduced twice daily v i a the water inlet to the aquaria at a ration o f approximately 2 % biomass/day.  A  summary o f the experiments completed, the experimental conditions and start date as well as the average starting length and weights for each population are provided i n Table 2.  29  Table 2 - Summary of Aquarium Experiment Conditions, Date and Fish Size Start Date  May 28/01 July 30/01 Oct. 3/01 Oct. 16/01 Aug. 6/01 Oct. 10/01  June 5/01 July 6/01 Sept. 11/01 June 21/01 July 13/01 Sept 17/01 Aug. 13/01 Oct. 23/01 Aug. 13/01 Oct. 30/01  #Days Observed  Population 1  Mean Mean Population 2 Mean Fork Fork Weight Length Length (g)±SD (mm) + (mm) SD Equal Contests - Size Matched, Simultaneously Introduced Intraspecific Coho Contests 3 or 5 40.8+1.8 0.65 ± Salmon R. 40.7 ± Street Creek 0.12 2.0 coho (wild) Coho (wild) 3 or 5 Salmon R. 54.6 ±2.9 1.83 ± 54.7 ± Kitimat Coho 2.9 coho (wild) 0.32 (farm) 3 or 5 Salmon R. 63.2 ±4.6 2.73 ± 63.1 ± Kitimat Coho 0.62 (farm) 4.7 coho (wild) 3 or 5 69.2 ± 6.5 69.1 ± Street Creek 3.62 ± Kitimat Coho Coho (wild) 0.98 (farm) 6.5 3 or 5 Street Creek 59.0 ± 1.9 2.26 ± Chilliwack 58.9 ± 0.25 1.8 Coho (wild) coho (hatchery) 3 or 5 67.3 ±4.5 3.38 ± Chilliwack 67.2 ± Street Creek Coho (wild) 0.72 coho 4.6 (hatchery) Atlantic/Pacific Salmonid Contests 3 or 5 Salmon R. 41.2 ± 1.5 0.69 ± 41.0± Atlantic coho (wild) 0.11 (farm) 1.5 3 or 5 48.4 ±2.9 1.22 ± 48.9 ± Street Creek Atlantic Coho (wild) 0.23 (farm) 3.0 3 or 5 Cutthroat 54.9 ±2.8 1.58 ± 54.7 ± Atlantic 0.32 3.0 trout (wild) (farm) Unequal Contests - Size Matched, Atlantic salmon 3.5 days prior residence 4 or 6 Salmon R. 0.80 ± 42.0 ±3.2 Atlantic 42.1 + 0.22 (farm) 3.4 coho (wild) 4 or 6 48.8 ±2.2 1.29 ± 48.8 ± Street Creek Atlantic 0.19 Coho (wild) (farm) 2.2 4 or 6 58.6 ±4.5 58.0 ± Cutthroat 1.85 ± Atlantic trout (wild) 0.45 (farm) 4.5 Unequal Contests - Simultaniously Introduced, Atlantic salmon size advantage 5 Salmon R. 57.7 ±4.4 2.13 ± 68.4 ± Atlantic 0.44 (farm) coho (wild) 6.2 5 Street Creek 65.1 ±8.0 3.12± Atlantic 75.6 ± Coho (wild) 1.11 (farm) 10.8  Mean Weight (g)  0.66 ± 0.11 1.79 ± 0.29 2.73 ± 0.59 3.75 ± 0.97 2.28 ± 0.24 3.41 ± 0.69 0.70 ± 0.09 1.18 ± 0.22 1.58 ± 0.27 0.78 ± 0.2 1.15 ± 0.18 1.88 ± 0.41 3.33 ± 1.04 4.63 ± 2.04  Behavioural observations were made over a five minute period, on introduction of the fish to the aquaria and just prior to and during each twice daily feeding over the course of the experiment. In equal contests, observations were made for three days at which time half of the aquaria, chosen at random were observed for two additional days to determine if dominance relationships were stable. In addition, any aquaria in which dominance had not been determined (i.e. one fish 30  consistently dominant over four observation periods, see below) were maintained for the full five days. Prior residence experiments were similarly conducted but observations were also made immediately after competitors were introduced (i.e. observations were completed for four or six days). All aquaria in size advantage experiments were observed for five days (due to limited (two) replicates at each size increment).  Observations recorded included number of food pellets obtained, position relative to the feed tube outlet, submissive posturing and agonistic behavioral interactions (nip, chase, charge, approach and threat) occurring during a five minute period. It was also noted whether these interactions caused displacement of the other fish.  Observational criteria are defined as follows: Nip - Any biting motion made by one fish towards another; contact does not necessarily occur and thus, "nip" includes the "threat nip" of Chapman (1962)(Taylor and Larkin 1986). Chase - One fish pursued another fish for two or more body lengths without making physical contact (Taylor and Larkin 1986; McMichael et al. 1999). Charge - Any rapid, direct movement by an aggressor towards another fish; this behavior is distinguished from Approach by rapid acceleration (Kratt and Smith 1979; Taylor and Larkin 1986). Approach - Any movement by one fish towards another by swimming or drifting; this behavior usually initiates contests and is often followed by a nip or lateral display (Kratt and Smith 1979; Taylor and Larkin 1986). Threat - overt sign of aggression, such as fin-flares and body arching (Taylor and Larkin 1986; Holtby et al. 1993), includes lateral display and wigwags of Taylor and Larkin (1986). Submissive posturing - dorsal fin fully depressed with anal fin usually depressed and caudal fin folded; fish is motionless often on bottom or near surface (Holtby et al. 1993).  The fish were again measured for fork length and weight at the end of each experiment. Any trials in which a fish died (2), escaped (2), or had visible abnormalities (6) were discarded from the analysis. At the end of the experiments all fish were euthanized by an overdose of anaesthetic. 31  3.2.3 Data Analysis Immediately following introduction, fish typically displayed signs of stress such as sinking to the substrate for two or three minutes of the five minute observation period. Data collected during this period were not used in assigning dominance or behavioural analyses. Agonistic interactions which caused displacement and those that did not were pooled because interactions which did not cause displacement were few in number and usually involved approaches by a dominant individual toward a submissive subordinate.  Relative Competitive Ability: Individual dominance was based on three equally weighted criteria: 1) directing a greater number of agonistic interactions toward the other fish, 2) maintaining a position (or upstream position) in the food delivery area for a greater portion of the observation period, and 3) eating more food (i.e., a minimum of 7 pellets and > 60% of total obtained). An individual was classified as dominant if it achieved the greater score on these measures of dominance over at least four consecutive observation periods, observation periods in which equal scores were obtained excluded. If individual dominance could not be assigned based on these criteria, the fish were classified as equal competitors. Population dominance was determined based on the frequency of dominant individuals within each population using Pearson's Chi-square test. Equal competitors were given scores of 0.5 each toward their overall population's score.  A Chi-square test (Pearson's) was used to test for effect of fin clip on dominance (i.e. if marking adversely affected the relative competitive ability of clipped individuals) and was not statistically significant for any of the experiments.  Behavioural Interactions: The number of individual and total agonistic interactions was determined for eachfishin the aquaria. Population behaviour was compared by analysis of variance (ANOVA) on total behaviours observed and multivariate analysis of variance (MANOVA) using the general linear model procedure in SAS (SAS 2001), Version 8.2 (Meredith and Stehman 1991). Population and time (observation period) were included as 32  factors. The MANOVA was conducted using Pillai's trace (Pillai 1955). Post hoc analysis of individual behaviours was carried out by ANOVAs to provide additional information on behavioural differences. Transformations were ineffective in normalizing the skewed data distributions and equalizing variances of residuals, however, the analyses were conducted using Blom transformed data (Blom 1958). MANOVA is robust to skewed distributions and departures from equal variance and variance-correlation assumptions, particularly if Pillai's trace is employed (Zar 1999). MANOVA results generally concurred with the results of the ANOVA on total behaviours observed.  Growth: Average daily length and weight change was determined for each fish in the aquaria. Population growth was compared by analysis of variance (ANOVA) using the general linear model (GLM) procedure in SAS (SAS 1989), Version 8.2. A single factor ANOVA including population as a factor (equivalent to a t-test) was conducted, on either the untransformed data or Blom transformed data (Blom 1958), in order of preference. If transformations did not normalize the data, a non-parametric Wilcoxon two-sample test was conducted.  In addition, a two factor ANOVA of growth using the same method above was completed to allow the effect of status to be analyzed. Factors included were population and status (i.e. dominant, equal and subordinate as determined above). Analyses were conducted, in order of preference, on the untransformed data, In (y+1) transformed data, Blom transformed data (Blom 1958), or rank transformed data as the distributional characteristics of the data would allow. If the original and transformed data did not meet the assumptions of ANOVA, the results of analysis on the untransformed data and rank transformed data were compared and if they concurred the results were reported based on the probability level of the nonparametric test (Zar 1999).  33  3.3 Results 3.3.1 Relative Competitive Ability In equal contests (Table 3), w i l d Salmon River coho salmon were dominant to Street Creek salmon. Both w i l d coho salmon populations were competitively equal to farmed coho salmon. Variable results were obtained for Street Creek and Chilliwack Hatchery coho salmon; Chilliwack Hatchery fish were marginally dominant i n the first experiment when larger individuals were chosen from this population and smaller individuals from the Street Creek population. B y contrast, neither population was dominant i n the second experiment when the populations were better matched for size and there was no size bias i n the individuals utilized.  In equal contests between farmed Atlantic salmon ( M o w i strain) and w i l d coho salmon or coastal cutthroat trout (Table 3), Atlantic salmon were subordinate i n all cases. When Atlantic salmon were given a 3.5 day residency advantage however, they were competitively equal to both w i l d coho salmon populations but they remained subordinate to coastal cutthroat trout. Contests i n which Atlantic salmon were given a size advantage indicate that Atlantic salmon juveniles with a 3 to 38% length advantage were competitively equal to Street Creek coho salmon, but remained subordinate to the Salmon River population. The results from the contest between Atlantic salmon and Street Creek coho salmon suggest that a size difference o f approximately 10% fork length represents a threshold at which Atlantic salmon become competitively equivalent to Street Creek coho salmon. N o size threshold difference was observed i n the results from the contest between Atlantic salmon and Salmon River coho salmon.  34  Table 3 - Summary of Aquarium Experiment Results on Relative Competitive Ability (Chi-Square Test) Exp # 1 6 10 12 7 11 2 4 8 3 5 9 13 14  Population  # Dominant Population 2 # Dominant Dominant Probability Population Individuals Individuals Equal Contests - Size Matched, Simultaneously Introduced Intraspecific Coho Contests 16/19 0.0029 Salmon R. Street Creek Coho 3/19 Salmon R. coho coho (wild) (wild) (wild) Salmon R. 7/19 Neither 0.2513 Kitimat Coho 12/19 (farm) coho (wild) Salmon R. 8.5/19 Neither 0.6374 Kitimat Coho 10.5/19 coho (wild) (farm) 0.6547 Street Creek 11/20 9/20 Neither Kitimat Coho (farm) Coho (wild) Street Creek 5/20 Chilliwack coho 15/20 Chilliwack coho 0.0253 Coho (wild) (hatchery) (hatchery) Street Creek 10/20 Chilliwack coho Neither 1.0000 10/20 (hatchery) Coho (wild) Atlantic/Pacific Salmonid Contests Salmon R. <0.0001 19/20 Atlantic (farm) 1/20 Salmon R. coho coho (wild) (wild) 0.0047 Street Creek 15.5/19 Atlantic (farm) 3.5/19 Street Creek Coho (wild) Coho (wild) Cutthroat 18.5/20 1.5/20 <0.0001 Atlantic (farm) Cutthroat trout trout (wild) (wild) Unequal Contests - Size Matched, Atlantic salmon 3.5 days prior residence 6/20 0.0736 Salmon R. 14/20 Niether Atlantic (farm) coho (wild) Street Creek 13.5/20 6.5/20 Neither 0.1083 Atlantic (farm) Coho (wild) 12.5/14 1.5/14 0.0023 Cutthroat Atlantic (farm) Cutthroat trout trout (wild) (wild) Unequal Contests - Simultaniously Introduced, Atlantic salmon size advantage Salmon R. 23/27 4/27 Atlantic (farm) Salmon R. coho 0.0003 coho (wild) (wild) 13.5/24 10.5/24 0.5316 Street Creek Atlantic (farm) Neither Coho (wild) 1  3.1.2 Behavioural Interactions  In coho versus coho salmon experiments and in residence advantage experiments, nip was usually the most frequent behaviour (16/18 experiments) followed by approach (12/18 experiments) (Table 4). In equal Atlantic versus coho salmon or coastal cutthroat trout experiments and in size advantage experiments, approaches were typically the most frequent interaction observed (9/10 experiments), followed by nipping (9/10 experiments). Charge was consistently the least common behaviour observed. 35  The total number of agonistic interactions initiated by both competitors per five minute observation period in intraspecific contests (mean 6.37) was consistently three to four times greater than in interspecific contests (mean 1.61) (Table 4). Table 4 - Mean Number Behavioural Interactions per Observation Period for each Population (averaged over the course of each experiment) Exp #  n  Population  Nip  Chase  Charge  Approach  Threat  Equal Contests - Size Matched, Simultaneously Introduced Intraspecific Coho Contests 1 19 Salmon River coho 1.45* 0.60* 0.24* 0.52* 0.36 Street Creek coho 0.75 0.21 0.05 0.35 0.32 6 19 Salmon River coho 0.73 0.22 0.40 0.02 0.48 2.17* Kitimat (farm) coho 0.94* 0.29* 0.83 0.44 10 19 Salmon River coho 1.61 0.81 0.21 0.75 0.38 Kitimat (farm) coho 2.49 0.90 0.19 0.41 0.82 12 20 Street Creek coho 0.96 0.35 0.06 0.69 0.12 Kitimat (farm) coho 2.52* 0.81* 0.13 0.95 0.15 7 20 0.54 0.22 0.34 Street Creek coho 0.02 0.60 1.81* Chilliwack Hatchery 0.66* 0.10 0.74* 0.58 Coho 11 20 0.81 Street Creek coho 1.24 0.31 0.08 0.21 Chilliwack Hatchery 0.47 0.07 1.65* 1.02 0.12 coho Atlantic/Pacific Salmonid Contests 2 20 Salmon River coho 0.68* 0.07* 0.02 0.74* 0.11* Atlantic (farm) 0.11 0.01 0.01 0.13 0.07 4 19 Street Creek coho 0.52* 0.06 0.03 0.76* 0.12 0.27 0.04 0.11 0.04 Atlantic (farm) 0.01 8 20 Cutthroat trout 0.48* 0.15* 0.13* 0.63* 0.03 Atlantic (farm) 0.01 0.00 0.00 0.04 0.00 Unequal Contests - Size Matched, Atlantic salmon given 3.5 days prior residence 3 20 Salmon River coho 0.01 0.09* 0.22 0.02 0.21* 0.48* 0.03* 0.03 0.07 0.04 Atlantic (farm) 5 20 Street Creek coho 0.29 0.02 0.54* 0.15* 0.02 Atlantic (farm) 0.59 0.03 0.00 0.10 0.03 14 9 0.72* 0.14 Cutthroat trout 0.18* 0.08* 0.46* 0.05 0.01 0.01 0.04 0.00 Atlantic (farm) Unequal Contests - Simultaniously Introduced, Atlantic salmon given size advantage 13 27 Salmon River coho 0.04 0.03 0.80* 0.04* 0.12 Atlantic (farm) 0.06 0.01 0.00 0.09 0.00 24 14 0.16 Street Creek coho 0.03 0.02 0.76* 0.13* 0.23 0.24 Atlantic (farm) 0.05 0.03 0.00 * Statistically significant difference based on ANOVA.  Total  3.17* 1.68 1.84 4.67* 3.76 4.81 2.18 4.56* 1.72 3.87* 2.64 3.33 1.62* 0.32 1.49* 0.48 1.42* 0.05 0.55 0.65 1.02* 0.75 1.58* 0.11 1.03* 0.16 1.08* 0.55  36  In intraspecific contests, the average number o f agonistic interactions performed by each population by observation period showed considerable variation (Figures 1 to 6). There were no differences between populations i n a given experiment for change i n the mean number o f agonistic interactions over time (i.e. there is no evidence that the trends for each population i n a given experiment are not parallel) (p<0.05, A N O V A for total behaviours and M A N O V A ) . There was a statistically significant change (decrease) i n the mean number o f agonistic interactions over time i n two o f the six experiments (p<0.05, A N O V A for total behaviours and M A N O V A ) (Figure 1 and 5). A significant difference in the mean number o f agonistic interactions over time was also observed i n two additional comparisons (Figure 2 and 6) for the A N O V A for total behaviours only (p<0.05). In four o f the six experiments the difference i n agonistic interactions between populations was significant. Salmon River coho salmon were more aggressive than those from Street Creek (p<0.05, A N O V A for total behaviours and M A N O V A ) (Figure 1). Post hoc A N O V A s on individual behaviours indicate that Salmon River coho salmon initiated a l l behaviours significantly more with the exception o f threats (Table 4). Farm-raised Kitimat coho salmon were more aggressive than Street Creek coho salmon (p<0.05, A N O V A for total behaviours and M A N O V A ) (Figure 2) and those from the Salmon River i n the first o f two experiments (p<0.05, A N O V A for total behaviours only) (Figure 6). Analysis o f individual behaviours indicated that more nips, chases and charges were initiated by farm Kitimat coho salmon. Chilliwack hatchery coho salmon were more aggressive than Street Creek coho salmon in the initial comparison o f the populations (p<0.05, A N O V A for total behaviours and M A N O V A ) (Figure 5). M o r e nips, chases and approaches were initiated by Chilliwack hatchery coho salmon.  In two o f these experiments Kitimat coho salmon (Figure 2 and 3) were found to be more aggressive despite the fact that the populations did not differ competitively. In experiments with farm Kitimat coho, dominant individuals were noted to be more overtly aggressive. Specifically, in the initial comparison o f Kitimat coho and Salmon River coho salmon, on average dominant Kitimat coho salmon initiated approximately twice as many agonistic acts as dominant Salmon River coho salmon. This was also true o f Kitimat coho relative to Street Creek coho salmon. O n average dominant Kitimat River coho salmon initiated approximately two and a half times as many agonistic acts as Street Creek coho salmon. 