Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Influence of physical habitat on the seasonal movement, growth, and habitat association of individual… Roberge, Michelle 2000

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

Item Metadata

Download

Media
831-ubc_2000-0551.pdf [ 6.24MB ]
Metadata
JSON: 831-1.0089598.json
JSON-LD: 831-1.0089598-ld.json
RDF/XML (Pretty): 831-1.0089598-rdf.xml
RDF/JSON: 831-1.0089598-rdf.json
Turtle: 831-1.0089598-turtle.txt
N-Triples: 831-1.0089598-rdf-ntriples.txt
Original Record: 831-1.0089598-source.json
Full Text
831-1.0089598-fulltext.txt
Citation
831-1.0089598.ris

Full Text

I N F L U E N C E O F P H Y S I C A L H A B I T A T O N T H E S E A S O N A L M O V E M E N T , G R O W T H A N D H A B I T A T A S S O C I A T I O N O F I N D I V I D U A L C O A S T A L C U T T H R O A T T R O U T by M I C H E L L E R O B E R G E B . S c , University of British Columbia, 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 L M E N T O F 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 O F M A S T E R O F S C I E N C E in T H E F A C U L T Y O F G R A D U A T E S T U D I E S Department of Zoology We accept this thesis as conforming to the re(i>rirety standard T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A October 2000 © Michelle Roberge, 2000 In presenting t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree at the University of B r i t i s h Columbia, I agree that the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r reference and study. I further agree that permission for extensive copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the head of my department or by h i s or her representatives. I t i s understood that copying or p u b l i c a t i o n of t h i s thesis f o r f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of The U n i v e r s i t y of B r i t i s h Columbia Vancouver, Canada Date ABSTRACT A multi-stream comparative study conducted in southwestern British Columbia revealed that juvenile coastal cutthroat trout (Oncorhynchus clarki clarki) were found at highest density in narrow (< 3 m), shallow (< 40 cm) streams with non-coarse substrate. To assess how cutthroat trout grow and behave within different physical habitats, I selected three streams for intensive examination from the physical structure gradient that I derived in the multi-stream study: (A) narrow width, shallow, fine substrate; (B) narrow width, shallow, coarse substrate; (C) wide width, deep, coarse substrate. I used radiotelemetry and bi-monthly recapturing of PIT tagged fish from set trap locations to examine individual growth and movement among the stream types over the fall, winter and spring of 1998/99. Individual PIT tagged fish tended to be heavier and grew faster during the spring than the winter or fall in all streams. Growth rate did not seem to be dependent on the physical habitat of the stream. During normal seasonal flood events, movement rate of fish from each of the three stream types was not affected. During a major flood event that occurred in stream types (C) and (B), fluctuations in water depth in stream type (C) displaced several radiotagged fish (> 100 m) and caused a general downstream movement in PIT tagged individuals. Almost no movement was detected during the winter in the three streams for PIT tagged fish (< 0.09 m/day). Density was lowest in stream (C) and highest in stream (A). Past and current provincial forestry practices usually result in removal of riparian vegetation on small streams. This can alter their physical habitat, usually by widening and aggrading the stream bed. Therefore, in streams where riparian logging has occurred, displacement of cutthroat trout during major floods may be a concern. i i TABLE OF CONTENTS ABSTRACT II TABLE OF CONTENTS III LIST OF TABLES V LIST OF FIGURES VII LIST OF APPENDICES IX ACKNOWLEDGMENTS X GENERAL INTRODUCTION 1 Ecology of Coastal Cutthroat Trout 2 Effects of Forestry on Stream Habitat and Fish 4 CHAPTER ONE 8 Introduction 8 Study Sites 9 Methods 10 Site Selection 10 Fish Survey 12 Habitat Survey 12 Data Analysis 13 Results 14 Discussion 22 Small-scale Study: Hypothesis and Site Selection 25 CHAPTER TWO 26 Introduction 26 Study Sites 30 Methods 33 Stream Habitat Surveys 33 iii Initial Set-up and Daily Sampling 33 Monthly Sampling 34 Fish Trapping and PIT Tagging 36 Radiotelemetry 38 Data Analysis 41 Stream Habitat Surveys 41 Fish Trapping and PIT Tagging 42 Radiotelemetry 45 Growth Rate 46 Habitat Association 47 Results 47 Stream Conditions 47 Number of Cutthroat Trout Sampled 57 Hypothesis 1: Seasonal and Short-term Movement Rate 58 Flood Events 67 Hypothesis 2: Growth Rate 68 Hypothesis 3: Habitat Association 75 PIT Tagged Fish 75 Radiotagged Fish 76 Population Density and Probability of Survival 76 Discussion 81 Movement 81 Growth 88 Habitat Association 92 Population Density and Probability of Survival 94 GENERAL CONCLUSION 96 LITERATURE CITED 98 APPENDIX 110 iv L I S T O F T A B L E S Table 1.1: Habitat characteristics of the study sites used in the synoptic survey in Chapter One. 15 Table 1.2: Density of fish species at the study sites. 16 Table 1.3: Structure coefficients of both the cutthroat trout and coho salmon principal component analyses. 17 Table 1.4: Multiple regression relationships between the principal component axes and cutthroat trout density. 18 Table 1.5: Average stream characteristics for the Sunshine Coast and West Coast of Vancouver Island regions. 18 Table 1.6: Average stream characteristics for cleat-cut, second growth and old growth sites. 18 Table 2.1: General morphological characteristics of the three study streams used in Chapter Two. 30 Table 2.2: Mean water and air temperature during fall and winter in the three streams. 34 Table 2.3: Details of radiotagged fish used during the study. Table 2.4: Description of average depth and flow rate of channel unit types. 40 43 Table 2.5: Average water depth, coefficient of variation and discharge at the stationary metre sticks during fall and winter. 48 Table 2.6: Summary of average channel unit dimensions during fall and winter of 1998 and summer of 1999. 54 Table 2.7: Summary of hardwood, softwood, canopy cover and riparian vegetation for fall and winter. 57 Table 2.8: Breakdown of the total number of fish implanted with PIT tags and the number of recaptures. 58 Table 2.9: Analysis of variance result table for movement rate among seasons, streams and size classes for PIT tagged fish. 59 v Table 2.10: Number of within-day observations at different distances from initial locations per day for radiotagged fish. 64 Table 2.11: Average home range size of radiotagged fish using both the 95% and 100% estimation methods. 67 Table 2.12: Analysis of covariance result table among seasons, streams and fish size for PIT tagged fish. 70 Table 2.13: Pearson's correlation coefficients between length or weight and weight growth rate for PIT tagged fish. 75 Table 2.14: Habitat association by channel units for PIT tagged fish. Table 2.15: Habitat association by cover type for PIT tagged fish. Table 2.16: Habitat association by substrate class for PIT tagged fish. Table 2.17: Habitat association by channel unit and cover type for radiotagged fish. 77 78 79 80 vi L I S T O F F I G U R E S Figure 0.1: Range of coastal cutthroat trout. Figure 1.1: Location of steams surveyed on the Sunshine Coast and West Coast of Vancouver Island regions for Chapter One. 11 Figure 1.2: Plots of principal component axis two versus one for both the cutthroat trout and coho salmon sites analyses. 19 Figure 1.3: Relationship between cutthroat trout density and principal component axis one. 20 Figure 1.4: Average density of cutthroat trout and coho salmon within each the Sunshine Coast and West Coast of Vancouver Island regions. 20 Figure 1.5: Average density of cutthroat trout and coho salmon in clear-cut, second growth and old growth sites. 21 Figure 1.6: Plot of the principal component axis two versus one for the cutthroat trout sites arranged by riparian status. 21 Figure 2.1: Geographical range of coastal cutthroat trout and location of study sites for Chapter Two. 27 Figure 2.2: Predictions for movement rate, growth rate and habitat association for Chapter Two. 29 Figure 2.3: General morphology, location of physical structures and channel units in the three study sites. 32 Figure 2.4: Placement of minnow traps within the three study sites. 37 Figure 2.5: Histograms of size distribution of cutthroat trout within the three streams. 44 Figure 2.6: Hydrograph of each stream from September 1998 to A p r i l 1999. 50 Figure 2.7: Change in water level indicating timing of floods in the three streams. 51 Figure 2.8: Displacement of marked large woody debris pieces within Ava lon and Husdon Creeks. 52 vn Figure 2.9: Summary of habitat survey data by month within each of the three streams. 55 Figure 2.10: Average seasonal movement rate of both PIT tagged and radiotagged fish within the three streams. 60 Figure 2.11: Average seasonal movement rate of each size class of PIT tagged fish within the three streams. 61 Figure 2.12: Movement pattern over the course of the study of individual PIT tagged fish from the three streams. 62 Figure 2.13: Movement pattern over the course of the study of individual radiotagged fish from the three streams. 63 Figure 2.14: Histograms and chi-squared results of the real-random populations comparisons of PIT tagged fish. 65 Figure 2.15: Location and size of home ranges of radiotagged fish within the three streams. 66 Figure 2.16: Average movement rate during flood and non-flood periods for PIT tagged and radiotagged fish. 71 Figure 2.17: Average movement rate of PIT tagged fish during the l-in-40 year flood event and non-flood periods in Avalon and Pipe Creeks. 72 Figure 2.18: Average movement rate of each size class of PIT tagged fish in Avalon Creek during flood and non-flood periods. 72 Figure 2.19: Average growth rate in length and width for PIT tagged fish in each study stream, 73 Figure 2.20: Average In weight for fish caught in minnow traps in each of the three streams over fall and winter 1998, and spring and summer 1999. 74 Figure 2.21: Average activity level of radiotagged fish at different times of day during fall and winter in the three streams. 74 Figure 2.22: Outcome of predictions made for Chapter Two. 89 viii LIST OF APPENDICES Appendix 1: Discharge water-depth relationships for the three streams. 110 Appendix 2: Activity level of dead fish carrying a transmitter. I l l Appendix 3: Example calculation of 95% and 100% home range size for radiotagged fish. 112 Appendix 4: Characteristics of marked large woody debris pieces in Avalon and Husdon Creeks. 113 ix A C K N O W L E D G M E N T S I would first like to thank my funding agency, the British Columbia Ministry of Environment, Lands and Parks. The generous contribution came from a Forest Renewal British Columbia project under the direction of Dr. Jordan Rosenfeld, B.C. Ministry of the Environment, Lands and Parks, called 'Sensitivity of anadromous cutthroat trout to the effects of logging'. I would also like to thank the Vancouver Natural History Society for awarding me the 2000 Scholarship Award, and the Society members for expressing interest in my work. This thesis would not have been completed if it were not for the patience and guidance provided by my supervisor Dr. Scott G. Hinch during the past three years. Many thanks for his understanding and dedication to this thesis and my personal well-being. Thank you to my committee members, Dr. Rick Taylor, Dr. Michael Healey and Dr. Jordan Rosenfeld for continued advice and support. Thank you again to Dr. Jordan Rosenfeld for his personal friendship and help throughout this project, and to his team of talented researchers and assistants (Dave Bates, the Sechelt Native Band, Marc Porter, Taja Lee, Dave Forsyth, and Brent Matsuda) who collected the data used in Chapter One. I would also like to thank my dedicated field assistant Andrew Lotto. Andrew taught me how to work efficiently in the field, laugh, take a practical joke, give a practical joke and cook - 1 will always be indebted. Kyle Young diligently edited early drafts of this thesis and provided a number of well received free meals, thank you. My father Georges Roberge, and friends Pier van Dishoeck, Ian Gazeley, Dr. Jordan Rosenfeld, Shelly Boss, and Eric Lotto all provided valuable help in the field carrying buckets, taking stream measurements and getting wet. Grant McBain (Department of Fisheries and Oceans), Jim Wilson (Department of Fisheries and Oceans), and Eric Parkinson (B.C. Ministry of the Environment, Lands and Parks) generously lent equipment and advice during the field season (sorry about the truck Eric). Finally, I would like to thank Suzanne Russell for her friendship and hospitality, the Chutter family for their wonderful ocean-side cabin, and Matthew McLeod for his support and patience while I finished writing this thesis. Hail Caesar! x G E N E R A L I N T R O D U C T I O N Coastal cutthroat trout, Oncorhynchus clarki clarki are found along the west coast of North America from the Eel River in northern California to Seward in southeastern Alaska (Figure 0.1). Cutthroat trout reside on the west slopes of the Coast Range mountains in British Columbia (B.C.) and the Cascade Mountains in the Pacific Northwest (Trotter 1989). Although cutthroat trout are widespread, information about population size and changes in life history are unknown for approximately 80% of the populations within B . C . (Slaney et al. 1996). The information that is available suggests a major decline in abundance within and among populations throughout their range over the past 20 years (Trotter 1989; Slaney et al. 1996). The decline has been linked to overfishing and to habitat loss largely from urbanization, agriculture and forestry practices (Slaney et al. 1996; Nehlsen et al. 1997). Changing ocean conditions has not yet been explored as a possible mechanism for decline for cutthroat trout, however it is a likely contributor to the decline. This thesis focuses on habitat loss in the freshwater environment. The main objective of this thesis is to determine how alterations to habitat affect the behaviour (movement, growth and habitat association) of cutthroat trout during the winter. Lower growth rate may lead to reduced fecundity while greater movement can cause higher mortality both of which would lead lower population density. In this thesis I use both a large- and small-scale approach to address my main objective. Large-scale (synoptic survey) studies are ideal for detecting patterns in species richness (Griffiths 1997), fish abundance, distribution, and diversity over a wide scale (Taylor and Lienesch 1995; Jowett et al. 1996, Maret et al. 1997). Small-scale studies can identify how abiotic and biotic factors influence the patterns seen in large-scale studies (Magnan et al. 1994) by mainly addressing individual characteristics such as behavioural or growth data from single populations (Cunjak and Power 1987; Quinn and Peterson 1996; Berg and Bremset 1998), or intra-interspecific interactions (Hartman 1965; Bisson et al. 1988). B y combining large and small-scale studies, results from large-scale studies can be used as the basis for small-scale studies, so that the overall conclusions are more reliable (see Magnan et al. 1994). Here I use a large-scale study to identify what abiotic factors (habitat types) are associated with cutthroat trout density (Chapter One). I used 1 this information as the basis for a small-scale study which documented the seasonal behaviour of cutthroat trout in the different habitat types (Chapter Two). In this general introduction I describe how forestry practices affect stream habitat and fish species, following a description of cutthroat trout ecology. Ecology of Coastal Cutthroat Trout Cutthroat trout typically live in small (< 5m wide), headwater and tributary streams (see Chapter One; Hartman and G i l l 1968; Swales et al. 1987; Bozek and Hubert 1991). Within the stream environment cutthroat trout can be found associated with specific channel units (backwater pools, deep pools, and glides; Wilzbach and Hal l 1985; Glova 1986; Bisson et al. 1988), physical cover structures (Heggenes et al. 1991b), and lateral habitat as defined by Moore and Gregory (1988a). Habitat association seems to change from summer to winter due to seasonal changes in abiotic and biotic environmental conditions. Heggenes et al. (1991b) found that coastal cutthroat trout use shallower areas with overhead cover during the winter (18 cm) compared to the summer (22 cm). Throughout their range, cutthroat trout populations are either diadromous (sea-going), freshwater stream resident, lake resident or lake migrating (Trotter 1989). The general life cycle of diadromous coastal cutthroat trout is similar to Pacific salmon, however cutthroat trout are iteroparous (spawn more than once), and can live up to 10 years (Scott and Crossman 1985). The life cycle of coastal cutthroat trout varies depending on latitude, with more northern populations spawning later in the year (Trotter 1989). In B . C . , spawning takes place during the spring (February to May) . Smoltification in diadromous cutthroat trout usually occurs after at least 2 years in freshwater at a size of between 130 and 200 mm in length (Trotter 1989). Downstream migration to the estuary takes place in late spring (Benke 1992). Cutthroat trout have been found to migrate up to 70 km from their natal stream, however these migrations do not include passage through large open ocean areas (Trotter 1989). In B . C . , salt water residency occurs during the summer (approximately four months), and migration back into freshwater occurs during the fall (August-November). Few cutthroat trout remain in the estuary during winter (Trotter 1989). Presumably all age classes of diadromous and stream resident cutthroat trout reside within the stream during the winter. 2 150° 145° 140° 135° 130° ^ — r 125° I lio- ns0 I Alaska a ) - J v Y u k o n h 60° • N PACIFIC O C E A N British Columbia fry. f Iff-4 i . - J " ' Vaiuouver Canada * '>X * " USA 1 Washington /• Oregon < California; 55° \- 50° \- 45° 40° Figure OJ: Range of coastal cutthroat trout (shaded area). M a p adapted from Scott and Crossman (1985). 3 Apart from the long distance migrations diadromous fish undertake, fish also make short distance movements within and between channel units and individual physical structures (cover) in search of food and to avoid predators or high velocity areas. Within-stream movement of cutthroat trout generally decrease as summer progresses and water level decrease (Heggenes et al. 1991a; Brown and MacKay 1995; Young 1996; Young et al. 1997). Winter is thought to be a critical period for fish (Cunjak 1988). In coastal systems, compared to summer base flows, flow during winter is relatively high and variable. In spring, melting of winter snow leads to high flow and freshet events which again cause large fluctuations in discharge. A s well during winter in all areas of cutthroat trout range, water temperatures decrease dramatically from summer. Interior (non-coastal) populations of cutthroat trout were found to reduce daily activity and movement during winter (Young 1998). A n y increase in movement during the winter may lead to an increase in predation risk, an increase in stress level (Cunjak and Power 1987; Cunjak 1988), or a reduction in growth rate and ultimately survival. Effects of Forestry on Stream Habitat and Fish Seventy-five percent of B .C . ' s human population lives within 60 km of the coast in two main regions of the province, the lower mainland, and the east coast of Vancouver Island, making the coastal regions important economically (Minister of Public Works and Government Services Canada 2000). Both marine areas and freshwater streams along the coast are important ecosystems for salmon and trout. Habitat loss due to urbanization, agriculture and forestry practices alter natural conditions which affect the life history and survival of salmonids. Some forestry practices in particular, such as clear-cut logging to the stream bank, can have negative affects on salmon and trout bearing streams (Hall and Lantz 1969; Ell iot 1986; Platts et al. 1989; Lamberti et al. 1991; Fausch and Northcote 1992; Slaney et al. 1996; Nehlsen et al. 1997). Before 1995 in B . C . , there was little legislated protection for streams from deleterious forestry practices. In 1995, the B . C . Forest Practices Code (FPC) was enacted by the provincial government to protect forest ecosystems, including freshwater streams and aquatic communities, from harmful forestry practices. The F P C guidelines require that buffer strips be left along stream edges at varying widths depending on the size of the stream and the presence/absence of fish. 4 The riparian are is divided into a Reserve Zone and a Management Zone. Within the Reserve Zone, no logging is permitted (ie. all tree are to be left standing). N o Reserve Zone is required for streams smaller than 1.5 m in width, even i f fish are present. The Management Zone is set outside the Reserve Zone or replaces the Reserve Zone i f one is not required. Here logging is permitted and the degree of logging is left up to the discretion of the logging contractor (Sierra Legal Defense Fund 1997). Because of the discretionary nature of Management Zone application and monitoring, this zone is often completely harvested of marketable timber. A s well , windfall is a problem in areas with only 20 m wide Reserve or Management Zones (Young et al. 1999). The Sierra Legal Defense Fund (1997) conducted an audit of the F P C in 1996, a year after its inception. Their main finding was that the guidelines were not being followed as 83% of the streams surveyed that were not supposed to be logged, according to the guidelines, were clear-cut to the stream bank. Cutthroat trout are typically found in streams < 5 m wide and Sierra Legal Defense Fund found that 79% of streams < 1.5 m wide were clear-cut to the stream bank. This finding suggests that cutthroat trout receive minimal protection from forestry practices in B . C . . The effect of forestry practices, such as clear-cut logging to the stream bank, on stream habitat is well studied in coastal systems (Slaney et al. 1977; Murphy and K o s k i 1989; Lamberti et al. 1991; Ralph et al. 1994; Hartman et al. 1996), but less is known for interior systems (Platts et al. 1989). Logging to the stream bank eliminates the riparian habitat surrounding and hanging over the stream (ie. canopy). With the loss of canopy cover, seasonal temperature fluctuation are more severe in small streams. During the summer, stream temperature in unshaded areas is warmer than shaded areas (Moore and Gregory 1988a; Smale and Rabeni 1997), sometimes by as much as 5-15°C (Hall and Lantz 1969; Hartman and Scrivener 1990). In Carnation Creek located in coastal B . C . , water temperature of the stream flowing through clear-cut areas increased in all months of the year (Holtby 1988). Hall and Lantz (1969) reported increases of 15°C in streams running through recently clear-cut areas in coastal Oregon, and Young et al. (1999) reported a similar increase in temperature in a coastal B . C . stream flowing through a clear-cut. Increasing stream temperature reduces the amount of dissolved oxygen within the water (Hall and Lantz 1969) which could be deleterious to fish. On the other hand, 5 increased light penetration resulting from an open canopy increases stream primary production (Murphy and Hal l 1981; Wilzbach and Hal l 1985; Moore and Gregory 1988a; Rosenfeld and Hudson 1997) and increases cutthroat trout foraging efficiency (Wilzbach and Hal l 1985; Wilzbach et al. 1986). A s a consequence of increased food availability and foraging efficiency, cutthroat trout growth rate is often higher in open, unshaded areas than shaded areas (Wilzbach et al. 1986; Moore and Gregory 1988b). The abundance of physical structures such as large woody debris ( L W D ) and root-wads are either reduced (Osborn 1980) or increased (Hartman and Scrivener 1990) within streams after logging occurs. Physical structures may decrease within the stream either due to the loss of L W D input from the riparian area (Ralph et al. 1994), or selective removal of L W D from streams during logging (Bilby and Ward 1991). Physical cover provides refuge for fish either from predators, or high flow conditions (enabling fish to conserve energy) (Hartman and G i l l 1968; Tschaplinski and Hartman 1983; Cunjak and Power 1986; Glova 1986; Heifetz et al. 1986; Bisson et al. 1988; M c M a h o n and Hartman 1989; Brown and M a c K a y 1995; Watson and Hillman 1997). Juvenile salmonids move out of areas that have reduced cover (Tschaplinski and Hartman 1983; El l iot 1986; M c M a h o n and Hartman 1989). The addition of slash, or logging debris such as L W D or root-wads increases the amount of physical in-stream structures. These structures are often unstable (Hartman and Scrivener 1990), short lived, and can negatively impact stream stability, channel morphology, and may increase the amount of sediment that enters a stream after a storm. With the removal of riparian habitat, the size distribution of L W D shifts to smaller pieces (Ralph et al. 1994) which can affect the structure and distribution of channel units. A s well , with the complete loss of an intact riparian area, the potential future input of L W D and other large physical structures into the stream decreases (Ralph et al. 1994), which can have implications for the future productivity and physical stability of the stream. Tree roots and logs from the riparian area stabilize stream banks. Loss of riparian habitat can lead to decreased bank stability as bank erosion and sedimentation into rivers increase (Murphy et al. 1986). A n increase in fine sediments in streams can be detrimental to salmonids because it reduces invertebrate food supply (Bergstedt and Bergersen 1996) potentially affecting growth. A s well , increased fine sediments fill in 6 inter-gravel/cobble spaces which are used by juvenile fish as refuge (Anderson 1998), reduce the amount of good quality spawning habitat within streams (Platts et al. 1989), and can lead to the widening and filling in of the stream channel and channel units (pools), making them shallower and less suitable for fish (Anderson 1998). A s well , coarse sediments can enter a stream after logging. Land slides and debris flows from unstable road construction or stream banks can add large amounts of cobble and boulders into the stream which can widen, aggrade and re-route the channel (Ricks 1995). Crushing of eggs or juvenile fish and displacement can occur when large sediments are transported downstream during heavy flow events (Seegrist and Gard 1972). 