Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Distribution, habitat use, and migratory life history of Southern Dolly Varden (Salvelinus malma lordi)… Heavyside, Julian 2021

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2021_november_heavyside_julian.pdf [ 2.56MB ]
Metadata
JSON: 24-1.0397105.json
JSON-LD: 24-1.0397105-ld.json
RDF/XML (Pretty): 24-1.0397105-rdf.xml
RDF/JSON: 24-1.0397105-rdf.json
Turtle: 24-1.0397105-turtle.txt
N-Triples: 24-1.0397105-rdf-ntriples.txt
Original Record: 24-1.0397105-source.json
Full Text
24-1.0397105-fulltext.txt
Citation
24-1.0397105.ris

Full Text

DISTRIBUTION, HABITAT USE, AND MIGRATORY LIFE HISTORY OF SOUTHERN DOLLY VARDEN (Salvelinus malma lordi) IN A SALMON-BEARING WATERSHED ON THE CENTRAL COAST, BRITISH COLUMBIA by  Julian Heavyside  B.Sc., The University of British Columbia, 2016  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Zoology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) April 2021 © Julian Heavyside, 2021   ii  The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, a thesis entitled:  DISTRIBUTION, HABITAT USE, AND MIGRATORY LIFE HISTORY OF SOUTHERN DOLLY VARDEN (Salvelinus malma lordi) IN A SALMON-BEARING WATERSHED ON THE CENTRAL COAST, BRITISH COLUMBIA   submitted by Julian Heavyside in partial fulfillment of the requirements for the degree of Master of Science in Zoology  Examining Committee: Eric B. Taylor, Department of Zoology, UBC Supervisor  Jill Jankowski, Department of Zoology, UBC Supervisory Committee Member  Michelle Tseng, Department of Botany and Zoology, UBC Additional Examiner   Additional Supervisory Committee Members: Scott Hinch, Department of Zoology and Faculty of Forestry, UBC Supervisory Committee Member iii  Abstract Anadromous freshwater fishes are among the most threatened vertebrates in North America. Population declines and collapses due to overfishing, habitat loss, and climate change have long been documented for socioeconomically important salmonids such as Pacific salmon (Oncorhynchus spp), but the conservation status of other species such as Dolly Varden char (Salvelinus malma) remains largely unknown. Even without the pressures of direct commercial harvest, many anadromous salmonids may still be vulnerable to the other threats faced by commercially-harvested species, but conservation efforts continue to be hampered by a lack of basic life history and demographic information. While the Northern Dolly Varden (S. m. malma) has been assessed as “Special Concern” by the Committee on the Status of Endangered Wildlife in Canada, the Southern Dolly Varden (S. m. lordi) has received little conservation attention despite near ubiquity in coastal watersheds from Alaska to Washington, partly due to a lack of data throughout the broad coastal range. I conducted surveys on a population of Southern Dolly Varden in a lake-headed coastal watershed on the mid coast of British Columbia to identify key spawning, rearing, and foraging habitat, to characterize seasonal patterns in their iteroparous anadromous migratory life history, and to assess the extent of migratory coupling among Pacific salmon and Dolly Varden throughout the year. I found that Dolly Varden used a wide range of habitats distributed throughout the entire watershed, and that there were distinct seasonal movements that were temporally associated with Pacific salmon migrations. My observations indicate that this population of Southern Dolly Varden requires the entire extent of the watershed, from lake tributaries to the estuarine river mouth, for individuals to express their complete migratory life history, and that this population is vulnerable to any disturbance leading to the degradation or loss of access to key habitats for spawning, rearing, and foraging. This iv  population could serve as a baseline for understanding the conservation status of Southern Dolly Varden throughout the rest of the range, and can be used as a rare model for studying predator-prey migratory coupling as it relates to socioeconomically important species of Pacific salmon. v  Lay Summary Salmon migrations are a defining feature of many coastal watersheds. Salmon often travel great distances to reach their spawning rivers, and ascend high into the watershed to spawn and die. This migration triggers similar migrations of their predators–– especially freshwater fishes that eat salmon eggs. Dolly Varden (a species of char) are one such fish that can migrate much like salmon do from marine habitats to stream habitats to eat salmon eggs, to spawn, and to shelter overwinter. Dolly Varden are common throughout coastal British Columbia, but very little is known about them. I studied Dolly Varden in the Koeye River on the Central Coast of British Columbia to determine which habitats were important for spawning, rearing, and foraging, and how closely their migrations mirrored those of salmon. This work will help inform the conservation of Dolly Varden by providing documenting habitat information about this poorly known species. vi  Preface All aspects of this work were conducted in collaboration with the QQs Projects Society (a Heiltsuk First Nation nonprofit group), the Heiltsuk Integrated Resource Management Department, the Tula Foundation/Hakai Institute, and the Salmon Watersheds Lab at Simon Fraser University. I conducted this research under the broad umbrella of the Koeye River Salmon Ecosystem Study, which is an ongoing Heiltsuk-led research initiative to implement sustainable fisheries management strategies in the Koeye River and elsewhere in Heiltsuk Traditional Territory. I shared responsibility for the study design and the collection of data with Dr. William Atlas and Karl Seitz, who were graduate