37  8 I  6  obs#  Figure 1 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 1, Salmon R. (SR) / Street Cr. (SC) coho salmon  8 £ 6  0-1  ,  ,  ,  ,  ,  -I  0  2  4  6 obs #  8  10  12  Figure 2 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 12, Street Cr. (SC) / Kitimat (farm) coho (KT) salmon  0-1 0  , 2  , 4  , 6 obs#  , 8  , 10  1 12  Figure 3 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 6, Salmon R. (SR) / Kitimat (farm) coho (KT) salmon (1 exp.) st  38  Figure 4 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 10, Salmon R. (SR) / Kitimat (farm) coho (KT) salmon (2 exp.) nd  8  0  4  ,  1  ,  ,  ,  1  0  2  4  6  8  10  12  obs #  Figure 5 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 7, Street Cr. (SC) / Chilliwack Hatchery (CH) coho salmon (1 exp.) st  B  6  obs #  Figure 6 - Mean number of behavioural interactions by observation period - Aquarium Exp. #11, Street Cr. (SC) / Chilliwack Hatchery (CH) coho salmon (2 exp.) nd  In interspecific contests, the average number of agonistic interactions performed by each population by observation period (Figures 7 to 14), indicate that Pacific salmonids were typically 39  more aggressive than Atlantic salmon. In three experiments (Figure 10,11 and 12), change i n the mean number o f behavioural interactions over time differed among populations (p < 0.05, A N O V A for total behaviours and M A N O V A for Figure 11 and 12; M A N O V A only for Figure 10). The difference was most dramatic in contests between coastal cutthroat trout and Atlantic salmon with trout increasing their agonistic behaviour sharply about half way through the experiment (Figure 11 and 12). The mean number o f agonistic behaviours was significantly greater for coho salmon or coastal cutthroat trout than Atlantic salmon i n which the former populations were determined to be dominant (i.e. equal contests, Figure 7, 9 and 11 and trout versus Atlantic salmon with a residence advantage, Figure 12) (p < 0.05, A N O V A for total behaviours and M A N O V A ; A N O V A for total behaviours only for Figure 7). Analysis o f individual behaviours indicated that coho salmon and coastal cutthroat trout initiated a variety o f behaviours more often, particularly nips, chases and approaches. The mean number o f agonistic behaviours was also significantly greater for coho salmon than Atlantic salmon i n comparisons in which Atlantic salmon were given a size advantage (Figure 13 and 14) (p < 0.05, A N O V A for total behaviours and M A N O V A ) . Comparison o f individual behaviours indicated that these differences were likely due to an increased incidence o f approaches and threats by coho salmon. In the two remaining experiments i n which Atlantic salmon were given a residence advantage over coho salmon competitors (Figure 8 and 10), populations differed i n the mean number o f agonistic interactions (p < 0.05, A N O V A for total behaviours and M A N O V A for Figure 10; M A N O V A only for Figure 8). Analysis o f individual behaviours indicates that i n the first comparison o f Salmon River coho salmon, Atlantic salmon with a residence advantage initiated more chases and nips, but coho salmon initiated more approaches and threats (Table 4). In the comparison o f Street Creek coho and Atlantic salmon with a residence advantage, Street Creek coho salmon initiated more approaches and threats only.  40  0 0  2  4  6 o b s  8  10  12  #  Figure 7 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 2, Salmon R. (SR) coho / Atlantic (farm) (AT) salmon  4 .£ 3  0  2  4  6  8  10  o b s #  Figure 8 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 3, Salmon R. (SR) coho / Atlantic (farm) (AT) salmon prior residence  4  0A 0  ,  ,  ,  1  ,  1  2  4  6  8  10  12  o b s #  Figure 9 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 4, Street Cr. (SC) coho / Atlantic (farm) (AT) salmon  c  Figure 10 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 5, Street Cr. (SC) coho / Atlantic (farm) (AT) salmon prior residence  Figure 11 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 8, Coastal Cutthroat trout (CT) / Atlantic (farm) (AT) salmon  Figure 12 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 9, Coastal Cutthroat trout (CT) / Atlantic (farm) (AT) salmon prior residence  42  Figure 13 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 13, Salmon R. (SR) coho / Atlantic (farm) (AT) salmon size advantage  4 .£ 3  0  2  4  6  8  10  12  obs #  Figure 14 - Mean number of behavioural interactions by observation period - Aquarium Exp. # 14, Street Cr. (SC) coho / Atlantic (farm) (AT) salmon size advantage  3.1.3 Growth  Patterns of length and weight change were consistent with the exception of the experiment comparing Street Creek coho salmon with those from Kitimat (farm), in which Kitimat coho salmon grew considerably in length but lost weight (Table 5). This may have occurred because the Kitimat coho salmon were temporarily on a high ration prior to the experiment and weighed consistently more than their competitors at the time of the experiment. Tests of differences in growth between the two competing populations indicated that in all cases dominant populations had a greater daily length and weight change than subordinate populations. In four out of the six experiments noted above in which a population was determined to be dominant this effect was statistically significant for length, weight or both measures of growth. 43  In the remaining experiments in which populations were determined to be competitively equal, most differences in growth were non-significant with the exception of three out of eight comparisons. In the exceptions (i.e. the first comparison of wild Salmon River coho salmon versus farm Kitimat coho salmon (Table 5) and comparison of both wild salmon populations versus Atlantic salmon with a residence advantage (Table 6)) the faster growing population had 63 to 70 % dominant individuals and, therefore, status contributed to the size of the differences observed as described below.  Table 5 - Intraspecific Coho Salmon Contests - Mean Daily Length and Weight Change Exp. #1  Population  M=19  Salmon R. coho (wild) Street Creek Coho (wild)  «=19 #6 «=19 M=19  #7 n=20  Population Salmon R. coho (wild) Kitimat Coho (farm)  Population  Length Change (mm) NS 0.15  Weight Change (g)  Exp.  NS 0.020  #12 w=20  0.14  0.017  n=20  •S 0.12  *S' -0.007  #10 n=19  0.28  0.033  w=19  2  1  NS 0.09 2  Street Creek coho (wild) 7i=20 Chilliwack coho 0.13 (hatchery) NS = non-significant (p=0.05 level) *S = significant (p=0.05 level) t-test using untransformed data Wilcoxon two-sample test  1  *  i  -0.014  #11 «=20  0.014  «=20  s  Population Street Creek Coho (wild) Kitimat Coho (farm)  Population Salmon R. coho (wild) Kitimat Coho (farm)  Population Street Creek coho (wild) Chilliwack coho (hatchery)  Length Change (mm) NS 0.10  Weight Change (g) NS 0.003  0.19  -0.019  NS 0.21  NS 0.042  0.28  0.045  NS 0.16  NS 0.016  0.14  -0.001  2  1  2  2  2  1  1  2  44  Table 6 - Interspecific Atlantic Salmon and Coho Salmon or Coastal Cutthroat Trout Contests - Mean Daily Length and Weight Change Exp.  #2  Population  w=20 «=20  Salmon R. coho (wild) Atlantic (farm)  #4  Population  «=19 n=19  Street Creek coho (wild) Atlantic (farm)  #8  Population  n=20 w=20  Cutthroat trout (wild) Atlantic (farm)  #13  Population  n=27  Length Change (mm) *S 0.34  Weight Change (8) *S 0.029  «=20  0.14  -0.000  «=20  NS 0.12  NS 0.010  #5 «=20  0.09  0.006  «=20  NS 0.18  •S 0.016  w=14  0.10  -0.005  «=14  3  3  3  NS  3  Salmon R 0.23 coho (wild) «=27 Atlantic (farm) 0.20 size advantage NS = non-significant (p=0.05 level) *S = significant (p=0.05 level) t-test using untransformed data t-test using Blom transformed data Wilcoxon two-sample test  2  1  1  Length Change (mm) NS 0.14  Weight Change (g)  0.19  0.009  Population  *s  Street Creek coho (wild) Atlantic (farm) res. advantage  0.14  NS 0.009  0.08  0.008  0.14  0.024  0.06  -0.000 NS  Exp.  #3  #9  Population Salmon R. coho (wild) Atlantic (farm) res. advantage  3  3  * i s  -0.004  1  Population Cutthroat trout (wild) Atlantic (farm) res. advantage  #14  Population  NS  0.040  «=24  0.20  0.015  -0.003  w=24  Street Creek coho (wild) Atlantic (farm) size advantage  0.25  0.025  •s  1  1  1  1  2 3  Two factor ANOVAs which included population and status as factors, allowed the effect of status to be analyzed separately from population effects. These analyses indicated that dominant fish gained more weight and grew more in length than subordinant fish in all comparisons. The greater weight gain was statistically significant for all experiments with the exception of the comparison between wild coho salmon from the Salmon River and Street Creek. Increased growth in length was statistically significant in eight of 14 experiments.  45  3.4 Discussion 3.4.1 Relative Competitive Ability  Overall the experiments indicated that native Pacific salmonid juveniles are competitively superior to farm Atlantic salmon although when Atlantic salmon were given a size or residence advantage their competitive ability was similar to w i l d coho salmon. A m o n g w i l d coho salmon the Salmon River population was competitively superior to the Street Creek population but both were competitive equals with farm coho salmon derived from a third population (Kitimat River). C h i l l i w a c k hatchery coho salmon were competitively superior to w i l d Street Creek salmon i n one experiment but equal i n a second. Significant differences i n agonistic behaviour or growth rates within individual experiments were generally consistent with the pattern o f competitive abilities.  Relative competitive ability was assessed based on three factors which were given equal weighting: agonistic interactions, position and feeding. Aggression and obtaining energetically favourable stream positions are associated with dominance (Fausch 1984; Holtby et al. 1993). The association o f higher status with greater growth i n all experiments, is evidence that this approach was effective in assigning higher status to individuals that had superior performance for a fitness related trait.  The dominance o f coho salmon from Salmon River relative to those from Street Creek indicates that competitive ability may vary between w i l d coho salmon populations. Varying intraspecific competitive ability and aggressiveness has been demonstrated with other coho salmon populations (Rosenau and M c P h a i l 1987) as well as other salmonid species (e.g. chinook salmon, Taylor and Larkin 1986). Rosenau and M c P h a i l ' s (1987) results suggest a genetic basis for the coho salmon population differences. They inferred that the differences could be due to differences i n selection pressure for agonism i n the two source streams where growth conditions and densities o f predators and competitors vary. Such differences also exist i n the source streams o f the two w i l d coho salmon populations investigated. Higher numbers o f competitors and higher flow in the Salmon River relative to Street Creek may favour increased aggression i n  the Salmon River, however, the basis o f the competitive differences between these two populations is unknown. Such intraspecific variation i n competitive ability implies there is potential for intraspecific differences i n competitive ability relative to an introduced species such as the Atlantic salmon.  Both w i l d coho salmon populations were competitively equal to farmed coho salmon. This result may be due to lack o f an effect o f culture on competitive ability or could reflect population differences. Cultured salmonids have been observed to differ from w i l d populations morphologically and behaviourally (e.g. aggression, predator avoidance). Some studies show differences relative to the population's w i l d counterparts (see studies cited E i n u m and Fleming 2001). In many other instances, however, this may be due to population rather than culture effects (see studies cited E i n u m and Fleming 2001). Results o f studies o f Atlantic salmon indicate that farm fish were competitively superior only relative to other w i l d populations but not to the primary founder population from which the strain was derived (Einum and Fleming 1997; Fleming and E i n u m 1997; M c G i n n i t y et al. 1997).  Variable results were obtained for comparisons between coho salmon from Street Creek and the Chilliwack Hatchery. In the first experiment involving these populations Chilliwack Hatchery coho salmon were dominant but i n a subsequent experiment they were competitively equal. These conflicting results could be due to a size selection bias i n the first trial, as larger hatchery salmon were required i n order to size match the populations. Overall, i f the results o f the two experiments are pooled, the populations were competitively equal.  In equal contests between farmed Atlantic salmon ( M o w i strain) and the w i l d coho salmon populations or coastal cutthroat trout, Atlantic salmon were subordinate i n a l l cases. A s the smallest Atlantic salmon were avoided and size is generally positively correlated with dominance (Huntingford et al. 1990), this may have resulted i n a bias favouring Atlantic salmon (i.e. larger individuals are more likely to be dominant than randomly selected individuals). For this reason, my results suggest that Pacific salmonid species w i l l outcompete randomly selected Atlantic salmon juveniles.  47  Comparison o f the results o f the various experiments i n terms o f relative competitive ability o f the populations provides insight into the effect o f prior residence and size advantage which are often associated with dominance i n salmonids (Berejikian et al. 1996; Cutts et al. 1999; Johnsson et al. 1999). Huntingford et al. (1990) notes that larger size o f dominant fish may be a consequence and not a cause o f high status. In the current study, Pacific salmon species dominated Atlantic salmon i f evenly matched, but i n some instances prior residence and size advantage rendered them equal competitors. Specifically, when Atlantic salmon were given residency advantage they were competitively equal to both w i l d coho salmon populations, but they remained subordinate to coastal cutthroat trout. When Atlantic salmon were given a size advantage they were competitively equal to Street Creek coho salmon but remained subordinate to those from the Salmon River. Such factors may affect competitive outcomes i n natural systems and highlights the importance o f the conditions under which the competitors encounter one another (see Implications below).  3.4.2 Growth  Tests o f differences i n growth between the two competing populations indicated that i n a l l cases dominant populations had a greater daily length and weight change than subordinate populations, although not a l l differences were significant. Status had a significant effect on change i n weight for a l l but one experiment and on change i n length i n eight o f 14 experiments. The higher number o f non-significant results for growth i n length may be due to reduced measurement accuracy (+/- 0.5 mm) relative to the mean daily length change being measured or due to the unidirectional nature o f length change compared to weight which may go up or down. The fact that growth was affected by status (i.e. dominant individuals grew more than subordinate individuals) indicates that growth differences between the populations was at least partly behaviourally mediated.  3.4.3.Behaviour  Statistically significant population differences i n levels o f aggression were observed for two cultured salmonid populations. Farm Kitimat coho salmon displayed more agonistic behaviour 48  than Street Creek and Salmon River coho salmon (in the first o f two experiments) as did Chilliwack hatchery coho salmon toward Street Creek coho salmon (in the first o f two experiments). In the comparisons o f Kitimat coho salmon versus Salmon River coho salmon and Chilliwack hatchery coho salmon versus Street Creek coho salmon mentioned above, this could, i n part, be due to a status effect as more individuals were dominant than i n the competing population. This would therefore increase the average number o f agonistic interactions per observation for the population as a whole. Though a population effect cannot be ruled out and some differences were not statistically significant, i n all cases cultured coho salmon exhibited higher levels o f aggression relative to w i l d conspecifics, consistent with the literature on cultured salmonids (Einum and Fleming 2001).  In two instances farm Kitimat coho salmon were significantly more aggressive than w i l d fish, despite the populations being found to be competitively equal. This may also be due i n part to overt aggression o f dominant individuals as well as a greater number o f dominant individuals as mentioned above. Farm Kitimat coho salmon were observed to be particularly aggressive, with dominant individuals initiating approximately twice as many agonistic interactions than dominant individuals i n the w i l d populations i n two comparisons. Greater overall aggression o f the population, therefore, is not necessarily associated with dominance o f the population.  In three interspecific contests, the mean number o f behavioural interactions changed over time with coho salmon or cutthroat trout being highly aggressive despite limited or zero aggressive behaviours on the part o f the Atlantic salmon. W i t h the exception o f the equal contest between Salmon River coho and Atlantic salmon, the level o f aggression exhibited by the Pacific salmonids remained constant or increased over time, evidence o f overt aggressiveness toward interspecific competitors. In the intraspecific contests however, where there was a significant change i n aggression with time, the number o f agonistic interactions decreased with time.  Population differences i n levels o f aggression were observed i n a l l equal interspecific contests, with Pacific salmon being more aggressive. The same pattern was observed when Atlantic salmon were given a size advantage over coho salmon competitors, however aggressive interactions that differed significantly were limited to approaches and threats, indicating that the  size advantage was affecting the behaviour of the coho salmon competitors. In experiments in which Atlantic salmon had a residence advantage, Pacific salmonids remained more aggressive than Atlantic salmon in two of three comparisons. In all residence advantage experiments however, Atlantic salmon conducted more aggressive interactions relative to their competitors during the first observation period and during some subsequent observation periods with coho competitors. In addition, Atlantic salmon conducted significantly more overtly aggressive behaviours (nips and chases) than Salmon River coho, but the coho salmon conducted more approaches and threats. The greater aggression of Atlantic salmon noted in early observation periods could potentially have implications in natural systems if competitive outcomes are more important during initial encounters rather than prolonged periods, as the encounter could be brief with the subordinate competitor migratingfromthe area. Nonetheless, Atlantic salmon exhibited low levels of aggression overall, relative to those observed in Pacific salmon species.  