7 C H A P T E R O N E Introduction The behaviour and distribution of fish in streams are influenced by both abiotic (habitat) and biotic (eg. predation, competition) factors (Magnan et al. 1994). Within a population, biotic factors are influenced by the state of the abiotic factors (Berg and Bremset 1998). Investigation of these processes in both lake (Hinch 1991) and stream systems (Naslund et al. 1998) has generally focussed on very large- (single to multiple watersheds) or very small- (within a single stream reach) scale. Large-scale studies are used to provide patterns of fish distribution over a large geographical or environmental scale, while small-scale studies determine what interactions are occurring within, and between species at the local scale (Hinch 1991). Large-scale (synoptic survey) studies are ideal for detecting patterns in species richness (Griffiths 1997), fish abundance, distribution, and diversity over a wide scale (Taylor and Lienesch 1995; Jowett et al. 1996, Maret et al. 1997). However, more detailed information, such as diet preference (Lacasse and Magnan 1992; Magnan et al. 1994), and life history variation (Young 1999) can also be detected. Abiot ic or habitat factors, such as stream width or gradient, are often the primary type of variable for this kind of study. For example, Watson and Hil lman (1997) surveyed 93 streams in Washington, Idaho and Montana for bull trout (Salvelinus confluentus) presence and 24 habitat variables. They found that bull trout were present more often in wide valley streams, with tree and shrub vegetation and boulder/cobble substrate. Watson and Hi l lman (1997) were able to correlate bull trout distribution with three habitat variables but no causal mechanism was identified for this association. A problem with the large-scale approach is that specific mechanisms regulating biotic patterns often cannot be directly tested from the data, but only speculated upon. Small-scale studies can identify how abiotic and biotic factors influence the patterns seen in large-scale studies (Magnan et al. 1994) by mainly addressing individual characteristics such as behavioural or growth data from single populations (Cunjak and Power 1987; Quinn and Peterson 1996; Berg and Bremset 1998), or intra-interspecific interactions (Hartman 1965; Bisson et al. 1988). Quinn and Petersen (1996), correlated 8 coho salmon survival to different habitat variables in B i g Beef Creek, Washington. They found that coho salmon survival was correlated with the quantity of large woody debris ( L W D ) , habitat type, and distance from estuary. Logistically, only a few variables can be studied at once in small-scale studies. Magnan et al. (1994) has suggested this method as being biased because some biologically important interactions between abiotic and biotic factors may be missed due to certain variables being excluded. Combining large and small-scale studies could reduce the bias suggested by Magnan et al. (1994) by using results from large-scale studies as the basis for small-scale studies. For example, Naslund et al. (1998) combined data from a nation-wide survey on brown trout (Salmo trutta) distribution, size and stream type with data from an intensive three stream study looking at the same variables. They found that the synoptic survey supported their results from the three stream study. This study is the first of two complimentary studies dealing primarily with coastal cutthroat trout (Oncorhynchus clarki clarki) habitat association and behaviour in coastal British Columbia. This study has three objectives. First, to use data collected during a synoptic survey in coastal British Columbia to determine which abiotic factors best characterize small coastal streams. The second objective is to determine i f there are any relationships between the abiotic factors assertained by objective one and (i) coastal cutthroat trout and coho salmon (O. kisutch) densities and (ii) anthropogenic activities (primarily logging). The third objective is to use the habitat information to generate hypotheses and select appropriate study streams for the complementary small-scale study which wi l l investigate coastal cutthroat trout habitat use, growth and movement in coastal streams that differ according to the habitat characterization determined in this study. Study Sites In the summers of 1997 and 1998,35 sites in 31 streams were selected for study in two regions of southwest British Columbia. The regions spanned an area > 1000 k m 2 (Figure 1.1). Eleven sites were located on the southwest coast of British Columbia's mainland (called the Sunshine Coast - SC region) and 18 on the west coast of Vancouver 9 Island ( W C region; Table 1.1). Both regions lie within the coastal western hemlock biogeoclimatic zone. The geology of the S C region is mostly granitic rock from the early Cretaceous period. The geology of the W C region is mixed, with basaltic and metamorphic formations from the Jurassic, late Cretaceous and early Tertiary periods. Streams that were chosen had to be small in size (< 12 m width), free flowing to the ocean (to assume that populations were diadromous), and in a variety of landscapes (ie. degree of riparian logging: termed riparian status) so that a broad range of habitats that these fish encounter were included. Streams were selected using 1:20000 topographic and forest cover maps and ground searching. The three classes of riparian status were 1) old growth (OG), no harvesting or harvesting over 100 years ago, 2) second growth (SG), harvesting between 15 and 100 years ago and, 3) clear-cut (CC) , harvesting within the last 15 years (Table 1.1). Methods Site Selection The survey site within a stream was selected on the basis of easy accessibility from a road or trail. The length of the study site was typically equal to 10 times the average width of the stream. This method was used to ensure that the study site represented the characteristics of the whole stream. Each site consisted of several contiguous channel units, with at least one pool, one riffle and one glide. Six stream systems were sampled at multiple sites (Table 1.1). There were two criteria for considering sites within a stream system as independent. The first was that one site could be a tributary to the other site (different streams), as long as the two sites were at least 500 m apart (less chance that fish would move between the sites). The second criteria was that one site could be sampled between years, so that at least the coho salmon were not sampled twice, and the cutthroat trout would be larger and presumably occupying different habitat (see General Introduction). 10 Figure 1.1. General location of study streams within the West Coast of Vancouver Island Region and Sunshine Coast Region. See Table 1.1 for the stream names that correspond to the presented stream numbers. 11 Fish Survey Channel spanning nets were used at the upstream and downstream ends of each channel unit to prevent fish emigrating from the channel unit. Each channel unit was sampled by electroshocking using the triple pass method (Seber and L e Cren 1967). This method is widely used because population abundance estimates can easily be obtained (Seber and L e Cren 1967; Zippen 1958; Schnute 1983). In this study however, sample size of fish within the majority (24 of 31) of streams was not sufficient to accurately estimate population abundance (Zippen 1958). Instead, density (fish/m 2) was estimated by dividing the number of fish caught in each site by the surface area of the site (average channel unit width times the site length). From each sample pass, fish were identified to species, measured for mass (wet weight) and fork length. Fish that could not be identified to species were recorded as 'unknown'. After the third pass was complete, all fish were returned to the channel unit of capture. Habitat Survey Beginning at the downstream end of each site, each channel unit was measured for maximum (thalweg) depth, average wetted and bankfull width, length, gradient (Suunto clinometer), and temperature (°C). Water velocity was measured at 40% of water depth (March-McBirney model 2000 FlowMate velocity meter) once per channel unit. Percent overhead canopy cover was estimated visually from the center of each channel unit. Large woody debris ( L W D ; wood pieces submerged or partially submerged and ^ 15 cm dbh), boulder, undercut bank, submerged and overhanging vegetation cover were estimated as a percentage of each channel unit. Substrate was categorized as fines (<, 1.0 mm), gravel (2.0 - 64.0 mm), cobble (64 mm - 256 mm), boulder (> 256 mm) and bedrock (continuous with the bank). Each substrate class was visually estimated as a percentage of the stream bed of each channel unit. 12 Data Analysis Before any statistical analyses were performed, all percentage variables were arcsine transformed (asin(sqrt(x))), and continuous variables were log transformed (log(x+l)). Data were averaged per site, so that a site represented a replicate in the analyses. Principal Components Analysis (PCA) was used to examine the stream habitat data and describe the physical variation among the sites. Two P C A s were performed. The first P C A included sites where cutthroat trout (and any other species) were present and excluded any sites where cutthroat trout were absent (n = 32). The same analysis was carried out on coho salmon sites (sites where coho salmon were absent were excluded; coho salmon sites, n = 26). Averages of eight habitat variables were used in both P C A s : wetted stream width, maximum depth, percent log cover, percent undercut bank cover, and percent fine, gravel, cobble and boulder substrates. The other variables were excluded from analyses because they were inconsistently measured during the surveys so had no value recorded in the data set. Mult iple linear regression was used to relate fish densities (dependant variables) to the P C axes (independent variables). The analysis examined the relationship between the dependant variable and each independent variable simultaneously for each species. The general model R 2 is presented along with the slope,F - value and P - value for each separate relationship. These analyses can be used for predictive purposes, however in this paper these analyses were used to identify general associations. A n unpaired t-test was used to compare coho salmon density between regions. Analysis of covariance ( A N C O V A ) was used to compare cutthroat trout density between regions using boulder substrate, thalweg depth and stream width as independent covariates (because they were found to be related to cutthroat trout density). Analysis of variance tests were performed to compare coho salmon density and cutthroat trout density among riparian status types. Eight unpaired t-tests were carried out to test i f each of the eight habitat variables differed between regions. Eight A N O V A tests were performed to determine i f habitat variables differed among the classes of riparian status. T o control for eight simultaneous t-tests and A N O V A s . I used the standard Bonferroni technique to 13 correct the alpha level. Calculated test statistics, and associated P and |3 (power) values are presented for each test. Results A total of 2260 fish were caught during the synoptic surveys. Coastal cutthroat trout and coho salmon were the two most common species caught during the surveys (32 and 29 sites respectively; Table 1.2). A l l coho salmon were either young of the year (0+) or one year olds (1+), while some cutthroat trout could have been several years of age (> 2+). Three other species of fish were caught with regularity: the coastrange sculpin (Cottus aleuticus), prickly sculpin (Cottus asper), and three-spined stickleback (Gasterosteous aculeatus) (Table 1.2). Table 1.3 shows the P C structure coefficients. These are the correlations between the original habitat variables and the new PC variables, which are used to determine which variables are most important in the P C A . The results were similar for the first two axes of both P C A s . The first axis (PCI) described a gradient in stream width, water depth and dominate substrate type (Table 1.3; Figure 1.2). Stream width accounted for the majority of the variation of P C I for the cutthroat trout and coho salmon analyses (Table 1.3). The second axis (PC2) of both P C A s represent a gradient in substrate size (Table 1.3; Figure 1.2). There was no linear relationship between coho salmon density and any P C axis, however cutthroat trout density was negatively correlated to P C I (multiple regression; Table 1.4; Figure 1.3). There was no difference in coho salmon density (t005{2)26 = -0.797; P = 0.4327), or cutthroat trout density (F0.o5<2).i.i5 = 0.701; P = 0.4157) between regions (Figure 1.4) or among riparian classes (coho salmon sites: P0.05(2)2,25 -2.730; P = 0.0847; p = 0.48; cutthroat trout sites: F0.o5(2),2,29= 1.608; P = 0.2177; p = 0.30; Figure 1.5). In general however, coho salmon density tended to be highest in clear-cut sites compared to old growth (Pearson's Least Squared Difference (PLSD) P = 0.0505) and second growth sites ( P L S D P = 0.0503; Figure 1.5). 14 There were no significant differences in habitat variables between regions and among riparian status (Table 1.5 and 1.6). Figure 1.6 is the same figure as Figure 1.2, however the sites are arranged by riparian status. Clear-cut sites seem to clump together, however the sample size is quite low, as is the case for old growth sites. Second growth sites are the most variable in habitat type encompassing the whole range of PC axes. Table 1.1: Habitat characteristics of the study sites. Tributary = Trib.. Riparian status: OG: old growth (> 100 years since last harvest, or no harvest); SG: second growth (15 100 years since last harvest); CC: clear-cut (< 15 years since last harvest). Region Sample Stream Name No. Site Thalweg Average Reach Riparian Year Depth (m) Width (m) Length (m) Status SC 97 Angus 1 1 0.41 4.6 21.2 S G SC 98 Angus 1 2 0.45 4.18 45.3 SG SC 97 Burnett 2 3 0.47 4.57 16.6 SG SC 98 Coho 3 4 0.15 1.68 23.2 SG SC 98 Coho S. Arm 3 5 0.15 1.24 21.8 SG SC 97 Colvin 4 6 0.14 1.45 19.7 SG SC 97 Cook 5 7 0.12 1.1-7 30.1 SG SC 97 Husdon 6 8 0.27 2.61 237.2 SG SC 98 Husdon 6 9 0.30 2.43 34.3 SG SC 97 Lawlers 7 10 0.21 2.60 216.2 O G SC 97 Scrub 8 11 0.27 3.07 18.9 S G SC 97 Snake 9 12 0.24 4.05 25.3 S G SC 98 Snake 9 13 0.16 1.97 75.2 SG SC 97 Tzoonie 10 14 0.41 3.00 21.7 SG SC 97 Tzoonie Trib 11 15 0.18 3.15 30.4 SG SC 98 Wee 12 16 0.17 1.48 22.6 SG s c 97 Wilson 13 17 0.31 4.36 60.2 S G w c 98 Car Wash 14 18 0.34 2.22 32.1 SG w c 98 Chess 15 19 0.27 1.84 22.5 O G w c 98 Clear-cut 16 20 0.35 3.30 22.4 CC w c 98 Early 17 21 0.29 1.72 24.0 S G w c 98 Fundy 18 22 0.64 6.60 174.1 SG w c 98 Kootowis S. Trib 19 23 0.20 1.36 40.6 CC w c 98 Kootowis 20 24 0.20 1.40 50.3 CC w c 98 Kootowis N. Trib 21 25 0.32 1.35 61.5 CC w c 98 Kootowis Trib 22 26 0.31 1.54 35.7 CC w c 98 Log Boom 23 27 0.19 1.18 17.9 O G w c 97 Lost Shoe 24 28 1.07 8.28 180.6 S G w c 98 Oyster Jim 25 29 0.37 2.59 40.3 O G w c 98 Schooner 26 30 0.47 3.07 56.0 SG w c 98 Sewage 27 31 0.24 0.92 34.0 SG w c 98 Staghorn N. Trib 28 32 0.29 1.42 29.4 SG w c 98 Staghorn S. Trib 29 33 0.24 2.21 46.4 S G w c 98 Ankle 30 34 0.32 3.10 28.4 O G w c 98 Ginnard W. Trib 31 35 0.21 1.49 68.5 O G 15 Table 1.2: Density (fish/m2) of fish species at the different sites. Scientific nomenclature for each species is given in the Results section of the text. Stream Site Density (fish/m2) Cutthroat Coho Coastrange Sculpin Prickly Three-spined Other trout salmon sculpin spp. sculpin stickleback Angus 1 0.10 0.08 Angus 2 0.16 0.52 0.91 0.01 Burnett 3 0.32 0.15 Coho 4 0.85 0.58 0.74 Coho S. Arm 5 2.05 0.25 Colvin 6 0.69 0.66 0.20 0.03 Cook 7 0.79 0.03 Husdon 8 0.32 0.13 Husdon 9 0.56 0.27 Lawlers 10 0.08 0.09 Scrub 11 0.05 0.09 Snake 12 0.15 0.44 Snake 13 0.13 0.22 0.41 Tzoonie 14 0.14 Tzoonie Trib 15 0.33 Wee 16 1.60 0.21 0.03 Wilson 17 0.06 Car Wash 18 1.19 0.36 Chess 19 0.20 0.02 0.37 Clear-cut 20 0.35 1.35 0.07 0.09 Early 21 0.55 0.27 Fundy 22 < 0.01 0.02 0.04 Kootowis S. Trib 23 0.21 1.71 0.02 Kootowis 24 0.39 1.55 Kootowis N. 25 0.54 0.10 Trib Kootowis Trib 26 0.27 0.19 0.29 Log Boom 27 0.14 0.10 0.72 Lost Shoe 28 0.03 0.05 0.01 0.01 0.00 Oyster Jim 29 0.08 0.11 Schooner 30 0.17 0.20 Sewage 31 1.32 Staghorn N. Trib 32 1.30 0.12 0.02 Staghorn S. Trib 33 0.53 0.27 0.07 Ankle 36 0.13 0.01 0.17 0.02 Ginnard W. Trib 35 0.02 0.31 0.02 0.05 0.05 Total Number 820 846 266 264 12 42 10 (%) (36) (37) (12) (12) (0.5) (2) (0.5) 16 a o S 13 CO O o o co •a .S ra M 'co c§ i S 5 8. 3 O 1? C£ ra O co co c .22 1 s c g o o G fe 2 •c g q C ' C o I s .2 b V §1 ° b ex -a 3 x — O <L> <U P <D fc a o o ^ i § ^ • ' P2 -=1 C*H CO CO © ^ H { N » — I © © © © o o o o d d d d « B N h —i O O N N 'fl d d d d c D O ^ o ' ' i o 8 d d © d d d d d o o o o o o o o — OQ &5 t<) r~ >?> J 00 VI J if) h M IT) d d d d o ' o ' O O * * © < s c o d r r > v o > 7 i c r \ © d o d o © © © * * * * * ' * 1 1 u & J •§ ! ! s 5 g « u - fl J? ! l l l l ^ ^ ^ ^ ^ ^ a o c o d © C N 8 c4 I <o 3 a .§ 81 £ 5 S 8 S S © d d o d d d © CM © VO CM 8 © - £ $ a 2 $ £ © C M - * ' * © - * © © d d d d d d d d C 7 \ © t ~ - - ; v O O \ © - H © © © © © © © o d d m N n o N f o © o d d © © © > £ ( f ) 2 2 $ »0 r f © © © © o o © © 9 9 ° * * * © © © © © &\OO©^HVOCOCO^T ^ ^ ^ ^ ^ ^ 00 © <2s v o CO c4 VO CM c o 8 it Table 1.4: Mult iple regression relationships with the slope, F -values and associated P values presented for the cutthroat trout analysis. The asterisk indicates a significant relationship between the P C axis and cutthroat trout density (alpha = 0.00625). Variable Slope F- value P-value R 2 Model 1.79 0.138 0.42 P C I -0.655 10.99 0.004* P C 2 -0.050 0.02 0.881 PC3 0.368 0.98 0.335 P C 4 0.304 0.19 0.666 P C 5 -0.518 0.33 0.573 P C 6 -1.054 1.06 0.315 P C 7 1.388 0.77 0.390 P C 8 0.033 0.00 0.989 Table 1.5: Average (± SE) of steam characteristics by region (SC: Sunshine Coast; W C : West Coast of Vancouver Island). Unpaired t-test values and associated P - values are also presented. Significance level was set at P < 0.00625. S C W C t-value P-value Stream Width (m) 2.69 ± 0.34 2.55 ± 0.55 0.219 0.8284 Maximum Depth (m) 0.25 ± 0.03 0.35 ± 0.06 -1.423 0.1661 % L o g Cover 10 .10± 2.80 6.70 ± 2.20 0.959 0.3462 % Undercut Bank Cover 6 . 1 0 ± 2.40 4.30 ± 1.60 0.621 0.5396 % Fine Substrate 13.30 ± 4.60 21.60 ± 7 . 1 0 -0.984 0.3340 % Gravel Substrate 44.00 ± 5.10 28.50 ± 4.00 2.402 0.0234 % Cobble Substrate 15.20 ± 3.30 19.40 ± 5.50 -0.643 0.5254 % Boulder Substrate 11 .80± 4.80 9.90 ± 4.20 0.299 0.7671 Table 1.6: Average (± SE) of stream characteristics in clear-cut (CC) , old growth (OG) and second growth (SG) streams. Analysis of variance F -values and associated P -values are also presented. C C O G S G F-value P-value Stream Width (m) 1.79 ± 0.38 1.92 ± 0.60 2.99 ± 0.45 1.197 0.3195 M a x . Depth (m) 0.28 ± 0.03 0.24 ± 0.04 0.32 ± 0.05 0.229 0.7971 % L o g Cover 4.80 ± 0.70 7.30 ± 5.90 9.30 ± 2.50 0.421 0.6609 % Undercut Bank 7.60 ± 3.20 2.00 ± 2.00 5.40 ± 1.90 0.782 0.6232 % Fine Substrate 27.60 ± 7.00 5.00 ± 4.00 17.40 ± 5.80 0.937 0.4056 % Gravel Substrate 30.60 ± 7.30 33.00 ± 10.70 37.90 ± 4.60 0.333 0.7203 % Cobble Substrate 13.40 ± 11.20 26.70 ± 10.80 15.90 ± 3 . 1 0 0.682 0.5149 % Boulder Substrate 2.80 ± 2.80 18.30 ± 11.70 10.70 ± 4.00 0.874 0.4303 18 • Sunshine Coast O West Coast of Vancouver Island Fine t oo 0> OS 2 Gravel Cobble Fine Gravel Cobble 1 •8H .6 .4 1 r? 5 .2 H i o --.2 --.4 a) CO O Q SO - i — | — i — | — i — | — i — | — i — | — i — r — -.8 -.6 -.4 -.2 0 .2 .4 PC 1 (54.6 %) t .9 •6H •3H <s a O H 1 -.3 1 b) O CO o T ' 1 > 1 r " .6 .8 1 1.2 T 1 1 1 1 1 1 1 1 1— .8 -.6 -.4 -.2 0 .2 Narrow Shallow PCI (56.0%) Stream Morphology and Substrate i—>—i—1—i—>—i—<-4 .6 .8 1 1.2 • Wide Deep Boulder Figure 1.2: Principal component axis one (PCI) versus two (PC2) for (a) the cutthroat trout sites analysis, and (b) the coho salmon sites analysis. The same variables loaded for PCI in the two analyses. 19 •a c H S 1.2 -1 -.8 -.6 -.4 -.2 P = 0.0035 r = -0.655 n = 28 • n. * -.6 -.4 -.2 0 .2 .4 .6 Nar row S h a l l o w PC 1 (54.6 %) 1.2 W i d e D e e p B o u l d e r Figure 1.3: Relationship between cutthroat trout density and P C I of cutthroat trout analysis. Pearson correlation (r), associated P -value, and sample size (n) are indicated. Q *3 5 .45 .40 .35 .30 .25 .20 0 .15 Cutthroat Trout Sites Coho Salmon Sites Figure 1.4: Mean density (± 1 SE) of cutthroat trout and coho salmon within each region ( • Sunshine Coast;0 West Coast of Vancouver Island). 20 .8 -.7 -.6 " .5 -.4 -.3 -.2 -.1 -0 -I O Clear-cut • Old growth 3 Second growth Cutthroat Trout Sites Coho Salmon Sites Figure 1.5: Average density (± 1 SE) of fish within streams of different riparian status: clear cut < 15 years since last harvest; old growth > 100 years since last harvest, or no harvest; second growth 15 - 100 years since last harvest. F i n e t 00 G r a v e l C o b b l e • O Clear-cut \M • Old growth 3 Second growth 3 O « 3 CO • 3 3 3 -.6 -.4 Nar row S h a l l o w -.2 0 .2 .4 .6 PCI (54.6 %) Stream Morphology and Substrate 1 1.2 W i d e D e e p B o u l d e r Figure 1.6: The different riparian status types arranged in the cutthroat trout PCA graph. Riparian status types include: clear cut < 15 years since last harvest; old growth > 100 years since last harvest, or no harvest; second growth 15 - 100 years since last harvest. 21 Discussion Data from the synoptic surveys were useful in characterizing small coastal streams. From the two P C A s , the combination of stream width, water depth and substrate size were found to be important abiotic variables that describe small coastal streams within the two regions sampled (Figure 1.2). Of these variables, stream width was the most important variable, accounting for over 96% of the variation of the first P C axis of both analyses. Coho salmon density was not related to any of the eight P C axes. Murphy et al. (1989) found that coho salmon were present across a range of habitats in the Taku River, Alaska, but density was greatest in off-channel habitat where water velocity was < 30 cm/s. Off-channel habitat was not sampled in this study, and velocity was not included in the analyses because it was sampled infrequently. Incorporating off-channel habitat and velocity data into the correlative analyses may have yielded a relationship. A s well , it is known that coho salmon are naturally found in lakes (Swales et al. 1987), ponds (Peterson 1982) and large streams (Sandercock 1991) which were not sampled either. Bradford et al. (1997) compiled data from several studies that estimated coho salmon smolt abundance specifically within rivers. They were able to correlate smolt abundance with various habitat variables specific to the regions sampled, however they concluded that there are no good general predictive models to estimate coho salmon abundance because spatial and temporal variability among streams is high, making it difficult to find a general model. Another approach may be to include biotic factors, such as the number of spawning coho salmon, the density of other fish species and food availability. A s Chapman (1966) stated, the abiotic condition of the stream wi l l legislate the density within a stream while the biological or biotic factors within the stream wi l l govern or determine what density wi l l prevail. L o w adult ocean survival over the past 20 years (Slaney et al. 1996) has lead to reduced numbers of spawning coho salmon within streams which in turn may account for low juvenile density. Perhaps i f juvenile coho salmon densities within streams were at higher historical levels, stronger associations might have been found between density and 22 habitat. Competition between coho salmon and other salmonids (Hartman 1965; Taylor 1991), and trout (Bisson et al. 1988; Sabo and Pauly 1997) has been found to influence coho salmon habitat association within the stream, and it may influence differences in coho salmon density between habitat types or streams. Coho salmon fry density tended to be higher in clear-cut streams as Thedinga et al. (1989) has also found. The removal of the forest canopy increases the amount of light that penetrates the stream which can cause an increase in invertebrate production (Murphy and Hal l , 1981) which may influence fish density. Mason (1976) found that food rather than space limited coho salmon production during the summer. The advantages of clear-cut streams during the summer on coho salmon fry may be negated during the winter as coho salmon fry often migrate out of these streams usually due to their lack of suitable overwintering habitat (Tschaplinski and Hartman 1983). Stream width was the most important variable in predicting cutthroat trout density. M y results are consistent with other studies which have shown that cutthroat trout densities are in higher in narrow, small streams (Hartman and G i l l 1968; De Leeuw and Stuart 1981; Johnson et al. 1986; Swales et al. 1987; Bozek and Hubert 1991). Cutthroat trout are thought to have been forced into these small streams due to competitive interactions with other species, particularly coho salmon (Glova 1986; Bisson et al. 1988) and rainbow trout or steelhead (O. mykiss) (Hartman and G i l l 1968) when present. I also found that cutthroat trout densities were higher in streams which had a low percentage of boulder substrate. Streams that have few boulders are often dominated by smaller substrates such as gravel and fines. Others have found that cutthroat trout associate positively with gravel substrates (Heggenes et al. 1991b). There were no differences in cutthroat trout densities among the riparian status classes. Others have found that cutthroat trout have higher biomass or density in streams that have been most recently logged (Murphy et al. 1981; Connolly and Hal l 1999). From the P C A for cutthroat trout sites, recently clear-cut streams were fairly closely clumped in terms of similar habitat types, as were old growth sites, however the sample size for old growth sites was very low. Second growth sites on the other hand had the widest range of habitats, (ie. found at each axes extreme). The majority of cutthroat trout streams 23 presumably flow through second growth forests, so having a more specific classification (such as years since logging) may be a better method for predicting cutthroat trout density. There are several possible mechanisms to explain why cutthroat trout were had higher density in narrow, shallow, and smaller substrate streams. There is evidence that higher invertebrate production leads to higher growth rates (Moore and Gregory 1988a) or abundance (Murphy and Hal l 1981; Wilzbach and Hal l 1985) of cutthroat trout. High density cutthroat trout streams (narrow, shallow, smaller substrate) may have higher invertebrate abundance than less favourable streams, which leads to higher growth rate and density. There may be a limit however, as it is well understood that invertebrate biomass or abundance decreases with increases in fine substrates (Slaney et al. 1977; Lester et al. 1994). Therefore, it is unlikely that streams dominated by purely fine substrate and not a combination of gravel and fines can provide adequate surface area for invertebrate production to influence fish density. A less obvious mechanism to explain the observed differences in cutthroat trout density between stream types may be that movement or swimming activity of fish influences growth rate and density between stream types, assuming that food is limiting. Increasing average daily or seasonal activity limits the amount of overall energy a fish has for normal digestion and growth (Priede 1985). In less favourable habitats, such as ones with limited refuges, fish may move more often in search of suitable habitat. Therefore, fish that move more often may grow less or have less energy for reproduction which may lead to lower density. The negative correlation between cutthroat trout density and P C I show that cutthroat trout are more abundant in narrow, shallow streams, with fewer boulder substrates. The relationship between cutthroat trout density and P C I is clearly not linear. Perhaps a negative exponential relationship would be more appropriate to devise a predictive model, however that was not the intention of this paper. Here, finding general associations between habitat variables and densities was the goal because the results w i l l be used to select stream habitat types that reflect the general habitat conditions in which cutthroat trout are found. 