students in the Moore Lab and Salmon Watersheds Lab at Simon Fraser University while I was conducting this research. I conducted field work in 2018, but I have used data collected between 2015 and 2019 in my analysis. As such, the majority of the PIT tag and occurrence data I have used in my analysis were collected by Dr. William Atlas, Karl Seitz, and numerous field technicians in the years prior to my involvement in the project. I held primary responsibility for the analysis of these data, and for the preparation of this manuscript for submission. I received guidance and supervision from my supervisor, Dr. Rick Taylor, and from my supervisory committee members, Dr. Jill Jankowski and Dr. Scott Hinch, and from my main research collaborators Dr. William Atlas and Karl Seitz. All fish capture, tagging, and handling procedures were approved by the University of British Columbia Animal Care Committee (certificate number A20-0014).   vii  Table of Contents  Abstract .................................................................................................................................... iii Lay Summary ............................................................................................................................ v Preface ...................................................................................................................................... vi Table of Contents .................................................................................................................... vii List of Tables ............................................................................................................................ ix List of Figures ........................................................................................................................... x Acknowledgements ................................................................................................................. xii Dedication ............................................................................................................................... xv Introduction: Conservation Ecology of Anadromous Freshwater Fishes in Salmon-bearing Watersheds in Western North America ................................................................................... 1 Conservation Ecology of Anadromous Freshwater Fishes ....................................................... 1 Salmon-bearing Watersheds in Western North America .......................................................... 3 Migratory Coupling Between Pacific Salmon (Oncorhynchus spp.) and Their Freshwater Iteroparous Salmonid Predators and Scavengers ...................................................................... 5 Life History Diversity of Dolly Varden (Salvelinus malma) Populations in Western North America .................................................................................................................................. 8 The Koeye River Watershed on the Central Coast of British Columbia ................................. 12 Thesis Objectives .................................................................................................................. 16 Materials and Methods ........................................................................................................... 17 Stream and Tributary Surveys ............................................................................................... 17 Downriver Movement Monitoring ......................................................................................... 18 viii  Estuary Monitoring ............................................................................................................... 19 PIT tagging protocol for stream, river, and estuary surveys ................................................... 20 Tracking Tagged Individuals ................................................................................................. 21 Statistical Methods ................................................................................................................ 22 Results ..................................................................................................................................... 25 Stream Surveys ..................................................................................................................... 25 Downriver Movement Monitoring ......................................................................................... 26 Estuary Monitoring ............................................................................................................... 29 Tagging and Detection Summary .......................................................................................... 31 Timing of Population-Scale Movement Throughout the Watershed ....................................... 38 Discussion ................................................................................................................................ 41 Dolly Varden life history ....................................................................................................... 41 Migratory coupling ............................................................................................................... 48 Conservation implications ..................................................................................................... 50 Limitations of my research .................................................................................................... 51 Future research...................................................................................................................... 