3.4.4.Implications  Currently, in British Columbia, hatchery supplementation programs are widespread and intense and hatchery fish, therefore, interact with wild salmonfrequentlyas juveniles infreshwaterand marine nursery areas and as adults in spawning areas. Total introductions of hatchery salmon in British Columbia are on the order of half a billion fish per year (Alverson and Ruggerone 1997). Farmed salmonids are introduced into the wild through accidental releases. Although estimated rates of escapefromwell run aquaculture operations are small (1-2%) (Alverson and Ruggerone 1997) the number of feral farmfishthat can interact with wildfishincreases with aquaculture production. In British Columbia, for the period 1992 to 1996 it is estimated that approximately 100,000 farm salmon escaped per year (Alverson and Ruggerone 1997). 2002 production has increased approximately four fold relative to this time frame (BCSFA 2003), however, and current numbers of escaped farmfishare, therefore, on the order of a few hundred thousand salmon per year. In parts of Europe where production is greater and where natural production has been compromised due to a variety of factors, numbers of farmfishmay rival or exceed those of wildfishin the natural environment (Gausen and Moen 1991; Saegrov et al. 1997). In addition, progenyfromthe spawning of feral Atlantic salmon have been observed in three British Columbiarivers(Volpe et al. 2000; DFO 2003). The long term consequences of these 50  introductions is currently the source o f much debate (Alverson and Ruggerone 1997; Gross 1998; E i n u m and Fleming 2001). Current estimates o f risk are based on limited information and additional research is needed (Alverson and Ruggerone 1997; Waples 2001; P F R C C 2003).  Cultured native species w i l l interact with conspecifics to a greater or lesser extent depending on their point o f release, survival and rate o f homing or straying to rivers. Based on the present study, although cultured coho salmon exhibited higher (although i n most cases not statistically significant) levels o f aggression, they were not found to dominate w i l d conspecifics. The outcome o f competitive interactions between w i l d and cultured native species i n natural systems then may not favour cultured species unless they have a size advantage. M y study, therefore, suggests that consideration must therefore be given to the size at release o f hatchery salmonids and potential effects o f salmon o f farm origin. Farm salmon are artificially selected for rapid growth i n the culture environment so that the progeny o f farm fish or hybrids with w i l d fish may outgrow w i l d conspecifics (Einum and Fleming 1997; M c G i n n i t y et al. 1997; but see Fleming and E i n u m 1997). In addition, dominant individuals i n cultured fish populations may be overtly aggressive and though the population may not be dominant as determined by statistical analyses, a portion o f the cultured population may dominate some w i l d individuals. I f cultured fish are introduced i n large numbers, these dominant individuals have the potential to adversely affect a population o f w i l d individuals i f the size o f the w i l d population is small relative to the introduced population. In this regard, attention has to be given to the number o f hatchery salmonids introduced to a system and the number o f feral native farm fish that can potentially be liberated as a percentage o f total production.  The potential exists for food and space competition between Atlantic salmon and coastal cutthroat trout or coho salmon juveniles as a l l feed to some extent on invertebrate drift (Wankowski and Thorpe 1979; G l o v a 1984; Sandercock 1991) and may co-occur i n some habitats. Typically Atlantic salmon utilize riffle habitat, coho salmon pools and coastal cutthroat trout intermediate habitats, though a l l habitats are used to some extent by each species (Sandercock 1991; Sabo and Pauley 1997; Heggenes et al. 1999). For example, Atlantic salmon may utilize pools i n the absence o f pool dwelling competitors, as they grow larger and during winter (Fausch 1988). The distribution and degree o f interaction o f a non-native species such as 51  the Atlantic salmon with w i l d Pacific salmonids depends on the factors mentioned above as well as interactive segregation with novel competitors. Micro-habitat use by Atlantic salmon w i l l depend upon their innate habitat preference and the species assemblage o f the system i n question.  Based on studies o f size-matched juveniles, coastal cutthroat trout (present study), steelhead trout (Volpe et al. 2001) and coho salmon (present study, Beall et al. 1989) dominate Atlantic salmon. T i m i n g o f emergence and consequent residence and potential size advantage o f early emerging species can, however, affect these dominance relationships.  Though potential variation i n developmental traits between populations (Murray et al. 1990) and limitations o f the degree day concept (Beacham and Murray 1990) are complicating factors i n making simple predictions, consideration o f the approximate timing o f emergence o f Atlantic salmon relative to Pacific salmonid species is warranted due to the importance o f residence effects.  The potential timing o f emergence o f progeny o f feral Atlantic salmon in British  Columbia is currently unknown. Based on timing o f egg-takes from farm raised broodstock and observations o f spawning colouration and ripeness o f feral Atlantic salmon i n B . C . estimated timing o f spawning is early November to late January (Alverston and Ruggerone 1997; V o l p e et al. 2001). This timing corresponds with coho salmon spawn timing (Scott and Crossman 1998) and precedes that o f steelhead/rainbow trout and coastal cutthroat trout (mid-April to late June and February to M a y , respectively (Scott and Crossman 1998). A s coho salmon spawn timing corresponds with that anticipated o f feral Atlantic salmon and coho salmon require approximately 420 degree days o f incubation to hatch (Laufle et al. 1996) they probably emerge prior to Atlantic salmon which require approximately 500 degree days o f incubation to hatch (Jobling 1995). Coastal cutthroat trout require approximately 320 degree days o f incubation to hatch but due to their later spawn timing they may emerge before or with Atlantic salmon. Rainbow trout require approximately 375 degree days o f incubation to hatch (Jobling 1995), but due to the later spawn timing o f this species they may emerge after feral Atlantic salmon.  Additional research is required to assess the potential effects o f Atlantic salmon juveniles i n sympatry with multiple Pacific salmon species. Underyearling offspring produced by feral 52  Atlantic salmon will be dominated by underyearling coho salmon and coastal cutthroat trout with a size and residence advantage. As a result, juvenile Atlantic salmon may be driven into riffle habitat where they would compete with and potentially dominate rainbow trout juveniles, which may be smaller than underyearling Atlantic salmon (Volpe 2001a). In addition, limited (Heland and Beall 1997) or no information is available regarding interactions between Pacific salmon species and Atlantic salmon other than underyearlings. Anadromous Atlantic salmon typically spend two to three years infreshwaterprior to migrating to the ocean (Scott and Crossman 1998). Should Atlantic salmon become established in Pacific coast drainages, underyearling Pacific salmon will potentially encounter yearling or older Atlantic salmon with a size and residence advantage, which improves their relative competitive ability.  Small scale experiments in simple environments may overestimate competitive effects. In natural systems, additional habitat complexity increases opportunities for resource partitioning and may reduce competitive pressure (Hearn 1987). Investigations such as the present series of experiments, allows determination of relative dominance in forced competition. Despite their limitations they have provided a context for more clearly formulated hypotheses about competition in the wild and assist in interpretation of larger scale investigations.  53  CHAPTER IV - COMPETITIVE INTERACTIONS BETWEEN FARMED ATLANTIC SALMON AND NATIVE COHO SALMON IN SEMI-NATURAL STREAM CHANNELS 4.1 Introduction Introduction of cultured fish, whether through accidental release of farm fish or stocking of hatchery producedfish,has the potential to affect wildfishthrough a variety of genetic and ecological processes (Hindar et al. 1991; Fleming et al. 2000; Einum and Fleming 2001). In British Columbia, the recent discovery of offspring produced by feral Atlantic salmon (Salmo salar) in three Vancouver Island rivers (Volpe et al. 2000; DFO 2003) has brought a high profile to the issue of farming of a non-native species and increased the urgency for information relating to potential consequences of free ranging Atlantic salmon on the Pacific coast.  The purpose of this study is to investigate the relative competitive ability of farm Atlantic salmon and native coho salmon (Oncorynchus  kisutch). This chapter addresses competition for  food and space between Atlantic salmon and coho salmon juveniles during their freshwater phase to provide insight into the potential effects of introduction of Atlantic salmon on the Pacific coast.  Limited research on juvenile competition between Atlantic salmon and Pacific salmon species has been conducted. Several studies have investigated rainbow/steelhead (O. mykiss) competition with Atlantic salmon (Gibson 1981; Hearn and Kynard 1986; Jones and Stanfield 1993; Volpe et al. 2001a). The juveniles of both species utilize similar ecological niches as they often are found in riffle habitat and have similar feeding habits (Hearn and Kynard 1986). These studies indicate that Atlantic salmon juveniles are competitively inferior to size-matched or larger rainbow/steelhead trout.  The potential also exists for food and space competition between juvenile coho and Atlantic salmon in streams. Both species feed largely on invertebrate drift (Wankowski and Thorpe 1979; Glova 1984; Sandercock 1991). Coho salmon typically utilize pool habitat (Hartman 1965; Glova 1984) but are also found in run and riffle habitat (Glova 1984) whereas Atlantic  54  salmon, while most often found i n riffle habitat (Heggenes et al. 1999), also utilize runs and pools (Rimmer et al. 1983; Hearn and Kynard 1986).  Previous investigations o f coho and Atlantic salmon competition were conducted on the Atlantic coast o f North America and Europe to test the possible effects o f introduced coho salmon on native Atlantic salmon using w i l d or hatchery populations. These studies generally indicate that coho salmon dominate size matched or smaller Atlantic salmon (Hearn 1978, cited i n Gibson 1981) and results suggest that juvenile coho salmon adversely affect the growth and survival o f Atlantic salmon (Beall et al. 1989; Jones and Stanfield 1993). Increased or hastened Atlantic salmon emigration has also been noted (Beall et al. 1989). In contrast, Gibson (1981) concluded that coho salmon were less aggressive than Atlantic salmon though they could displace Atlantic salmon smaller than themselves. These trials were unreplicated, however, and should therefore be interpreted with caution.  A l l previous work on Atlantic / coho salmon juvenile competition, has utilized hatchery or w i l d salmon. Several studies have demonstrated that Atlantic salmon behavior and morphology can differ depending on the extent o f culture (Einum and Fleming 1997; Fleming and E i n u m 1997). Farm salmon used in the British Columbia aquaculture industry have been subjected to artificial selection for multiple generations and may differ substantially from w i l d conspecifics i n traits such as aggression and risk aversion (Einum and Fleming 1997). For this reason it is important to test competition between coho salmon and farm Atlantic salmon specifically, i n order to assess potential ecological effects o f feral Atlantic salmon on coho salmon.  In this chapter I present the results o f experiments designed to test the relative competitive ability o f the two species in semi-natural stream channels. The experiments were designed to compare the growth and behavior o f each species i n sympatry and allopatry. Comparisons involved three groups o f juveniles; w i l d coho salmon, hatchery coho salmon, and farm Atlantic salmon.  55  4.2 Methods 4.2.1 Source and Rearing Conditions for Juvenile Salmonids  Underyearling w i l d coho salmon were obtained by beach seining at the Salmon River (a tributary to the Fraser River) near its confluence with Coglan Creek i n Langley, B . C . on M a y 3,2001. N o coho salmon stocking has been conducted i n the Salmon River system with the exception o f approximately 150,000 unfed fry released between 1980 and 1988 (R. Cooke, Fisheries and Oceans, personal communication) so I infer that Salmon River coho salmon are essentially wild. Underyearling hatchery coho salmon were obtained from the Chilliwack Hatchery, on M a y 2, 2001. The broodstock for this hatchery population was originally obtained from Salwein Creek, a tributary to the ChiHiwack/Vedder River (B. Stanton, Fisheries and Oceans, personal communication). Juvenile Atlantic salmon ( M o w i strain) were obtained from Target Marine Ltd., a commercial aquaculture facility i n Sechelt, B . C . on M a y 10,2001. A l l fish were immediately transported to the Department o f Fisheries and Oceans' Cultus Lake facility. E a c h population was held i n 4 m by 0.4 m troughs until they were used i n experiments, a minimum o f six weeks.  A l l fish were fed commercial starter feed. Coho salmon were maintained on a  ration o f 1-2 percent biomass/day and Atlantic salmon were fed to satiation. Prior to use i n experiments, coho salmon were maintained at 8 to 9°C. Atlantic salmon were maintained at 10 to 14°C to encourage growth i n order to size match the populations. Fish were acclimated to experimental temperatures as required by adjusting temperature by a maximum o f 1°C per hour.  4.2.2 Experimental Setup and Procedure  Tests were conducted i n 4,5 m long sections o f artificial stream channels. E a c h section was divided from the upstream to downstream end, into a 0.3 m long by 0.9 m wide drop pool (to maximize water flow), a 2.4 m long by 0.7 m wide riffle section, a 1.2 m long by 0.9 m wide pool section and a 0.6 m long by 0.9 m drop pool. The fish were confined to the riffle and pool area using l / 8 inch vexar mesh screen. Substrate consisted o f half inch gravel. M e a n riffle m  depth (+/- S D ) was 6.4 +/- 0.6 c m and mean pool depth (+/- S D ) was 27.0 +/- 0.8 cm. Water was circulated through the channels at approximately 180 L / m i n resulting i n a mean riffle velocity 56  •(+/- SD) of 15.0 +/- 4.0 cm/sec and variable pool velocities with a mean (+/- SD) of 1.9 +/-1.8 cm/sec. Eight cobbles were placed in each riffle area to provide velocity refuges. Water temperature varied over the course of the experiments between 10 and 12.25°C with a mean temperature of approximately 11°C (Table 7). Three sets of experiments were completed between June 19 and July 30,2001 (Table 7). The first two sets included the four treatment groups: 12 Atlantic salmon alone; 12 wild coho salmon alone; 6 Atlantic plus 6 wild coho salmon; 12 Atlantic plus 12 wild coho salmon.  Table 7 - Start Size of Juvenile Salmon, Experimental Dates and Mean Water Temperature  Mean start length (cm) (+/-SD) Mean start weight (g) (+/- SD)  At  1 42.6 +/- 2.8  Experiment # 2 45.2 +/- 3.0  3 46.2 +/- 2.7  SRCo CHCo At  42.6 +/- 2.7 0.75+/-0.15  45.2 +/- 3.0 0.96+/-0.19  48.5 +/- 2.3 45.9 +/- 2.5 1.01 +/- 0.21  SRCo CHCo  0.76+/-0.17 June 19 June 30 11.1 +/-0.6  1.00+/-0.21 July 4 July 15 10.8+/-0.5  1.22+/-0.20 1.00 +/-0.20 July 16 July 30 11.0+/-0.4  Exp. Start Exp. End Mean Temp (°C) (+/-SD)  The density for each section was approximately 4.5 or 9 fish per square metre, the lower density being representative of natural stream densities of juveniles of the size utilized (Wankowski and Thorpe 1979; Dill et al. 1981). Six stream channels provided 16 sections which allowed four replicates of each treatment per set (two sections in which velocities differed the most from other sections were not utilized). The treatments were randomly assigned among sections within the channels with the constraint that three of the replicates had to be distributed in an upstream, midstream and downstream section.  57  A third experiment was conducted to address the effects o f prior residence (PR) on the competitive ability o f Atlantic salmon ( A ) relative to both w i l d coho salmon ( W C ) and hatchery coho salmon (HC). The treatments were: 12A; 1 2 W C ; 12HC; 12A(PR) + 12WC; 6 A ( P R ) + 6 W C ; 12A(PR) + 12HC; 6 A ( P R ) + 6 H C , 12A + 12HC; 6 A + 6 H C . Atlantic salmon were introduced to the stream channel three days prior to introduction o f the coho salmon. A s the hatchery coho salmon had not previously been assessed under conditions o f simultaneous introduction, treatment groups for these conditions were completed. W i l d coho salmon were from the Salmon River and hatchery coho salmon were from the Chilliwack Hatchery. T w o replicates were completed for each treatment.  Average starting size for the fish used i n each set o f experiments ranged from 42.6 m m (0.75 g) in the first set to 48.5 m m (1.22 g) i n the third and species were matched for size within each set (Table 7). Size matching required that the smallest Atlantic salmon not be used (see Discussion). Fish were not fed the day that they were introduced to the artificial stream channel. Fish were collected from holding troughs and anaesthetized using clove o i l (1 part clove o i l dissolved i n 10 parts ethanol as per Keene et al. 1998, final concentration 40 ppm based on Taylor and Roberts 1999), measured, weighed and clipped (on adipose and/or caudal fins) so that individual fish could be identified when recovered from the channel. Fry were paired according to size (i.e. equal fork length (+/-1 mm) and weight (+/-10%) i n both s y m p a t i c and allopatric comparisons and paired fry were given similar fin clips. Fish o f each population for specific sections were placed separately i n 10 L holding containers to recover from anaesthetic before introduction to the artificial stream channel.  