24 Small-scale Study: Hypothesis and Site Selection I have presented two possible mechanisms which could answer the question of why cutthroat trout density is related to stream width, depth and substrate composition. First, food availability is higher in the preferred habitat which leads to higher growth rate. Direct testing of this hypothesis is needed. Second, the amount a fish moves in different stream types is inversely related to growth rate which directly influences density. Because there is virtually no research relating habitat type to movement and growth rate of cutthroat trout, I designed a small-scale comparative study based on the results of this large-scale study. The study was conducted during the winter because stream conditions are presumably most adverse (Cunjak 1988), even though the results presented here are from the summer. I w i l l compare cutthroat trout movement and growth rates between different stream types based on the P C and regression analysis results (Figure 1.2; Figure 1.3). In comparative ecological studies on fish populations, streams are often classified by their width (Hartman and G i l l 1968), gradient (De Leeuw and Stuart 1981), or a combination of both (Bozek and Hubert 1991). Stream width (inferred from watershed size) and gradient are probably used because data for these variables are easily obtainable from topographic maps and aerial photographs. From my results, substrate composition and stream depth along with stream width may be more important variables when sampling streams, especially for the south coast of British Columbia. A s a result I have classified my study streams for the small-scale study as: relatively wide, deep, with coarse substrate (low density); relatively narrow, shallow, with fine substrate (high density) and; relatively narrow, shallow, with coarse substrate (intermediate density). Among-stream trends in abiotic conditions that were identified from the large-scale study were used to select sites for my small-scale study (Chapter Two). I am confident that the results from the small-scale study wi l l have less bias and present a more realistic portrayal of the factors influencing coastal cutthroat trout density in small coastal streams. 25 C H A P T E R T W O Introduction Coastal cutthroat trout inhabit aquatic systems along the northwest coast of North America (Figure 2.1). Although cutthroat trout are widespread, information on population abundance and life history trends is unknown for approximately 80% of the populations within B . C . (Slaney et al. 1996). The information that is available suggests a major decline i n abundance within and among populations throughout their range over the past 20 years (Trotter 1989; Slaney et al. 1996). Habitat loss due to stream side logging is thought to be one cause for the observed range-wide decline in cutthroat trout population abundance (Trotter 1989; Slaney et al. 1996). Coastal cutthroat trout typically live in small (< 5 m wide), headwater and tributary streams (see Chapter One; Hartman and G i l l 1968; Swales et al. 1987; Bozek and Hubert 1991), which under the current forestry guidelines may not get adequate protection from streamside clear-cut logging. The effects of forestry practices on streams habitat are well studied in coastal systems (Slaney et al. 1977; Murphy and Kosk i 1989; Lamberti et al. 1991; Ralph et al. 1994; Hartman et al. 1996), but less is known for interior systems (Platts et al. 1989). In general, logging to the stream bank eliminates the riparian habitat surrounding and hanging over the stream (ie. canopy), reduces the usefulness of L W D within the stream, and increases the chance of added sediment load into the stream via erosion and land slides (see General Introduction). I found in Chapter One that cutthroat trout density was negatively correlated with a gradient in stream habitat. The focus of this chapter is to further the investigation into the mechanism(s) behind this correlation. I compared the movement, growth and habitat association of cutthroat trout in three streams, each of which has been altered by logging (second growth streams), but to different degrees as each stream fits the habitat and density criteria outlined in Chapter One. From now on I w i l l refer to streams that had some level of habitat degradation to be in an "altered" state, ranging from "heavily-altered" to "moderately-altered" to "minimally-altered". 26 a) 145" 14IP J-Alaska <% 'S 1 \ Yukon • N PACIFIC OCEAN British . Columbia ( m anada VYashihgtoni / \ . Oregon ; ^ Strait of Georgia km 5 10 Vancouver 123° 45' 123° 30' 123° 15' h 49° 30' Figure 2.1: The shaded area in (a) represents the range of coastal cutthroat trout (adapted from Scott and Crossman 1985). The black square identifies the study area which is shown in more detail in (b). The arrows in (b) indicate the location of the study sites on each stream. 27 This study was conducted during the winter as there is a lack of information about the general ecology of cutthroat trout during this season. In general, winter stream conditions are thought to be harsher than summer because water temperatures are lower, discharge is higher, and fluctuation in water levels greater. Cutthroat trout are thought to change their habitat use from summer to winter, however this shift has not been adequately documented and needs further investigation. As well , there are conflicting views in the literature about how short-term flow changes (floods) affect fish in streams (Hanson and Waters 1974; Harvey 1987; Heggenes 1988; Todd and Rabeni 1989; Harvey et al. 1999). The main objective of this study is to examine how cutthroat trout respond, as measured by movement, growth and habitat association, to changes in seasonal and short-term (ie. flood event) flow conditions in streams of differing physical characteristics. I measured movement, growth and habitat association of cutthroat trout in the three stream types from August 1998 to A pr i l 1999. The specific objectives were to (1) contrast the rates and directions of cutthroat trout movement among the stream types, during seasonal and short-term changes to stream discharge, (2) contrast seasonal growth rates among the stream types, and (3) contrast seasonal habitat associations for cutthroat trout among the stream types. I made three predictions based on the assumption that, during the fall, stream discharge (water level) would be relatively low and constant and, during the winter, stream discharge would be relatively high and variable (Figure 2.2). The first prediction was that during the fall, movement rates within the streams would be similar among stream types whereas during the winter movement would be greater in the heavily-altered stream compared to the minimally-altered stream and intermediate in the moderately-altered stream. The second prediction was that during the fall , growth rates would be equal among streams but that during the winter growth rates would be least in the heavily-altered stream, intermediate in the moderately-altered stream and highest in the minimally-altered stream (the ability to acquire food has been reduced assuming that fish need to spend their energy avoiding high flows as the habitat has been altered) The final prediction was that there would be no difference in seasonal habitat association by cutthroat trout among the streams types. 28 F A L L s tab le d i s c h a r g e l o w w a t e r l e v e l Minimally-Altered Moderately-Altered Heavily-Altered = m o v e m e n t = g r o w t h = a s s o c i a t i o n = m o v e m e n t = g r o w t h = a s s o c i a t i o n = m o v e m e n t = g r o w t h = a s s o c i a t i o n W I N T E R v a r i a b l e d i s c h a r g e h i g h w a t e r l e v e l Minimally-Altered T|T m o v e m e n t • g r o w t h = a s s o c i a t i o n Moderately-Altered ^ J r m o v e m e n t g r o w t h = a s s o c i a t i o n Heavily-Altered ^ m o v e m e n t ^ g r o w t h = a s s o c i a t i o n Figure 2.2: P r e d i c t i o n s f o r m o v e m e n t ra te , g r o w t h ra te a n d h a b i t a t a s s o c i a t i o n d u r i n g the t w o s e a s o n s . " = " m e a n s n o d i f f e r e n c e a m o n g s t r e a m t y p e s ; " m e a n s a n i n c r e a s e o r g r e a t e r t h a n the o t h e r s ; " m e a n s a d e c r e a s e o r l ess t h a n the o t h e r s ; a n d "f^  " m e a n s i n t e r m e d i a t e t o t he o t h e r s . D i s c h a r g e a n d w a t e r l e v e l s s ta ted f o r e a c h s e a s o n a r e a s s u m p t i o n s ( s e e T a b l e 7 f o r m o r e d e t a i l ) . 29 Study Sites The study was conducted on the Sunshine Coast between Port Melon and Sechelt, B . C . (Figure 2.1). Three streams were selected on the basis of their average width and dominant substrate class (see Chapter One for rationale): relatively narrow, shallow with fine substrate (minimally-altered); relatively wide, deep with coarse substrate (heavily-altered); and narrow, intermediate depth and coarse substrate (moderately-altered; Table 2.1). Table 2.1: General morphological characteristics of the three study streams. Slope, bankfull width, boulder/cobble substrate, and sand/gravel substrate are averages (per channel unit) over the e ^ Avalon Husdon Pipe Heavily-Altered Minimally-Altered Moderately-Altered Stream Order 2 2 1 Study Reach Length (m) 220 180 120 Slope (%) 4.22 ± 0 . 3 1 , 173 1.76 ± 0.14,251 3.47 ± 0 . 4 9 , 6 5 Bankfull Width (m) 13.08 ± 0.37, 182 4.16 ± 0.11, 302 5.28 ± 0.16, 69 Boulder/cobble (%) 63.48 ± 0.91, 178 9.54 ± 0.44, 268 45.31 ± 1.46, 65 Sand/Gravel (%) 34.70 ± 0.62, 178 77.58 ± 0.86, 268 49.62 ± 1.04, 65 The minimally-altered stream, Husdon Creek, is a tributary to Wilson Creek which flows into the Strait of Georgia (Figure 2.1). The study site is approximately 2.5 km from the Wilson Creek estuary and was about 180 m long (Figure 2.3). The area surrounding the study site is second growth forest (81 - 100 years old), dominated by western hemlock (Tsuga heterophylla), western red cedar (Thuja plicata) and red alder (Alnus rubra). A deactivated logging road and bridge run through the middle of the study site (Figure 2.3). The headwaters of Husdon Creek contain fairly low gradient small tributary streams which run through predominately young (< 20 year old) lodgepole pine (Pinus contorta) and Douglas fir (Pseudotsuga menziesii), and western red cedar forests. In 1983, a 17.3 ha. area was clear-cut logged and then burned two kilometers upstream of the study sight. The sight now has a stand of < 20 year old western red cedar and western hemlock with some spruce (Picea spp.), red alder and balsam poplar (Poplus balsamifera). In 1988 a 9.7 ha. area was clear-cut logged and subsequently burned in 1989 approximately one kilometer upstream of the study site. There has been no evidence 30 of major post-logging environmental changes to the lower sections of Husdon Creek, however there may be a small increase in fine substrate accumulation within the stream since logging occurred (Grant M c B a i n , Department of Fisheries and Oceans, Sunshine Coast, B . C . , P .O. Box 10, Madera Park, B . C . , V O N 2H0, personal communication). The heavily-altered stream, Avalon Creek, flows directly into Howe Sound (Figure 2.1). The study site is appoximately 400 m from the estuary and is approximately 220 m in length (Figure 2.3). The area surrounding the study section is second growth (41 - 60 years old) forest consisting of western hemlock and bigleaf maple (Acer macrophyllum). There are two distinct sections of Avalon Creek: the upper, steep headwater section that is dry for most of the year, and the lower fish bearing section. The transition between these two sections occurs at a culvert under a highway. The headwaters of Ava lon Creek originate in a large, high gradient gully. Within a six month period between the end of 1984 and mid-1985, three areas within the headwater section of Avalon Creek were clear-cut. A total of 107.1 ha of forest was clear-cut in the upper headwater section, while two clear-cuts of 86.1 ha and 20.1 ha each were logged lower down. Since 1984, there have been several landslides and debris flows which have originated within the gully (Ian Gazeley, local resident, Sunshine Coast, B . C . , P.O. Box 1318, Gibsons, B . C . , V O N 1V0, personal communication). Some debris flows (including one in November of 1998) have washed over the highway and into the downstream, fish bearing reach (study site), causing extensive widening and accumulation of coarse substrate. I selected an unnamed tributary to Avalon Creek to represent the moderately-altered study stream. Pipe Creek was initially a logging skidder trail, but over time it has eroded and now maintains a low level of water throughout the year. Coho salmon (O. kisutch) and cutthroat trout persist there throughout the year. Pipe Creek is less than 500 m in length when fully wet. The study site was approximately 120 m in length (Figure 2.3). The stream is fed mostly from a roadside ditch and from a small groundwater feeder tributary. A t low summer flows, there is no surface water connection between Ava lon and Pipe Creeks. Red alder and bigleaf maple are the predominant tree species surrounding the stream. 31 -220 m -200 m 180 m r—160 m 140 m r—120 m I—100 m -80 m -60 m —40 m -20 m 0 m 6 m 12 m •0m l—I—I—I—I Avalon Creek L e g e n d — • L W D - o r i g i n a l i p / I P I s o l a t e d P o o l = L W D - n e w s p / S P S i d e P o o l L W D - l o s t p / P P o o l = • U C B a n k p p / P P P l u n g e P o o l • R o o t - w a d g / G G l i d e • R o o t - w a d - n e w r / R R i f f l e • B o u l d e r n / N R u n O T r e e c / C C a s c a d e Q- M a r k e d L W D t / T S t e p 9 D r y B a r d / D D r y © P o s t capital leters indicate final conditions * M e t e r - s t i c k smal ieacrs indicate initinl conditions p/N p/N survey area C \ "a ti £ W S P i t F/^ wre 2J: Morphology and the location of physical structures and channel units within the three study sites. Shown are the average bankfull widths. Wetted width was only at bankfull width during high flows. Bankful width, channel unit type and location of physical stream changed during the study. These diagrams attempt to show the average morphology and change of physical structures and channel units over time. 3 2 Methods Data collection began on August 9, 1998 and ended on March 28, 1999, with extra collection during the weeks of Apr i l 11, June 21, and August 25, 1999.1 conducted monthly and some daily sampling of flow conditions and stream habitat to monitor the changing characteristics of the stream over the seasons. Movement of cutthroat trout was assessed using two methods: mark-recapture using minnow trapping of fish implanted with passive induction transmitter tages (known as PIT tags), and radiotelemetry. The two methods of observation were used in order to assess movement rates over the full size range of cutthroat trout within these streams. Movement rates were calculated using all data, however growth rate, population density and survival estimates were calculated using the PIT tagged data only. Stream Habitat Surveys Initial Set-up and Daily Sampling Seasons defined by water temperature changes rather than by calendar dates are more realistic when dealing with fish (Cunjak and Power 1986). Thus I defined seasons by significant transitions in water temperature. I measured air and water temperature on days when present at each stream using a hand held thermometer. I determined the seasons after all the data were collected. Fall began in September, 1998, when stream water temperatures began to decrease from the steady high summer temperatures. Fal l ended at the end of December, 1998, once the water temperature started to maintain a constant low level. During the winter water temperatures were constant and low. This season began in early January, 1999 and ended in mid A p r i l , 1999 (when water temperatures began to increase again). I only took one temperature reading in June 1999 and assumed that this period was spring as the temperature was still relatively cool (range: 6.0°C to 12°C; A p r i l , 1999 to June 1999). I took one temperature reading in August 1999 and since the temperature was quite warm (> 10.0 °C) I assumed that was summer (June, 1999 to August, 1999; Table 2.2). 33 T o measure daily water depth I stationed metal metre sticks (0 cm = ground level) within the study site in Husdon and Pipe Creeks (Figure 2.3) and downstream of the study site in Avalon Creek. Daily rainfall was collected in 100 ml graduated cylinders placed near each stream in areas free of overhanging vegetation or trees. Stream discharge was estimated by calculating the channel rating curve (R.W. Newbury, Consulting Engineer, Newbury Hydraulics, P.O. Box 1173, Gibsons, B . C . , V O N 1 VO, personal communication) for each stream. T o predict stream discharge, I carried out repeated water velocity transects at the stationary metre sticks using a March-McBi rney model 2000 FlowMate velocity meter. Water depth was plotted against the discharge calculated from the velocity transect so that stream discharge could be inferred from the depth on the metre sticks (Gore 1996; see Appendix 1). Table 2.2: Mean water temperature and air temperature (°C; ± SE , n) of fall and winter in each stream. Temperatures for spring and summer are not averages, but one time measurements (n = 1). (Note: summer may have commenced earlier in Husdon Creek than the others as stream temperature was at 12°C in June.) Stream Fall Winter Spring Summer Water Temperature Avalon 8.91 ± 0.34,38 4.38 ± 0.06, 54 7.0 11.5 Husdon 8.97 ± 0.26, 45 5.86 ± 0 . 1 0 , 4 7 12.0 13.5 Pipe 8.19 ± 0 . 2 6 , 22 5.10 ± 0 . 0 5 , 5 2 6.0 10.0 Air Temperature Avalon 9.35 ± 0 . 5 8 , 3 9 5.01 ± 0.38, 54 12.5 -Husdon 9.08 ± 0.59,46 5.63 ± 0.29, 48 16.0 -Pipe 7.32 ± 0.55, 23 5.31 ± 0 . 4 2 , 52 12.5 -In Ava lon and Husdon Creeks, I placed numbered wooden posts every 15 m along the stream bank for the length of the study site as reference points of linear stream distance (Figure 2.3). To quantify the stability of the physical structures within the streams, I randomly selected and individually marked 16 pieces of L W D within Avalon and Husdon Creeks (Figure 2.3). The distance from the nearest downstream wooden post to the L W D was recorded at the beginning of the study. A t different times throughout the study I recorded the position of each of these pieces of L W D . 34 Monthly Sampling I conducted habitat surveys (procedure modified from the Methods for Stream Habitat surveys, Oregon Department of Fish and Wildlife, 1997) approximately every 28 days to characterize the within-stream physical habitat of each stream. This sampling schedule allowed me to detect temporal changes in the morphology and physical structure of the streams thereby better enabling me to determine the available habitat for fish (Herger et al. 1996; Giberson and Caissie 1998; Hilderbrand et al. 1999). During each habitat survey, I estimated the length, average wetted and bankfull width, average and maximum depth, percent substrate, percent physical cover, percent canopy cover, slope and the number of trees and percentage of vegetation in the designated riparian area of each channel unit within the entire study site. Widths and lengths of channel units were measured using a 50 m measuring tape. Depths were estimated using a measuring stick. Substrate was categorized according to the Methods of Stream Habitat surveys (Oregon Department of Fish and Wildlife, 1997) as organic (sticks, leaves, fine organic detritus etc.), silt (< 0.0625 mm), sand (1.0 -0.0625 mm), small gravel (2.0 - 4.0 mm), large gravel (4.0 - 64 mm), cobble (64 - 256 mm), boulder (> 256 mm), bedrock (continuous with bank). Physical cover included L W D (> 15 cm diameter), small woody debris (SWD: < 15 cm diameter, or upturned shrub roots), undercut banks (< 20 cm deep), root-wads (upturned or imbedded tree roots), vegetation (submerged to ^ 1.0 m above water) and boulders. A l l physical cover types excluding undercut banks were determined by visually estimating the percentage of the channel unit area covered by each structure. The remainder percentage of the channel unit was considered as open water. Percent undercut bank was determined by estimating how much of the total length of the channel unit bank was undercut (eg. the complete length of the left bank would equal 50%). Canopy cover was estimated by standing in the center of each channel unit and looking up to estimate how much of the sky was blocked out by trees branches and leaves (ie. 75% canopy cover equals 25% open sky). Slope was estimated using a hand held Sunnto clinometer. The designated riparian area was defined as the area covering 10 m x the length of the channel unit. Woody (shrubs) and non-woody (grasses and ferns) plants were estimated as a percentage of the riparian area 35 ground cover. A s well I counted the number of hardwood and softwood trees within the riparian area. After a storm event (defined as > 10 cm/day change in water depth) or when there was not sufficient time to complete a full habitat survey, 1 would conduct a "partial" habitat survey. The partial surveys were done within a 30 m permanently marked section of the study site (Figure 2.3). Usually only length, average wetted and bankfull width, and average and thalweg depth were measured for each channel unit during these surveys. Fish Trapping and PIT Tagging I used minnow traps, baited with salmon roe or wet cat food in perforated canisters, to capture individual fish for purposes of implanting PIT tags or transmitters and measuring movement rates of recaptured PIT tagged fish. The traps were effective in capturing fish between 30 and 150 mm in length. I placed the traps < 5 m apart in a variety of channel units and cover types using a standardized random pattern, however low water levels in Avalon Creek during August, September and October limited the placement of traps to < 10 m apart (Figure 2.4). I marked the location of each trap with flagging tape so that the distance between each adjacent trap would be constant. Over the course of the study, as water levels rose and as the stream channels changed in shape and location, new traps were added to fi l l in areas where large gaps occurred. The number of traps set between trapping sessions was seldom consistent because traps could not be set in areas that were too shallow or fast flowing. Overall I successfully marked a total of 65 ,47 and 24 traps in Husdon, Avalon and Pipe Creeks respectively (Figure 2.4). Ten additional traps were flagged within an area 25 to 50 m upstream and downstream of the study sites. These traps were set once a month to catch fish that were potentially emigrating from or immigrating into the study site. Minnow traps were set every two weeks (termed a trapping session) during the 34 week study, plus three additional trapping sessions in A p r i l , June and August of 1999. Trapping began in mid to late August in Husdon and Avalon Creeks and mid October in Pipe Creek. Traps were fished for at least a 4 hours set during each sampling session in 36 Avalon Figure 2.4: Placement of minnow traps within the three study sites. Diagrams include all traps that were flagged during the study. Not all traps were set during a single trapping session. "Addit ional" traps are not shown, as they were set between 25 and 50 m downstream and upstream of the study sites (see Methods: Fish Trapping and PIT Tagging). 