54 References ............................................................................................................................... 56 Appendix ................................................................................................................................. 63  ix  List of Tables Table 1…………………………………………………………………………………………...25  Summary of variation in presence, abundance, and size classes of Dolly Varden in streams throughout the Koeye River watershed.  Table 2…………………………………………………………………………………………...34 Tagging and detection summary of Dolly Varden in the Koeye River watershed.  Table 3…………………………………………………………………………………………...35 Detection summary over the duration of the study for Dolly Varden initially tagged in the estuary.  Table 4…………………………………………………………………………………………...36 Detection summary over the duration of the study for Dolly Varden initially tagged at the smolt trap.  Table 5……………………………………………………………………………………………………………...37 Select capture histories for individual Dolly Varden in the Koeye River watershed  Table 6……………………………………………………………………………………………………………...40 Summary of date of passage of Dolly Varden, sockeye salmon, and coho salmon at detection sites throughout the Koeye River watershed (mean date with associated standard deviation in days).  Appendix Table A1…..….……………………………………………...………………………63 AIC scores from mixture analysis of size distributions for each survey type of Dolly Varden where k represents the number of potential size classes in the mixture model. The most highly supported value of k is shaded in grey. x  List of Figures Figure 1…………………………………………………………………………………….……..9 Southern Dolly Varden (Salvelinus malma lordi), lateral view.  Figure 2………………………………………………………………………………………….14 Location of the Koeye River watershed on the Central Coast of British Columbia.  Figure 3………………………………………………………………………………………….18 Location of streams surveyed throughout the Koeye River watershed. Pink lines indicate the surveyed extent of each stream.  Figure 4………………………………………………………………………………………….19 Location of Dolly Varden sampling sites throughout the Koeye River watershed. Black points and the black square represent beach seine sites throughout the estuary and the location of the smolt trap, respectively. Black bars represent the location of the four Radio Frequency Identification (RFID) antennae deployed throughout the watershed.  Figure 5………………………………………………………………………………………….26 Size distribution of juvenile Dolly Varden sampled in nine streams in the Koeye River watershed (N = 117). Vertical dashed lines represent size thresholds of age-classes (in years) documented for Southern Dolly Varden in the Campbell Lake watershed, British Columbia (Michalski 2006).  Figure 6………………………………………………………………………………………….27 Size distribution of Dolly Varden sampled at the smolt trap (N = 431). Horizontal lines represent size ranges of age-classes (in years) documented for southern Dolly Varden in the Campbell Lake watershed, British Columbia (Michalski 2006).  Figure 7………………………………………………………………………………………….28 Body length of Dolly Varden sampled at the smolt trap as the sampling season progressed (N = 431, pooled across 2015-2019 seasons). Horizontal line represents the presumed size threshold for smolts.  Figure 8………………………………………………………………………………………….29 Body length of Dolly Varden smolts sampled at the smolt trap throughout the season (N = 364, pooled across 2015-2019 seasons).  Figure 9………………………………………………………………………………………….30 Size distribution of Dolly Varden sampled in the Koeye River estuary, pooled across all years (N = 680). Vertical dashed lines represent mean length of each size class. Horizontal lines represent size ranges of age-classes (in years) documented for southern Dolly Varden in the Campbell Lake watershed, British Columbia (Michalski 2006).    xi  Figure 10………………………………………………………………………………………31 Size distribution of Dolly Varden sampled in the Koeye River estuary for each year. Vertical dashed lines represent mean body length of each size-class identified by mixture analysis of individuals pooled across all years (Figure 9).  Figure 11………………………………………………………………………………………39 Date of passage of Dolly Varden and Pacific salmon (sockeye salmon and coho salmon) at detection sites throughout the Koeye River watershed. The upper panel represents downriver movement in lower river (A) and occupation of the estuary (B). The middle panel represents upriver movement in the lower river (C) and at the lake outlet (D). The lower panel represents movement into tributaries above the lake (E).   xii  Acknowledgements I want to start by acknowledging that this research was conducted in the traditional, ancestral, and unceded territories of the Heitlsuk (Haíłzaqv) and Oweekeno (Wuikinuxv) Nations, and the traditional, ancestral, and unceded territories of the Coast Salish peoples–– the Sḵwx̱wú7mesh (Squamish), Stó:lō, Səl̓ílwətaʔ/Selilwitulh (Tsleil-Waututh), and xʷməθkʷəy̓əm (Musqueam) Nations. As a Western scientist, I am a participant in the current ongoing process of colonialism, which here on the coast has been marked by genocide, land theft, forced removal and relocation of people, exclusion of Indigenous communities from resource management, and the attempted dismantling of the traditional practices that have sustained cultures and ecosystems since time immemorial. I want to celebrate the community of Bella Bella and all the people there who have showed me how to thrive on the coast. Much of my field time was spent wading through remote salmon streams, leaving precious few moments in town, but the days in Bella Bella have left me with a lifetime of cherished memories. I am particularly grateful for my relationship with Cal Humchitt. Cal has been an amazing collaborator, guide, mentor, and friend, and has opened my eyes to the