Fry for each section were introduced to the head o f the pool and allowed to acclimate overnight for approximately 10 to 15 hours prior to first observation. Starting on the second day, food (commercial starter) was introduced into each section by automatic feeders at the head o f the riffle and pool. The total ration was approximately 2 % body mass/day, divided equally between riffle and pool feeders and was introduced over approximately 10 hours during the day. For set 1 and 2, observations continued for ten days (days 2-11 o f the experiment) and on day 12 the fish were removed and the sections prepared for the next set o f experiments. For set 3, observations  58  continued for 13 days (days 2-4 while Atlantic salmon had prior residence and 5-14 after coho salmon were introduced) and on day 15 thefishwere removed. Afinemesh screen was placed at the head of each section to screen out any additional food in the form of plankton or commercial feed from upstream sections. Each section was maintained by daily cleaning of screens. Behavioural observations were made for one focalfishof each species in each section once daily through small openings in burlap screens which covered the channel. The focalfishwas choosen arbitrarily within each section with the constraint that afishpositioned upstream, midstream or downstream relative to conspecifics was observed in the replicate treatments. Recorded observations included number of feeding attempts and agonistic behavioral interactions (nip, chase, charge, approach) of a focalfishin the pool sections overfiveminutes (see Chapter 3 for behavioural interaction definitions). The number offishof each species in the pool was counted and the number occurring in the riffle section was estimated as the difference between the number stocked and the number in the pool. On day 12 (Set 1 and 2) and day 15 (Set 3), thefishwere recovered from the stream channel sections by minnow trapping and dip netting, measured for fork length and weighed. At the end of the experiments allfishwere euthanized by an overdose of anaesthetic. 4.2.3 Data analysis Average growth (both length and weight change), behavioural interactions per observation period and distribution per observation period (percentage offishusing pool habitat) were determined for each channel section. Data for experimental set 1 and 2 were pooled for analysis. Results for each population of coho salmon in experimental set 3 were analysed separately. Treatments were compared by analysis of variance (ANOVA) using the general linear model (GLM) procedure in SAS (SAS 2001), Version 8.2. Paired comparisons among treatments in the ANOVAs were made by Tukey's method. For experimental set 1 and 2, factors included were species and assembly (i.e. allopatric, sympatric high density, sympatric low density) and set was 59  included in the model as a block. For experimental set 3 analysis factors included were species, assembly for each coho salmon population, as well as residence for the Chilliwack Hatchery coho salmon population. Analyses were conducted, in order of preference, on the untransformed data, In (y+1) transformed data, Blom transformed data (Blom 1958) or rank transformed data as the distributional characteristics of the data would allow. If the original and transformed data did not meet the assumptions of ANOVA, the results of analysis on the untransformed data and rank transformed data were compared and if they concurred the results were reported according to the probability level of the nonparametric test (Zar 1999). For set 3, Atlantic salmon growth change is for 13 days of growth rather than 10 as in set 1 and 2. Atlantic salmon that were simultaneously introduced in set 3 (i.e. growth change was for 10 days) was scaled to 13 days for comparison purposes. Tests for effect of section position on growth using experimental set 1 / 2 length and weight data (i.e. upstream, midstream or downstream) was not significant (4 factor ANOVA). 4.3 Results Over the course of the experiments, one coho salmon died (the smallest salmon in an allopatric treatment group, set 1, day 8 of the experiment) and one Atlantic salmon died (one of the largest in an allopatric treatment group, set 3, day 13 of the experiment). No correction was made for the density difference resulting from the death of these fish. 4.3.1 Growth  Set 1 and 2 (Farm Atlantic salmon versus Wild coho salmon - simultaneous  introduction):  Coho salmon grew significantly faster than Atlantic salmon (Table 8; Tukey test for length, p < 0.05). On average (+/- S.E.) coho salmon grew 3.82 +/- 0.14 mm, gaining 0.29 +/- 0.015 g and Atlantic salmon grew 0.92 +/- 0.07 mm, losing -0.014 +/- 0.006 g. Analysis of the length-weight relationship for each species prior to and after the experiment indicates that the length to weight ratio for both species increased over the course of the experiment. There was also an effect of assembly and set (Table 8), salmon growing more in sympatric treatments and during set 2 than set 1 (Tukey test for length, p < 0.05). ANOVA results were consistent for length and weight 60  change data, with the exception o f a significant interaction between species and assembly for weight data (Table 8). The original length change means by treatment have a distribution consistent with that o f the weight change means, though the interaction was not significant. The transformation o f the length change data altered the relative distribution o f the means thereby reducing the probability o f an interaction effect. The interaction o f weight change data is present because allopatric coho salmon grew less relative to conspecifics i n s y m p a t i c treatments, than allopatric Atlantic salmon grew under a similar comparison. M a i n effects were not interpreted for weight change data due to the magnitude o f a significant interaction effect (Christensen 1996).  Table 8 - ANOVA Results for Growth - Set 1 and 2, Farm Atlantic salmon versus Wild coho salmon, simultaneous introduction. M e a n Length Change* 2 Factor A N O V A randomizec block design Source df F 0.05.6.41 Species 1 823.40 2 30.34 Assemblage Set (block) 1 5.75 Species*Assemblage 2 0.61 * Analysis conducted using ln+1 transformed data **Analysis conducted using untransformed data  P O.0001 <0.0001 0.0212 0.5490  M e a n Weight Change** Df 1 2 1 2  F 0.05.6.41  987.68 25.64 41.94 4.30  P O.0001 <0.0001 <0.0001 0.0201  M e a n growth o f both species o f salmon i n sympatry at high and l o w densities did not vary significantly (Table 9, Figure 15). Both species o f salmon grew significantly faster i n the presence o f the other species than when alone (Tukey test for length and weight change for coho salmon, length change only for Atlantic salmon, p < 0.05) (Figure 15). Length and weight data provided consistent results for all comparisons with the exception that the weight change o f Atlantic salmon i n the allopatric treatment was not significantly different than the l o w density s y m p a t i c treatment (Tukey test, p > 0.05).  61  Table 9 - Mean growth of Atlantic and Coho salmon by Assemblage - Set 1 and 2, Farm Atlantic salmon versus Wild coho salmon, simultaneous introduction Treatment  M e a n Length Change + S.E. (mm) 0.55 + 0.05 1.09 + 0.11 1.12 + 0.07 3.17 + 0.22 4.07 + 0.18 4.23 + 0.18  (»=8) At At At Co Co Co  with C o (low density) with C o (high density) with A t (low density) with A t (high density)  M e a n Weight Change + S.E. fe) -0.04 + 0.01 -0.01+0.01 0.01 + 0.01 0.23 + 0.02 0.31 + 0.02 0.33 + 0.02  5  B &  K +  o  4.5  4^  12Co(wthAt) 6Co(v«thAt)  3.5  3  12 Co  o  f  2  1 0 . 5 -1 0  « ^  12At(vuthCo)  i 6At(v«lhCo)  12At  Tieatment  *Vertical lines indicate significantly different growth (Tukey test, p < 0.05) Figure 15 - Mean Length Change by Treatment, Set 1 and 2, Farm Atlantic salmon versus Wild coho salmon, simultaneous introduction  Set 3 (Farm Atlantic salmon versus Wild coho salmon - Atlantic salmon residence  advantage)  Coho salmon grew significantly faster than Atlantic salmon (Two factor A N O V A for length and weight change, p < 0.05). O n average (+/- S.E.) coho grew 5.01 +/- 0.13 m m , gaining 0.462 +/0.025 g and Atlantic salmon grew 2.27 +/- 0.30 m m gaining 0.078 +/- 0.026 g. Length but not weight change data indicated assembly had a significant effect on growth (Two factor A N O V A , p < 0.05) although results were consistent. Allopatric treatments showed the least growth followed by l o w density sympatric treatments and then high density sympatric treatments. 62  Tukey tests did not differentiate consistently between assembly in terms of length change data for untransformed and rank transformed data. Analysis of growth of salmon by treatment group (Table 10) did not differentiate intraspecific differences in growth for either species (power = 0.47 and 0.22 for length and weight, respectively). Table 10 - Mean growth of Atlantic and Coho salmon by Assemblage - Set 3, Farm Atlantic salmon versus Wild coho salmon, Atlantic salmon residence advantage Treatment (n=2)  At At with Co (low density) PR At with Co (high density) PR Co Co with At (low density) PR Co with At (high density) PR  Mean Length Change + S.E. (mm) 1.44 + 0.19 2.33 + 0.00 3.04 + 0.13 4.77 + 0.31 5.00 + 0.17 5.27 + 0.15  Mean Weight Change + S.E. (g) 0.01+0.01 0.09 + 0.04 0.14 + 0.01 0.44 + 0.01 0.45 + 0.06 0.50 + 0.05  PR=Atlantic salmon given 3 days prior residence Set 3 (Farm Atlantic salmon versus Hatchery coho salmon - Atlantic salmon residence advantage):  Hatchery coho salmon grew significantly faster than Atlantic salmon (Table 11,  Three factor ANOVA for length and weight change, p < 0.05). On average (+/- S.E.) coho salmon grew 4.87 +/- 0.11 mm, gaining 0.435 +/- 0.029 g and Atlantic salmon grew 2.27 +/- 0.24 mm, gaining 0.070 +/- 0.019 g. Length but not weight change data indicated residence had a significant effect on growth (Three factor ANOVA, p < 0.05) though results were consistent and in order of increasing growth were allopatric treatments, simultaneous treatments and prior residence treatments. Assembly and residence did not have a significant effect on growth (Three factor ANOVA, p > 0.05). A Tukey test (p>0.05) did not differentiate between treatment groups for each species.  63  Table 11 - Mean growth of Atlantic and Coho salmon by Assemblage - Set 3, Farm Atlantic salmon versus Hatchery coho salmon, Atlantic salmon residence advantage Treatment (n=2)  At At with Co (low density) At with Co (high density) At with Co (low density) PR At with Co (high density) PR Co Co with At (low density) Co with At (high density) Co with At (low density) PR Co with At (high density) PR  Mean Length Change + S.E. (mm) 1.44 + 0.19 1.68 + 0.38 2.49 + 0.33 2.79 + 0.63 2.96 + 0.12 4.56 + 0.15 4.71 + 0.21 5.06 + 0.06 4.79 + 0.21 5.23 + 0.31  Mean Weight Change + S.E. (g) 0.01 + 0.01 0.02 + 0.05 0.10 + 0.01 0.09 + 0.04 0.13 + 0.05 0.39 + 0.05 0.37 + 0.02 0.44 + 0.03 0.44 + 0.06 0.54 + 0.10  PR=Atlannc salmon given 3 days prior residence  43.2 Behavioural Interactions Set 1 and 2 (Farm Atlantic salmon versus Wild coho salmon - simultaneous introduction):  The  average number of intraspecific interactions per observation period was significantly higher than interspecific interactions for coho salmon (Tukey test, p < 0.05). The average number of intraspecific behavioral interactions per observation period varied significantly by species and set, being highest amongst coho salmon and in Set 1 (Table 12). There was, however, a significant interaction between species and assembly (Table 12), apparently due to allopatric Atlantic salmon engaging in the fewest interactions and allopatric coho salmon engaging in the most (Table 13). The Tukey test (p >0.05) lacked power to differentiate between treatment groups by species (power=0.21). The most common type of intraspecific interaction for each species was nipping (68 and 43% of intraspecific interactions for Atlantic and coho salmon, respectively). The average number of interspecific behavioral interactions per observation period did not significantly differ between species or by assemblage; however, there was a significant difference between Set 1 and 2 with more interactions occurring during Set 1. The most common type of interspecific interaction for both species was approaches (59% and 55% of interspecific interactions initiated by Atlantic and coho salmon, respectively). Total interaction analysis was consistent with intraspecific results based on rank transformed data. The analysis  64  of the transformed data revealed a significant interaction that was not present i n the analysis o f untransformed data.  Table 12 - ANOVA Results for Behavioural Interactions - Set 1 and 2, Farm Atlantic salmon versus Wild coho salmon, simultaneous introduction Intraspecific*  Total*  Interspecific**  Two Factor A N O V A randomized jlock desiip i Source df Fo.05.6.41 df Fo.05.4.27 P 1 39.48 <0.0001 1 0.13 Species Assemblage 2 0.80 0.4575 1 0.01 Set (block) 1 5.18 0.0282 1 12.43 5.37 0.0085 1 1.00 Species*Assembly 2 * Analysis conducted using rank transformed data ** Analysis conducted using ln+1 transformed data  P 0.7229 0.9288 0.0015 0.3252  df 1 2 1 2  P O.0001 0.6741 0.0042 0.0013  Fo.05.6.41  38.88 0.40 9.21 7.81  Table 13 - Mean number of Behavioural Interactions per Observation Period - Set 1 and 2, Farm Atlantic salmon versus Wild coho salmon, simultaneous introduction Treatment («=8) At A t with C o (low density) A t with C o (high density) Co C o with A t (low density) C o with A t (high density)  Intraspecific 0.3 0.7 0.4 3.0 2.2 1.3  + + + + + +  0.2 0.3 0.1 0.5 0.9 0.3  Interspecific  0.3 + 0.1 0.3 + 0.1 0.3 + 0.1 0.2 + 0.1  Total 0.3 + 0.2 0.9 + 0.2 0.7 + 0.1 3.0 + 0.5 2.5+1.0 1.5 + 0.3  Set 3 (Farm Atlantic salmon versus Wild coho salmon - Atlantic salmon residence  advantage):  The average number o f intraspecific interactions per observation period was significantly higher than interspecific interactions (Three factor A N O V A , p < 0.05). The average number o f intraspecific, interspecific or total agonistic interactions per observation period did not vary significantly between the two species or by assembly (Two factor A N O V A , p > 0.05) (Table 14). The most common type o f intraspecific interaction between Atlantic salmon was nipping (56%) and between coho salmon was approaches (39%). The most common type o f interspecific interaction for both species was nipping (67 and 44% o f interspecific interactions initiated by Atlantic and coho salmon, respectively).  65  Table 14 - Mean number of Behavioural Interactions per Observation Period - Set 3, Farm Atlantic salmon versus Wild coho salmon, Atlantic salmon residence advantage Treatment (»=2) At At with Co (low density) PR At with Co (high density) PR Co Co with At (low density) PR Co with At (high density) PR  Intraspecific 1.7+1.5 1.4 + 0.7 0.4 + 0.2 1.9 + 0.2 1.6 + 0.1 1.1 + 0.8  Interspecific 1.1 + 0.7 0.2 + 0.1 0.1+0.1 0.1+0.1  Total 1.7+1.5 2.5+1.4 0.5 + 0.1 1.9 + 0.2 1.7 + 0.0 1.1 + 0.8  PR=Atlantic salmon given 3 days prior residence Set 3 (Farm Atlantic salmon versus Hatchery coho salmon - Atlantic salmon residence advantage):  The average number of intraspecific interactions per observation period was  significantly higher than interspecific interactions (Four factor ANOVA, p < 0.05). The average number of intraspecific or interspecific agonistic interactions per observation period did not vary significantly between the factors (Three factor ANOVA, p > 0.05). Total agonistic interactions varied significantly by assembly and residence but a Tukey test (p>0.05) failed to differentiate between treatments. Interactions in allopatric treatments and low density treatments in which competitors were simultaneously introduced were higher than other treatments (Table 15). As in experimental set 1 and 2, the most common type of intraspecific interaction for each species was nipping (54 and 45% of intraspecific interactions for Atlantic and coho salmon, respectively). The most common type of interspecific interaction for both species was approaches (56 and 48% of interspecific interactions initiated by Atlantic and coho salmon, respectively).  66  Table 15 - Mean number of Behavioural Interactions per Observation Period - Set 3, Farm Atlantic salmon versus Hatchery coho salmon, Atlantic salmon residence advantage Treatment Intraspecific (»=2) At 1.7+1.5 At with Co (low density) 0.9 + 0.8 At with Co (high density) 0.6 + 0.2 0.3 + 0.1 At with Co (low density) PR At with Co (high density) PR 0.4 + 0.4 1.9 + 0.5 Co Co with At (low density) 1.4 + 0.4 Co with At (high density) 0.5 + 0.4 Co with At (low density) PR 0.8 + 0.1 Co with At (high density) PR 0.5 + 0.1 PR=Atlantic salmon given 3 days prior residence  Interspecific 0.8 + 0.0 0.1 + 0.1 0.4 ±0.1 0.2 + 0.2 0.3 + 0.2 0.3 + 0.2 0.2 + 0.1 0.1+0.1  Total 1.7+1.5 1.7 + 0.8 0.6 + 0.1 0.7 + 0.2 0.6 + 0.3 1.9 + 0.5 1.7 + 0.6 0.8 + 0.2 1.0 + 0.1 0.5 + 0.1  4.3.3 Distribution and Activity Set 1 and 2 (Farm Atlantic salmon versus Wild coho salmon - simultaneous introduction):  Both  species were most abundant in the pool habitat. A significant difference in distribution was observed between treatment groups for both species and assemblage (Table 16). Coho salmon utilized the pool habitat significantly more than Atlantic salmon (Tukey test p<0.05). Significantly fewer fish utilized the pool habitat in allopatric treatments than sympatric treatments (Tukey test p<0.05). On average, approximately 65% of Atlantic salmon in isolation were observed in the pool habitat (Table 17), significantly less than all other treatments groups in which on average 81 to 89 % of the fish utilized pool habitat (Tukey test p<0.05). Table 16 - ANOVA Results for Distribution - Set 1 and 2, Farm Atlantic salmon versus Wild coho salmon, simultaneous introduction Distribution 2 Factor ANOVA randomized block design Source Df F 0.05.6.41 24.82 Species 1 Assemblage 2 14.95 Set (block) 1 0.11 Species* Assemblage 2 2.68  P <0.0001 <0.0001 0.7468 0.0805  67  Table 17 - Distribution of Atlantic and Coho Salmon - Set 1 and 2, Farm Atlantic salmon versus Wild coho salmon, simultaneous introduction Treatment («=8) At A t with C o (low density) A t with C o (high density) Co C o with A t (low density) C o with A t (high density)  M e a n % Fish i n Pool Habitat ± S.E. (mm) 65.5+ 1.4 82.0+ 3.3 81.0+ 2.8 81.6+2.1 87.3+ 2.6 89.1+ 1.4  Coho salmon were always observed i n the water column. Atlantic salmon were observed i n the water column, stationary on the substrate or conducting both o f these activities intermittently during an observation period (Table 18). Atlantic salmon tended to remain stationary on the substrate more often when i n isolation then when i n the presence o f coho salmon. Atlantic salmon (i.e., focal fish) activity varied significantly with presence or absence o f coho salmon (x  2  = 83.7, d.f. = 2, p < 0.0001). In addition, Atlantic salmon with coho salmon fed approximately 10 times as much as Atlantic salmon i n isolation.  