37 the fall, spring and summer when water temperatures were > 6°C and overnight during the winter when water temperatures were < 6°C. A l l fish caught were measured for length to the nearest millimeter (fork length) and wet weight to the nearest 100 t h of a gram (0.01 g). Each captured fish received a unique fin clip depending on i f it was caught within the study reach or the extra upstream or downstream areas (anal and caudal clips only). A l l fish greater than 65 mm long were implanted with a PIT tag, which was inserted into the abdominal cavity posterior to the pectoral girdle (Prentice et al. 1990). Immediately following a trapping session, I measured habitat characteristics for each trap location, including the type of channel unit, water depth, distance to nearest bank, water velocity (at 40% depth), dominate substrate class and physical cover type and combinations of these. During some weeks it was impossible to record all the habitat information due to hazardous water conditions or short day length. Radiotelemetry I captured fish for implantation of transmitters primarily during the trapping sessions, however the first five fish were caught using an electroshocker (Smith-Root Model 12A). I used radiotelemetry to track the movements of 18 fish (Table 2.3). Because of the size and weight of the transmitters, only relatively large fish exceeding 140 mm in length and 35 g in weight (average length = 175 mm, range = 147 - 263; average weight = 62.69 g, range = 35.37 - > 200 g) could be implanted (Table 2.3). Surgical implantation of transmitters into the abdominal cavity is thought to have little affect on the natural behaviour and physiology of the fish (Martin et al. 1995) as long as the transmitters do not exceed 4% of their body weight (Adams et al. 1998a, Adams et al. 1998b). I used either Holohi l B D - 2 G transmitters (1.75 g, 18x9x6 mm body with a 19 cm antennae, battery life of 4 months) or Holohil B D - 2 H transmitters (1.35 g, 18x8x7 mm with an internal antennae, battery life of 2 months), depending on the weight of the fish (smaller fish received the lighter transmitter). Each transmitter was programmed with a unique frequency thereby identifying each individual fish. I anaesthetized fish in 11 of water with approximately a half a tablespoon of sodium bicarbonate. Once the ventilation rate of the fish reached approximately one operculum 38 movement per two seconds, I placed it on its dorsal side in a trough lined with neoprene which could be adjusted to match the width of each fish. Water was manually poured over the mouth and gills during the surgery. A 10 mm long incision was made off the midline of the ventral surface mid way between the pelvic and pectoral girdles. The transmitter body and external antenna (if present) were placed within the abdominal cavity. Three sutures were used to close the incision, then the wound was cleaned with an antibacterial agent and sealed with a veterinarian glue. The surgical procedure took 3 to 5 minutes to complete. Fish were then placed into a 101 recovery bucket filled with fresh water. Before returning a fish back to the capture site, it was left in a recovery bucket until it displayed an upright position and was actively swimming (usually took 3 0 - 6 0 minutes). Fish were tracked using a Lotek Suretrack STR1000 radio receiver and H-antenna on at least three days per week (each day is termed an observation day). I located fish at least twice within each observation day on an hourly basis. Within a week, a minimum of six observations were made per fish. Fish were tracked at different times during the day: M : morning, dawn ti l l noon; A : afternoon, noon till dusk; N : night, dusk till midnight. I attempted to vary the time of day I was at a stream so that observations could be made at each time period. Less observations were made during the night period due to safety reasons. During a storm event I tried to locate all radiotagged fish at least once; however, this was not always possible due to hazardous water conditions. Fish were located when the signal on the radio receiver was strong and fairly constant. I tested the accuracy of the receiver by locating a dead fish implanted with a transmitter and tied to a stick in the water column. I found that during normal flow conditions fish position was estimated within 1 m of the actual location, whereas during high flow conditions (which caused high turbidity) fish position could be assertained within 5 m. When I tracked live fish, each new location was marked with flagging tape and given a unique code. I also recorded the activity level of fish during each observation. Act ivi ty level could be estimated by pointing and holding the antenna in the direction of the fish and then recording the change in signal strength every three seconds for one minute (20 measurements). Large changes in signal strength were interpreted as 39 greater fish activity, whereas little or no change in signal strength meant less fish activity (Cook and Bergersen 1988, Clapp et al. 1990). To insure that large changes in signal strength corresponded with changes in fish activity, I implanted a transmitter into a dead fish, tied it to a stick, submerged it under water and recorded the change in signal strength as I held the fish stil l , twisted it in a side to side motion and moved it in a back and forth motion between 30 - 60 cm, imitating fish holding and feeding behaviours (see Appendix 2). Table 2.3: Details of radiotagged fish used during the study. Cristy was implanted with a transmitter on Sept 9, 1998 but was not found within the study area until Oct 22, 1998 for unknown reasons. Cristy traveled between Avalon and Pipe Creeks and is treated independently for each stream (as 2 fish). The possible reasons why radiotracking stopped for each fish are: known predation, battery (death or faulty), flood ( l - in-40 year flood event), or unknown (flood, battery or predation). Asterisks indicate fish that were excluded from any analyses. Fish Stream Length (mm) Weight (g) % o f body wt by tag Start Stop Reason Helen* Avalon 169 72.67 2.4 Sept 9 Sept 9 predation Cristy Avalon/ 186 76.67 2.3 (Sept 9) Jan 20 battery Pipe Oct 22 J im Avalon 165 52.72 3.3 Sept 9 Nov 10 flood Little B i l l y Ava lon 147 35.37 3.5 Oct 21 Dec 16 battery Darell Avalon 168 52.50 3.3 Oct 21 Dec 16 battery Sanjie Ava lon 191 84.06 2.1 Oct 21 Feb 25 battery Bernie Husdon 178 74.61 2.3 Sept 10 Jan 28 battery Glosh Husdon 162 60.60 2.9 Oct 21 Feb 25 battery Chester Husdon 172 56.43 2.7 Oct 30 Dec 16 battery Clark Husdon 142 37.00 3.4 Oct 30 Dec 16 battery Marina Husdon 142 39.35 3.2 Nov 12 Feb 4 battery Jake Husdon 159 48.09 3.6 Dec 9 A p r 2 battery Harry Husdon 193 91.03 1.9 Dec 11 A p r 15 battery Lujay Husdon 170 55.44 3.2 Jan 7 A p r 15 battery Pisces Pipe 171 56.64 3.1 Oct 22 Nov 12 flood L o u L o u Pipe 153 39.29 3.2 Oct 29 Dec 16 battery Chance* Pipe 1 7 5 - 60.29 2.9 Nov 12 Nov 12 flood/ unknown Prince Pipe 167 59.47 2.9 Dec 10 A p r 15 battery Calum* Pipe 172 60.01 2.9 Jan 22 Jan 22 unknown Mar io Pipe 263 - - Feb 25 A p r 15 battery 40 Once a radiotagged fish was located during an observation day, I recorded the type of channel unit, physical cover type, depth of the water and distance to the nearest bank for each fish's location. Data Analysis A l l data used in statistical analyses were first visually inspected for normality using box-plots. A n y variables found to be highly non-normal were transformed using either the natural or base 10 logarithmic transformation. I primarily used parametric 2-way analysis of variance ( A N O V A ) tests with season (fall, winter, and spring) and stream (Husdon, Ava lon and Pipe) as the two main factors. Other factors, such as fish size class and flood/non-flood periods, were included in the analyses when applicable. In some cases where sample size was low or data were not available, spring and/or Pipe Creek data were excluded from the analysis. In all analyses, each fish was considered as the unit of replication (n) within a treatment (stream, season, size class or flood/non-flood period). I averaged the individual observations per fish (both radiotagged and PIT tagged fish) to obtain one value per fish and these were then used to calculate the factor mean and standard error. A l l tests, unless otherwise noted, were considered significant to an a = 0.05 which is commonly used as the level at which to reject the null hypothesis. For most tests, a power value (P) was included along with the calculated test statistic and associated P - value. Power, is the probability that the null hypothesis was rejected when it in fact it should have been rejected. Power can be used to infer the biological significance of a test when statistical significance is not present (P > 0.05; Thomas and Juanes 1996). I considered a power level of ^ 0.80 to represent a reliable test (Thomas and Juanes 1996). Stream Habitat Surveys I used descriptive statistics to summarize the abiotic "treatment" conditions within the three study streams. Individual channel units were assigned to ten types, ranging from a dry channel unit (containing no water - Code 0), to slow flowing channel units (Codes 1 - 4), to fast flowing channel units (Codes 5 - 8), to a braid or secondary water channel 41 within the main stream banks (Code 9; Table 2.4). Means and/or percentages per channel unit were calculated for each habitat variable studied during the monthly habitat surveys (see Methods: Monthly Sampling). Fish Trapping and PIT Tagging Population abundance (± 95% confidence limits) and survival estimates were calculated for each site using the Jolly-Seber method (see Krebs 1989). Site density (fish/m 2) was calculated as the abundance divided by the area of the site. Based on length-frequency distributions of all PIT tagged fish, I lumped the fish into three main size classes that were comparable among streams (Figure 2.5) and easily interpretable. I considered a fish to have moved i f it was caught in a trap different from that of a previous trapping session. Movements downstream from the previous trap location were considered negative distances and upstream movements were positive distances. Movement rate was defined as the distance in meters between the two traps (including the direction), divided by the number of days between trapping sessions. These between trap movements are highly conservative (compared to radiotelemetry measurements) as they actually represent a change in location over a 14 day time span, rather than a constant daily movement rate. Many movements made by the PIT tagged fish went undetected, however I felt that the time between trapping sessions was long enough to allow fish to return to normal behaviour (after the intrusion of trapping), but short enough to detect short-term movements made by these fish. Movement rate was averaged per fish per season and tested among the seasons and streams using A N O V A . T o test for differences in short-term movement rate, I compared the movement rate during non-flood periods to flood periods. Movement rate during non-flood periods was calculated as the average movement rate of individual fish between trapping sessions before, after or between flood events. Movement rate during a flood was the movement between the trapping session immediately before a flood to immediately after a flood. Movement pattern of each individual fish is the net distance and direction of movement over time relative to the "initial capture location". "Initial capture location" is the location within the stream in which a fish was first captured (trapped) (considered 42 metre "0"). Movements upstream from the initial capture location were positive values and movements downstream were negative values. The last location in which each fish was recaptured was referred to as the "final recapture location". To test i f fish were moving randomly within the streams I created three computer generated populations (one per actual population) that had the same trapping pressure and movement rate as each of the actual populations, however the computer generated populations were able to move randomly. Each computer generated population consisted of 700 fish that could either move 0 m or move upstream or downstream the same average distance between the 19 trapping sessions as the real fish did in each of the three study streams (36.8 m, 10.4 m, 11.2 m for Avalon, Husdon and Pipe Creeks respectively). I compared the distribution of the final recapture location of the computer generated population to that of its respective real population using the Kolmogorov-Smirnov test. Table 2.4: Average depth and flow rate (± SE , n) of channel units types (code) used in analyses. Average depth and flow rate (m/s) were estimated from individual channel units measured during the trapping sessions. One measurement was taken per channel unit per trapping session. Some channel units were missed during any one trapping session because of lack of time or dangerous water conditions. Negative flows occur when there are eddies or circular flows caused from obstructions in the water. Coefficient of variation ( C V ) is calculated for the mean flow rate and shows the relative fluctuation in mean flow rate among the channel unit types (eg. side pool flow rate varies four times as much as in glides). Channel Unit Type Code Mean Depth (cm) R o w Rate (m/s) C V Dry Slow 0 Backwater/isolated pool 1 37.96 ± 1 . 6 6 , 77 -0.013 ± 0.008, 69 -5.297 Side pool 2 34.09 ± 0 . 8 0 , 3 5 2 0.009 ± 0.002, 285 4.056 Pool 3 34.10 ± 0 . 7 1 , 3 9 8 0.019 ± 0.002,333 1.803 Plunge pool/cascade pool 4 41.18 ± 1.54, 198 0.022 ± 0.008, 175 4.828 Fast Glide 5 31.12 ± 0 . 6 4 , 4 1 3 0.088 ± 0.005, 354 1.048 Run 6 27.85 ± 0.54, 515 0.112 ± 0.006, 477 1.062 Riffle 7 22.21 ± 0.60, 260 0.086 ± 0.009, 182 1.461 Cascade 8 27.52 ± 0.79, 136 0.038 ± 0.008, 107 2.284 Side channel 9 25.33 ± 1.50, 136 0.025 ± 0 . 0 1 3 , 27 2.802 43 a 3 e U 20 n 18 H 16 14 " 12 -10 -8 " 6 4 2 i 0 Husdon n . . . 60 80 100 120 140 160 180 9 " 8 " 7 " 6 " 5 " 4 -3 " 2 " 1 -0 -Avalon ft hrm 60 80 100 120 140 160 180 10 -9 " 8 " 7 -6 " 5 " 4 -3 -2 " 1 " Pipe 60 80 100 120 140 160 S i z e I C l a s s 65 - 99 mm 100- 119 mm > 120 mm 180 Length (mm) Figure 2.5: Histograms for size distribution of cutthroat trout from the streams. Three size classes that are present in all three streams were chosen: 65 - 99 mm; 100 - 119 mm and; > 120 mm. There is a fourth size class, < 65 mm, that does not show up in these data but were captured throughout the study. 44 Radiotelemetry Most fish were observed during two seasons, therefore their observations were split into fall and winter and the particular fish was considered as two different individual fish for between season comparisons (this explains why some tests have n values that are greater than the actual number of fish that were implanted with a transmitter). I calculated movement of radiotagged fish using two methods: net within-day movement and among day movement rate. Net within-day movement was the distance gained or lost within an observation day. Net within-day movement during the three different time periods of the day were compared using A N O V A with seasons, streams and time of day (morning, afternoon and night) as the factors. Among day movement or movement rate was the distance moved from the last relocation of an observation day to the first relocation of the next observation day divided by the number of days between the observations. Differences in movement rate were assessed using A N O V A with streams and seasons as the factors. T o estimate home range size I modified the "center of activity" equation from Hayne (1949): N = r * 2 * J t * w / d 2 , where N is the average number of sites a fish was relocated, r is the distance between the most frequently used relocation site to other relocation sites (+ 0.1 m), w is the width of the stream and d is the distance between possible relocation sites. I chose d to equal 5 m, a distance of stream in which I could have possibly located a fish. A n example of the calculation of home range is shown in Appendix 3 or see Hayne (1949). I calculated home range for 100% and 95% of the relocations. Calculating home range of the innermost 95% of relocations is a commonly used method as it eliminates possible outliner points such as one time excursions (Snedden et al. 1999). Estimation of home range was limited to fish that had greater than 20 individual observations (Snedden et al. 1999). I used a paired t-test to determine i f there was a difference between the 95% and 100% home range estimates of fish in Husdon Creek and another test for Avalon and Pipe Creeks to see i f fish were moving outside of their home range. I pooled Avalon and Pipe Creeks to increase the sample size. I calculated, for each fish separately, home ranges for fall and winter seasons. I used 45 A N O V A to contrast home range size between seasons and streams. If there were < 20 observations per fish for a season, it was eliminated from the analysis. Movement rate during non-flood and flood periods was compared using A N O V A with stream and timing of flood as the factors. Growth Rate Individual sizes and growth rates were only determined for recaptured PIT tagged fish because the radiotagged fish were typically not recaptured. To get a snapshot of fish size per season and to compare fish size among streams and seasons I conducted a two-factor analysis of covariance ( A N C O V A ) on average weight (log transformed) with length as the covariate, streams and seasons as the factors and all fish caught in minnow traps as the replicates (including non-PIT tagged fish, but excluding radiotagged fish). I employed two criteria to the data to eliminate the chance of multiple measurements of a fish within a season, as some PIT tagged and non-PIT tagged fish were caught more than once per season. For PIT tagged fish that were caught more than once within a season I averaged the size (length and weight) of the fish for that season and included this average in the analysis. I excluded all recaptured non-PIT tagged fish from the analysis since I did not know what the size of the fish was the first time it was captured as they were not individually marked. Growth rate was calculated by two methods. The first method calculated growth rate as the change in length (mm) or weight (g) from one recapture to the next divided by the number of days between the recaptures (mm/day). In an attempt to correct for initial size within each season, I also calculated growth rate relative to initial size (%/day): ((S t + 1 -So I S,) * 100% / (day t + 1 - day,), where S, + 1 is the size (either length or weight) of the fish at recapture, S, is the size from the previous capture, day t + 1 is the day of the recapture and day, is the day of the previous capture. Total and seasonal average growth rates for both methods were calculated for each fish and these were then used to compare average growth rate among seasons and streams using an A N O V A . Correlations between either length and length growth rate, weight and length growth rate, length and weight growth rate or weight and weight growth rate were performed to 46 check i f different sized fish grew at different rates among the streams and seasons. A s wel l , correlations were performed between fish length, weight, growth rate (length and weight) and movement rate for each season within each stream. I used the sequential Bonferroni technique to correct a for multiple comparisons (Avalon Creek fall, winter, spring, Husdon Creek fall, winter, spring, and Pipe Creek fall and winter). Coefficient of variation (CV) was calculated and then In transformed for each set of activity level measurements (receiver readings) taken during an observation of a radiotagged fish. To estimate activity at different times of day during each season, individual observations per fish were pooled within a stream and A N O V A test was performed with stream, season and time of day as the three factors. The In of the coefficient of variation was done for the single test fish and A N O V A was used to test i f there were any differences in activity level among the three movement patterns. Habitat Association Association with channel units, physical cover structures and substrate were estimated using the "habitat availability equation" outlined in Matthews (1996): D = (r-p) / ((r+p) - 2 * r * p), where r is the proportion of the habitat characteristic used by the fish (estimated from direct fish observations), p is the proportion of the habitat characteristic available to the fish (estimated from habitat survey data), and D is the habitat association value. Habitat association ranges from -1.0 to 1.0, where -1.0 to -0.26 shows strong disassociation; -0.25 to 0.25 shows no association; 0.26 to 0.49 shows moderate association; and 0.50 to 1.0 shows strong association. Habitat association with channel unit type, physical cover type and substrate type was calculated for PIT tagged individuals for each month, each season and averaged among months for an overall association value. For radiotagged fish I calculated habitat association values for combined channel unit and physical cover type for each stream and season pooled. 47 Results Stream Conditions I made several predictions before starting the study and these were based on the assumption that during the fall, discharge would be stable and water level low, while the opposite conditions would exist during the winter (Figure 2.2). I found that my assumption for fall conditions better explained winter conditions and vise versa (Table 2.5). The hydrograph of each stream is shown in Figure 2.6. Average water depth was not different between fall and winter (streams combined: F005f2)1232 = 2.182; P= 0.1410; p = 0.30), but it was different among the streams (seasons combined: F005(2)2232 - 28.838; P < 0.0001; p = 1.00). There was an interaction between season and stream (F005(2)2232 = 4.138; P = 0.0172; p = 0.73) because Avalon Creek had higher average water depth during the fall than the winter, whereas in the other two streams, average water depth was highest during the winter. Avalon Creek had the highest water level among the streams for both fall and winter. A s well , on average it rained more often at Avalon and Pipe Creeks (24 ml/24 hrs) compared to Husdon Creek (9 ml/24 hrs). Table 2.5: Mean recorded water depth (± SE , n) at the stationary metre stick. The fluctuation in water depth (coefficient of variation = C V ) and calculated discharge for the three streams for fall and winter are also shown. A review of the assumed water condition from before the study started and the actual outcome are included (see Figure 2 and 23 for details). Fal l Winter Depth (m) C V Discharge (m3/sec) Depth (m) C V Discharge (m3/sec) Ava lon 0.35 ± 0.03,32 58.19 2.100 0.31 ± 0.02,48 29.59 0.747 Husdon 0.22 ± 0.01,37 19.84 0.082 0.27 ± 0.01, 54 12.57 0.177 Pipe 0 . 1 4 ± 0.02, 13 46.46 0.000 0.21 ± 0.01, 54 14.18 0.051 Assumption Actual low low/high stable variable low low/high high ...high variable stable high high 48 Fluctuations in water level (stable versus variable) were opposite to my predicted pattern (Table 2.5). During the fall, water level fluctuated more than during the winter. Avalon Creek had the greatest fluctuation in water depth for both seasons, whereas Husdon Creek had the lowest and Pipe Creek was intermediate. A flood event occurs when water level increases over a short period of time (ie. hours, minutes) above average water depth (Heggenes 1988; Pearson et al. 1992) and usually coincides with heavy rain fall or rapid snow melt. I characterized a flood event within the three study streams as a period when the water depth rose (or fell, as some increases in water depth were missed) by at least 10 cm per day (Figure 2.7). Husdon Creek experienced at least three high water events (4, 6, 9 ; Figure 2.7): the first in late November, the second in mid-January, and the third at the beginning of March. The water depth in Husdon Creek never rose above 45 cm in height at the metre stick. A t flooding the water would gently breach the stream banks and flood the floodplain. Ava lon Creek on the other hand experienced at least four major floods ( 1 , 3 , 6 , 7 ; Figure 2.7) and several minor flood events (2, 5, 8, 10; Figure 2.7). The flood in mid-November (3; Figure 2.7) was recorded as being a l-in-40 year flood event for the Sunshine Coast (Grant M c B a i n , Department of Fisheries and Oceans, Sunshine Coast, B . C . , P .O. B o x 10, Madera Park, B . C . , V O N 2H0, personal communication). The steep, deforested headwaters section made Avalon Creek vulnerable to debris flows causing the stream to be highly unstable. During the l-in-40 year flood event, and to a lesser degree during the mid-October flood (1; Figure 2.7), L W D and course substrate were transported downstream from the upper reach and headwater section and partially rerouted the stream channel in the lower reach, including the study site. A s evidence of the instability in Avalon Creek, 9 of the 16 pieces of marked L W D were displaced out of the study site into the far downstream section or into the estuary during the l-in-40 year flood event (Figure 2.8), and many new pieces entered the study site from upstream over the course of the study (Figure 2.3). Overall, all but one of the marked L W D pieces were displaced at least half a metre downstream by the end of winter (Figure 2.8). Of the seven pieces which were not lost during the study, four were classified as "dry" meaning that they were located within the stream banks but out of the water, and three were straddled across 49 0 I i i i i i i < i i i i i i i i i i i i i i i i i i i i i i i i i i i i—IIII—(—i—i i i i 250 275 300 325 350 375 400 425 450 47 0 " T • T" T' 1—T" "1—•! •• -T— I™1 1 1---I"1 T | 1 I—I I | 1 1 I 1 | 1 I I 1 | 1 f * '1 | ' 1 1 |—1 1 I 1 I T \ 250 275 300 325 350 375 400 425 450 475 250 275 300 325 350 375 400 425 450 475 Time (days) September October November December January February March April 1998 1999 Figure 2.6: Hydrographs of each stream over the course of the study. The dotted lines represent suggested water depth at the time of the l-in-40 year flood event which occurred on November 14 - 16, 1998 (days 318 - 320). 50 40 n § 30 H 1 o I 6 £ .1 a -20 H -30 O Avalon • Husdon 3 Pipe O O 2 o 8 O e 0 7 o -1—1—1—1—I—I—I I I I I I I—I—p I I I I I I I I I ~r 250 275 300 325 350 375 400 425 450 475 Time (days) S e p t e m b e r O c t o b e r N o v e m b e r D e c e m b e r J a n u a r y F e b r u a r y M a r c h A p r i l 1 9 9 8 1 9 9 9 Figure 2.7: Change in water depth (cm/day) for each stream. Water depths were measured at the stationary meter-sticks. Changes > ± 10 cm/day were considered food events. The numbers indicate separate flood events. 51 250 n Avalon 200 H 150 H 8 s as • c 100 H 50 H -50 220 240 — i — 260 Husdon 160 i 10 16 14 '15 280 300 — i — 320 * 340 360 — i — 380 e 140 ; 120 : 100 H 80 H 60 H 40 -j 20 H 0d -20 .1 •2 3 •4 • 5 7 "6 .8 .9 .10 .11 .12 •13 .14 • 15 .16 8 •9 .11 .12 •13 T 400 420 440 —i— i •—i 1—i 1—i 1—i •—i 1—i •—i 1—i • — i 1—i 220 240 260 280 300 320 340 360 380 400 420 440 Time (days) Figure 2.8: Displacement of selected pieces of L W D within Avalon and Husdon Creeks. Each stream had 16 pieces of L W D individually marked at the beginning of the study. Each piece of L W D is represented by a line and a number. Meter 0 represents the downstream end of study site. The line indicates the location of the L W D over the course of the study. Nine of the 16 pieces were lost from Avalon Creek at the time of the l-in-40 year flood event in Avalon Creek (*). See Appendix 1 for details of each L W D . 52 the stream channel suspended over the water (see Appendix 4). A s wel l , nine of the fifteen pieces that moved in Avalon Creek were oriented nearly perpendicular to the flow of water, while the other six pieces were nearly parallel to the flow of water. In Husdon Creek, only two pieces of L W D , number 6 and 16, were moved downstream (Figure 2.8). The total movement of the Husdon Creek pieces totaled 8.4 m, while in Avalon Creek the totaled distance was over 1000 m (see Appendix 4). The l-in-40 year flood event also partially rerouted the water course in Pipe Creek. This flood was the most severe high water event that occurred in Pipe Creek, even though it did not show as a flood event in Figure 2.7 (due to missing data). Pipe Creek also flooded during mid-January and February (6, 8 respectively; Figure 2.7). The three streams experienced similar changes in channel unit size and depth over time (fall, winter and summer; Table 2.6). The bankfull width, depth, length, and area per channel unit plus the total area of the study site increased from fall to winter (increasing water level) and decreased from winter to summer (decreasing water level; Table 2.6). Avalon Creek had the widest bankfull width and largest area per channel unit of the three streams. Bankfull width remained wide from winter to summer in Avalon Creek because the stream banks had eroded approximately three meters, primarily during the l-in-40 year flood event in November of 1998. Because the stream banks eroded during this flood, several deciduous trees were uprooted and fell across the stream, providing new habitat and widening the stream (Figure 2.3). Area, percent physical cover and substrate per channel unit for each month is summarized in Figure 2.9. The total area of fast water channel unit types (glides, runs, riffles, and cascades) increased relative to slow water channel unit types (isolated, backwater, side, and plunge pools) from fall to winter as water levels rose (Figure 2.9). During the winter months, over 89% of the total area within Avalon Creek was riffle and cascade habitat. During the same time period, less than 65% of the total area within Husdon Creek was glide and run habitat, with less than 20% riffle or cascade habitat. 