inner workings of the coast. Cal was the first to show me Koeye, and he encouraged me to find a way to work there. He knew it would be an opportunity to work with youth. I owe a great deal to my supervisor, Rick Taylor, who guided me with endless patience and encouragement through the often-troubled waters of academic research. Rick has taught me so much about how to solve problems that seem at first to have no solutions, and the value of asking for help. My committee members Jill Jankowski and Scott Hinch were there to help when I needed it most, and gave me a diversity of very useful perspectives on my research questions. This thesis truly would not have been possible without the guidance and support of Will Atlas, xiii  who warmly welcomed me into the weir crew and helped me with every aspect of this project. Will is a brilliant scientist and it has been a pleasure to spend so much time in the field talking about salmon rivers. I also thank Karl Seitz for showing me how to work hard and have fun in the field and for keeping all the systems running smoothly. Will and Karl had many years of field experience at Koeye before I showed up, and I am so thankful for everything they shared with me. Richard Wilson-Hall, Jared Reid, Jeff Brown, Jeremy Jorgenson, and Kai Humchitt were awesome technicians and great company on the trails and in the streams.  This project started out as a BRITE internship, and I thank Larry Jorgenson, Jess Housty, William Housty, and Louisa Housty of QQs Projects Society for the opportunity to join the Koeye salmon research program during my BRITE internship and to be a part of the incredible work that is being done at Koeye. I thank Eric Peterson and Christina Munck of the Tula Foundation for generously providing internship funding. I thank Katie Beall for helping me put together a successful internship application, and for generally getting me out of all kinds of trouble around the Biodiversity Research Centre throughout my time at UBC. Sometimes grad school feels like a wish come true (especially in an amazing research centre like Biodiv), and sometimes (well, a lot of the time) it feels like a desperate struggle just to keep your head above the water (especially in an amazing research centre like Biodiv). Many people have made it feel like a wish come true, and I am particularly grateful for everyone involved with the ZGSA, Biodiversity’s Enviable Excuse for Researchers to Socialize, The Botany and Zoology Wellness Initiative, and, of course, Huts. My office mates Emily, Mairin, John, Manny, Mariana, and Libby made me look forward to every day on campus. To all the mischievous, nature-loving pals I made throughout my time at UBC, I can’t name you all here xiv  but you know who you are! Thank you for offering your friendship, you’ve filled my life with so much love and enjoyment. xv  Dedication I dedicate this thesis to my parents, Jane and Michael, and my sister, Paige. And to Kaleigh, for lovingly supporting and encouraging me throughout this degree. I love you all.  1  Introduction: Conservation Ecology of Anadromous Freshwater Fishes in Salmon-bearing Watersheds in Western North America  Conservation Ecology of Anadromous Freshwater Fishes Fresh water represents less than 1% of the Earth’s available aquatic habitat by volume, yet it supports more than half of the global taxonomic diversity of ray-finned fishes (Dawson 2012; Carrete Vega and Wiens 2012). This diversity is geographically distributed across countless bodies of water on all continents except for Antarctica. Freshwater habitats are edged by terrestrial and marine habitats which inherently create barriers that promote the evolutionary processes that generate diversity. Historical hydrological events, such as watershed exchanges and marine incursions, have triggered episodes of population expansions and invasions, isolation, divergence due to drift or adaptation, and post-isolation contact, which have ultimately resulted in the accumulation of freshwater species richness at local, regional, and global scales (May 1994; Barraclough et al. 1998; Bloom et al. 2013). The ecological characteristics that have been shaped during the evolution of freshwater fishes also make them vulnerable to climate and land use change, which now occurs at unprecedented rates (Kundzewicz et al. 2008). For example, metapopulation dynamics that allow regional persistence of species across a network of localities can be disrupted when dispersal among localities is hampered by physical barriers to dispersal or suboptimal water conditions among localities. Similarly, species that are represented by populations at very few or even single localities are at high risk of global extinction (e.g., Hatfield 2001). Land use change and introductions can lead to novel interspecific interactions like competition or hybridization (e.g., 2  Taylor et al. 2006) and water conditions can change beyond the range of tolerable conditions (e.g. Pearson 1999). Barriers can prevent migratory species from moving among spatially distinct habitats to complete life history transitions, and the widespread use of dams and dykes to generate hydroelectricity and control flow and sedimentation in navigable waterways is of special concern for diadromous fishes, which undertake migrations between marine and freshwater habitats throughout their life cycles (Baras and Lucas 2001). Anadromous fishes are those that reproduce in fresh water where the young live and feed for variable time periods, then migrate to the sea where they achieve most of their growth before returning as maturing adults to fresh water. Anadromous life histories are marked by events that coincide with seasonal climatic and hydrological events