Table 18 - Atlantic Salmon Activity - Set 1 and 2, Farm Atlantic salmon versus Wild coho salmon, simultaneous introduction  A t alone A t with C o  Count («=8) Percentage Count («=32) Percentage  Swimming 8 10.5 83 54.6  Intermittent 2 2.6 34 22.4  O n substrate 66 86.8 35 23.0  Set 3 (Farm Atlantic salmon versus Wild or Hatchery coho salmon - Atlantic Salmon  residence  advantage): A s i n the previous experimental sets, both species were most abundant i n the pool habitat (Table 19). For Salmon River coho salmon data, a significant difference i n distribution was observed between species (Two factor A N O V A ) for both untransformed and rank transformed data, coho salmon utilizing the pool habitat significantly more than Atlantic salmon. There was, however, also a significant interaction between species and assembly for the rank transformed data, as pool usage was highest for Atlantic salmon i n l o w density sympatric treatments but lowest for coho salmon i n l o w density sympatric treatments. A Tukey test (p>0.05) lacked power to differentiate between treatments for each species (power = 0.20). For  Chilliwack Hatchery coho salmon data there was also a significant difference i n distribution by species with coho salmon utilizing pool habitat more than Atlantic salmon. There were significant assembly and residence effects for rank transformed but not untransformed data. Again, a Tukey test did not differentiate between treatments for each species.  Table 19 - Distribution of Atlantic and Coho Salmon - Set 3, Farm Atlantic salmon versus Wild or Hatchery coho salmon, Atlantic Salmon residence advantage  Treatment (*=2) At A t with C o (low density) A t with C o (high density) A t with C o (low density) P R A t with C o (high density) P R Co C o with A t (low density) C o with A t (high density) C o with A t (low density) P R C o with A t (high density) P R PR = prior residence na = not applicable  M e a n % Fish i n Pool Habitat + S.E. Salmon River Coho Chilliwack Hatchery Coho 67.3 + 8.5 na na 84.2 + 4.2 80.4 + 2.9 91.2 + 0.5 na na 88.3 + 0.0 93.3+1.7  67.3 + 8.5 61.7+8.3 82.9 + 0.4 73.3 + 25.0 86.7 + 0.8 90.0 + 0.0 86.7 + 3.3 89.2 + 5.0 91.3 + 2.1 90.8 + 3.3  A s i n the previous experiments, Atlantic salmon tended to remain stationary on the substrate more often when i n isolation then when i n the presence o f coho salmon (Table 20) (post hoc comparison % = 76.9, d.f. = 2, p < 0.0001), activity increasing even when comparing treatment 2  sections before and after introduction o f coho salmon.  Table 20 - Atlantic Salmon Activity - Set 3, Farm Atlantic salmon versus Wild or Hatchery coho salmon, Atlantic Salmon residence advantage.  A t alone A t with C o  Count («=2) Percentage Count («=12) Percentage  Swimming 5 10.0 98 81.7  Intermittent 13 26.0 9 7.5  O n substrate 32 64.0 13 10.8  69  4.4 Discussion 4.4.1 G r o w t h  This study was designed to determine the relative competitive ability o f juvenile coho and Atlantic salmon. The expectation o f such a design was that a superior competitor w i l l exhibit improved growth i n sympatry relative to growth i n the presence o f conspecific competitors only. In the present study density differences i n sympatry d i d not appreciably affect the growth o f either species. In experimental set 1 and 2, both coho and Atlantic salmon grew significantly faster i n the presence o f the other species than when alone. This apparent conflict is resolved based on behavioural observations. Atlantic salmon i n isolation typically remained stationary on the substrate throughout the observation period (86.8 and 64% o f observations i n experimental set 1 / 2 and 3, respectively) whereas in sympatry with coho salmon they were typically observed in the water column (54.6 and 81.7% o f observations i n experimental set 1 / 2 and 3, respectively). Apparently, the presence o f coho salmon stimulated the Atlantic salmon to move up into the water column, swimming and feeding more actively. Based on these observations and dominance relationships between the species (Chapter 3), it appears that coho salmon obtain additional food ration by out-competing Atlantic salmon. A s the smallest Atlantic salmon were avoided and size is generally positively correlated with dominance (Huntingford et al. 1990), coho salmon would be expected to outcompete randomly selected Atlantic salmon juveniles. Atlantic salmon exhibit reduced growth i n isolation as food ration was not utilized due to lack o f activity relative to sympatric treatments. Their growth, therefore, is not reflective o f competitive pressures. Overall, coho salmon grew significantly faster than Atlantic salmon, even when Atlantic salmon were given a residence advantage. These results are consistent with investigations o f the relative competitive ability o f w i l d coho salmon and w i l d Atlantic salmon by Beall et al. (1989), in which coho salmon were observed to rapidly obtain a size advantage and reduced the growth o f Atlantic salmon fry, especially i n pools.  70  Cultured fish have been observed i n some instances to have reduced feeding rates relative to w i l d conspecifics (Einum and Fleming 2001; but also see M e s a 1991). Fenderson et al. (1968) observed that hatchery and w i l d Atlantic salmon o f the same social dominance level did not differ i n feeding rate, however i n separate groups o f hatchery and w i l d parr, hatchery salmon fed less than their w i l d counterparts. Fenderson et al. (1968) inferred that this was due to increased agonistic interactions between hatchery fish, however, rather than a lack o f activity as i n the current study (see below). Sosiak et al. (1979) and Bachman (1984) also observed reduced feeding i n hatchery fish after release but a specific cause was not determined. Differences between cultured and w i l d fish i n experimental settings could be due to relative differences i n stress levels or i n release experiments reflective o f differences that are temporary and w i l l be reduced or eliminated over time through acclimation or learning (Bachman 1984). Regardless o f the cause o f the reduced feeding i n allopatric treatments o f Atlantic salmon, this effect was reduced i n the presence o f coho salmon (see below).  Increased growth o f both species was observed with each subsequent experimental set and growth was significantly higher i n experimental set 2 than set 1. Greater absolute growth was anticipated i n the second experiment. The ration was consistent between the experiments (2% body weight daily), however, as the fish i n the second experiment weighed approximately 2 3 % more they received more feed by mass.  In experimental set 3, treatments involving Salmon River coho and Atlantic salmon with a residence advantage, did not show statistically significant intraspecific differences i n growth for each species. This result could be a statistical artifact due to l o w power. Because, however, the coho salmon growth was approximately the same for allopatric and sympatic treatments, the results suggest that the coho and Atlantic salmon were more equal competitors when the latter had a residence advantage. This interpretation is consistent with results obtained i n a series o f aquarium experiments (Chapter 3), but further work would be required to confirm this observation. Atlantic salmon growth was higher for a l l treatments but this could be due to greater activity i n allopatric treatments, a set effect or a residence effect. Again, comparison o f Atlantic salmon performance between allopatric and sympatic treatments does not reflect competitive ability due to lack o f activity i n allopatric treatments.  Results from treatments involving C h i l l i w a c k Hatchery coho and Atlantic salmon generally indicate the lowest growth for each species i n allopatric treatments and suggest that coho salmon are an equal or perhaps a superior competitor. Prior residence may have improved Atlantic salmon growth but a residence disadvantage did not appear to adversely affect the coho salmon suggesting additional work is required to resolve competitive relationships between these populations.  4.4.2 B e h a v i o r a l Interactions  The number o f intraspecific interactions was consistently higher than interspecific interactions. There was insufficient power to differentiate between treatment groups i n experimental set three. A s the number o f replicate treatments was l o w and behavioural interaction levels are often quite variable, behavioral interactions for experimental set 3 are not considered particularly well characterized, therefore, discussion is limited to results from experimental set 1 / 2.  Intraspecific interaction levels were only slightly higher than interspecific interaction levels i n Atlantic salmon but were approximately seven times higher i n coho salmon. Interspecific aggression levels did not vary between the two species but intraspecific aggression was higher between coho salmon than Atlantic salmon. Aggressiveness can vary with many factors including time o f day (Volpe et al. 2001a), season (Hartman 1965), food availability (Slaney and Northcote 1974; D i l l et al. 1981) and habitat characteristics (Hartman 1965). Observations for this study were made during the day and i n pool habitat only, as the experimental apparatus did not allow observations i n the riffles. Coho salmon i n sympatry with rainbow trout were observed by Hartman (1965) to be more aggressive i n pools than riffles, so i n the present study, this species aggressiveness may have been overestimated. Atlantic salmon activity was potentially underestimated as they have been observed to exhibit increased aggressiveness i n riffles (Kalleberg 1958) and at night (Volpe et al. 2001a).  Some Atlantic salmon i n sympatric treatments were observed to exhibit a school-type social structure i n the pool habitat, rather than defending individual territories. This type o f behaviour 72  is considered atypical of pre-smolt Atlantic salmon but has been observed previously (Kalleberg 1958; Wankowski and Thorpe 1979). This type of behaviour has been associated with reductions in current velocity (Kalleberg 1958) and could potentially be an artifact of the experimental apparatus (also see Distribution and Activity). 4.4.3 Distribution and Activity It was anticipated that Atlantic salmon would utilize the riffle habitat, particularly in the presence of a pool dwelling competitor due to interactive segregation (Fausch 1988) and as they are thought to be better adapted to defend territories in this type of habitat (Gibson 1973,1981). However, both coho and Atlantic salmon, particularly Atlantic salmon in sympatry, were observed to predominantly utilize pool habitat. This result could reflect a preference of Atlantic salmon for pool habitat when in sympatry, however, the presence of Atlantic salmon in pool habitat could have been underestimated, particularly in allopatric treatments. In allopatric treatments Atlantic salmon were more often stationary on the substrate than other treatments. Their cryptic colouration could have resulted in underestimation of Atlantic salmon presence in pools. Regardless of the potential for underestimating the Atlantic salmon presence in the pool habitat, the results indicate that coho salmon did not cause considerable, if any, competitive exclusion of Atlantic salmonfromthe pool habitat to the riffles as generally over 80% of the Atlantic salmon were present in the pools in sympatic treatments.  The presence of the majority of Atlantic salmon in the pool habitat could be reflective of behavioural differences of farmed salmon relative to wild Atlantic salmon. It has been noted that cultured Atlantic salmon, specifically hatchery salmon, swim higher above the substrate and are in contact with the substrate less often then wild Atlantic salmon (Sosiak 1979; Dickson and MacCrimmon 1982). In addition, farm Atlantic salmon have been observed to utilize habitat with lower current velocity (Einum and Fleming 1997). Alternatively, this observation could be an artifact of the experimental apparatus. Though small (i.e. < 7 cm) Atlantic salmon juveniles have been observed to utilize riffle habitat of the depth and velocity available in the experimental stream channels (Heggenes et al. 1999), the availability of slightly deeper (i.e. 10 20 cm) and higher velocity (i.e. 20 - 30 cm/sec) habitat may have increased riffle usage. 73  Although juvenile Atlantic salmon often maintain positions on the substrate with only minimal excursions to midwater (Kalleberg 1958), fish observed i n allopatric treatments were usually stationary throughout the observation period and feeding was reduced. Apparently the presence o f coho salmon stimulated activity and feeding o f the Atlantic salmon juveniles. The observation o f increased feeding i n the presence o f coho salmon is consistent with observations for some individuals i n previous aquaria experiments (Chapter 3). Atlantic salmon, that had not fed during a prior residence period i n which they were i n isolation began feeding after the introduction o f a coho salmon competitor. The presence o f active competitors, swimming and feeding, may have provided a visual cue to increase feeding motivation or reduce perceived risk of feeding. In contrast, Beall et al. (1989) concluded based on pseudo-replicated data that coho salmon juveniles suppressed the activity o f Atlantic salmon juveniles. A s i n the current study, more Atlantic salmon were motionless when alone than with coho salmon, although the results were not significant. Atlantic salmon predominantly swam intermittently and swam throughout the observation period more often when alone (although not significantly), however. The disparity i n the results could potentially be due to behavioural differences between cultured and w i l d Atlantic salmon, used i n the current study and B e a l l et al.'s (1989) study, respectively. Cultured salmonids have been observed to differ i n their feeding habits and risk avoidance (see studies cited i n E i n u m and Fleming 2001).  4.4.4 Implications  O n the north Pacific coast, there is concern that Atlantic salmon w i l l establish self-reproducing populations i n coastal rivers with negative consequences for native species. Atlantic salmon that escape from farming operations are introduced into the environment year after year, both as adults and to a lesser extent, juveniles. Despite recent improvements to limit escapes, it is estimated that 1-2% o f fish raised i n net cages escape annually (Alverson and Ruggerone 1997). In British Columbia, for the period 1992 to 1996 it is estimated that approximately 100,000 farm salmon escaped per year (Alverson and Ruggerone 1997). A s farming activity using similar practices expands, there w i l l be more feral salmon introduced annually. 2002 production has increased approximately four fold relative to the early 1990s ( B C S F A 2003) and therefore, 74  current numbers of escaped farm fish are on the order of a few hundred thousand salmon per year. With increased escapes there is greater potential for feral Atlantic salmon to establish selfreproducing populations, although many other factors also come into play.  Generally, the risk and effects of an introduced species establishing self-reproducing populations is difficult to assess (Crawley 1986; Hindar et al. 1991; Kareiva 1996) and currently the source of much debate (e.g. Alverson and Ruggerone 1997; Gross 1998).  In addition, the extent of  establishment can varyfroma few localized populations to many large, extensive populations and their effects on native species will vary. Due to the complexity of interactions that can potentially occur in the receiving community and the multiplicity of factors dictating population level outcomes (e.g. environmental conditions, size and status of wild populations, characteristics of introduced population), empirical risk assessment studies are required (Hindar etal. 1991).  The poor performance of Atlantic salmon in attempted introductions is cited as evidence that the risk of establishment of self-reproducing populations is very low (MacCrimmon and Gots 1979; Alverson and Ruggerone 1997). Failed deliberate attempts to introduce Atlantic salmon in B.C. and the Pacific Northwest (MacCrimmon and Gots 1979; Alverson and Ruggerone 1997), does add to the weight of evidence that risk of establishment is low; however, repeated attempts at introduction do not always produce the same result (Fleming and Petersson 2001; Todd 2001). In addition, it has been noted that under changing conditions the potential for establishment by Atlantic salmon may increase (Alverson and Ruggerone 1997; Gross 1998). Currently, in the north Pacific, progeny from the spawning of feral fish have been observed in three British Columbia rivers (Volpe et al. .2000; DFO 2003) indicating that concern is justified.  The potential exists for food and space competition between Atlantic and coho salmon juveniles. Both feed largely on invertebrate drift (i.e. Atlantic salmon both surface and suspended drift, Wankowski and Thorpe 1979; coho salmon primarily surface drift, Glova 1984; Sandercock 1991) and may co-occur in some habitats. Fausch (1988) indicates that Atlantic salmon are more abundant in riffles during summer, especially in the presence of pool dwelling competitors (Gibson 1966,1978; Kennedy and Strange 1986). Fausch (1988), however, noted that Atlantic 75  salmon may also shift to pools during winter, as they grow larger or when competitors are absent (see also Gibson 1981,1993; M a c C r i m m o n et al. 1983; Heggenes et al. 1999; Fjellheim and Johnson 2001). A s agonistic motivation has been observed to be weak during winter when rainbow trout and coho salmon co-occur i n pools (Hartman 1965), co-occurrence o f coho and Atlantic salmon and potential for. interaction i n winter may not be critical to population level outcomes.  