53 Table 2.6: Summary of average channel unit dimensions (± SE, n) within the three streams for the fall and winter of 1998 and the summer of 1999. A l l values are means per channel unit, while total area is the mean per habitat surveys. Fall Winter Summer Avalon Bankfull Width (m) 12.26 ± 0.52,91 14.72 ± 0.63, 65 15.72 ± 1.17,5 Average Depth (m) 0.22 ± 0.02,85 0.28 ± 0.02,67 0.23 ± 0.02,5 Length (m) Area (m2) 11.60 ± 1.03,90 11.30± 1.32,69 7.32 ± 3.46,5 42.42 ± 5.12,85 55.59 ± 8.21,69 22.43 ± 12.84, 5 Total Area (m2) 1281.86 ±29.95, 2 1278.57 ± 103.08, 3 Husdon Bankfull Width (m) 3.58 ±0.13, 151 5.20 ± 0.22, 110 3.39 ±0.19, 13 Average Depth (m) 0.18 ±0 .01 , 146 0.33 ± 0.01, 109 0.12 ±0.02, 13 Length (m) Area (m2) 4.42 ± 0.24, 148 6.12± 0.47, 111 3.50 ± 0.42, 13 9.86 ± 0.60, 148 17.22 ± 1.61, 111 6.92 ± 1.01, 13 Total Area (m2) 453.26 ± 2.54, 2 637.06 ± 16.56, 3 Pipe Bankfull Width (m) 5.14 ±0.16, 18 5.45 ±0 .21 ,47 3.91 ± 0 . 1 3 , 4 Average Depth (m) 0.15 ±0.02, 18 0.19 ± 0.01, 47 0.15 ±0.02, 4 Length (m) Area (m2) 4.95 ± 1.37, 18 6.76 ± 0.64,47 7.28 ± 3.35,4 11.59 ±4.70, 18 18.25 ± 2.25, 47 15.00 ±7.15, 4 Total Area (m2) 208.66, 1 286.56 ±7.56, 3 The percent structural cover (LWD, root-wads etc.) per channel unit within each stream did not change drastically over the course of the study (Figure 2.9). During all months, Husdon Creek had the highest percentage of LWD, and undercut bank cover of all three streams. The percentage of root-wad cover was highest in Avalon Creek throughout the study. Percent vegetative cover was higher in Avalon and Pipe Creeks during early fall, however it switched during late fall (November, December) and winter as leaves fell from the shrubs lining the stream banks. There was virtually no boulder cover in Husdon Creek, while boulders covered an average of 20% of channel units in Avalon and Pipe Creeks. The percentage of the channel unit that was left open, or free of physical cover was approximately equal among the streams (Pipe Creek was slightly higher) and among the months. 54 B 3 5 0 - August - , * r | Loa In 5 0 4 0 3 0 H 0 2 3 4 5 6 7 9 2 0 " J n U R L S V B O 5 0 4 0 ~\ 3 0 2 0 10 i 1^ I -T O S g G C B • Avalon Husdon Pipe 1 8 0 1 5 0 -1 2 0 -9 0 " 6 0 " 3 0 -September 0 1 2 3 5 6 7 8 9 U R L S V B O T O S g G C B 244 254 264 274 9 0 0 6 0 0 I 3 0 0 i 1 2 3 4 5 6 7 8 9 U R L S V B O T O S g G C B 274 284 294 304 4 5 0 3 6 0 J November 2 7 0 -1 8 0 " 9 0 " I I. 1 1 2 3 4 5 6 7 8 9 -ULUJ .1.. T O S G C B 304 314 ' 324 334 5 0 4 0 H 3 0 2 0 1 10 1 1 2 4 6 7 8 9 Channel Unit T O Physical Cover S g G C B Substrate 334 344 354 364 Time ( d a y s ) Figure 2.9: Summary of habitat survey data. Column A is the total area (m 2) of each channel unit type within each stream. The numbers on the X-axis indicate different channel unit types (see Table 5 for legend). Column B is percent in-stream physical cover per channel unit as undercut bank, U ; R: rootwad; L W D , L ; S W D , S; vegetation, V ; boulder, B ; open (no cover), O. Column C is percent substrate class per channel unit as silt, T ; organic matter, O; sand, S; small gravel, g; large gravel, G ; cobble, C ; boulder, B . Column D is water depth (cm) over time (days). The asterick in November represents the timing of the l- in-40 year flood event (days 318-320). The dark horizontal bars along the X-axis ( • ) represent the days in which the habitat surveys were done in each month. 55 - January 1 2 4 5 6 7 8 9 February Jit-1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 Channel Unit U R L S V B O Physical Cover T O S g G C B Substrate 425 435 445 455 Time (days) Figure 2.9: continued. 56 Husdon Creek was dominated by finer substrates (sand and gravel) than the other two streams, and Ava lon and Pipe Creeks were dominated by relatively coarse substrate (cobble and boulder; Figure 2.9). A s water level increased the percentage of silt and organic material decreased within the streams (eg. September verses October in Ava lon Creek). Sand substrate remained a dominate substrate within Husdon Creek each month. Canopy cover and riparian area characteristics are summarized in Table 2.7. In general, Husdon Creek had the highest number of softwood trees, and greatest percentage of canopy cover and non-woody (fern) plants compared to the other streams. Table 2.7: Riparian area characteristics of the three streams. Presented are the number of hardwood (bigleaf maple, red alder) and softwood (western redcedar, western hemlock, Douglas fir) trees present within the study site, mean percent (± S E , n) per channel unit of woody plants (eg. shrubs) and non-woody plants (eg. ferns, grasses) located within the riparian area for fall and winter, and mean percent (± SE , n) per channel unit canopy cover per channel unit for fall and winter. Avalon Husdon # Hardwoods 91 57 74 # Softwoods 38 50 1 Fall % Woody Plants 44.01 ± 3 . 5 3 , 9 0 7.86 ± 0.96, 126 67.22 ± 3 . 0 6 , 18 % Non-Woody Plants 16.53 ± 2.44, 90 22.89 ± 1.76, 126 11.39 ± 1.50, 18 % Canopy Cover 39.39 ± 3.29, 91 74.69 ± 5.49, 127 78.33 ± 2.06, 18 Winter % Woody Plants 43.87 ± 4 . 1 5 , 5 5 4.57 ± 0.58, 102 55.11 ± 2.84, 47 % Non-Woody Plants 8.36 ± 1.21,55 27.34 ± 2.43, 102 7.36 ± 0.80, 47 % Canopy Cover 19.45 ± 3.07, 67 41.82 ± 2 . 9 , 109 21.23 ± 1.15,47 Number of Cutthroat Trout Sampled A total of 864 fish were caught in minnow traps between August 1998 and August 1999. O f the 864 fish caught, 510 were implanted with PIT tags, of which 273 (53.5%) were recaptured at least once (Table 2.8). For radiotelemetry measurements, 18 fish were implanted with transmitters and repeatedly relocated during the eight month study (Table 2.3): 8 from Husdon Creek, 5 from Avalon Creek, and 5 from Pipe Creek. One fish, Cristy, moved between Avalon and Pipe Creeks, and was counted independently for each stream. A total of 1124 observations were made on radiotagged fish, 742 during the fall and 382 during the winter. 57 Table 2.8: Breakdown of the total number of fish implanted (tagged) with PIT tags and the total number recaptured at least once during the study (% of total tagged shown in brackets). The total number of fish recaptured at least once is broken down into three categories, recaptured once, recaptured twice and recaptured > 3 times during the study (% of total recaptured shown in brackets). Additional recaptures represents the number (% of total recaptured shown in brackets) of fish that were recaptured in the upstream or downstream "additional" traps. Stream Total Total Recaptured Recaptured Recaptured Additional Tagged Recaptured Once Twice > Three Recaptures Times Upstream / Downstream Avalon 110 57 (51.8) 32 (56.1) 14 (24.6) 11 (19.3) 0 (0.0) / 4 (7.0) Husdon 311 182 (58.5) 78 (42.9) 51 (28.0) 53 (28.3) 2 (1 .1) /3 (1.6) Pipe 89 39 (43.8) 19 (48.7) 11 (28.2) 9 (23.0) 2 ( 5 . 1 ) / 1 (2.6) Total 510 273 (53.5) 129 (47.3) 76 (27.8) 73 (26.7) 4 (1.5) / 8 (2.9) Hypothesis 1: Seasonal and Short-term Movement Rate The first prediction was only partially met. Movement rate was different for PIT tagged fish (Figure 2.10a), but not radiotagged fish (Figure 2.10b). There was a significant interaction between season and stream for average movement rate (F005(2)4286 = 4.967; P = 0.0007; p* = 0.97) of PIT tagged fish resulting from the large negative movement rate during the fall in Avalon Creek compared to Husdon and Pipe Creeks (Figure 2.10a). Fish in Husdon Creek moved relatively little (< 0.11 m/day) during any season. There was no difference in movement rate among seasons, however movement rate during the winter (<, 0.088 m/day) was relatively less compared to the fall (<, 0.78 m/day) and spring (:S 0.330 m/day) in all streams. There was an interaction for movement rate among seasons, streams and size classes of PIT tagged fish (Table 2.9; Figure 2.11). A s well , there were no differences in net within-day movement among the different time periods of the day for radiotagged fish. Movement patterns over time (from their initial capture locations = 0 m) of recaptured PIT tagged fish and radiotagged fish from the three streams are shown in Figures 2.12 and 2.13 respectively. Two patterns emerge, both PIT tagged and radiotagged fish seemed to remain within the same area for extended periods of time, and/or many fish from all three streams made long distance movements, either upstream 58 or downstream. Unlike radiotagged fish in Husdon Creek, radiotagged fish in Avalon and Pipe Creeks made return trips after their long distance movements. As well, long distance movements in Avalon Creek and somewhat in Pipe Creek occurred simultaneously (during the l-in-40 year flood event), while long distance movements in Husdon Creek were scattered through time (Figure 2.13). Bernie from Husdon Creek, was the only fish observed to make a continuous, gradual migration upstream. Cristy was the only radiotagged fish to travel between Avalon and Pipe Creeks (Figure 2.13). There were five PIT tagged fish that moved between Avalon to Pipe, three during the fall and two during the winter (Avalon to Pipe), and one in the spring and the one in the summer (Pipe to Avalon). Table 2.9: Analysis of variance table for movement rate (m/day) among seasons (fall, winter, spring), streams (Husdon and Avalon) and size classes (65 - 99 mm, 100 - 119 mm, > 120 mm) of PIT tagged fish. Factors DF Sum of Squares Mean Square F-value P-value R - value Stream 1 3.040 3.040 4.759 0.0301 0.58 Season 2 11.405 5.702 8.927 0.0002 0.98 Size Class 2 1.028 0.514 0.805 0.4484 0.18 Stream x Season 2 11.209 5.605 8.773 0.0002 0.98 Stream x Size Class 2 0.957 0.478 0.749 0.4741 0.17 Season x Size Class 4 9.169 2.292 3.588 0.0073 0.88 Stream x Season x Size Class 4 10.617 2.654 4.155 0.0029 0.93 Residual 235 150.122 0.639 In general I observed very little within-day change in location for radiotagged fish. Greater than 90% of the net within-day movements were within 5 m of the first relocation for that day (Table 2.10). There was only one observation > 30 m (Table 2.10). Many PIT tagged fish remained within 25 m of their initial capture location (Figure 2.14). Movement of PIT tagged fish within each stream was not random since the frequency 59 1.25 1.00 0.75 0.50 0.25 0.00 -0.25 -0.50 -0.75 -1.00 -1.25 a) • Fal l H Winter HI Spring Avalon Husdon Pipe Avalon Husdon Figure 2.10: Average movement rate (+ 1 SE) among seasons for (a) PIT tagged fish and, (b) radiotagged fish from the three streams. Positive values indicate average movement rate in the upstream direction and negative values indicate average movement rate in the downstream direction. 60 •3 i S e 1 -. 8 -. 6 -. 4 -. 2 -0 . 2 -. 4 -Pipe • 65 - 99 mm • 100 -119 mm HI > 120 mm Fall Winte r Spr ing Figure 2.11: Average movement rate (+ 1 SE) of each size class of PIT tagged fish for each season and stream. Positive values indicate average movement rate in the upstream direction and negative values indicate average movement rate in the downstream direction. 61 Figure 2.12: Movement patterns of PIT tagged fish from the three streams. Metre "0" represents the initial capture location. A l l movement distances are relative to that location. Positive distance moved is upstream and negative distance moved is downstream. The arrows indicate the timing of the l-in-40 year flood that occurred in Avalon and Pipe Creeks in November 1998 (days 318-320). 62 1 Husdon —•— Bernie —0~ Chester -200 -• -m- Clark - • - Glosh -300 - _o— Harry - * — Jake -400 " —"— Lujay —°— Marina » » • t • m m 500 i 1 1 ' 1 1 • — ' 1 1 « > 1 1 1 — ' 1 « > ' 1 • • ' 1 240 280 320 360 400 440 480 360 400 Time (days) Figure 2.13: Movement patterns of radiotagged fish from the three streams. Each fish is identified by a unique symbol. Metre "0" represents the initial capture location. A l l movement distances are relative to that location. Positive distance moved is upstream and negative distance moved is downstream. The arrows indicate the timing of the l- in-40 year flood that occurred in Avalon and Pipe Creeks in November 1998 (days 318-320). 63 distribution of the final locations between the random populations and the real populations were different (Figure 2.14). More fish were found downstream of their initial capture location than upstream in Avalon Creek (32 and 24 respectively). The opposite was true in both Husdon and Pipe Creeks where the final recapture location of more fish was upstream rather than downstream (121 and 61; 27 and 11 respectively). One hundred percent and 95% home range size of all radiotagged fish, except Jake (Husdon Creek) and Pisces (Pipe Creek) are shown in Figure 2.15. Home range size was larger using the 100% estimate compared to the 95% estimate for the Husdon Creek test (10.05(2).? = -2.791; P = 0.0269) and the Avalon/Pipe Creek test (t005(2)7 = -4.082; P = 0.0047). Average home range size was larger in Pipe Creek compared to the other two streams (Table 2.11). There was no significant difference in home range size between seasons (fall and winter). Table 2.10: Number of within day observations per distance class pooled within streams (253 observations). Distance refers to the distance gained (positive) or lost (negative) relative to the first relocation within a day of observation on a radiotagged fish. Observations are totaled within streams and within distance classes. Distance (m) Fall Winter Total Avalon Husdon Pipe Avalon Husdon Pipe >-30 0 0 0 1 0 0 1 30 to-5 0 5 0 0 1 0 6 -5 to 5 38 77 20 11 61 34 241 5 to 30 1 1 0 1 1 0 4 > 3 0 0 0 0 0 0 0 0 Total 39 83 20 13 64 34 253 64 120 Husdon 100 H n = 182 80 " 60 -40 -20 -0 3 3 . 5 % -200 -100 6 6 . 5 % 100 300 200 • • • r -200 -100 200 18 -16 -14 -12 -10 -8 -6 -4 -2 -0 -Avalon n = 56 57.1 % • i i | i i l~] i -400 -200 42.9 % 1 I I | I » 1 | I I I | 200 400 -400 -200 i 1 1 • I 200 400 225 i 1 • • 100 200 -200 -100 0 100 200 Final Relocation Distance (m) Final Relocation Distance (m) Figure 2.14: Histograms of the real and simulated random populations of PIT tagged fish. Chi-squared (x2) values and P - values are shown for each real-random population comparison, as are the observed (actual population) and expected (random population) percentages of fish that ended up downstream (left side of zero) and upstream (right side of zero) of their initial capture location. 6 5 Om 6m 12m -220 m l I I l i Avalon Creek Legend • L W D - o r i g i n a l = L W D - n e w ^ » L W D - l o s t U C B a n k • R o o t - w a d • R o o t - w a d - n e w • B o u l d e r O T r e e 9 D r y B a r @ P o s t < t M e t e r - s t i c k Figure 2.15: Home range length using 95% and 100% estimation methods for most radiotagged fish. Two fish were excluded, the first because insufficient data was available to calculate home range (Pisces, Pipe Creek), and the second because the entire home range was outside the study site (Jake, Husdon Creek). Some home ranges extend outside the boundaries of the study sites and are not shown in entirety on this diagram. See Table 2.3 for the time period in which each fish was radiotracked. 66 Table 2.1 T. Mean home range size in meters (± SE) for the 95% and 100% estimation methods. Home range is calculated for each season (fall and winter), and pooled across seasons to include the whole study (total). Some fish had mean home range size calculated for both seasons as well as in the total (Avalon and Husdon Creeks) or for neither season but in the total (Pipe Creek) thereby explaining why fall and winter "n -values" do not equal the total number of fish included in the total. Stream Season 95% 100% n Home Range (m) n Home Range (m) Avalon Fal l 4 5.00 ± 1.35 4 34.20 ± 12.99 Winter 1 32.80 1 46.00 Total 4 12.20 ± 6.99 4 40.00 ± 12.58 Husdon Fal l 4 23.03 ± 13.24 4 33.35 ± 16.12 Winter 4 10.35 ± 2.71 4 16.58 ± 4.79 Total 7 31.34 ± 18.30 7 47.30 ± 18.19 Pipe Fal l 1 40.70 1 118.70 Winter 2 35.25 ± 29.95 2 86.75 ± 55.45 Total 4 50.13 4 , 105.43 ± 2 5 . 7 1 R o o d Events There was no difference in movement rate of cutthroat trout between flood and non-flood periods (excluding the l-in-40 year flood event) in all streams for PIT tagged fish (streams combined: F005f2)iU49 = 2.734; P = 0.100; |3 = 0.36; Figure 2.16a), and radiotagged fish (streams combined: F005(2) 1 2 5 = 0.043; P = 0.8368; p = 0.10; Figure 2.16b). A separate analysis comparing the movement rate during the l-in-40 year flood event and non-flood periods for PIT tagged fish in Avalon and Pipe Creeks only, showed an interaction between the two factors (F005(2)133 = 44.576; P < 0.0001; (3 = 1.00; Figure 2.17). There was no statistical relationship between movement rate and the l-in-40 year flood event for radiotagged fish (t005(2h4 = -8.729; P = 0.2969), however the actual observed individual movements present interesting findings. From Figures 2.12 and 2.13, there seems to be a link between the timing of the l-in-40 year flood event and the downstream movement of 6 of 7 radiotagged fish and many PIT tagged fish from Avalon and Pipe Creeks. The PIT tagged fish that moved downstream during the flood did not return upstream after the flood. Alternatively, four radiotagged fish, Little B i l l y , Darell and Cristy from Avalon Creek and Lou Lou from Pipe Creek, moved between 5 0 - 1 0 0 67 m downstream during the flood and returned back to the location where they were found before the flood occurred (Little B i l l y and Darell), exceeded this distance by over 20 m (Lou Lou), or entered a different stream (Cristy moved into Pipe Creek). T w o fish, J im from Ava lon Creek and Pisces from Pipe Creek were completely lost during the l-in-40 year flood event. T w o days before the flood occurred both fish were successfully located and the day after the flood, neither of these fish were found. I assumed these two fish were either in estuary, where they would be virtually impossible to relocate using the radio receiver, or buried under cobbles and boulders that moved during the flood. Only one fish Sanjie (Avalon Creek) remained within the same location (plunge pool) during and after the flood. Cristy, who moved into Pipe Creek during the flood and remained there for several weeks made other between stream movements. On one occasion (day 378), Cristy moved downstream in Pipe Creek to the confluence with Avalon Creek, then moved 60 m upstream in Avalon Creek at which point the water depth rose to flood levels (flood event 6, Figure 2.7) and Cristy returned downstream and reentered Pipe Creek, which was also in flood, but at a lower discharge (Figure 2.12 and Figure 2.13). Movement rate was variable among size classes between flood and non-flood periods in the three streams. N o statistical analysis could be done on these data due to low sample sizes. However, all size classes in Avalon Creek seemed to have a downstream movement rate (negative) during flooding periods (excluding the l-in-40 year flood event), while no distinct pattern could be detected in the other two streams (Figure 2.18). During the l-in-40 year flood event, movement rate was downstream regardless of fish size. Hypothesis 2: Growth Rate The second prediction that growth rate would be equal during the low flow conditions and different during high flow conditions was only partly met. Growth rate (in length) differed among streams (seasons combined: F005(2U279 = 4.732; P = 0.0095; p = 0.79) and seasons (streams combined: F005(2U279 = 19.705; P < 0.0001; p = 1.00; Figure 2.19). There was no interaction between the two factors {F005(2)4279 = 1.163; P = 0.3273; p = 0.36). Growth rate was greater in Pipe Creek than in Husdon Creek (Pearson's Least 68 Squared Difference (PLSD) P = 0.0065), but not Avalon Creek ( P L S D P = 0.2829), which is not what was predicted (Figure 2.2). There was also a difference in growth rate i n weight among the seasons (streams combined: F005(2)2273 - 9.160; P = 0.0001; P = 0.99) but not streams (seasons combined: F 0 M ( 2 U 2 7 3 = 1.085; P = 0.3392; p = 0.23) or an interaction between the two factors (F005(2jA273 = 1.169; P = 0.3246; p = 0.36). Growth rate in both length and weight was greater in the spring than in the winter or fall with fall having the lowest growth rate of all in all three streams (Figure 2.19). The results for growth rate using the second method (%/day) were similar to the first method as percent growth rate in length was greatest during the spring, least during the fall and intermediate during the winter (streams combined: F005(2)2239 = 9.275; P = 0.0001; p = 0.99). Percent growth rate in weight was greater in the spring (streams combined: F00S(2)Z228 = 3.821; P = 0.0233; p = 0.69). There were no differences among streams for length (seasons combined: F005(2)2239 = 1.620; P = 0.2001; p = 0.33) or weight percent growth rate (seasons combined: F005(2)2228 = 1.023; P = 0.3611; p = 0.22). There was no correlation between fish size (length or weight) and average movement rate for any of the streams in fall or winter. Husdon Creek displayed a significant negative correlation between average movement rate and length growth rate during the fall (ta o5(2),9o = -4.348; P < 0.0001; correlation - 0.435). There was no correlation in Ava lon or Pipe Creeks in either season for length or weight growth rates with average movement rate. The size of fish caught in minnow traps was different among the streams (seasons combined: F005(2)Z1016 = 6.118; P = 0.0023; p = 0.92) and seasons (streams combined: F0Q5(2)jJ0](~ 13.655; P < 0.0001; p = 1.00; Figure 2.20) with interactions between seasons and streams, streams and length, seasons and length and streams, seasons and length (Table 2.12). Fish in Avalon Creek were larger during all seasons except for summer when Husdon Creek fish were larger and fish from Pipe Creek were smallest during all seasons (Figure 2.20). In general, weight of fish increased in all streams from fall to spring, but then decreased again in summer (Figure 2.20). During the winter in Avalon and Husdon Creeks, larger fish (in length and weight) were gaining more weight (higher growth rate) than smaller fish (Table 2.13). Wi th fish 69 size split into the three size classes, there was no difference in growth rate in length (with weight growth rate as a covariate) among the streams. Act ivi ty level was different among the streams (seasons combined: F005(2)282g - 8.274; P = 0.0003; p = 0.97), and time of day (streams combined: F005(2)Z829 = 3.498; P = 0.0307; p = 0.65). There were no interactions between and among factors in this test. Fish in Ava lon Creek were the most active and fish in Pipe Creek were the least active. Act ivi ty at night (dusk till midnight) was greater than the other times of the day (Figure 2.21). Table 2.12: Analysis of covariance table for fish size (log weight in grams) among seasons (fall, winter, spring and summer) and streams (Husdon, Avalon and Pipe) with length as the covariate. Factors D F Sum of Squares Mean Square F - value P - value B - value Stream 2 0.096 0.048 6.118 0.0023 0.90 Season 3 0.321 0.107 13.655 <0.0001 1.00 Length 1 16.674 16.674 2130.893 <0.0001 1.00 Stream x Season 6 0.139 0.023 2.967 0.0071 0.91 Stream x Length 2 0.098 0.049 6.292 0.0019 0.91 Season x Length 3 0.122 0.037 4.750 0.0027 0.91 Stream x Season x Length 6 0.114 0.019 2.423 0.0249 0.83 Residual 1332 7.826 0.006 70 •3 1 I I o 2 .4 -.2 -0 -.2 --.4 --.6 --.8 --1 ( a ) 1 5 n 10 5 i -5 H -10 H -15 •tSSKSSSS KSKSSSS w w w Avalon Husdon Pipe Flood Non-Flood (b) Flood Non-Flood Figure 2.16: Average movement rate (+ 1 SE) during flood and non-flood periods for (a) PIT tagged and, (b) radiotagged fish (both excluding the l-in-40 year flood event). Positive values indicate average movement rate was in an upstream direction and negative values indicate average movement rate was in a downstream direction. 71 s O 2 2 o--2 -4' -6 -8 -10 --EH Ava lon £1 Pipe l-in-40 Year Flood Non-Flood Figure 2.17: Average movement rate (+ 1 SE) of PIT tagged fish during the l- in-40 year flood event and during non-flood periods for Avalon and Pipe Creeks. Positive values indicate average movement rate was in an upstream direction and negative values indicate average movement rate was in a downstream direction. A n interaction between flood/non-flood periods and streams was present (P < 0.0001). 1 2.5" 2" 1.5" 1-.5" • 69 - 99 mm • 100-119 mm El > 120 mm Avalon Husdon Pipe Avalon Husdon Pipe Flood Non-Flood Figure 2.18: Average movement rate (+ 1 SE) of different sized PIT tagged fish during flood and non-flood periods (excluding the l-in-40 year flood). Positive values indicate average movement rate was in an upstream direction and negative values indicate average movement rate was in a downstream direction. 72 • Fall • Winter Spring Avalon Husdon Pipe Avalon Husdon Figure 2.19: Average growth rate (+ 1 SE) in length and weight for PIT tagged fish from the three streams over the three seasons (the S E bar is missing for spring in Pipe Creek because n = 1 for that sample). Growth rate was greatest during the spring, and slowest during the fall , with winter in between. Growth rate between Husdon and Avalon Creeks was not different. Growth rate was greater in Pipe Creek compared to Husdon Creek. 73 •2H 0 - 1 1 1 : • > Fall Winter Spring Summer Figure 2.20. Mean log weight (± 1 SE) for fish caught in minnow traps in the three streams over four seasons. Fall Winter Fall Winter Fall Winter Avalon Husdon Pipe Figure 2.21: Activity level (+ 1 SE) of radiotagged fish at different times of the day for fall and winter. The activity level is different among the streams and between night time (N) and the two other time periods of the day. 74 Table 2.13: Correlation coefficients and associated P values from Pearson correlations between length or weight and weight growth rate. Results from tests between length and weight with growth rate in length are not presented because all correlations were not significant. Critical P - values were calculated using the sequential Bonferroni technique. A n asterisk (*) indicates a significant result. Stream Season Weight Length Critical Correlation P - value Correlation P - value P - value Avalon Fall 0.045 0.8366 0.077 0.7221 0.0500 Winter 0.824 < 0.0001* 0.806 < 0.0001* 0.0063 Spring 0.917 0.0265 0.857 0.0696 0.0125 Husdon Fall 0.102 0.3525 0.143 0.1889 0.0250 Winter 0.544 < 0.0001* 0.550 < 0.0001* 0.0710 Spring 0.536 0.0061* 0.582 0.0023* 0.0100 Pipe Fall 0.332 0.2324 0.380 0.1655 0.0170 Winter 0.613 0.0008* 0.649 0.0003* 0.0080 Hypothesis 3: Habitat Association PIT Tagged Fish Seasonal and monthly changes in habitat association were observed in all three streams. However, among streams cutthroat trout were associated with similar habitat features, as I predicted. In general, cutthroat trout were strongly associated with slow water channel unit types (backwater/isolated pools, side pools, lateral scour pools and plunge/cascade pools) and glides (Table 2.14). Fish in all three streams were strongly disassociated with side channels, cascades and for the most part riffles. In Avalon Creek, fish were associated with boulder cover during each month, each season and when all monthly data were combined (Table 2.15). During the fall and when all monthly data were combined, fish in Husdon Creek were strongly associated with boulder cover but during the winter they were strongly associated with L W D . Physical cover association varied from month to month for fish in Pipe Creek, but generally they too were strongly associated with boulder cover. For the most part fish in all three streams were not associated with vegetation cover and root-wads and were strongly disassociated with open habitat Association with substrate type was less strong. Fish in Pipe Creek were strongly associated with large gravel. Surprisingly, fish did not associate with boulder substrate except in Husdon Creek in November 1998 (Table 2.16). 75 Radiotagged Fish All radiotagged fish in Avalon Creek, with the exception of Cristy, were most often found within a large plunge pool located approximately in the middle of the study site (Figure 2.3). This plunge pool consisted of a large fallen tree creating the plunge, a large (> 4 m wide) root-wad that occupied one bank of the pool and several large boulders and cobbles within the deepest part of the plunge pool. This plunge pool ranged in size between 2x2mto6x9m depending on the water level. Radiotagged fish used the whole plunge pool, but usually each fish remained within a general area and was associated with one cover type (eg. Little Billy with root-wad). In Pipe Creek, radiotagged fish used a variety of locations, however each fish was generally found associated with a boulder or cluster of boulders and cobbles, within a variety bf channel unit types. In Husdon Creek, radiotagged fish used primarily LWD in pools. Calculated habitat association values corresponded fairly well to the direct observations (Table 2.17). The radiotagged fish seemed to be strongly associated with faster channel units, particularly runs, compared to the smaller PIT tagged fish which were mostly associated with slow channel units. Population Density and Probability of Survival Population densities (fish/m2) with 95% confidence intervals were 0.04 (0.03 - 0.11), 0.25 (0.22 - 0.44), and 0.16 (0.15 - 0.53) for Avalon, Husdon and Pipe Creeks respectively. Population density was higher in Husdon Creek and Pipe Creeks than in Avalon Creek. The confidence intervals for Pipe Creek were large and overlapped the population density in Husdon Creek. Probability of survival (surviving to the next time period) estimates within streams were highly variable. Confidence intervals were not calculated for the survival estimates because recaptures within the three streams were too few. Recapturing a larger number of individuals would have yielded more realistic probability of survival estimates. Probability of survival for Avalon, Husdon and Pipe Creeks were 1.03,0.97, and 0.92 respectively. 76 C O "8 cs •a 9 o -S ~ jn tu .tS CD O 2.S 1 8 a -S -2 * S 8 2 * CO > . 3 is a ^ all •8.1 8 > c ^ .2 ^ e <- 3 2o o .2 3 J2 2 c K -2 13 8 i » 8 > w \2 co JR TJ S ? II s -s § fi. I" 60.2 U <U c» 2 3 " a 3 3 „ TJ O J co S is to .s ° * (3 cj O C O CO 00 • — £ TJ •§ cu cs F"1 cj IS — d 8. . es • £ a £ -5 8 a « «2 §i S w. »5 TJ S O 8 3 « fl .2 «n § 8 .fit; co .O — pj co ia '8 C O tn ca I s 9 2 (j > > w 8 S I II § a « 88 Si "a s d 8 d 8 d 8 d R d d S d 8 d 8 5 8 B 8 8 8 S 8 8 o o ^ 88 9 "3.3 S d 8 d 8 d a 8 d 8 d 8 d d s d 8 000 SO'O 00»-000 600 O.OS 0.07 0.00 0.00 0.01 0.00 •JM SS.«) 805 ° °5 i 0 0 9 l 8 8 8 o o q ° ° 9 l 8 3 ; " N O O JJj t s •» d d 3 88 d d 88.8 ° ° 2| S ? 3 o o ^ 5.8 5 " f l a a 5 fl S « d o g a s p 88 d o 88 9 d o Q o o | 0.00 0.31 8 8 9 d d 5 83 9 do ^ O O K © o ^ 0.02 0.29 0-92 0.00 0.00 88 d d 88 d d 88 d o 88 d o 3 S 8 0 0 § 1 '3 o. => 3 d s d 8 8 2 , d d 1 3 t d o 1 88 d o 8 8 8 S , d 0 < 88 O O 8 8 , . 8 8 ; d 0 2 88 d d 3 8 8 0 0 TI 83 9 O O O 8 8 d d 88 d d 5 8 8 8 8 8 ; § s . I a a •c S Q 8 .2 Z " *S 5 g d S C TJ "Ti .2 S 9 c3 OS s « n •S 'g 5 I -« TJ co O 6 ^3 > a § 60 i3 ll ca o Id 5 § las.» l i s d d r t OOw o o « I 88 0 o o o "5 ! 8 8ott d o O 88 O O ? 8 ; as « d d 85 d d 1 |8 8 ! 8 -88 d o 9 5.8 3 a a : d d 5 8 -o o • 0 0 3 8 8 8 o o 9| 88 d d d d | 8 = s d o 5 88g o o 3 8 8 8 0 0 'rl S. 8 8 V - o Q do I t 8 a 8 3 5 3 S « d d ^ 38 d d 2 2 Si , 0 R 8 5 , d o < 88 d d 8 8 9 d d a o o 2 Q a n 8 88 ; 889 OO ^ 0 8 52 d d 58 8 •a ° ° < ? l 3 8 ° ° <? d d ? | 7 33S 3 S • 9 8 K B 88 s 0 ?l 88 d d 8 8 S d d ^ 8S9 00^ P o 3 |o o ^ 83 • d o 1 8 ? « d 0 1 88 d o 88 d d 1 a j t; d 0 s ; 8 « 8S i 88 d o F88 ' ° 0 ?l |S 8 91 Is 5, °3 d d 2 5 8 fc d d <0 BI5 i i 3 - 8 8 : 2 BC fi O 3 11 a 8 -s j cn cS <U a fl 3 » . 2 5 13 8 ? < d d 1 3t8: 1 0 ° M pS 8 • a s 8 g | s d o g o d I d 0 8 8 8 0 ?l 8 5a 8 S1 d d 2 o o 1 8S d o 88 d d 88 ; d d 8 2 , o o < 88; d o ' I P 18 8 O cj to O o fl ! C O 58 a 88 60^ 00 ft 8 a d d ^ | s a. 3 s a < o 0, 88 d d 8 8 9 o o O S S fir d o g 8 ? 8 8 = 9 o o Q 889 0 d Q 88 d o SS R do § o o § s s a o o «8 8 8 8 0 0 ?l 8S 5 0 a V-8 o j 58, d d ' 88 d o 8 8 3 o o ^ | o o § o o ^ | 8 8 S 88 8$ d d 8 8 8 0 -H 9 5 a d d | d d » SS 1 0 o < ) 88 d o 88 . 0 d 1 8.8.8 ft8s ° 0 2| ° ° 9 1 889 d d 9 8? . d d < s a : 88 d d O N * 9 88 d d 83 9 d o ^ 0 0 3 83 9 d d Q 88 d d 88 d o I 3 pi 11] 111 < D - < < 3 < < 3 < 111 111 111 < D < 111 o <; ? 1 111 1° 5 j l o o $ I d d ^ d d 2 80; d d o o ' 8 8.8 | o o 9 l 111 < D < 111 O < 1 8-s a d d s s 0 o 83 o o S S ; d d 9 -8S5 d o § * S S 0 0 ° S3 , d d 1 sa 1 o o »» S 5 8 0 0 91 88 fc = 0^ a a s do 9 . o o S o 3 , 0 d 1 8? s do 5 5 S « 8 8 8 00^ 11« 0 0 8 8 s O 0 8 » ; 00*5 8 3 0 o " 8 8 ° ° 9 l °°s>l s . " 5 83 o d o 9 n 9 P r b 3 ° ° 3 0 0 3 S S , d d I 8 5 ; 55 d o s s . a d o 3 III .9 111 •a }s 1) c .2 60 i I co a, S CO in "3 T 3 a g co CU U CO II «s § 1 CO IT) w 3 u i O -"3 O T •5 T 3 "O M 3 « u S ea •S Q « i a - ° S e c > eu eu > <u p <u s .3 CO J > .a CO II O co II « M -2 e j3 •cs co S _ S .2 o _ a co 93 u co co g I 3 § 5 H a j 2? 8 o ^ Q vo u. "— fN 1 _ cj a -o ^ » c £ 0 u CO '0 <3\ CD :3 CS > 1 C I CU CO CO CO CO '•3 a I "3 S <=> > ea I § c> a u I -- S ea s JJ« co CO •a 2 U 13 co U UH M 8 8i co ea . tu cs cs '-2 •ft S a S .2 -2 ts 2 O co a . ea is K3 s 3$ ; S O 8 O $ b 3 O s b s b s 18 5 * 0 0 5 o S 8 8 O O ^| N T -S 8 8 O O ^| S S 5 8 8 O O 8 8 O O 3 O S o 3 O 8 O S o 8 b 8 8 O 8 O 8 O 8 8 ° 3 0 0 91 S 8 8 0 0 5 8 8 0 0 0 8 2 0 0 ? 0 0 91 0 0 ^ 8 3m 8 0 S 0 0 * Q S 8 « 8 8 <N 88 O O 1 5 8 g 00 S| l § a l i s 83s O 6 * 8 8 8 OO *0 |2 5! Bis 8 8 ci o 8 8 ci b r 8 8 O O ^| 5 8 8 1 0 0 "rl 18S, q 3 3 0 0 * !5S ? 11-0 0 o . o o K 0 b s s 5 Soft 8 5 ;o 8 | 8 e 8 a O 1 <u CU co J > a ea 8 CM CO t »- o 0 b § 8 8 § 8 9 " " 9 o o , o o , o" o ' ° ° 9 l 8 S t 0 0 ^ o r ! 0 0 3 Q O b ? 0 0 o 8 8 5 ob S ;9l 8 S 0 b b Q o o 3 ^ IV* ° ° 9 § 0 " 8 2 S 0 0 91 r - CO _ o o d O O " « 9 00 § b b 5 ° ° 9 l 8 8 5 o o ^ 00*1 5 R « 003 11-° ° 9 85« b b Z ^ ^ >p 00 JJ 0 0 5 r- co _ §55 •a < |oo 5| go8 Ms Ml " » a s o g d d S bb ^ b 3 9las? < b b 1 85 SS .8 o ? 8 o o ij 0 0 ^ S 3 o b o •> S 8 » o o 9 . 5 » « o 8 8 SS« b b 2| o 0 b 8 ? b b lis 00 •» ° . 8 S 0 0 9l S 8 § o o S <n V= v- ; 9 S 8 g 58 s 8 S a o 0 0 ^ — a b b 2 So g b b ^ 5 8 s 0 0 ^ 3 8 ; b b , 0 0 S S 3 b b 8.8. S o o 5 5 8 s S 8 8 0 0 *r! i l l I < => < 0 0 9l 8 8 a : b o * 8 S K b o * Q o o , b o , 8 3 ^ ° ° 9 So 3 0 0 91 t s 9 ° ° 9 8 S 5 8 8 g 00 s loo ^ i s 0 0 b 8 S S lo b ! ^ 0 0 9 | 8 8 S o O o o ; b 0 , 0 5 «, 0 0 ^ b o « 0 0 ° 8 8 sj M 2 3S<8 ° ° 9 l ° ° 9 " «o b b w a s s 00 5 S S g 3 8 -b b , a s * 8 8 8 ° o l o b o " S 8 t 28o - 8 5 ^ V rid _ . 5 8 8 0 0 T1 8 8 t b b 9 -° ° 9' < = < 1 1 1 < D TJ c S .5 o S ^ o o "8 •c 8 a s CL, CO i t s .S «3 "cL, <* a r S II c3 <» § 1 CO * " 03 fl q> -a a •§ p ca ea 3 ea TJ J5 a 2 CQ fl U O J3 -a CO u £. O •a & ej CJ flH w .fl 4) cn co CO f i r CO M H CO >> <— X) O « 12 .2 S 2 S? co .O co <<-, 2 U to .2? -§ * va w '•*•' ^ ,a> 2 is « "S > 8 I CO H-> ea op 8 fl co | _ C o co O ca fi I I J3 J3 co >/-> 8 © o "2 TJ S fi ca ea P CN — o S3 fi CJ oj I | OJ OJ •2 J2 > > i e "* e o TJ o -is o \S ca c •S 'g o <« co « CO C8 < a ~ « e 11 S .2 S a 8 f£ CO O <L> cj - f l TJ m VD CN d TJ fl ca d A <u I co CJ J 3 "ca > TJ fl ca VD cs © e CJ i S^ T J ^ -O CJ co TJ ^ S3 co ?5 ca 3 t> > 1 A) 0* s<5 ca 60 fl O * - CO „ O O Kl d «0 o o d d 1° a "Ho o «i 1 ca Q o •a -5 o 0.48 0.52 0.90 -0.86 0.92 0.38 -0.90 0.91 0.27 0.02 0.06 0.03 0.00 0.04 0.04 -0.03 0.03 0.01 0.43 0.42 0.18 0.53 0.67 0.22 0.46 0.50 0.24 0.51 0.00 0.00 0.08 0.02 o| 0.07 0.09 0.74 0.08 0.09 0.08 0.00 0.11 fi fiO 072 0.11 0.04 8. <? 0.07 0.12 8 d 5 8 a d o g 88 & ° ° ° | a ! SI 3 8 S 28 d d 8 8 s 8dfc 88 2 d o * d o g CN CN ^ do j ? 5 § 0 0 d '2 g oo si 8! £8 2 -dl 58 «M 53 a ° ° o | ^2 <o o o d d CM CM d d ° ° o v: CM d d <0 Q d d ^ - S » ° ° d 8 ° 2 , 0 0 o | CO lO _ do r ° ° d CM S d d 8.83 S8a o o 3 o o J 88 c d d V 83 § d.ScM o d d 18, ! § 8 , CM C O d o <0| ° ° d l i s , C O d o 5 8 d d o S8 fM ° ° ' d | ••- °> O O 28 & ood l do 2 2 CM CM x ^ cv ° ° d 88 d d S.CMfM 0 0 dl CM d 8§s d o g 5 i » i ° ° d ° ° d ^ 3 o o 3 8S CM o ° d | ° ° d| 83 d d 83 d d 8: '2 3 o d 3 o d CM 0 0 dl 8 8 S oo § ° ° o ^8<g 1 1 z a l l i < D < I M I 88 as K d d "J Q So d d 2 8 8 ° ° d o o § LO CM d d d8 d d i t CM ^ dor; I s - <n O O ^ 88 8 5 $ CM CO d d 8 8 -^8 & ° ° dl 2 i ° ° d | S 8 S °o 'd | 8 8 S oo § 8!* CO ° ° d - 8 8 . 8 CM o d 28JM d o o 8 8 j> Id d S8s o o S 88 j d d s • « ^ cn d d <J O l i s ° ° d ° ° d 58 !2 CM o o si 8 Sco ° ° d ss d d o d d o r ; do si d o CM« ddr: d d § ° ° o ?9 6 0 6 0 S a, o . t i ~ Q s 2 - H O 0 co CO CO co CS < «> 8 8 S CO CO > J3 O • £ "8 * ~ CD d ao, H d io. to oo, o CN O r» e e CM T - O ° CO CM s CO o a e ee e • • o lO, CM, O) CO in m ee N • • o a o> CM in d oo CM O CO O e to o e •«»• o r» e ee o> • • e e a e e e CO ao o 9 co o o e o <o co in j -<? . ee i - <? 9 d d 9 o o to o. M o CO d o>, co 00 ci to O O T O O o " • co d to CM, tO, o> a>\ CM, to o» o io ee 0) o> to en w °> cn N •* en • P O gj gj d O ii) n a N ce en ee ee • • • • O O O O co o> o d o CN d o o o o CM o oo d CN, O) CO d en d to T- CO | s . (fi o o ° o i - ee in 0 CO (fi (0 •r 9 O O (0 CO eo n en co CM d S CO d 9 8s d l a g < co J H ^ Q O £ £ u w O Oi co D > ^ Q Q § . Q Q S ^ i u S O 05 5 D > 80 Discussion Few studies have investigated coastal cutthroat trout winter ecology and behaviour at the individual level in streams of differing physical habitat types (but see Harvey et al. 1999). M u c h is known about cutthroat trout summer ecology (Moore and Gregory 1988a; Bozek and Hubert 1991; Rosenfeld et al. 2000) and movements (Harvey 1998) as most studies are done during the summer when weather conditions are more favourable for working outdoors. Other studies that have been conducted in the winter on cutthroat trout have reported abundance (Johnson et al. 1986; Swales et al. 1987) or habitat association (Bustard and Narver 1975b; Murphy et al. 1986; Heggenes et al. 1991b; Solazzi et al. 1997), but in the past have not focussed on individual fish behaviours. Recently, Harvey et al. (1999) published a paper on the winter movement of cutthroat trout in pools with and without L W D . Their study was specific to the effects of L W D on fish movement in one type of stream whereas my study focused on general cutthroat trout behaviours over a longer time frame (three seasons) in different stream types (heavily-, moderately-, and minimally-altered). A few other studies have documented cutthroat trout winter movements, however these studies were done on interior populations (Brown and M a c K a y 1995; Jakober et al. 1998) which experience different winter conditions to coastal populations. Movement During the fall when water levels were most variable, PIT tagged fish in Husdon Creek moved the least amount while fish in Avalon Creek moved the most and fish in Pipe Creek moved intermediately. The high movement rate during the fall in Ava lon Creek seemed to be influenced by a high water phenomenon, the l-in-40 year flood event. Lack of movement during the fall has been attributed to sufficient habitat availability (Tschaplinski and Hartman 1983; Young 1998). Husdon Creek has abundant overwintering habitat that cutthroat trout are known to use (deep, covered pools: Glova 1986; Heifetz et al. 1986; and sand, gravel substrates: Brown 1987; Brown and M a c K a y 1995), which could account for the relatively low movement rate in Husdon Creek during 81 the fall. The interaction between movement rate in Avalon and Pipe Creeks during the fall is directly due to four fish that moved down and out of Avalon Creek, and up into Pipe Creek. During the winter, conditions were fairly similar among the stream types: water level was high and stable and water temperatures were cold. Movement rate during the winter was minimal in all streams. Hartman (1963) found that as water level increased fish tended to become more associated with within-stream physical structures. A s well , as water temperatures decrease fish associate with the substrate (Rimmer et al. 1983) and deeper, slower habitats (Bustard and Narver 1975a). I found very little or no association with substrate in any of the streams during the winter, but there seemed to be a greater association with slower channel unit types by PIT tagged fish during the winter compared to the fall in the streams. Physiological constraints on fish during the winter (Cunjak 1988) may limit their ability to move freely through the stream. In general, different sized PIT tagged fish moved in slightly different directions and magnitudes during the three seasons. N o single size class of fish behaved the same among the streams, except for the largest size class (> 120 mm) which moved downstream during the fall and spring. During the spring these fish may have been starting their downstream migration to the ocean. The movement during the fall is opposite to what would be expected, as these fish would be returning upstream from the ocean at this time. Fish could have been moving downstream at the end of summer (Herger et al. 1996) in search of deeper water. The restricted movement paradigm states that resident fish, including cutthroat trout (Heggenes et al. 1991a), are sedentary. However recently this paradigm has been questioned (Gowan et al. 1994, Fausch and Young 1995). There is ample evidence that stream fishes are able to move long distances (Gowan et al. 1994; Adams et al. 2000; Smithson and Johnston 1999). Smithson and Johnston (1999) found that stream fishes from the Ouachita Highlands exercised both sedentary and motile hehaviours. The dispute over this paradigm stems from movement studies which have reported that the majority of marked fish are never recaptured (see Gowan et al. 1994). In this study, almost 50% of the PIT tagged fish from each stream were not recaptured. In many past 82 studies, any non-recaptures were assumed to have died, however in my study I suspected that they either died, successfully evaded capture (for eight months in this study) or moved out of the study site. Less than 3% of the recaptured fish from this study were caught in the additional traps that were set outside the study site to catch emigrating fish. Tagged fish may have moved farther outside the study area than I sampled, thereby evading recapture and causing the percentage of emigrating fish to be artificially low. Nevertheless, the low percentage of observed emigrating fish and the high number of fish that remained within small home ranges seems to provide evidence that during the fall and winter, under normal flow conditions, these coastal cutthroat trout populations are fairly sedentary. Later, I wi l l discuss the presence of the highly mobile fish present within these populations. Within-day movement of radiotagged fish was minimal (^ 5 m) in all streams. Other studies have shown similar results with cutthroat trout (mean 5 m: Heggenes et al. 1991a) and California golden trout (Oncorhynchus mykiss aguabonita) of similar size (< 5 m; Matthews 1996). Among-day movements may have occurred during the night after I finished radiotracking as there was no difference in movement among the three time periods I sampled during the day. Most PIT tagged fish moved between 0 - 25 m from their initial capture location. There was a definite non-random pattern of movement for PIT tagged fish as more fish in Husdon and Pipe Creeks moved upstream of their initial capture location. Upstream movement in Pipe Creek may be inflated by the five fish that were detected to have moved from Avalon Creek up into Pipe Creek. Selection may favour upstream movement in resident populations of fish as downstream movements could would result in those fish being lost from the genetic pool (Northcote 1992). However, fish in this study are assumed to be diadromous as there are no barriers to the ocean downstream of the study sites, so selective pressures for upstream movement may not be occurring here. Heggenes et al. (1991a) demonstrated a net downstream movement of resident cutthroat trout. More fish in Avalon Creek ended up downstream of their initial capture location. This directionality may be influenced by fish leaving Ava lon Creek to enter Pipe Creek, as well as by the l-in-40 year flood event where a large number of fish moved downstream. Movement upstream in Avalon Creek may have been 83 restricted due to a large step above the plunge pool (Figure 2.3) that may have prevented smaller fish from passing, however larger fish were still able to pass (personal observations; Adams et al. 2000). Many fish in Avalon Creek moved down over the step, and did not recover, or did not attempt to recover. This lack of movement upstream may eventually cause the elimination of cutthroat trout from the upper reaches of Ava lon Creek. Home range is the area in which an animal undertakes normal activities (Hayne 1949) and is influenced by abiotic and biotic factors (Winter 1977). Although fish were intensively sampled over an eight month period, average 95% home range size of radiotagged fish in this study was found to be fairly small in all streams (average < 50 m), a result comparable to other studies which have found cutthroat trout home range to vary in size from 4 to 20 m in length (Miller 1957; Heggenes et al. 1991a). In contrast, Colorado River cutthroat trout (O. clarki pleuriticus) have home ranges of over 900 m in length (Young 1996). M y results may differ from Young's because of the different life histories of these two closely related subspecies. Colorado River cutthroat trout are freshwater resident and inhabit large river systems allowing them to explore great distances within freshwater whereas the fish in this study are restricted to small stream systems and between stream movements would involve passing through ocean water. Ninety-five percent home range size was larger in Husdon Creek than Ava lon Creek (31.34 m and 12.2 m respectively). I suggest that radiotagged fish in Avalon Creek primarily remained within one channel unit, the plunge pool, presumably because it was the deepest and most diverse habitat available. The large home range size by Husdon Creek fish may be due to the diverse habitat structures available throughout the stream, which would allow fish to utilize a larger area. Matthews (1996) reported that California golden trout had greater cumulative movement in preferred sedge habitat over less favourable areas. During the winter, cutthroat trout are known to inhabit small streams (Swales et al. 1987) and areas of slow (Brown and MacKay 1995; Solazzi et al. 1997) and shallow water (Heggenes et al. 1991b). Pipe Creek was slower, shallower and smaller than Ava lon Creek, so radiotagged fish like Cristy presumably used Pipe Creek as an 84 overwintering area. This could account for the large average home range size within Pipe Creek. Cutthroat trout are territorial fish (Glova 1986), and so movement may be limited in streams were density is high, such as in Husdon Creek. Radiotagged fish in Husdon Creek were found throughout the stream and their movements overlapped at the edges of their home ranges. Radiotagged fish in Avalon Creek were found within the same channel unit (plunge pool) for most of the study period. Aggression (Hartman 1965; Cunjak and Power 1986) and territory size (Glova 1986) are known to collapse during the winter when water levels are cold, but the common territories in Avalon Creek were present in the fall when temperatures were well above the coldest winter temperatures. Bachman (1984) suggested that when water velocities are high, fish cannot afford to attack one another because venturing from cover may make them vulnerable to the high flow conditions. Water levels and velocity were high in Avalon Creek during the fall and winter which may have forced fish to inhabit common habitat. Contrary to the restricted movement paradigm, there were fish from each stream which made long distance movements. Cutthroat trout are known to make long distance movements during summer in coastal streams (Miller 1957; Heggenes et al. 1991a) and during winter (Brown and MacKay 1995) and summer (Young 1996) in interior streams. Six PIT tagged and one radiotagged fish in Avalon Creek and three PIT tagged and two radiotagged fish in Husdon Creek moved over 100 m from their original capture location. A s well , the 100% home range size was larger than the average 95% home range size in all streams. This means that the outlying 5% of relocations were made well outside the normal home range. Usually, long distance movements for many fish species can be attributed to movements to overwintering areas of warmer water (Brown and M a c K a y 1995) or slower, deeper habitats (Chisholm et al. 1987; Clapp et al. 1990). Long distance movements are also made before and after spawning (Clapp et al. 1990; Northcote 1992; Modde and Irving 1998). The smaller PIT tagged fish were probably not moving for reasons related to spawning, however one radiotagged fish, Mario from Pipe Creek, was in obvious breeding condition when captured, so the movement patterns made in late winter by this fish could have been due to pre-spawning behaviours. Unlike fish from 85 Avalon Creek, movement of fish in Husdon Creek did not seem to be related to changes in long or short-term flow conditions. Their movements may have been due to either random explorations, inter- or intra-specific competition with larger fish, shifts in food or habitat requirements, or preparation for the spring seaward migration (eg. Lujay). Another more encompassing mechanism to explain fish movement may be that some individuals are genetically programmed (possess a "movement gene") to move long distances from their natal areas. This phenomenon would reduce the chance of inbreeding of related individuals, reduce competition for resources among individuals and increase the spread of individuals to new areas (in the long-term would result in range expansion). Although this idea has not been previously explored, it may warrant some future consideration. Most of the long distance movements made by both PIT tagged and radiotagged fish in Ava lon and Pipe Creek occurred simultaneously during the l-in-40 year flood event in November of 1998. This flood caused a marked, immediate downstream movement by all size classes (PIT tagged and radiotagged) of cutthroat trout in Avalon Creek and downstream movement of some fish from Pipe Creek. During normal flood conditions, movement rate was not statistically different among the three stream types or with non-flood periods, however the power of these tests were low. To increase the power of the tests I needed to have a larger sample size or lower variance within my treatments (Thomas and Juanes 1996). However, the current analysis suggests that under normal seasonal flow conditions, differences in habitat type do not cause differences in movement rate of cutthroat trout. Others have found that floods do not necessarily cause displacement of other fish species (Heggenes 1988; Todd and Rabeni 1989; Harvey et al. 1999). The only statistical differences in movement rate for radiotagged fish in this study were during the l-in-40 year flood event. One hundred percent home range size in Avalon Creek was 7 times longer than 95% home range size during the fall, when the flood occurred. The outlying relocations within this stream were made during the l-in-40 year flood event when Little B i l l y , Darell and Cristy were found over 50 m downstream from their previous location. A distance of between 50 and 100 m has been found to be normal for other species of comparable size 86 (Whitworth and Strange 1983; Matthews 1996), however Little B i l l y and Darell were found consistently within a 5 m range, and immediately after their sudden downstream movement returned to their original home unit (the plunge pool). L o u L o u exceeded the return distance by 20 m. Mi l l e r (1957) suggested that fish that have been displaced during high flows wi l l return to their home location once stream conditions are suitable. Smaller PIT tagged fish did not recover their original location within the stream after the flood, rather, they remained at their new location. Their smaller size and poorer swimming ability may have prevented them from returning to their original location. The suddenness of the l-in-40 year flood followed by the simultaneous downstream movement of all size classes of fish in Avalon Creek and the loss of two fish from Avalon and Pipe Creeks suggest that these fish were displaced by the flood event. Heggenes (1988) did not find that brown trout, of comparable size, were displaced during artificial and natural flood events in an experimental stream. However, that experiment was conducted during the summer when fish presumably have more energy to allocate to swimming against fast currents, and the water temperature is presumably warm enough not to hinder swimming ability. A s well , the experimental flood events lasted 20 minutes, whereas the l-in-40 year flood event lasted for over 12 hours. Fish find refuge under rocks or L W D , but as flow and turbulence increase and persist a fish may lose the ability to maintain a constant position within the water column, so the likelihood of displacement increases. Weiss and Kummer (1999) found that brown trout in a small Austrian stream were displaced downstream following a large flood event, but they concluded that the movement downstream was caused by other factors. Downstream movements into calmer waters, such as small tributary streams or off channel habitat, during the onset of flood events has been documented before in brook trout (Salvelinus fontinalis: Chisholm et al. 1987), coho salmon (Peterson 1982) and cutthroat trout (Hartman and Brown 1987; Harvey et al. 1999). There is evidence from this study that fish in Avalon Creek did use Pipe Creek as a refuge during the major flood. Pipe Creek was mainly used as a temporary winter refuge, as there was movement back into Avalon Creek during the spring and summer once stream conditions in Avalon Creek were more favourable. 