It is important to differentiate between selective segregation which refers to innate differences i n resource use (e.g. habitat use) that precludes interaction and interactive segregation which refers to niche shifts by one or both species that results from interspecific competition (Nilsson 1967; Fausch 1988). Habitat use by Atlantic salmon w i l l depend upon their innate habitat preference and the species assemblage o f the system i n question. The optimal habitat for Atlantic salmon i n the presence o f multiple Pacific salmonid species is unknown. Previous studies have indicated that size matched juvenile rainbow/steelhead, coastal cutthroat trout and coho salmon are more aggressive than Atlantic salmon (Chapter 3; Volpe et al. 2001a). The common interactions between Atlantic salmon and Pacific salmon species may be dictated by interactive segregation among several species.  M y results suggest that coho salmon outcompete underyearling Atlantic salmon o f a similar size for food and space. O n the Pacific coast of North America underyearling coho salmon i n natural systems are anticipated to emerge prior to progeny o f feral Atlantic salmon. Coho salmon typically spawn between November and January i n North America (Scott and Crossman 1998) after approximately 420 degree days o f incubation to hatch (Laufle et al. 1986). Hatching usually takes place in early spring and alevins emerge after 2-3 weeks i n the gravel (Scott and Crossman 1998). In contrast, w i l d Atlantic salmon spawn i n October and November i n Atlantic Canada (Scott and Crossman 1998) and often as late as January i n Europe (Alverston and Ruggerone 1997). In eastern Canada, hatching usually takes place i n A p r i l and requires approximately 500 degree days o f incubation (Jobling 1995) and alevins emerge i n M a y or June (Scott and Crossman 1998). Based on timing o f egg-takes from farm raised broodstock and observations o f spawning colouration and ripeness o f feral Atlantic salmon i n B . C . estimated timing o f spawning is early November to late January (Alverston and Ruggerone 1997; V o l p e et  al. 2001a).  In addition, farm Atlantic salmon observed during an experiment did not display  physical or behavioural spawning characteristics until mid-January (Volpe et al. 2001b). A s their spawning time is anticipated to roughly coincide on the Pacific coast o f North America and coho salmon require approximately 80 fewer degree days o f incubation to hatching and spend less time i n the gravel after hatching, underyearling coho salmon are anticipated to have a residence advantage relative to underyearling Atlantic salmon.  Underyearling coho salmon are also anticipated to have a size advantage over underyearling Atlantic salmon due to their earlier emergence and more rapid growth relative to w i l d underyearling Atlantic salmon. This scenario is supported by the findings o f this study and those o f Beall et al. (1989) i n which emerging coho salmon obtained a size advantage o f approximately 5 to 10 m m length relative to emerging Atlantic salmon when eyed eggs at the same stage o f development were planted. These factors suggest that underyearling coho salmon w i l l outcompete underyearling Atlantic salmon produced from feral fish. It is important to remember, however, that growth for other populations o f coho salmon and Atlantic salmon, particularly Atlantic salmon o f farm origin which are artificially selected for rapid growth, may differ from the populations previously studied.  Additional research is therefore required to assess the potential effects o f Atlantic salmon juveniles i n sympatry with multiple Pacific salmon species. M y results suggest, that underyearling offspring produced by feral Atlantic salmon may, i n the presence o f underyearling coho salmon with a size and residence advantage, utilize riffle habitat, where they w i l l come into contact with rainbow trout juveniles which may be smaller than underyearling Atlantic salmon (Volpe et al. 2000) as well as other species. Additional research is also required to assess the effects o f underyearling Atlantic salmon on rainbow trout juveniles with a size disadvantage as well as other species of Pacific salmon juveniles that have not yet been investigated.  A s the Atlantic salmon is an introduced species locally, large scale experiments i n natural habitats are not feasible on the Pacific coast. The use o f artificial stream channels allowed us to increase the level o f habitat complexity over that i n most laboratory environments. Nevertheless, the results presented here may overestimate competitive effects as additional  habitat complexity may increase opportunities for resource partitioning and reduce competitive pressures (Hearn 1987). In addition, this study did not allow emigrationfromthe experimental sections so the differences in growth due to intraspecific and interspecific competitive effects would not be confounded by salmon density changes. Additional work should consider the role of migration in determining the ecological outcome of interaction between Atlantic salmon and Pacific salmonid juveniles. Typically coho salmon spend one or sometimes two years infreshwaterwhile anadromous Atlantic salmon spend two to three years infreshwater(Scott and Crossman 1998), prior to migrating to the ocean. Coho salmon typically return to freshwater to spawn at three or four years of age and are semelparous, dying after they spawn (Scott and Crossman 1998). In contrast, Atlantic salmon typically return to freshwater to spawn after one or more years at sea and are iteroparous, often not dying after spawning and having the opportunity to spawn more than once (Fleming 1998; Scott and Crossman 1998). As a consequence, juvenile coho salmon will potentially encounter Atlantic salmon of various age classes, some of which will have a size and residence advantage. Prior residence of Atlantic salmon improved their competitive ability relative to coho salmon (see also Chapter 3). Habitat overlap may occur, particularly as older Atlantic salmon enter deeper water as they grow and when they seek sheltered habitat during winter (Fausch 1988). The effects of competition between older Atlantic salmon and juvenile coho salmon has not been studied and additional investigation is required to address potential effects relevant to population response. Limited research suggests that predation by yearling Atlantic salmon on emerging or juvenile coho salmon could potentially occur (Heland and Beall 1997).  78  CHAPTER V - GENERAL DISCUSSION AND CONCLUSIONS This thesis has explored aspects of competition for food and space between cultured and wild salmonids during their juvenile freshwater phase with a view to assessing the risk to wild populations from cultured salmonids. Particular emphasis was given to competitive interactions between farmed Atlantic salmon and wild coho salmon or coastal cutthroat trout to assess the potential effects of feral Atlantic salmon on native Pacific coast species.  As described by Hearn (1987), during their freshwater, stream resident stage, juvenile salmonids compete for space rather than directly for food, cover or other resources (Chapman 1966). Individuals compete for favourable stream positions based on their value as feeding sites (Fausch and White 1981; Bachman 1984; Fausch 1984). Individuals that are not successful in competing for territories may be displaced downstream (Chapman 1962) and displaced individuals are thought to have poor survival (Dill et al. 1980). Competition, therefore, is important to individual fitness and can influence local characteristics of salmonid populations through natural selection over time.  Research conducted for this thesis included both intra- and interspecific tests of relative competitive ability. Intraspecific results indicate that competitive ability differed between the two wild coho salmon populations tested. Farmed coho did not differ in competitive ability from either wild coho salmon population. In addition, pooled results of two experiments comparing hatchery coho salmon and the least competitive wild coho salmon population suggested that the populations were competitively equal. Cultured salmonids have been observed to differ from wild populations morphologically and behaviourally (e.g. aggression, predator avoidance) (Taylor 1986; Einum and Fleming 2001). In the present study, cultured coho salmon displayed more agonistic behaviour (although not always significantly) but this did not translate into dominance in competition for food. Additional work would have to be conducted to determine if both levels of aggression and equal relative competitive ability of the populations in this study were primarily due to culture or population differences.  79  Interspecific contests indicated that in equal contests, farmed Atlantic salmon (Mowi strain) were subordinate to wild coho salmon and coastal cutthroat trout in all experiments. This result was consistent in both aquaria and larger scale channel experiments. Pacific salmon species were also more aggressive in aquaria experiments. In channel experiments, which allowed increased partitioning of competitors spatially, the level of interspecific aggression was the same for Atlantic and coho salmon. Prior residence and greater relative size are often associated with dominance in salmonids (Berejikian et al. 1996; Cutts et al. 1999; Johnsson et al. 1999). In this study, when Atlantic salmon were given residency advantage they were competitively equal to both wild coho populations but they remained subordinate to coastal cutthroat trout. When Atlantic salmon were given a size advantage they were competitively equal to Street Creek coho salmon but remained subordinate to Salmon River coho. Such factors may affect competitive outcomes in natural systems and highlights the importance of the conditions under which the competitors encounter one another.  Hatchery salmonids:  Currently, on the North Pacific coast, hatchery supplementation programs  are widespread and intense, producing all species of native Oncorhynchus for release, with annual introductions on the order of half a billion salmonids per year. Hatchery fish, therefore, may interact with wild salmon frequently, depending on their point of release, survival and rate of homing or straying to rivers. In the current study, size matched hatchery fish were, overall, competitively equal to a wild population. As hatchery juveniles are often larger for a given age, however, they could dominate wild salmon if their larger size outweighs any residence advantage that wild fish may have (Berejikian et al. 1996; Rhodes and Quinn 1998; McMichael et al. 1999), highlighting the importance of the conditions under which hatchery introductions are made. In terms of determining whether hatchery salmonids pose a threat to wild conspecifics at a population level, competition as well as genetic and other ecological factors including disease, predation and mix stock fishing, come into play. A review of previous studies suggests that hatchery fish typically have lower survival rates (Leider et al. 1990; Einum and Fleming 2001; 80  but see Berejikian et al. 1999; Rhodes and Quinn 1999) and lower reproductive success (Chilcote et al. 1986; Leidar et al. 1990; M c L e a n 2003). There is high variability i n level o f introgression o f hatchery alleles into w i l d populations (Hindar et al. 1991; Fleming and Petersson 2001; Utter 2001) and l o w levels may be observed despite large scale introductions (Isaksson 1988; Hilborn and Winton 1993). Long-term hatchery program evaluations are typically measured i n terms o f juvenile survival to time o f release, contribution to the commercial catch and adult returns o f fish o f hatchery origin (Winton and Hilborn 1994; Finstad and Jonsson 2001; Fjellheim and Johnsen 2001; Waples et al. 2001). This demographic boost is necessary but not sufficient for improved natural production (Waples et al. 2001). Evaluations o f supplementation programs rarely assess numbers o f natural spawners or natural productivity. When such evaluations o f natural productivity are conducted, results often indicate that desired increases are not being obtained (Winton and Hilborn 1994; Nickelson et al. 1986; U n w i n and G l o v a 1997; Fleming and Petersson 2001; Vollestad and Hesthagen 2001; Waples et al. 2001).  Furthermore, published reviews o f the literature on supplementation and much needed assessments o f natural productivity often do not differentiate between multigeneration or single generation hatchery stocks, the quality o f hatchery salmonids introduced or management practices which can affect the survival and homing o f hatchery salmonids (e.g. studies cited i n Jonsson 1997; Finstand and Jonsson 2001; Vollestad and Hesthagen 2001). Attention to these factors w i l l likely improve the effectiveness o f supplementation programs (Campton 1995; E i n u m and Fleming 2001) and must be considered i f we are to eliminate confounding factors that could be responsible for failed efforts to increase natural productivity. Despite a large body o f research on this topic, Waples et al. (2001) concluded that i f only single generation hatchery broodstock from a local native population are assessed, the premise that supplementation can be used to provide a net long-term benefit to natural populations remains an untested hypothesis. Despite this limitation i n our knowledge, the poor reproductive success and limited introgression o f hatchery alleles are indications that programs are likely not working i n terms o f increasing natural production although harvests can increase. The goal o f conventional hatcheries, i.e. to increase fish availability for harvest should be differentiated from the goal o f supplementation hatcheries, i.e. to increase production over the long-term, as they pose different risks and benefits (Fleming and Petersson 2001; Waples et al. 2001). The effectiveness o f hatchery 81  programs, their risk (Hindar et al. 1991; Busack and Currens 1995) and their cost (Hilborn and Winton 1993; Winton and Hilborn 1994; Naylor et al. 1998) must be weighed against alternative methods of achieving improved natural production such as habitat improvements and increased escapement (Hilborn and Winton 1993; V0llestad and Hesthagen 2001).  Farmed Coho Salmon:  In the current study, size matched farmed coho salmon were often more  aggressive but, overall, competitively equal to two wild coho salmon populations. This may be due, in part, to overt aggressiveness of dominant individuals. Farm Kitimat coho salmon were observed to be particularly aggressive in two comparisons, with dominant individuals initiating approximately twice as many agonistic interactions than dominant individuals in the wild populations. Greater overall aggression of the population therefore, is not necessarily associated with dominance of the population. Higher aggression is consistent with the literature on farm Atlantic salmon (Einum and Fleming 1997; Fleming and Einum 1997). Results of previous studies on relative competitive ability of farm relative to wild Atlantic salmon suggest population effects may also be important. Previous studies have indicated that farm fish were competitively superior in natural or semi-natural environments only relative to wild populations other than those from which they were derived (Einum and Fleming 1997; McGinnity et al. 1997). In the present study potential population effects cannot be ruled out. There is also concern that the progeny of farm fish or hybrids with wild fish may outgrow wild conspecifics as they have been selected for rapid growth in the farm environment. This rapid growth may confer a competitive advantage (Einum and Fleming 1997; McGinnity et al. 1997; but see Fleming and Einum 1997). Additional research is required to determine if farm coho salmon outgrow wild coho salmon in natural environments. As with hatchery salmonids, the importance of competition between farm and wild salmon at the population level will depend upon the number of feral spawning adults relative to wild adults, their relative competitive ability and reproductive success as well as the relative competitive ability and fitness of their offspring. In this regard, reference to interactions of farmed and wild Atlantic salmon in their native range may provide insight. ResultsfromEuropean studies indicate that sea-run farm salmon have lower reproductive success (Fleming et al. 1996; Fleming 82  et al. 2000), sometimes even in the absence of competition with wild conspecifics (Fleming et al. 1996; but see Lura 1995, cited in Saegrov et al. 1997). Fleming et al. (2000) found that the genetic contribution or farmed salmon was primarily through hybridization of farm females with wild males due to indiscriminant choice of mates on the part of male salmon. Recent findings however, (Garant et al. 2003) show that precocious male parr achieved greater reproductive success than wild parr and may hasten introgession of farm alleles into wild populations of Atlantic salmon. These results indicate there are two potential pathways for interbreeding between farm salmon and wild conspecifics which could have adverse effects on the fitness of wild salmon.  Currently in British Columbia, farming of native species (coho and chinook salmon) comprises approximately 20% of production in open net-cages (BCSFA 2003). Although estimated rates of escape from well run aquaculture operations are small as a percentage of production (1-2%) (Alverson and Ruggerone 1997) the number of feral farmfishthat can interact with wild fish increases with aquaculture production. In British Columbia, current numbers of escaped farm fish are on the order of a few hundred thousand salmon per year, the majority being Atlantic salmon. In some locations in eastern Canada and Europe farming of native species (i.e. Atlantic salmon) is particularly intense. Farm salmon that have escaped into the wild, spawn successfully (Lura and Saegrov 1991; Webb et al. 1993) and in some rivers farm spawners outnumber wild spawners (Gausen and Moen 1991; see studies cited in Heggberget et al. 1993; Saegrov et al. 1997). It has been confirmed that farmfishcontribute progeny to juvenile salmon populations  (Crozier 1993), often through hybridization with wild salmon (Crozier 1993; Fleming et al. 1996). The level of introgression of alleles of farmedfishis variable (Heggberget et al. 1993), however, in some systems a substantial percentage of the genetic makeup of the population is derived from farmfish(e.g. Saegrov et al. 1997).  