87 Stream conditions that cause displacement of individual cutthroat trout from their home ranges may have direct negative consequences on the stability of the population. Marschall and Crowder (1996) suggested that randomly occurring floods may have stronger negative affects on populations size than direct anthropogenic affects. Hoods that have a frequency of as little as 5 - 10 years can cause damage to stream habitat (from debris flows; Hartman et al. 1996) similar to the affect from the l-in-40 year flood event on Avalon Creek. Thus, displacement or loss of fish from Avalon Creek (and Pipe Creek) may occur more frequently than once every 40 years. Continual displacement and loss of individuals from the stream or sections of the stream (eg. above the step in Avalon Creek; see Figure 2.3), may be what is causing the low population density. The prediction that movement rate would be equal during stable flow conditions and different during variable flow conditions was upheld by PIT tagged fish only (Figure 2.22). Radiotagged fish showed no difference in movement rate between the seasons, however the power of the analyses were fairly low, reducing the reliability of the tests. I suspect that sampling error and low sample size may account for this result. From individual movement patterns, it is clear that radiotagged fish experienced major downstream movements, however the recorded movement was in the upstream direction. The inability to predict when flooding is going to occur and be present at each stream simultaneously during the storm event, limited the resolution of the data. A minimum of 10 fish per stream was the project goal, however there were not enough large fish caught during the study to meet that goal. For future studies I would suggest selecting a longer section of stream to sample so that (1) longer distance movements (of PIT tagged fish) can be detected and (2) the probability of catching a larger sample size of large fish is high. Growth Prediction two, that growth rate would be greatest in the minimally-altered stream during high flow condition was not met as growth rate was greatest in the moderately-altered stream (Pipe Creek) and higher during the winter and spring than fall or summer (Figure 2.22). Because growth rate was greatest in the moderately-altered 88 WINTER stable discharge high water level Minimally-Altered V = movement V = growth = association t Moderately-Altered V = movement V = growth Vx = association Heavily-Altered V = movement V = growth Vx = association FALL variable discharge lower water level Minimally-Altered V ^ movement x A. growth Vx = association t Moderately-Altered V movement x ^ growth association Heavily-Altered V T* movement x ^ growth Vx = association Figure 2.22: Outcome of predictions made at the beginning of the study. Symbols include "=", no difference between stream types; an increase or greater than the others; ", a decrease or less than the others; "AL", intermediate to the others; "V', indicates that the prediction was upheld, and "x" indicates that the prediction was incorrect. Discharge and water levels stated for each season are the actual water conditions during the study. Notice that the words in italics have been changed from the initial prediction figure (Figure 2.2) to match the actual conditions that were observed in the streams over the course of the study. 89 stream and there were no relationships between growth rate and movement, higher flow conditions (that influenced movement of fish in Avalon and Pipe Creeks) had little or no affect on growth rate. Other abiotic or biotic factors must explain the observed differences. Growth rate was equal among streams of different densities, suggesting that growth rate did not depend on fish density. Moore and Gregory (1988b) did not witness a change in growth with increased density of cutthroat trout. During the winter in all streams and during the spring in Husdon Creek, larger fish had a higher growth rate than smaller fish. These larger fish during the winter were presumably more likely to venture from cover to catch drift because during high flows smaller fish have a higher chance of displacement (Seegrist and Gard 1972; Harvey 1987); however at the highest flows, I found that all size classes of fish were equally susceptible to displacement. The water level in Pipe Creek was the lowest of the three streams throughout the study. These low flow conditions would have allowed the smaller fish to leave cover during feeding and thus have equal or higher opportunities for feeding and growth which could account for the higher overall growth rate. During the summer, growth rate in clear-cut streams is often higher compared to second growth or old growth streams because of increased light penetration causing water temperature to be warmer (Slaney et al. 1977) and invertebrate production or biomass to be higher (Wilzbach et al. 1986). Because I did not measure invertebrate biomass in this study I cannot determine i f water temperature or light penetration had an affect on food availability. It seems however that temperature directly did not influence growth rate as Husdon Creek had the warmest water temperature but lowest growth rate throughout the study. The low percentage of canopy cover in Pipe and Avalon Creeks could have extended the day length in these streams thereby allowing fish to feed for longer each day which may account for the higher growth rate in Pipe Creek (and somewhat in Ava lon Creek). Seasonal differences in growth rate and size were observed. Growth rate and average size of fish were greatest in the spring and lowest during the summer and fall and intermediate during the winter in all streams. Similar patterns of growth and fish 90 condition were shown by Ell is and Gowing (1957). They found that the pattern of fish condition was coincident with seasonal changes in water temperature. Others have found that fish have higher growth rates at particular temperature ranges (see El l i s and Gowing 1957). Lower and upper lethal temperature limits for cutthroat trout are 0.6°C and 22.8°C respectively (Meehan 1991). The preferred temperature range, or alternatively the optimal temperature range for growth of cutthroat trout may rest between 6-12°C, well within the lethal boundaries. This range is wide compared to other salmonids, such as steelhead trout which have a preferred range between 10-13°C (see Meehan 1991). Another explanation may come from Whitworth and Strange (1983) who found that brook trout and rainbow trout grew faster in the spring than other seasons. They attributed the higher spring growth rate to the potential competitive gain rainbow trout would have over brook trout. Competition between coho salmon and cutthroat trout exists (Bisson et al. 1988; Sabo and Pauley 1997) but is thought to be minimal during the winter when water levels are cold (Hartman 1965; Glova 1986). When water temperatures increase in the spring competition between coho salmon and cutthroat trout may increase, therefore cutthroat trout may grow faster to be competitive against coho salmon, as smaller cutthroat trout are known to be dominated by larger coho salmon (Sabo and Pauley 1997). Cunjak and Power (1987) found that brown trout and brook trout maintained feeding during the winter, even though metabolic rates were low. Not only was activity level of cutthroat trout in my study not different between fall and winter but fish grew during the winter in all streams implying that fish fed even though water temperatures reached were low (eg. 3.5°C Avalon Creek). Brown trout in a Michigan Creek, were more active during high flows because there were more invertebrates present in the water (Clapp et al. 1990). F low conditions and activity levels were higher in Avalon Creek, however, no relationships between these variables and food availability could be made as food availability was not measured in this study. The higher activity level in Avalon Creek could suggest that the fish were forced to swim more actively against the high flow conditions. Pipe Creek fish were the least active. Radiotagged fish in Pipe Creek were mostly confined to small channel units which may have limited the amount of small scale 91 movement these fish could do. This pattern was especially seen in winter when water levels sometimes dropped very low thereby confining fish even further. Young (1998) found that daily activity in cutthroat trout reduced by about 50% from summer to October. I found no difference in activity level between September and April. Activity level was usually greater during the night than the rest of the day. Fish are known to forage more during dusk, which could account for the observed difference (Todd and Rabeni 1989). Habitat Association In general habitat association of cutthroat trout was with slow channel unit types (pools) which corresponds with findings from winter (Murphy et al. 1986; Heggenes et al. 1991b), summer (Wilzbach and Hall 1985; Glova 1986; Bisson et al. 1988), and interior (Young 1996; Young 1998) studies on cutthroat trout. In particular, fish in Avalon Creek were associated with plunge pools. Others have found high biomass of cutthroat trout in plunge pools during the summer (Bisson et al. 1988; Herger et al. 1996), but this is the first study to document a high association with plunge pools during the winter. Differences in habitat association within specific channel units may reflect intraspecific competition among cutthroat trout. For example, three radiotagged fish in Avalon Creek were found predominately within one plunge pool. Chapman (1966) suggested that hierarchies rather than territories develop in pools, compared to shallower channel units. Within the plunge pool at any one time, the three fish were never located in the same general location, but were associated with different physical structures within the pool (eg. rootwad, open/boulder, LWD, personal observations). Nickelson et al. (1992) stated that plunge pools are not ideal habitats for fish during the winter, possibly because water velocity and turbidity increase tremendously as water level increases, thereby reducing the suitability of the habitat as a refuge. This could possibly account for the observed displacement of all but one radiotagged fish from the plunge pool during the l-in-40 year flood event. The displaced radiotagged fish were associated with either boulder or open habitat whereas the one radiotagged fish that remained in the plunge pool (Sanjie) was associated with the imbedded root-wad during the flood event. 92 There seemed to be a slight shift to slower water channel units during the winter for PIT tagged fish in the streams. A s well radiotagged fish seemed to be associated with a wider range of habitats during the fall than the winter. Monthly and seasonal shifts in habitat association can be caused by several factors. Herger et al. (1996) found that a shift in habitat association from July to August by Colorado cutthroat trout was related to changes in flow. Changes in water depth can alter the number of discrete channel units available to fish (Herger et al. 1996). Water levels fluctuated monthly and daily in each stream which may account for the variability in habitat association. Other factors that could cause monthly and seasonal changes in habitat association include inter- or intra-specific competition, localized changes in food availability (Wilzbach 1985) or changes to the configuration of microhabitat structures (Meyer and Griffith 1997). In general, cutthroat trout were associated with boulder cover, even in Husdon Creek where boulders accounted for < 3% of the total available cover. Boulders provide micro-scale refuges against high flow conditions (Heggenes 1988), and many species of fish use boulders as cover during normal flow conditions (Meyer and Griffith 1997) and during flood conditions (Hillman et al. 1987). However during major flood events cobbles, boulders and L W D can get transported downstream (Major 1997). This happened in Avalon and Pipe Creeks during the l-in-40 year flood event when large amounts of L W D , boulders and cobbles moved downstream scouring out the streams and rearranging the path of water. The effectiveness of boulders as refuge is eliminated when they are transported downstream during flood events. Association with boulders in Ava lon and Pipe Creeks may have added to the large number of fish that were displaced during the 1-in-40 year flood. The relationship between L W D and fish movement (Sundbaum and Naslund 1997; Harvey et al. 1999), and the usefulness of L W D as cover (Crispin et al. 1993; Heifetz et al. 1986; Fausch and Northcote 1992) has been extensively studied and is widely understood. Large woody debris was used disproportionately in all months and seasons in Husdon Creek. Large woody debris did not seem to play as an important roll for cutthroat trout as would be expected, although Berg et al. (1998) found that L W D was not influential as cover in Central Sierra Nevada, California. Large woody debris was largely 93 ignored by cutthroat trout in Avalon and Pipe Creeks possibly because it was unstable and not as abundant as boulders, which may have made it less desirable as cover. It seems as though cutthroat trout within the three streams were associated with physical cover that was most abundant within the stream. Substrate association again reflected availability of substrate type, excluding boulder substrate. Cutthroat trout were not preferentially associated with a particular substrate size class, except for large gravel in Pipe Creek. This may be because fish were mostly associated with a combination of substrate size classes, not one discrete size class. M y sampling method did not allow me to test habitat association with combinations of substrate size classes. Population Density and Probability of Survival A report conducted in the early 1980's stated that Avalon Creek was an important stream for salmon rearing and was one of the best locations in Howe Sound to angle cutthroat trout (Environment and Land Use Committee Secretariat 1980). Since 1980 Ava lon Creek has had extensive logging in its watershed which is now prone to landslides and debris flows (Ian Gazeley, local resident, Sunshine Coast, B . C . , P.O. Box 1318, Gibsons, B . C . , V O N 1V0, personal communication). In this study, population density was lowest in Avalon Creek compared to the other streams. Ell iot (1986) found that larger Dol ly Varden (Salvenlinus malma) emigrated from a stream that had high L W D removal. Large cutthroat trout from Avalon Creek may have left in search of a more stable stream after the first major debris flow (post 1983). This emigration may have initially reduced the effective population size in Avalon Creek, while displacement during subsequent debris flows (like the one that occurred in 1998) may be maintaining the density at a low level. Seegrist and Gard (1972) found that densities of rainbow trout and brook trout were depressed resulting from floods and that these populations were at low levels for several subsequent years. Isolated populations have a higher chance of extinction from localized disturbances like floods (Jones et al. 1998). Re-colonization into Ava lon Creek could occur via the estuary, as there are no streams that Ava lon Creek flows into, or flow into it (other than Pipe Creek), making Avalon Creek fairly isolated. 94 Bisson et al. (1988) found that cutthroat trout immigrated into a heavily altered stream that was devastated by the Mount St. Helen's eruption of 1980 because the stream held a large invertebrate population. Invertebrate production was not evaluated in this study, however i f the invertebrate density was relatively high in Avalon Creek, it may explain why a population still remains in this stream even though the habitat is heavily-altered and subject to debris flows. Emigration during the winter in search of temporary refuge in Avalon Creek could also explain why that stream had a low density of cutthroat trout During the winter, density within clear-cut streams is often reduced compared to summer as fish migrate to other streams in search of better habitat (Johnson et al. 1986). Ten percent of recaptured PIT tagged and radiotagged fish moved out of Avalon Creek into Pipe Creek, especially during the l-in-40 year flood, which may have influenced the density estimate in Ava lon Creek. Pipe Creek had relatively less variable and lower water levels than Avalon Creek, making it a less harsh environment. During the summer, Pipe Creek may not be as favourable because water levels return to low levels, thereby limiting the available habitat for fish (Herger et al. 1996). In the spring two fish moved from Pipe Creek back into Avalon Creek possibly in anticipation of low water levels. Probability of survival estimates were surprisingly high and equal among stream types, which is contrary to another study that showed higher mortality rate during the winter in a logged stream compared to an unlogged stream (0.62 and 0.24 respectively, Osborn 1980). Michael (1989) found overwinter survival of cutthroat trout to range between 40% and 62% in two Washington State streams while Martin et al. (1986) found juvenile coho salmon overwinter mortality in a devastated stream lacking streamside cover to reach as high as 97% and 100%. 95 G E N E R A L CONCLUSION The restricted movement paradigm states that resident fish are sedentary and movement is limited to small sections of stream. This study supports the claim by Smithson and Johnston (1999) that both non-mobile and mobile individuals exist within a population. Fausch and Young (1995) suggest that a more complex classification of movement be developed to accurately define movement within and among fish populations, rather than sedentary versus mobile. A l l individual fish are capable of long distance movements, however the reason or mechanism responsible for a fish to move may be what causes the different observed behaviours. Voluntary (habitat exploration, competition) versus involuntary movements (displacement, genetic) would be more appropriate terms for addressing sedentary versus mobile observations, as some fish from this study that would have been classified as sedentary (small home range) experienced large scale movements due to displacement which artificially increased their home range size. Cutthroat trout have been declining in population abundance across their range for the past 20 years (Trotter 1989). Habitat loss in the freshwater environment is only one possible mechanism for the decline. Overfishing and changing marine conditions (leading to poor marine survival) are other mechanisms, however their contribution to the overall decline is largely unknown. I investigated habitat loss, primarily due to logging practices because cutthroat trout spend large portions of their lives within the freshwater environment. I initially hypothesized that differences in movement between streams of differing habitat alteration influenced population density. In general, I found that the winter ecology of coastal cutthroat trout was similar to summer in all stream types. Movement rate of cutthroat trout within streams of differing physical habitat was low during the winter when stream flows were high and water temperatures were cold. Even during regular seasonal flood events movement rate was similar to non-flood periods in all stream types. The main difference observed in this study was during a major storm event where many fish from the heavily-altered stream were displaced downstream. This displacement could have been caused by the lack of suitable refuge available as these fish 96 were primarily associated with boulders. Boulders are transported downstream during extreme high flow conditions eliminating their usefulness as refuge by exposing fish to high velocities or causing death due to crushing. Although extreme weather conditions are rare and random (Marschall and Crowder 1996), they may occur often enough to have a negative affect on fish population density and the ability of a population to persist, even when average probability of survival and growth rate are not affected. Unlike interior systems, coastal streams are subject to flooding during the colder months of the year when fish are presumably most vulnerable to added energy costs. A s well , coastal streams that house cutthroat trout are often not part of a larger river system, but rather drain directly into the ocean. Generally, cutthroat trout do not venture far from their natal stream, so fish living in small semi-isolated streams have a higher chance of extinction i f the risk of extinction exists. L o w density, isolated populations in unstable streams, such as that in Avalon Creek, are the most vulnerable to stochastic events that can lead to loss of individuals, or ultimately extinction. Habitat degradation that causes changes to the stability and normal flow patterns within coastal streams often result from poor logging practices. The combination of the known affects of logging on streams and the knowledge that fish can become displaced during major floods provides evidence that populations could be maintained at low density. I believe that the first major debris flow that occurred within Avalon Creek caused an initial reduction in population density. Over time the density has either been reduced more, or maintained at a low level from repeated displacement events during subsequent floods. The displacement of individuals from the population during major flooding may be one mechanism behind the decline of within population density for cutthroat trout in B C . This is yet more evidence that land management and stream habitat restoration projects are worthy conservation practices for coastal cutthroat trout. The method of using a large-scale study as the baseline for an intensive small-scale study was effective in this thesis. I used the habitat gradient generated from Chapter One as an unbiased guideline for selecting streams for Chapter Two. I feel that the results obtained in Chapter Two are a fairly realistic representation of what is occurring in small streams along the southwest coast of B C . 97 LITERATURE CITED Adams, N .S . , D . W . Rondorf, S.D. Evans, and J.E. Ke l ly . 1998a. Effects of surgically and gastrically implanted radio transmitters on growth and feeding behaviour of juvenile chinook salmon. Transactions of the American Fisheries Society. 127: 128-136. Adams, N .S . , D . W . Rondorf, S.D. Evans, J .E. Ke l ly , and R . W . Perry. 1998b. Effects of surgically and gastrically implanted radio transmitters on swimming performance and predator avoidance of juvenile chinook salmon (Oncorhynchus tshawytscha). Canadian Journal Of Fisheries A n d Aquatic Sciences. 55: 781-787. Adams, S.B, C A . Frissell, and B . E . Rieman. 2000. Movements of nonnoative brook trout in relation to stream channel slope. Transactions Of The American Fisheries Society. 623-638. Allendorf, F . W . , D.Bayles, D . L . Bottom, K . P . Currens, C A . Frissell, D . Hankin, J .A . Lichatowich, W . Nehlsen, P .C. Trotter, and T . H . Williams. 1997. Prioritizing Pacific salmon stocks for conservation. Conservation Biology. 11(1): 140-152. Anderson, P . G . 1998. Sediment generation from foresty operations and associated effects on aquatic ecosystems. Pages 491-508 in M . K . Brewin and D . M . A . Monita, tech. coords. Forest-fish conference: land management pracices affecting aquatic ecosystems. Proc. Forest-Fish Conf., May 1-4, 1996, Calgary, Alberta. Nat. Resour. Can., Can. For. Serv., North. For. Cent., Edmonton, Alberta. Inf. Rep. N O R - X - 3 5 6 . Bachman, R . A . 1984. Foraging behaviour of free-ranging wi ld and hatchery brown trout in a stream. Transactions of the American Fisheries Society. 113 (1): 1-32. Benke, R.J . 1992. Native trout of western North America. American Fisheries Society Monograph 6. Berg, N . , A . Carlson, and D . Azuma. 1998. Function and dynamics of woody debris in stream reaches in the central Sierra Nevada, California. Canadian Journal of Fisheries and Aquatic Sciences. 55: 1807-1820. Berg, O . K . , and G . Bremset. 1998. Seasonal changes in the body composition of young riverine Atlantic salmon and brown trout. Journal of Fish Biology. 52: 1272-1288. Bergstedt, L . C . , and E.P. Bergersen. 1996. Health and movement of fish in response to sediment sluicing in the Wind river, Wyoming. Canadian Journal of Fisheries and Aquatic Sciences. 54: 312-319. B i lby , R . E . , and J .W. Ward. 1991. Characteristics and function of large woody debris in streams draining old-growth, clear-cut and second-growth forets in southwestern 98 Washington. Canadian Journal of Fisheries and Aquatic Sciences. 48: 2499-2508. Bisson, P.A., K . Sullivan, and J.L. Nielsen. 1988. Channel hydraulics, habitat use, and body form of juvenile coho salmon, steelhead, and cutthroat trout in streams. Transactions of the American Fisheries Society. 117: 262-273. Boisclair, D. 1992. Relationship between feeding and activity rates for activity foraging juvenile brook trout (Salvelinus fontinalis). Canadian Journal Of Fisheries And Aquatic Sciences. 49: 2566-2573. Bozek, M . A . , and W.A. Hubert. 1991. Segregation of resident trout in streams as predicted by three habitat dimensions. Canandian Journal Of Zoology. 70: 886-890. Bradford, M.J . , G.C. Taylor, and J. A . Allan. 1997. Empirical review of coho salmon smolt abundance and the prediciton of smolt produciton at the regional level. Transactions of American Fisheries Society. 126: 49-64. Brown, R.S., and W.C. MacKay. 1995. Fall and winter movements of and habitat use by cutthroat trout in the Ram River, Alberta. Transactions of the American Fisheries Society. 124: 873-885. Brown, T.G. 1987. Characterization salmonid over-wintering habitat within seasonally flooded land on the Carnation Creek flood-plain. British Columbia Ministry of Forests. Bustard, D.R., and D.W. Narver. 1975a. Aspects of the winter ecology of juvenile coho salmon (Oncorhynchus kisutch) and steelhead trout (Salmo gairdneri). Journal of Fisheries Research Board Canada. 32. 667-680. 1975b. Preferences of juvenile coho salmon (Oncorhynchus kisutch) and cutthroat trout (Salmo clarki) relative to simluated alteration of winter habitat. Journal of Fisheries Research Board. 32: 681-687. Campton, D.E, and F . M . Utter. 1987. Genetic structure of anadromous cutthroat trout (Salmo clarki clarki) populations in the Puget Sound area: edivence for resticted gene flow. Canadian Journal Of Fisheries And Aquatic Sciences. 44(3): 573-582. Chapman, D.W. 1966. Food and space as regulators of salmonid populations in streams. American Naturalist. 100: 345-357. Chisholm, I .M., W.A. Hubert, and T A . Wesche. 1987. Winter stream conditions and use of habitat by brook trout in high-elevation Wyoming streams. Transactions of the American Fisheries Society. 116: 176-184. 99 Clapp, D.F., R.D. Clark Jr., and J.S. Diana. 1990. Range, activity, and habitat of large, free-ranging brown trout in a Michigan stream. Transactions of the American Fisheries Society. 119: 1022-1034. Connolly, P.J., and J.D. Hall. 1999. Biomass of coastal cutthroat trout in unlogged and previously clear-cut basins in the central coast range of Oregon. Transactions Of The American Fisheries Society. 128: 890-899. Cook, M.F. , and E.P. Bergersen. 1988. Movements, habitat selection, and activity periods of northern pike in Eleven Mile Reservoir, Colorado. Transactions of the American Fisheries Society. 117: 495-502. Crispin, V . , R. House, and D. Roberts. 1993. Changes in instream habitat, large woody debris, and salmon habitat after the restructuring of a coastal Oregon stream. North American Journal Of Fisheries Management. 13: 96-102. Cunjak, R.A. 1988. Physiological consequences of overwintering in streams: the cost of acclimitization. Canadian Journal of Fisheries and Aquatic Sciences. 45: 443-452. Cunjak, R.A. , and G. Power. 1986. Winter habitat utilization by stream resident brook trout (Salvenlinus fontinalis) and brown trout (Salmo trutta). Canadian Journal of Fisheries and Aquatic Sciences. 43: 1970-1981. 1987. The feeding and energetics of stream-resident trout in winter. Journal of Fish Biology. 31: 493-511. De Leeuw, A .D . , and K . M . Stuart. 1981. Small stream enhancement possibilities for sea-run cutthroat trout in the Lower Mainland and Sechelt Peninsula. British Columbia Ministry of Environment, i : 99. Elliott, S.T. 1986. Reduction of a Dolly Varden population and macrobenthos after removal of logging debris. Transactions of the American Fisheries Society. 115: 392-400. Ellis, R.J., and H. Gowing. 1957. Relationship between food supply and condition of wild brown trout, Salmon trutta Linnaeus, in a Michigan stream. Limnology and Oceanography. 2(4): 299-308. Environment and Land Use Committee Secretariat. 1980. Howe Sound Overview Study. Province of British Columbia. Fausch, K . D . , and T.G. Northcote. 1992. Large woody debrs and samonid habitat in a small coastal British Columbia stream. Canadian Journal of Fisheries and Aquatic Sciences. 49: 682-693. 100 Fausch, K . D . , and M . K . Young. 1995. Evolutionary significant units and movement of resident stream fishes: a cautionary tale. American Fisheries Society Symposium. 17: 360-370. Giberson, D.J., and D. Caissie. 1998. Stream habitat hydraulics: interannual variability in three reaches of Catamaran Brook, New Brunswick. Canadian Journal of Fisheries and Aquatic Sciences. 55(2): 485-494. Glova, G.J. 1986. Interaction for food and space between experimental populations of juvenile coho salmon (Oncorhynchus kisutch) and coastal cutthroat trout (Salmo clarki) in a laboratory stream. Hydrobiologia. 131: 155-168. Gore, J.A. 1996. Discharge measurements and streamflow analysis. In Chapter 3 of Methods in Stream Ecology. Ed. Hauer, F.R., and Lamberti G.A. Academic Press. Gowan, C , M . L . Young, K . D . Fausch, and S.C. Riley. 1994. Restricted movement in resident stream salmonids: a paradigm lost? Canadian Journal of Fisheries and Aquatic Sciences. 51: 2626-2637. Griffiths, D. 1997. Local and regional species richness in North American lacustrine fish. Journal Of Animal Ecology. 66: 49-56. Hall, J.D., and R.L. Lantz. 1969. Effects of logging on the habitat of coho salmon and cutthroat trout in coastal streams, p. 355-375. In T.G. Northcote [ed.] Symposium on salmon and trout in streams, H.R. MacMillian Lectures in Fisheries, University of British Columbia, 388 p. 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 Research Board of Canada. 22(4): 1035-1078. Hartman, G.F., and C A . Gil l . 1968. Distribution of juvenile steelhead and cutthroat trout (Salmo gairdneri and (S. clarki clarki) within streams in Southwestern British Columbia. Journal of Fisheries Research Board. 25(1): 33-48. Hartman, G.F., and J.C. Shrivener. 1990. Impacts of forestry practices on a coastal stream ecosystem, Carnation Creek, British Columbia. Chapter 8. Canadian Bulletin of Fisheries and Aquatic Sciences. 223: 95-122. Hartman, G.F., J.C. Scrivener, and M.J. Miles. 1996. Impacts of logging in Carnation Creek, a high-energy coastal stream in British Columbia, and their implications for restoring fish habitat. Canadian Journal of Fisheries and Aquatic Sciences. 53(Suppl. 1): 237-251. Harvey, B.C. 1987. Susceptibility of young-of-year fishes to downstream displacement by flooding. Transactions of the American Fisheries Society. 116: 851-855. 101 Harvey, B.C. , R.J. Nakamoto, and J.L. White. 1999. Influence of large woody debris and a bankfull flood on movement of adult resident coastal cutthroat trout 'Oncorhynchus clarki) during fall and winter. Canadian Journal of Fisheries and Aquatic Sciences Canadian Journal of Fisheries and Aquatic Sciences. 56: 2161-2166. Hayne, D.W. 1949. Calculation of size of home range. Journal of Mammology. 30(1): 1-17. Heggenes, J. 1988. Effects of short-term flow fluctuations on displacement of, and habitat use by, brown trout in a small stream. Transactions of the American Fisheries Society. 117: 336-344. Heggenes, J., T.G. Northcote, and A . Peter. 1991a. Spatial stability of cutthroat trout (Oncorhynchus clarki) in a small, coastal stream. Canadian Journal of Fisheries and Aquatic Sciences. 48: 757-762. 1991b. Seasonal habitat selection and preferences by cutthroat trout (Oncorhynchus clarki) in a small, coastal stream. Canadian Journal of Fisheries and Aquatic Sciences. 48: 1364-1370. Heifetz, J. M . L . Murphy, and K . V . Koski. 1986. Effects of logging on winter habitat of juvenile salmonids in Alaskan streams. North American Journal of Fisheries Management. 6: 52-58. Herger, L .G. , W.A. Hubert, and M . K . Young. 1996. Comparison of habitat composition and cutthroat trout abundance at two flows in small mountain streams. North American Journal of Fisheries Management. 16: 294-301. Hilderbrand, R.H. , A . D . Lemly, and C A . Dolloff. 1999. Habitat sequencing and the importance of discharge in inferences. North American Journal Of Fisheries Management. 19: 198-202. Hillman, T.W., J.S. Griffith, and W.S. Platts. 1987. Summer and winter habitat selection by juvenile chinook salmon in a highly sedimented Idaho stream. Transactions of the American Fisheries Society. 116: 185-195. Hinch, S. 1991. Small- and large-scale studies in fisheries ecology: the need for cooperation among researchers. Fisheries. 16(3): 22-27. Holtby, L . B . 1988. Effects of logging on stream temperatures in Carnation Creek, British Columbia, and associated impacts on the coho salmon (Oncorhynchus kisutch). Canadian Journal of Fisheries and Aquatic Sciences. 45: 502-515. Jakober, M.J . , T.E. McMahon, R.F. Thurow, and C G . Clancy. 1998. Role of stream ice on fall and winter movements and habitat use by bull trout and cutthroat trout in 102 Montana headwater streams. Transactions of the American Fisheries Society. 127: 223-235. Johnson, S.W., J. Keifets, and K . V . Koski. 1986. Effects of logging on the abundance and seasonal distribution of juvenile steelhead in some southeastern Alaska streams. North American Journal of Fisheries Management. 6: 532-537. Jones, K . K . , J .M. Dambacher, B.G. Lovatt, A .G . Talabere, and W. Bowers. 1998. Status of Lahontan cutthroat trout in the Coyote Lake Basin, southeast Oregon. North American Journal Of Fisheries Management. 18: 308-317. Jowett, I.G., J. Richardson, and R . M . McDowall. 1996. Relative effects of in-stream habitat and land use on fish distribution and abundance in tributaries of the Grey River, New Zealand. New Zealand Journal of Marine and Freshwater Research. 30: 462-475. Krebs, C.J. 1989. Ecological Methodology. Harper Collins. New York. Lacasse, S., and P. Magnan. 1992. Biotic and abiotic determinants of the diet of brook trout, Salvelinus fontinalis, in lakes of the Laurentian Shield. Canadian Journal of Fisheries and Aquatic Science. 49: 1001-1009. Lamberti, G.A., S.V. Gregory, L.R. Ashkenas, R.C. Wildman, and K.M.S . Moore. 1991. Stream ecosystem recovery following a catastrophic debris flow. Canadian Journal of Fisheries and Aquatic Sciences. 48: 196-208. Lester, P.J., S.F. Mitchell, and D. Scott. 1994. Effects of riparian willow trees (Salix fragilis) on macroinvertebrate densities in two small Central Otago, New Zealand, streams. New Zealand Journal of Marina and Freshwater Research. 28: 267-276. Magnan, P., M . A . Rodriguez, P. Legendre, and S. Lacasse. 1994. Dietary variation in a freshwater fish species: relative contributions of biotic interactions, abiotic factors, and spatial structures. Canadian Journal of Fisheries and Aquatic Science. 51: 2856-2865. Major, J.J. 1997. Depositional processes in large-scale debris-flow experiments. The Journal of Geology. 105: 345-366. Maret, T.R., C.T. Robinson, and G.W. Minshall. 1997. Fish Assemblages and environmental correlates in least-disturbed streams of the Upper Snake River Basin. Transactions Of The American Fisheries Society. 126: 200-216. Marschall, E.A. , and L .B . Crowder. 1996. Assessing population responses to multiple anthrompogenic effects: a case study with brook trout. Ecological Applications. 6(1): 152-167. 103 Martin, D.J., L.J. Wasserman, and V . H . Dale. 1986. Influence of riparian vegetation on posteruption survival of coho salmon fingerlings on the west-side streams of Mount St. Helens, Washington. North American Journal Of Fisheries Management. 6:1-8. Martin, S.W., J.A. Long, and T.N. Pearsons. 1995. Comparison of survival, gonad development, and growth between rainbow trout with and without surgically implanted dummy radio transmitters. North American Journal Of Fisheries Management. 15: 494-498. Mason, J.C. 1976. Response of underyearling coho salmon to supplemental feeding in a natural stream. Journal Of Wildlife Management. 40(4): 775-788. Matthews, K.R. 1996. Habitat selection and movement patterns of California golden trout in degraded and recovering stream sections in the Golden Trout Wilderness. California. North American Journal of Fisheries Management. 16: 579-590. McMahon, T.E., and G.F. Hartman 1989. Influence of cover complexity and current velocity on winter habitat use by juvenile coho salmon (Oncorhynchus kisutch). Canadian Journal of Fisheries and Aquatic Sciences. 46: 1551-1557. Meehan, W.R., editor. 1991. Influences of forest and rangeland management on salmonid fishes and their habitats. American Fisheries Society Special Publication 19. Meyer, K . A . , and J.S. Griffith. 1997. Effects of cobble-boulder substrate configuration on winter residency of juvenile rainbow trout. North American Journal Of Fisheries Management. 17: 77-84. Michael, J.H. 1989. Life history of anadromous coastal cutthroat trout in Snow and Salmon Creeks, Jefferson County, Washington, with implications for management. California Fish and Game. 75(4): 188-203. Miller, R.B. 1957. Permanence and size of home territory in stream-dwelling cutthroat trout Journal of Fisheries Research Board. 14: 687-691. Ministry of Public Works and Government Services Canada. 2000. Canada's national programme of action for the protection of the marine environment from land-based activities (NPA). Government of Canada. Modde, T., and D. B. Irving. 1998. Use of multiple spawning sites and seasonal movement by razorback suckers in the Middle Green River, Utah. North American Journal Of Fisheries Management. 18: 318-326. Moore, K .M.S . , and S.V. Gregory. 1988a. Summer habitat utilization and ecology of cutthroat trout fry (Salmo clarki) in Cascade Mountain streams. Canadian Journal of Fisheries and Aquatic Sciences. 45: 1921-1930. 104 1988b. Response of young-of-the-year cutthroat trout to manipulation of habitat structure in a small stream. Transactions of the American Fisheries Society. 117: 162-170. Murphy, M . L . , and J.D. Hall. 1981. Varied effects of clear-cut logging on predators and their habitat in small streams of the Cascade Mountains, Oregon. Canadian Journal of Fisheries and Aquatic Sciences. 38: 137-145. Murphy, M . L . , J. Heifetz, S.W. Johnson, K . V . Koski, and J.F. Thedinga. 1986. Effects of clear-cut logging with and without buffer strips on juvenile salmonids in Alaskan streams. Canadian Journal of Fisheries and Aquatic Sciences. 43: 1521-1533. Murphy, M . L . , and K . V . Koski. 1989. Input and depletion of woody debris in Alaska streams and implications for streamside management. North American Journal Of Fisheries Management. 9: 427-436. Murphy, M . L . , J. Heifetz, J.F. Thedinga, S.W. Johnson, and K . V . Koski. 1989. Habitat utilization by juvenile Pacific salmon (Oncrhynchus) in the Glacial Taku River, southeast Alaska. Canadian Journal Of Fisheries And Aquatic Sciences. 46: 1677-1685. Naslund, I., E. Degerman, and F. Nordwall. 1998. Brown trout (Salmo trutta) habitat use and life history in Swedish streams: possible effects of biotic interactions. Canadian Journal of Fisheries and Aquatic Science. 55: 1034-1042. Nehlsen, W., J.E. Williams, and J.A. Lichatowich. 1997. Pacific salmon at the crossroads: stocks at risk from California, Oregon, Idaho, and Washington. Fisheries. 16: 4-21. Nickelson, T.E., M.F. Solazzi, S.L. Johnson, and J.D. Rodgers. 1992. Effectiveness of selected stream improvement techniques to create suitable summer and winter rearing habitat for juvenile coho salmon (Oncorhynchus kisutch) in Oregon coastal streams. Canadian Journal of Fisheries and Aquatic Sciences. 49: 790-794. Northcote, T.G. 1992. Migration and residency in stream salmonids - some ecological considerations and evolutionary consequences. Nordic Journal of Freshwater Research. 67: 5-17. Osborn J.G. 1980. Effects of logging on resident and searun populations of cutthroat trout (Salmo Clarki) in small tributaries of the Clearwater River, Jefferson County, Washington, 1978-1979. U of W. College of Fisheries. Fish. Res. Inst. 8018. Pearson, T.N. , H.W. L i , and G.A. Lamberti. 1992. Influence of habitat complexity on resistance to flooding and resilience of stream fish assemblages. Transactions of the American Fisheries Society. 121: 427-436. 105 Peterson, N.P. 1982. Immigration of juvenile coho salmon (Oncorhynchus kisutch) into riverine ponds. Canadian Journal of Fisheries and Aquatic Sciences. 39: 1308-1310. Platts, W.S., R.J. Torquemada, M . L . McHenry, and C.K. Graham. 1989. Changes in salmon spawning and rearing habitat from increased delivery of fine sediment to the south fork Salmon river, Idaho. Transactions of the American Fisheries Society. 118: 274-283. Prentice, E.F., T.A. Flagg, C.S. McCutcheon, D.F. Braistow, and D.C. Cross. 1990. Equipment, methods, and an automated data-entry station for PIT tagging. Pages 335-340 in N.C. Parker and five coeditors. Fish-marking techniques. American Fisheries Society, Symposium 7, Bethesda, Maryland. Priede, I.G. 1985. Metabolic scope in fishes. In: Tytler P, and Calow P. Fish Energetics., Chapter Two. Croom Helm, Great Britain. Quinn, T.P., and N.P. Peterson. 1996. The influence of habitat complexity and fish size on over-winter survival and growth of individually marked juvenile coho salmon (Oncorhynchus kisutch) in Big Beef Creek, Washington. Canadian Journal of Fisheries and Aquatic Science. 53: 1555-1564. Ralph, S.C., G.C. Poole, L .L . Conquest, and R.J. Naiman. 1994. Stream channel morphology and woody debris in logged and unlogged basins of western Washington. Canadian Journal of Fisheries and Aquatic Sciences. 51: 37-51. Ricks, C L . 1995. Effects of channelization on sediment distribution and aquatic habitat at the mouth of Redwook Creek, Northwestern California. US Geological survey professional paper 1454-Q. Q1-Q17. Rimmer, D . M . , U . Paim, and R.L. Saunders. 1983. Autumnal habitat shift of juvenile Atlantic salmon (Salmo solar) in a small river. Canadian Journal of Fisheries and Aquatic Sciences. 40: 671-680. Rosenfeld, J.S., and J.J. Hudson. 1997. Primary production, bacterial production, and invertebrate biomass in pools and riffles in southern Ontario streams. Archiv Hydrobiologia. 139(3): 301-316. Rosenfeld, J.S., M . Porter, and E Parkinson. 2000. Habitat factors affecting the abundance and distribution of juvenile cutthroat trout (Oncorhynchus clarki) and coho salmon (Oncorhynchus kisutch). Canadian Journal of Fisheries and Aquatic Management. 57:166-11A. Sabo, J.L., and G.B. Pauley. 1997. Competition between stream-dwelling cutthroat trout (Oncorhynchus clarki) and coho salmon (O. kisutch): effects of relative size and 106 population origin. Canadian Journal of Fisheries and Aquatic Sciences. 54: 2609-2617. Sandercock, F .K. 1991. Life history of coho salmon (Oncorhynchus kisutch). In Pacific salmon life histories, ed. C. Groot, and L. Margolis. UB.C. Press. Schnute, J. 1983. A new approach to estimating populations by the removal method. Canadian Journal of Fisheries and Aquatic Sciences. 40: 2153-2169. Scott, W.B., and E.J. Crossman. 1985. Freshwater fishes of Canada. Fisheries Research Board of Canada. Seber, G.A.F. , and E.D Le Cren. 1967. Estimating population parameters from catches large relative to the population. Journal Of Animal Ecology. 36: 631-643. Seegrist, D.W. and R. Gard. 1972. Effects of floods on trout in Sagehen Creek, California. Transactions Of The American Fisheries Society. 3: 478-482. Sierra Legal Defence Fund. 1997. Stream protection under the code: the destruction continues. Sierra Legal Defence Fund. Slaney, P.A., T.G. Halsey, and H.A. Smith. 1977. Some effects of forest harvesting on salmonid rearing habitat in two streams in the central interior of British Columbia. Fisheries Management Report No. 17. Slaney, T.L. , K . D . Hyatt, T.G. Northcote, and R.J. Fielden. 1996. Status of anadromous salmon and trout in British Columbia and Yukon. Fisheries. 21: 20-35. Smale, M . A . , and C F . Rabeni. 1997. Influences of hypoxia and hyperthermia on fish species composition in headwater streams. Transactions of the American Fisheries Society. 124:711-725. Smithson, E.B. , and C.E. Johnston. 1999. Movement patterns of stream fishes in a Ouachita highlands stream: an examination of the restricted movement paradigm. Transactions Of The American Fisheries Society. 128: 847-853. Snedden, G.A., W.E. Kelso, and D.A. Rutherford. 1999. Diel and seaonal patterns of spotted gar movement and habitat use in the Lower Atchafalaya river Basin, Louisiana. Transactions of the American Fisheries Society. 128: 144-154. Solazzi, M.F. , T.E. Nickelson, S.L. Johnson, S. and van de Wetering. 1997. Juvenile sea-run cutthroat trout: habitat utilization, smolt production, and response to habitat modification. Pages 148-150 in J.D. Hall, P.A. Bisson, and R.E. Gresswell, editors. Sea-run cutthroat trout: biology, management, and future conservation. Oregon Chapter, American Fisheries Society, Corvallis. 107 Sundbaum, K . , and I. Naslund. 1997. Effects of woody debris on the growth and behaviour of brown trout in experimental stream channels. Canandian Journal Of Zoology. 76: 56-61. Swales, F.C., J.R. Irvine, and C D . Levings. 1987. Overwintering habitats of coho salmon (Oncorhynchus kisutch) and other salmonids in the Keogh River system, British Columbia. Canandian Journal of Zoology. 66: 254-261. Taylor, E.B. 1991. Behavioural interaction and habitat use in juvenile chinook, Oncorhynchus tshawytscha, and coho, O. kisutch, salmon. Animal Behaviour. 42: 729-744. Taylor, C M . , and P.W. Lienesch. 1995. Environmental correlates of distribution and abundance for Lythrurus snelsoni: a range-wide analysis of an endemic fish species. The Southwestern Naturalist. 40(4): 373-378. Thedinga, J.F., M . L . Murphy, J. Heifetz, K . V . Koski, and S.W. Jonhson. 1989. Effects of logging on size and age composition of juvenile coho salmon (Oncorhynchus kisutch) and density of presmolts in southeast Alaska streams. Canadian Journal Of Fisheries And Aquatic Sciences. 46: 1383-1391. Thomas, L . , and F. Juanes. 1996. The importance of statistical power analysis: an example from Animal Behaviour. Animal Behaviour. 52: 856-859. Todd, B.L. , and C F . Rabeni. 1989. Movement and habitat use by stream dwelling smallmouth bass. Transactions of the American Fisheries Society. 118: 229-242. Trotter, P . C 1989. Coastal cutthroat trout: a life history compendium. Transactions of the American Fisheries Society. 118: 463-473. Tschaplinski, P.J., and G.F. Hartman. 1983. Winter distribution of juvenile coho slamon (Oncorhynchus kisutch) before and after logging in Carnation Creek, British Columbia, and some implications for overwinter survival. Canadian Journal of Fisheries and Aquatic Sciences. 40: 452-461. Watson, G., and T.W. Hillman. 1997. Factors affecting the distribution and abundance of bull trout: an investigation at hierarchical scales. North American Journal of Fisheries Management. 17(2): 237-252. Weiss, S., and H. Kummer. 1999. Movement of wild and experimentally stocked brown trout, Salmo trutta L . , before and after a flood in a small Austrian stream. Fisheries Management and Ecology. 6: 34-45. Whitworth, W.E., and R.J. Strange. 1983. Growth and production of sympatric brook and rainbow trout in an Applachian stream. Transactions of the American Fisheries Society. 112:469-475. 108 Wilzbach, M . A . 1985. Relative roles of food abundance and cover in determining the habitat distribution of stream-dwelling cutthroat trout (Salmo clarki). Canadian Journal of Fisheries and Aquatic Sciences. 42: 1668-1672. Wilzbach, M . A . , and J.D. Hal l . 1985. Prey availability and foraging behaviour of cutthroat trout in an open and forested section of stream. Verh. Internat. Verein. L imnol . 22: 2516-2522. Wilzbach M . A . , K . W . Cummins, and J.D. Hal l . 1986. Influence of habitat manipulations on interactions between cutthroat trout and invertebrate drift. Ecology. 67(4): 898-911. Winter, J .D. Summer home range movements and habitat use by four largemouth bass in Mary Lake, Minnesota. Transactions of the American Fisheries Society. 106(4): 323-330. Young, K . A . 1999. Environmental correlates of male life history variation among coho salmon populations from two Oregon coastal basins. Transactions of the American Fisheries Society. 128: 1-16. Young, K . A . , S .G. Hinch, and T . G . Northcote. 1999. Status of resident coastal cutthroat trout and their habitat twenty-five years after riparian logging. North American Journal of Fisheries Management. 19: 901-911. Young, M . K . 1996. Summer movements and habitat use by Colorado River cutthroat trout (Oncorhynchus clarkipleuriticus) in small, montane streams. Canadian Journal of Fisheries and Aquatic Sciences. 53: 1403-1408. Young, M . K . 1998. Absence of autumnal changes in habitat use and location of adult Colorado River cutthroat trout in a small stream. Transactions of the American Fisheries Society. 127: 147-151. Young, M . K . , R . B . Rader, and T . A . Belish. 1997. Influence of macroinvertebrate drift and light on the activity and movement of Colorado River cutthroat trout. Transactions of the American Fisheries Society. 126: 428-437. Zippen, C. 1958. The removal method of population estimation. Journal of Wildl i fe Management. 22(1): 82-90. 109 APPENDIX 0.22 0.20 0.18 0.16 0.14 0.12 0.10 0.08 0.06 0.04 0.005 0.015 0.025 0.035 0.045 0.055 Y = 0.365 + 0.057 * ln(X) R2 = 0.425 Husdon T—'—I—'—I—1—I—1—I—'—I—'—I—'—I—•—I Y = 0.321 + 0.038 *ln(X) R2 = 0.679 Ava lon -<—i—'—r~ 0.0 0.2 0.4 0.6 - I — 1 — i — 0.8 1.0 — i — 1.2 I— 1.4 1.6 Y = 0.137 + 1.342 * X R2 = 0.443 -i—I—i—|—i—|—r—|—i—|—i—|—i—|—i—|—i—|—i—| .000 0.020 0.040 0.060 0.080 0.100 Discharge (m3/sec) Appendix 1: Discharge/water-depth relationships for the three streams. Linear regression relationships and r 2 are indicated for each water-depth relationship. Water depth and discharge measurements waere taken at the stationary meter-sticks in each stream. 110 f a 1 mm CO £ -a > Back and Forth Side to Side Motionless Appendix 2: Act ivi ty levels recorded from a dead fish carrying a radiotransmitter. "Back and Forth" and "side to side" resembles fish movement and "Motionless" resembles a fish that is holding. The asterick indicates a significant difference (P < 0.05) between Motionless and the other two activities. I l l Appendix 3: Example of home range calculation (fish = L o u Lou). Relocation site (r) is the location within the stream relative to the modal site (0 ± 0.1 m; most frequently used site) that the fish was found. Negative values indicate that the site was downstream of the modal site, and positive values indicate that the site was upstream of the modal site. Number of relocations is the number of times Lou L o u was found at that site. N is the theoretical number of relocations that need to be observed at the relocation site based on its distance from the modal site (N = r * n * w / d 2 ; where d = 5 m and w = width of the stream). The index shows the relative importance of the relocation site (higher index numbers indicate that the relocation site are used frequently and the lower index numbers indicate relocation sites that are almost never used). Index is simply the number of relocations divided by N . A t this point, the relocation sites are sorted from most important (highest index value) to the least important (lowest index value) (already shown in this example). After the sorting is complete, the proportion of relocations is calculated. It is the number of relocations for each site divided by the total number of relocations (this example = 31). The cumulative proportion is used to estimate the 95% cutoff for estimating home range. Home range is then estimated as the distance (m) between the most upstream and downstream relocation sites within the 95% or 100% boundaries (eg. for 95% = 19.1 + 21.6 m = 40.7 m). Relocation Number of N Index Absolute Proportion Cumulative 95% 100% Site Relocations (relocations/ Value of of Home Home N) Index Relocations Range (m) Range (m) -0.1 13 -0.06 -206.901 206.901 0.4194 0.4194 0.9 1 0.57 1.768 1.768 0.0323 0.4517 -1.6 1 -1.01 . -0.995 0.995 0.0323 0.4840 -21.6 11 13.57 -0.811 0.811 0.3548 0.8388 -2.1 1 -1.32 -0.758 0.758 0.0323 0.8711 -17.0 1 -10.68 -0.094 0.094 0.0323 0.9034 19.1 1 12.00 0.083 0.083 0.0323 0.9357 40.7 -83.1 1 -52.21 -0.019 0.019 0.0323 0.9680 -99.6 1 -62.58 -0.016 0.016 0.0323 1.000 118.7 112 Appendix 4: Characteristics of the marked L W D pieces in Husdon and Avalon Creeks. The configuration of each piece of wood is described by its orientation to the stream bank (the number) and its placement within the stream banks (the letters). The number is the degree (°) the piece of wood is oriented to the stream bank (the base of the L W D is the origin). For example a piece of wood which is 180°would be oriented parallel to the stream bank with the base of the L W D at the upstream end. The letters represent F = wood spanning from bank to bank (full), H = wood touching one bank (half), M = wood not touching any bank (middle), I = wood embedded within the substrate, D = wood within the stream banks but out of the water (dry), o = wood suspended over the water. Values in brackets indicate a piece of L W D that could not be identified with absolute confidence (the tag could not be found, yet the wood looked similar in shape and size). LWD DBH Length Initial Final Initial In- Final In- Total (cm) (m) Configuration Configuration Stream Stream Distance Distance (m) Distance (m) Moved (m) Husdon 1 30-60 9-12 170-Fo 160-Fo 155.0 155.0 0.0 2 15-30 6 - 9 170-Fo 160-H 152.0 152.0 0.0 3 15-30 3 - 6 190-Ho 190-H 133.0 133.0 0.0 4 30-60 3 - 6 180-M 180-H 129.0 129.0 0.0 5 30-60 3 - 6 160-H 150-H 120.0 120.0 0.0 6 15-30 1- 3 180-H 240-H 106.5 103.2 -3.3 7 15-30 3 - 6 180-Ho 190-H 104.0 104.0 0.0 8 15-30 3 - 6 180-1 180-1 88.0 88.0 0.0 9 15-30 1- 3 120-1 130-HI 68.5 68.5 0.0 10 30-60 1- 3 100-H 90-HI 60.0 60.0 0.0 11 15-30 3 - 6 90-Fo 90-F 46.0 46.0 0.0 12 30-60 1- 3 200-F 220-H 38.0 38.0 0.0 13 15-30 3 - 6 45-Fo 60-Fo 33.0 33.0 0.0 14 15-30 1- 3 120-HI 110-MI 22.5 22.5 0.0 15 15-30 6 - 9 180-H 180-H 14.0 14.0 0.0 16 15-30 1- 3 270-Ho 180-H 1.5 Total: -3.6 -5.1 8.4 Avalon 1 15-30 3 - 6 120-Ho 221.0 >-221.0 2 30-60 9 - 12 180-D 180-D 203.0 203.0 0.0 3 30-60 3 - 6 270-Ho 184.5 > -184.5 4 rootwad <3 360-ID 175.5 > -175.5 5 15-30 3 - 6 210-H 162.0 > - 162.0 6 30-60 3 - 6 190-D 142.5 >-142.5 7 15-30 6 - 9 310-ID 270-IH 123.0 120.0 -3.0 8 30-60 6 - 9 160-ID 150-D 118.0 117.5 -0.5 9 15-30 9-12 260-Fo 180-1 103.0 96.0 -7.0 10 15-30 1- 3 110-Ho 82.0 >-82 11 30-60 1- 3 190-ID (180-ID) 74.0 (65) (-9) 12 30-60 6 - 9 180-Fo 190-Ho 57.0 52.0 -5.0 13 15-30 12-15 270-Fo 220-Fo 45.0 42.5 -2.5 14 30-60 6 - 9 160-K 37.0 > -37.0 15 30-60 3 - 6 190-H 15.0 >-15.0 16 15-30 3 - 6 200-D 7.0 Total: >-7.0 > 1000.0 113 

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.831.1-0089598/manifest

Comment

Related Items