Whether a similar outcome will result on the Pacific coast will depend on the intensity of the farming of native species, farm management and other factors affecting the health of local populations. The results of this study indicate that dominant individuals in cultured fish populations may be overtly aggressive and although the population may not be dominant as determined by statistical analyses, a portion of the cultured population may dominate some wild 83  individuals. If cultured fish are released in large numbers, these dominant individuals have the potential to adversely affect a wild population, particularly if the size of the wild population is small relative to the introduced population. In this regard, attention has to be given to the number of feral native farm fish that can potentially be liberated as a percentage of total production. In addition, the potential growth advantage of farm salmon, the theoretical concerns relating to farming such as disruption of local adaptation (Hindar et al. 1991) and the potential pathways for interbreeding of farm and wild salmon is sufficient to warrant serious concern regarding farming of native species. Farmed Atiantic Salmon:  Atlantic salmon, an introduced species to the Pacific coast of North  America, is currently being farmed in open net-cages in Washington state and in British Columbia. Feral Atlantic salmon have been captured from Washington state to Alaska and have been observed in approximately 80 rivers in British Columbia (DFO 2003). In addition, progeny from the spawning of feral Atlantic salmon have been observed in three British Columbia rivers (Volpe et al. 2000; DFO 2003). Current estimates of the likelihood that Atlantic salmon will establish self-sustaining populations on the Pacific coast of North America and the risk this poses to native salmonids are based on limited information and additional research is needed (Alverson and Ruggerone 1997). My thesis has contributed to increasing this information base by providing insight into the interactions between farm Atlantic salmon and wild coho salmon or coastal cutthroat trout.  Micro-habitat use by Atlantic salmon will depend upon their innate habitat preference and their interactions with the local species assemblage. The potential exists for food and space competition between Atlantic salmon and cutthroat trout or coho salmon juveniles as all feed to some extent on invertebrate drift and they may co-occur in some habitats. Based on studies of size-matched juveniles, coastal cutthroat trout, steelhead trout (Volpe et al. 2001) and coho salmon dominate Atlantic salmon. Timing of emergence and consequent residence and potential size advantage of early emerging species can affect these dominance relationships however. Offspring produced by feral Atlantic salmon, when in the presence of underyearling coho salmon with a size and residence advantage, may utilize riffle habitat or run habitat, thereby competing with rainbow and coastal cutthroat trout juveniles. As underyearlings, these latter species may 84  be smaller than underyearling Atlantic salmon. The freshwater residency patterns o f these four species differ so that as they grow older and larger their respective advantages are likely to change. T o date, however, only underyearling Atlantic salmon have been studied so the effects o f yearling Atlantic salmon need to be assessed. The outcome o f competitive interactions between introduced Atlantic salmon and the Pacific salmonid species w i l l however depend on the effects o f interaction o f all age classes o f the various species i n a given system.  The risk o f Atlantic salmon establishing self-sustaining populations remains uncertain as does the effects o f any such population on native species. Based on the results o f this study and literature reviewed, the greater competitive ability of Pacific salmon species may be, i n part, responsible for the failure o f attempts to introduce Atlantic salmon to the Pacific coast i n the past (Alverston and Ruggerone 1997). Other factors responsible for l o w levels o f introgression o f many cultured fish introductions may be responsible as well (Hindar et al. 1991; Fleming and Petersson 2001).  Despite the failure o f past attempts o f introduction o f Atlantic salmon on the Pacific coast, attention must be given to changing conditions which may favour the establishment o f Atlantic salmon as well as the scale o f aquaculture operations. Current conditions, such as reduced Pacific salmon abundance relative to the early 1900's when most introduction attempts were made, may create space that Atlantic salmon can exploit. A l s o , salmon farming provides for continuous introduction o f many more Atlantic salmon than i n the past at various stages o f development, at various times o f year and in multiple locations, particularly when operations are conducted on a large scale. This greatly increases the likelihood o f that chance combination o f events occurring that allows a species to establish, i n step with production expansion. I f farming o f Atlantic salmon is continued and our goal is to prevent risk o f competitive displacement o f native salmonids by Atlantic salmon, then sufficient resources w i l l be required to reduce the number o f escapees and destroy feral Atlantic salmon i n the waters o f the Pacific coast of North America.  If we are to maintain healthy salmonid populations i n the long term, great care w i l l have to be given to how culture operations are conducted and i n some instances operations may prove 85  ecologically unsustainable. In addition, the health of salmonid populations will depend on timely action being taken if adverse effects are observed. In many instances, clear demonstration of adverse effects may only come after irreversible changes have occurred or the explanation for adverse effects may be confounded by multiple potential sources of change. The level of precaution taken in British Columbia, has been reduced with the recent lifting of moratorium on salmon farming. This is unfortunate as our understanding of sources of risk will improve with time as well as technologies to mitigate undesirable effects and the type of technologies that are economically viable.  86  REFERENCES  Alverson, D.L., and G.T. Ruggerone. 1997. Escaped farm salmon: environmental and ecological concerns. Discussion paper part B. In Salmon aquaculture review. Technical advisory team discussion papers vol. 3. Environmental Assessment Office, Victoria, B.C. Bachman, R.A. 1984. Foraging behavior of free-ranging wild and hatchery brown trout in a stream. Transactions of the American Fisheries Society. 113:1-32. BC Fisheries. 2002. Fisheries Information Summary System (FISS) database, stocking query. Available:www.bcfisheries.gov.bc.ca/fishinv/db/default.asp (November 2002). Beacham, T.D. and C.B. Murray. 1990. Temperature, egg size, and development of embryo and alevins of five species of Pacific salmon: a comparative analysis. Transactions of the American Fisheries Society. 119:927-945. Beall, E., M. Heland, and C. Marty. 1989. Interspecific relationships between emerging Atlantic salmon, Salmo salar, and coho salmon, Oncorhynchus kisutch, juveniles. Journal of Fish Biology. 35(Suppl. A):285-293. Beckman, B.R., W.W. Dickhoff, W.S. Zaugg, C. Sharpe, S. Hirtzel, R. Schrock, D.A. Larsen, R.D. Ewing, A. Palmisano, C.B. Schreck, and C.V.W. Manken. 1999. Growth, smoltification and smolt-to adult return of spring chinook salmon from hatcheries on the Deschutes River, Oregon. Transactions of the American Fisheries Society. 128:11251150. Berejikian, B.A., E.P. Tezak, S.L. Schroder, C M . Knudsen, and J.J. Hard. 1997. Reproductive behavioral interactions between wild and captively reared coho salmon (Oncorhynchus kisutch). ICES Journal of Marine Science. 54:1040-1050. Berejikian, B.A., R.J.F. Smith, E.P. Tezak, S.L. Schroder, and C M . Knudsen. 1999. Chemical alarm signals and complex hatchery rearing habitats affect antipredator behavior and survival of chimook salmon (Oncorhynchus tshawytscha) juveniles. Canadian Journal of Fisheries and Aquatic Sciences. 56:830-838. Berejikian, B.A., S.B. Mathews, and T.P. Quinn. 1996. Effects of hatchery and wild ancestry and rearing environments on the development of agonistic behavior in steelhead trout (Oncorhynchus mykiss) fry. Canadian Journal of Fisheries and Aquatic Sciences. 53:2004-2014. Beveridge, M. 1984. Cage and pen farming: carrying capacity models and environmental impact. FAO Fisheries Technical Paper No. 255. UN/FAO, Rome, Italy. Birch, L . C 1957. The meanings of competition. American Naturalist. 91:5-18. Blom, G. 1958. Statistical estimates and transformed beta variables. John Wiley & Sons. New York. 87  Bonnell, G. 1999. Supplementation workshop presentation by Fisheries and Oceans Canada, Pacific Region, for the American Fisheries Society Western Division Meeting, Moscow, Idaho. BCSFA (British Columbia Salmon Farmers Association). 2003. Available: www.salmonfarmers.org/library/farm.html (January 2003). Busack, C.A., and K.P. Currens. 1995. Genetic risks and hazards in hatchery operations: fundamental concepts and issues. Pages 71-80 in H.L. Schramm, Jr. and R.G. Piper, editors. Uses and effects of cultured fishes in aquatic ecosystems. American Fisheries Society, Symposium 15, Bethesda, Maryland. Campton, D.E. 1995. Genetic effects of hatchery fish on wild populations of Pacific salmon and steelhead: what do we really know. Pages 337-353 in H.L. Schramm, Jr. and R.G. Piper, editors. Uses and effects of cultured fishes in aquatic ecosystems. American Fisheries Society, Symposium 15, Bethesda, Maryland. Carr, J.W., G.L. Lacroix, J.M. Anderson, and T. Dilworth. 1997. Movements of non-maturing cultured Atlantic salmon (Salmo salar) in a Canadian river. ICES Journal of Marine Science. 54:1082-1085. Chapman, D.W. 1962. Aggressive behavior in juvenile coho salmon as a cause of emigration. Journal of the Fisheries Research Board of Canada. 19:1047-1080. Chapman, D.W. 1966. Food and space as regulators of salmonid populations in streams. American Naturalist. 100:345-357. Chilcote, M.W., S.A. Leider, and J.J. Loch. 1986. Differential reproductive success of hatchery and wild summer-run steelhead under natural conditions. Transactions of the American Fisheries Society. 115:726-735. Christensen, R. 1996. Analysis of variance, design and regression; applied statistical methods. Chapman & Hall. New York. Cooke, R. 2003. Fisheries and Oceans Canada. Assessment Biologist, Habitat and Enhancement Branch, Vancouver. Personal communication with author. Crawford, S.S. 2001. Salmonid introductions to the Laurentian Great Lakes: an historical review and evaluation of ecological effects. Canadian Special Publication of Fisheries and Aquatic Sciences. 132. NRC Research Press, Ottawa, Ontario. Crawley, M.J. 1986. The population biology of invaders. Philosophical Transactions of the Royal Society of London. B 314:711-731. Crossman, E.J., and P.A. Larkin. 1959. Yearling liberations and change of food as effecting rainbow trout yield in Paul Lake, British Columbia. Transactions of the American Fisheries Society. 88:36-44.  88  Crozier, W . W . 1993. Evidence o f genetic interaction between escaped farmed salmon and w i l d Atlantic salmon (Salmo salar, L . ) i n a northern Irish river. Aquaculture. 113:19-29. Cunjak, R . A . , and J . M . Green. 1986. Influence o f water temperature on behavioural interactions between juvenile brook charr, Salvelinus fontinalis, and rainbow trout, Salmo gairdneri. Canadian Journal o f Zoology. 64:1288-1291. Cutts, C.J., B . Brembs, N . B . Metcalfe, and A . C . Taylor. 1999. Prior residence, territory quality and life-history strategies i n juvenile Atlantic salmon (Salmo salar L.). Journal o f Fish Biology. 55:784-794. D F O (Department o f Fisheries and Oceans). 2003. Atlantic Salmon Watch Program, freshwater Atlantic salmon sighting and capture data (preliminary). Available: www.pac.dfompo.gc.ca/sci/aqua/ASWP_e.htm. (April 2003). Dickson, T . A . , and H . R . M a c C r i m m o n . Influence o f hatchery experience on growth and behavior o f juvenile Atlantic salmon (Salmo salar) within allopatric and sympatric stream populations. Canadian Journal o f Fisheries and Aquatic Science. 39:1453-1458. D i l l , L . M . , R . C . Ydenberg, and A . G . H . Fraser. 1981. Food abundance and territory size i n juvenile coho salmon (Oncorhynchus kisutch). Canadian Journal o f Zoology. 59:18011809. Drake, J.A., H . A . Mooney, F. di Castir, R . H . Groves, F..J. Kruger, M . Rejmanek, and M . Williamson, editors. 1989. Biological invasions: a global perspective. John W i l e y & Sons, Toronto, Ontario. E i n u m , S., and L A . Fleming. 1997. Genetic divergence and interactions i n the w i l d among native, farmed and hybrid Atlantic salmon. Journal o f Fish Biology. 50:634-651. Einum, S., and L A . Fleming. 2001. Implications o f stocking: ecological interactions between w i l d and released salmonids. Nordic Journal o f Freshwater Research. 75:56-70. Fausch, K . D . 1984. Profitable stream positions for salmonids: relating specific growth rate to net energy gain. Canadian Journal o f Zoology. 62:441-451. Fausch, K . D . 1988. Tests o f competition between native and introduced salmonids i n streams: what have we learned? Canadian Journal o f Fisheries and Aquatic Sciences. 45:22382246. Fausch, K . D . 1998. Interspecific competition and juvenile Atlantic salmon (Salmo salar): on testing effects and evaluating the evidence across scales. Canadian Journal o f Fisheries and Aquatic Sciences. 55(Suppl. 1):218-231. Fausch, K . D . , and R . J . White. 1981. Competition between brook trout and brown trout for positions i n a Michagan stream. Canadian Journal o f Fisheries and Aquatic Sciences. 38:1220-1227.  89  Fenderson, O.C., W.H. Everhart, and K.M. Muth. 1968. Comparative agonistic and feeding behavior of hatchery-reared and wild salmon in aquaria. Journal of the Fisheries Research Board of Canada. 25(1): 1-14. Finstad, B., and N. Jonsson. 2001. Factors influencing the yield of smolt releases in Norway. Nordic Journal of Freshwater Research. 75:37-55. Fjellheim, A. and B.O. Johnsen. 2001. Experiences from stocking salmonidfryand fingerlings in Norway. Nordic Journal of Freshwater Research. 75:20-36. Fleming, LA. 1994. Captive breeding and the conservation of wild salmon populations. Conservation Biology. 8:886-888. Fleming, LA. 1998. Pattern and variability in the breeding system of Atlantic salmon (Salmo salar), with comparisons to other salmonids. Canadian Journal of Fisheries and Aquatic Sciences. 55(Suppl. l):59-76. Fleming, LA. and M.R. Gross. 1992. Reproductive behavior of hatchery and wild coho salmon (Oncorhynchus kisutch): does it differ? Aquaculture. 103:101-121. Fleming, LA. and M.R. Gross. 1993. Breeding success of hatchery and wild coho salmon (Oncorhynchus kisutch) in competition. Ecological Applications. 3(2):230-245. Fleming, LA., and E. Petersson. 2001. The ability of released, hatchery salmonids to breed and contribute to the natural productivity of wild populations. Nordic Journal of Freshwater Research. 75:71-98. Fleming, I. A., and S. Einum. 1997. Experimental tests of genetic divergence of farmedfromwild Atlantic salmon due to domestication. ICES Journal of Marine Science. 54:1051-1063. Fleming, I.A., B. Jonsson, and M.R. Gross. 1994. Phenotypic divergence of sea-ranched, farmed and wild salmon. Canadian Journal of Fisheries and Aquatic Sciences. 51:2808-2824. Fleming, I. A., B. Jonsson, M.R. Gross, and A. Lamberg. 1996. An experimental study of the reproductive behaviour and success of farmed and wild Atlantic salmon (Salmo salar). Journal of Applied Ecology. 33:893-905. Fleming, I. A., K. Hindar, LB. Mjolnerad, B. Jonsson, T. Balstad, and A. Lamberg. 2000. Lifetime success and interactions of farm salmon invading a native population. Proceedings of the Royal Society of London B. 267:1517-1523.  90  Gardner, J. andD.L. Peterson. 2003. Making sense of the salmon aquaculture debate: Analysis of issues related to netcage salmon farming and wild salmon in British Columbia. Prepared for the Pacific Fisheries Resource Conservation Council. Available: www.fish.bc.ca/reports/pfrcc making sense report.pdf (March 2003). Garant, D., LA. Fleming, S. Einum and L. Bernatchez. 2003. Alternative male life-history tactics as potential vehicles for speeding introgression of farm salmon traits into wild populations. Ecology Letters. 6:541-549. Gause, G.F. 1934. The struggle for existence. Williams and Wilkins Co., Baltimore. Gausen, D., and V. Moen. 1991. Large-scale escapes of farmed Atlantic salmon (Salmo salar) into Norwegian rivers threaten natural populations. Canadian Journal of Fisheries and Aquatic Sciences. 48:426-428. Gibson, R.J. 1966. Some factors influencing the distributions of brook trout and young Atlantic salmon. Journal of the Fisheries Research Board of Canada. 23:1977-1980. Gibson, R.J. 1973. Interactions of juvenile Atlantic salmon and brook trout. International Atlantic Salmon Foundation Special Publication. 4:181-202. Gibson, R.J. 1978. The behavior of juvenile Atlantic salmon (Salmo salar) and brook trout (Salvelinus fontinalis) with regard to temperature and water velocity. Transactions of the American Fisheries Society. 107:703-712. Gibson, R.J. 1981. Behavioural interactions between coho salmon (Oncorhynchus kisutch), Atlantic salmon (Salmo salar), brook trout (Salvelinus fontinalis) and steelhead trout (Salmo gairdneri) at the juvenilefluviatilestages. Canadian Technical Report of Fisheries and Aquatic Sciences. 1029. Gibson, R.J. 1993. The Atlantic salmon in freshwater: spawning, rearing and production. Reviews in Fish Biology and Fisheries. 3:39-73. Gjedrem, T., H.M. Gjoen, and B. Gjerde. 1991. Genetic origin of Norwegian farmed Atlantic salmon. Aquaculture. 98:41-50. Glova, G.J. 1984. Management implications of the distribution and diet of sympatric populations of juvenile coho salmon and coastal cutthroat trout in small streams in British Columbia, Canada. Progressive Fish Culturalist. 46:269-277. Grant, J.W. A. 1990. Aggressiveness and the foraging behaviour of young-of-the-year brook charr (Salvelinus fontinalis). Canadian Journal of Fisheries and Aquatic Sciences. 47:915-920. Gross, M.R. 1998. One species with two biologies: Atlantic salmon (Salmo salar) in the wild and in aquaculture. Canadian Journal of Fisheries and Aquatic Sciences. 55(Suppl. 1):131-144.  91  Gustafson, R.G., T.C. Wainwright, G.A. Winans, F.W. Waknitz, L.T. Parker, andR.S. Waples. 1997. Status review of sockeye salmon from Washington and Oregon. U.S. Department of Commerce. NOAA Tech. Memo. NMFS-NWFSC-33. Available: www.nwfsc.noaa.gov/pubs/tm/tm33/tm33.html. (January 2003). Hardin, G. 1960. The competitive exclusion principle. Science. 131:1292-1297. Hartman, G.F. 1965. The role of behavior in the ecology and interaction of underyearling coho salmon (Oncorhynchus kisutch) and steelhead trout (Salmo gairdneri). Journal of the Fisheries Reseach Board of Canada. 22:1035-1081. Healey, M.C. 1982. Timing and relative intensity of size selective mortality of juvenile chum salmon (Oncorhynchus keta) during early sea life. Canadian Journal of Fisheries and Aquatic Sciences. 38:952-957. Hearn, W.E. 1978. Behavioral interactions between juvenile coho salmon (Oncorhynchus kisutch) and juvenile Atlantic salmon (Salmo salar) in a laboratory still water environment. Master's thesis. Southwestern Massachusetts University, Amherst. Hearn, W.E. 1987. Interspecific competition and habitat segregation among stream-dwelling trout and salmon: a review. Fisheries. 12(5):24-31. Hearn, W.E. and B.E. Kynard. 1986. Habitat utilization and behavioral interaction of juvenile Atlantic salmon (Salmo salar) and Rainbow trout (S. gairdneri) in tributaries of the White River of Vermont. Canadian Journal of Fisheries and Aquatic Sciences. 43:19881998. Heggberget, T.G., B.O. Johnsen, K. Hindar, B. Jonsson, L.P. Hansen, N.A. Hvidsten, and A.J. Jensen. 1993. Interactions between wild and cultured Atlantic salmon: a review of the Norwegian experience. Fisheries Research. 18:123-146. Heggberget, T.G., F. Okland, and O. Ugedal. 1996. Prespawning migratory behaviour of wild and farmed Atlantic salmon, Salmo salar L., in a north Norwegian river. Aquaculture Research. 27:313-322. Heggenes, J., J.L. Bagliniere, and R.A. Cunjak. 1999. Spatial niche variability for young Atlantic salmon (Salmo salar) and brown trout (S. trutta) in heterogeneous stream. Ecology of Freshwater Fish. 8:1-21. Heland, M., and E. Beall. 1997. Experimental study of interspecific competition between coho salmon, Oncorhynchus kisutch, and Atlantic salmon, Salmo salar, juveniles in freshwater [French]. Bulletin Francais de la Peche et de la Pisciculture. 0(344-345):241252. Henry, J. 2002. Target Marine Products. Operations Manager, Madeira Park, British Columbia. Personal communication with author.  92  Herbold, B., and P.B. Moyle. 1986. Introduced species and vacant niches. American Naturalist. 128:751-760. Hershberger, W.K., Myers, J.M., Iwamoto, R.N., Mcauley, W.C., and A.M. Saxton. 1990. Genetic changes in the growth of coho salmon (Oncorhynchus kisutch) in marine netpens, produced in ten years of selection. Aquaculture. 85:187-197. Hilborn, R., and J. Winton. 1993. Learning to enhance salmon production: lessons from the Salmonid Enhancement Program. Canadian Journal of Fisheries and Aquatic Sciences. 50:2043-2056. Hinch, S. 2003. University of British Columbia, Vancouver, B.C. Associate Professor, Department of Forest Sciences / Institute for Resources, Environment and Sustainability. Personal communication with author. Hindar, K., N. Ryman, and F. Utter. 1991. Genetic effects of cultured fish on natural fish populations. Canadian Journal of Fisheries and Aquatic Sciences. 48:945-957. Holtby, L.B., D.P. Swain, and G.M. Allan. 1993. Mirror-elicited agonistic behaviour and body morphology as predictors of dominance status in juvenile coho salmon (Oncorhynchus kisutch). Canadian Journal of Fisheries and Aquatic Sciences. 50:676-684. Huntingford, F.A., N.B. Metcalfe, J.E. Thorpe, W.D. Graham, and C.E. Adams. 1990. Social dominance and body size in Atlantic salmon pan, Salmo salar L. Journal of Fish Biology. 36:877-881. Isaksson, A. 1988. Salmon ranching. A world review. Aquaculture. 75:1-33. Isaksson, A., S. Oskarsson, S.M. Einarsson, and J. Jonasson. 1997. Atlantic salmon ranching: past problems and future management. ICES Journal of Marine Science. 54:1188-1199. Jobling, M. 1995. Environmental biology of fishes. Chapman and Hall, New York. Johnsson, J.I., F. Nobbelin, and T. Bohlin. 1999. Territorial competition among wild brown trout fry: effects of ownership and body size. Journal of Fish Biology. 54:469-472. Jones, MX., and L.W. Stanfield. 1993. Effects of exotic juvenile salmonids on growth and survival of juvenile Atlantic salmon (Salmo salar) in a Lake Ontario tributary. Pages 7179 in R.J. Gibson and R.E. Cutting, editors. Production of juvenile Atlantic salmon, Salmo salar, in natural waters. Canadian Special Publication of Fisheries and Aquatic Sciences. 118. National Research Council of Canada, Ottawa, Ontario. Jonsson, B. 1997. A review of ecological and behavioural interactions between cultured and wild Atlantic salmon. ICES Journal of Marine Science. 54:1031-1039. Jonsson, B., N. Jonsson, and L.P. Hansen. 1990. Does juvenile experience affect migration and spawning of adult Atlantic salmon? Behavioural Ecology and Sociobiology. 26:225-230.  93  Kalleberg, H. 1958. Observations in a stream tank of territoriality and competition in juvenile salmon and trout. Report of the Institute of Freshwater Research Board of Canada. 19:55-98. Kareiva, P. 1996. Developing a predictive ecology for non-indigenous species and ecological invasions. Ecology. 77:1651-1652. Keene, J.L., D.L.G. Noakes, R.D. Moccia and C.G.Soto. 1998. The efficacy of clove oil as an anaesthetic for rainbow trout, Oncorhynchus mykiss (Walbaum). Aquaculture Research. 29:89-101. Kennedy, G.J.A. and C D . Strange. 1986. The effects on intra- and inter-specific competition on the distribution of stocked juvenile Atlantic salmon, Salmo salar L., in relation to depth and gradient in an upland trout, Salmo trutta L., stream. Journal of Fish Biology. 29:199214. Kratt, L.F. and Smith, R.F.J. 1979. Agonistic behavior of age 0, age 1 and non-breeding adult Arctic grayling, Thymallus arcticus (Pallas). Journal of Fish Biology. 15:389-404. Krebs, C.J. 2001. Ecology: the experimental analysis of distribution and abundance. 5 Edition. Benjamin Cummings, San Francisco. th  Kwain, W. 1983. Downstream migration, population size, and feeding of juvenile rainbow trout. Journal of Great Lakes Research. 9:52-59. Larson, G.L., and S.E. Moore. 1985. Encroachment of exotic rainbow trout into stream populations of native brook trout in the southern Appalachian Mountains. Transactions of the American Fisheries Society. 114:195-203. Lassuy, D.R. 1995. Introduced species as a factor in extinction and endangerment of native fish species. Pages 391-396 in H.L. Schramm, Jr. and R.G. Piper, editors. Uses and effects of cultured fishes in aquatic ecosystems. American Fisheries Society, Symposium 15, Bethesda, Maryland. Laufle, J.C, G.B. Pauley, and M.F. Shepard. 1986. Species profiles: life histories and environmental requirements of coastal fishes and invertebrates (Pacific Northwest) coho salmon. U.S. Fish and Wildlife Service Biological Report 82 (11.48). U.S. Army Corps of Engineers. TREL-82-4. Law, R., and A.R. Watkinson. 1989. Competition. Pages 243-284 in J.M. Cherrett, A.D. Bradshaw, F.B. Goldsmith, P.J. Grubb, and J.R. Krebs, editors. Ecological concepts. 29th Symposium of the British Ecological Society. Blackwell Scientific Publications, Oxford. Leider, S.A., P.L. Hulett, J.J. Loch, and M.W. Chilcote. 1990. Electrophoretic comparison of the reproductive success of naturally spawning transplanted and wild steelhead trout through the returning adult stage. Aquaculture. 88:239-252.  94  Lichatowich, J., L. Mobrand, and A. Lestelle. 1999. Depletion and extinction of Pacific salmon (Oncorhynchus spp.): a different perspective. ICES Journal of Marine Science. 56:467472. Lotka, A.J. 1925. Elements of Physical Biology. Hopkins & Williams, Baltimore. Lura, H. 1995. Domesticated female Atlantic salmon in the wild: spawning success and contribution to local populations. Dr. scient thesis. Institute of Zoology, University of Bergen, Norway. Lura, H., and H. Saegrov. 1991. Documentation of successful spawning of escaped farmed female Atlantic salmon, Salmo salar, in Norwegian rivers. Aquaculture. 98:151-159. MacArthur, R.H. 1972. Geographical ecology, patterns in the distribution of species. Harper and Row, New York. MacCrimmon, H , and B. Gots. 1979. World distribution of Atlantic salmon, Salmo salar. Journal of the Fisheries Research Board of Canada. 36:423-457. MacCrimmon, H.R., T.A. Dickson, and R.J. Gibson. 1983. Implications of differences in emergent times on growth and behaviour of juvenile Atlantic salmon (Salmo salar) and brook charr (Salvelinus fontinalis) in sympatric stream populations. Naturaliste Canadiene. (Reviews in Ecology and Systematics) 110:379-384. McGinnity, P., C. Stone, J.B. Taggart, D. Cooke, D. Cotter, R. Hynes, C. McCamley, T. Cross, and A. Ferguson. 1997. Genetic impact of escaped farmed Atlantic salmon (Salmo salar L.) on native populations: use of DNA profiling to assess freshwater performance of wild, farmed, and hybrid progeny in a naturalriverenvironment. ICES Journal of Marine Science. 54:998-1008. McMichael, G.A., T. N. Pearsons, and S. A. Leider. 1999. Behavioral interactions among hatchery-reared steelhead smolts and wild Oncorhynchus mykiss in natural streams. North American Journal of Fisheries Management. 19:948-956. Meredith, M.P. and S.V. Stehman. 1991. Repeated measures experiments in forestry : focus on analysis of response curves. Canadian Jouranl of Forestry Research. 21 :957-965. Mesa, M.G. 1991. Variation in feeding, aggression and position choice between hatchery and wild cutthroat trout in an artificial stream. Transactions of the American Fisheries Society. 120:723-727. Mills, EX., J.H. Leach, J.T. Carlton, and C L . Secor. 1994. Exotic species and the integrity of the Great Lakes: lessons from the past. Bioscience. 44(10):666-676. Moyle, P.B., and B. Vondracek. 1985. Persistence and structure of the fish assemblage in a small California stream. Ecology. 66:1 -13.  95  Murray, C.B., T.D. Beacham and J.D. McPhail. 1990. Influence of parental stock and incubation temperature on the early development of coho salmon (Oncorhynchus kisutch) in British Columbia. Canadian Journal of Zoology. 68:347-358. Naylor, R.L., Goldburg, R.J., Mooney, H., Beveridge, M., Clay, J., Folke, C , Kautsky, N., Lubchenco, J., Primavera, J. and M. Williams. 1998. Nature's subsidies to shrimp and salmon farming. Science. 282:883-884. Nickelson, T.E., M.F. Solazzi, and S.L. Johnson. 1986. Use of hatchery coho salmon (Oncorhynchus kisutch) presmolts to rebuild wild populations in Oregon coastal streams. Canadian Journal of Fisheries and Aquatic Sciences. 43:2443-2449. Nielsen, J.L. 1992. Microhabitat-specific foraging behavior, diet and growth of juvenile coho salmon. Transactions of the American Fisheries Society. 121:617-634. Nilsson, N.A. 1967. Interactive segregation between fish species. Pages 295-313 in S.D. Gerking, editor. The biological basis of freshwater fish production. Blackwell Scientific Publications, Oxford. 0kland, F., T.G. Heggberget and B. Jonsson, B. 1995. Migratory behaviour of wild and farmed Atlantic salmon (Salmo salar) during spawning. Jounal of Fish Biology. 46:1-7. Petrosky, C E . and T . C Bjornn. 1988. Response of wild rainbow (Salmo gairdneri) and cutthroat trout (S. clarki) to stocked rainbow trout in fertile and infertile streams. Canadian Journal of Fisheries and Aquatic Sciences. 45:2087-2105. PFRCC (Pacific Fisheries Resource Conservation Council) 2003. Advisory: Wild salmon and aquaculture in British Columbia. Available: www.fish.bc.ca/reports/pfrcc wild salmon and aquaculture_2.pdf (March 2003). Pillai, K.C.S. 1955. Some new test criteria in multivariate analysis. Annals of Mathematical Statistics. 26:117-121. Reisenbichler, R.R., and J.D. Mclntyre. 1977. Genetic differences in growth and survival of juvenile hatchery and wild steelhead trout, Salmo gairdneri. Journal of the Fisheries Research Board of Canada. 34:123-128. Rhodes, J.S., and T.P. Quinn. 1998. Factors affecting the outcome of territorial contests between hatchery and naturally reared coho salmon parr in the laboratory. Journal of Fish Biology. 53:1220-1230. Rhodes, J.S., and T.P. Quinn. 1999. Comparative performance of genetically similar hatchery and naturally reared juvenile coho salmon in streams. North American Journal of Fisheries Management. 19:670-677. Rimmer, D.M., U. Paim, and R.L. Saunders. 1983. Autumnal habitat shift of juvenile Atlantic salmon (Salmo salar) in a small river. Canadian Journal of Fisheries and Aquatic Sciences. 40:671-680. 96  Rosenau, M.L., and J.D. McPhail. 1987. Inherited differences in agonistic behavior between two populations of coho salmon. Transactions of the American Fisheries Society. 116:646654. Ruggerone, G.T., S. Kuchta, D. Bregar, and H. Senn. 1995. Database of propagated anadromous Pacific salmon, 1950-1993. Northwest and Alaska Fisheries Science Center, National Marine Fisheries Service, Seattle, Washington. Sabo, J.L., and G.B. Pauley. 1997. Competition between stream-dwelling cutthroat trout (Oncorhynchus clarki) and coho salmon (Oncorhynchus kisutch): effects of relative size and population origin. Canadian Journal of Fisheries and Aquatic Science. 54:2609-2617. Sasgrov, H., K. Hindar, S. Kalas, and H. Lura. 1997. Escaped farmed Atlantic salmon replace the original salmon stock in the River Vosso, western Norway. ICES Journal of Marine Science. 54:1166-1172. Sandercock, F.K. 1991. The life history of coho salmon (Oncorhynchus kisutch). Pages 395 to 445 in C. Groot and L. Margolis, editors. Life Histories of Pacific Salmon. University of British Columbia Press, Vancouver, British Columbia. SAS. 2001. Release 8.2. SAS Institute Inc. Cary, North Carolina, USA. Scott, W.B., and E.J. Crossman. 1998. Freshwater fishes of Canada. Gait House Publications, Oakville, Ontario. Shafland, P.K., and W.M. Lewis. 1984. Terminology associated with introduced organisms. Fisheries. 9(4): 17-18. Slaney, P.A., and T.G. Northcote. 1974. Effects of prey abundance on density and territorial behavior of young rainbow trout (Salmo gairdneri) in laboratory stream channels. Journal of the Fisheries Research Board of Canada. 31:1201-1209. Sosiak, A.J., R.G. Randal, and J.A. McKenzie. 1979. Feeding by hatchery-reared and wild Atlantic salmon (Salmo salar) parr in streams. Journal of the Fisheries Research Board of Canada. 36:1408-1412. Stanton, B. 2003. Fisheries and Oceans Canada. Manager, Chilliwack Hatchery. Personal communication with author. Swain, D.P, B.E. Riddell, and C.B. Murray. 1991. Morphological differences between hatchery and wild populations of coho salmon (Oncorhynchus kisutch): environmental versus genetic origin. Canadian Journal of Fisheries and Aquatic Sciences. 48:1783-1791. Taylor, E.B. 1986. Differences in body morphology between wild and hatchery populations of juvenile coho salmon. Progressive Fish Culturist. 48:171-176.  97  Taylor, E.B., and P.A. Larkin. 1986. Current response and agonistic behavior in newly emerged fry of Chinook salmon, Oncorhychus tshawytscha, from ocean- and stream-type populations. Canadian Journal of Fisheries and Aquatic Science. 43:565-573. Taylor, P.W., and S.D. Roberts. 1999. Clove oil: an alternative anaesthetic for aquaculture. North American Journal of Aquaculture. 61:150-155. Thompson, C.E., W.R. Poole, M.A. Matthews, and A. Ferguson. 1998. Comparison, using minisatellite DNA profiling, of secondary male contribution in the fertilization of wild and ranched Atlantic salmon (Salmo salar) ova. Canadian Journal of Fisheries and Aquatic Sciences. 55:2011-2018. Todd, K. 2001. Tinkering with Eden: a natural history of exotics in America. Norton, New York. Unwin, M.J. and G.J. Glova. 1997. Changes in life history parameters in a naturally spawning populations of chinook salmon (Oncorhynchus tshawytscha) associated with releases of hatchery-reared fish. Canadian Journal of Fisheries and Aquatic Sciences. 54:1235-1245. Utter, F. 2001. Patterns of subspecific anthropogentic introgression in two salmonid genera. Reviews in Fish Biology and Fisheries. 10:265-279. Vallestad, L.A. and T. Hesthagen. 2001. Stocking offreshwaterfish in Norway: management goals and effects. Nordic Journal of Freshwater Research. 75:143-152. Volpe, J.P. 2000. University of Alberta, Edmonton. Personal communication with the author. Volpe, J.P., E B . Taylor, D.W. Rimmer, and B.W. Glickman. 2000. Evidence of natural reproducation of aquaculture-escaped Atlantic salmon in a coastal British Columbia river. Conservation Biology. 14(3):899-903. Volpe, J.P., B.R. Anholt, and B.W. Glickman. 2001a. Competition among juvenile Atlantic salmon (Salmo salar) and steelhead (Oncorhynchus mykiss): relevance to invasion potential in British Columbia. Canadian Journal of Fisheries and Aquatic Sciences. 58:197-207. Volpe, J.P., B.W. Glickman, and B.R. Anholt. 2001b. Reproduction of aquaculture Atlantic salmon in a controlled stream channel on Vancouver Island, British Columbia. Transactions of the American Fisheries Society. 130:489-494. Volterra, V. 1926. Variazioni e fluttuazioni del numero d'individui in specie animali conviventi. Mem. R. Acad. Naz. Dei Lincei. 2:31-113. Wankowski, J.W., and J. E. Thorpe. 1979. Spatial distribution and feeding in Atlantic salmon. Journal of Fish Biology. 14:239-247. Waples, R.S., M.J. Ford, and D. Schmitt. 2001. Empirical results of salmon supplementation: a preliminary assessment. In T.M. Bert, editor. Ecological and genetic implications of aquaculture activities. Kluwer Academic Publishers, New York. 98  Webb, J.H., A.F. Youngson, IS. McLaren, and M.J. Donaghy. 1993. Spawning of farmed Atlantic salmon, Salmo salar, L., in the second year after their escape. Aquaculture and Fisheries Management. 24:557-561. Webb, J.H., D.W. Hay, P.D. Cunningham, and A.F. Youngson. 1991. The spawning behaviour of escaped farmed and wild adult Atlantic salmon (Salmo salar L.) in a northern Scottish river. Aquaculture. 98:97-110. Weins, J.A. 1989. The Ecology of Bird Communities. Volume 2. Processes and variations. Cambridge University Press, Cambridge, England. Werner, E.E. 1986. Species interactions in freshwater fish communities. Pages 344-357 in J. Diamond, and T.J. Case, editors. Community Ecology. Harper and Row, San Francisco. Winton, J., and R. Hilborn. 1994. Lessons from supplementation of Chinook salmon in British Columbia. North American Journal of Fisheries Management. 14(1): 1-13. Youngson, A.F., and E. Verspoor. 1998. Interactions between wild and introduced Atlantic salmon (Salmo salar). Canadian Journal of Fisheries and Aquatic Sciences. 55(Suppl. 1):153-160. Zar, J.H. 1999. Biostatistical Analysis. 4 edition. Prentice Hall, New Jersey. th  99  

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.831.1-0091402/manifest

Comment

Related Items