occurring at multiple spatial scales in marine and fresh waters and their interface (Quinn and Adams 1996; Spence and Hall 2010; Moore et al. 2012; Bond and Quinn 2013). Climate and land use change can alter behavioural cues, disrupt the timing of life cycle transitions, or block movement among critical habitats. Estuaries provide the critical link between marine and freshwater habitats for anadromous fishes and provide productive and safe areas for feeding (e.g. Beck et al. 2001) or physiologically transitioning to different salinities during migrations, but these critical habitats are now among the most ecologically disturbed on Earth due to widespread human occupation and industrial activity in coastal regions (Waycott et al. 2009). Conservation biology aims to identify the threats to the persistence of organisms throughout their natural ranges in order to inform protection and restoration policies and actions. Geographic distribution and habitat use are the foundations of understanding basic ecological characteristics such as life history and population health, and are tightly linked to conservation assessment of freshwater fishes in Canada (COSEWIC 2019). The area of occupancy at the regional scale and 3  the number of associated locations that contain key habitats at the local scale are key components to developing proper habitat protection strategies. For freshwater fishes, spawning and overwintering sites are often designated as critical habitat under Canada’s Species at Risk Act as “habitat that is necessary for the survival or recovery of a listed wildlife species and that is identified as the species' critical habitat in the recovery strategy or in an action plan for the species" (e.g., Species at Risk Act s2 2002). Such habitats may be used for reproduction, feeding, sheltering, or a combination of activities necessary for the persistence of a species. Animal movement links foraging, spawning, and sheltering habitats, and life history trade-offs can emerge when these habitats are separated in space or time (Stearns 1992), and single harmful events that disrupt movement create additional vulnerabilities for a population. For anadromous species, the locations of critical habitats can span many kilometres across marine and fresh waters. Identification of critical habitat for a given population, and understanding the factors that influence timing and extent of movement among these habitats, are required for successful conservation of populations of anadromous freshwater fishes.   Salmon-bearing Watersheds in Western North America  For many people, Pacific salmon and trout (Salmonidae: Oncorhynchus)-bearing watersheds in western North America conjure images of large, powerful rivers flowing hundreds of kilometres from inland headwaters where millions of Pacific salmon begin and end their lives each year. Despite their individual grandeur, such systems are quite rare, and their more common alternative– small to medium coastal watersheds– collectively contain the majority of freshwater Pacific salmon spawning habitat and populations in British Columbia (Levy and Slaney 1993; Connors et al. 2018). The average small to medium (< 300 km2) coastal watershed in British 4  Columbia is typically rain and snow-dominated with low freshwater productivity (Stockner and McIsaac 1996). Such watersheds experience more variable hydrology than large watersheds (Moore et al. 2015), and seasonal pulses of high-flow rates in late fall and spring contrast with low-flow rates and elevated water temperature in summer droughts that can inhibit movement of adult salmonids (Quinn et al. 1997). This in part has resulted in diversification across small watersheds as populations have evolved to suit local hydrological conditions (e.g., Taylor 1991). Small watersheds generally support higher rates of salmon-derived nutrient transfer into terrestrial food webs due to the high perimeter-area ratios that provide more access to adult salmon flesh to predators and scavengers (Hocking and Reimchen 2008). Larger river systems often have high flow rates in their lower reaches, which can reduce capture efficiencies of predators such as bears and eagles, important players in the transfer of nutrients from aquatic to terrestrial food webs (Gende et al. 2004).  Salmon-bearing watersheds also support an array of other species of freshwater fishes, including iteroparous anadromous salmonids such as coastal cutthroat trout (Oncorhynchus clarkii clarkii), steelhead trout (Oncorhynchus mykiss), and various species of char (Salvelinus spp.) such as Dolly Varden (S. malma) and bull trout (S. confluentus). The freshwater fishes that co-occur with Pacific salmon in coastal watersheds can be predators or scavengers of Pacific salmon, or benefit indirectly from the seasonal pulse of nutrients provided by the eggs and flesh of senescent salmon (Denton et al. 2009).  Pacific salmon populations in small to medium coastal watersheds of western North America have experienced the most proportionally severe declines and collapses over the past century, primarily due to widespread overfishing, negative interactions with hatchery-produced conspecifics, construction of hydroelectric dams, and degradation and loss of habitat (Nehlsen et 5  al. 1991; Quinn 2018). Another factor that has probably contributed to salmon declines is the mismanagement of large-scale mixed-stock fisheries (Levy 2006). The widespread dismantling of small-scale fisheries that had been managed by Indigenous fishers using local knowledge and place-based fishing methods for thousands of years prior to the establishment of colonial governments and industries likely also greatly contributed to the decline of wild Pacific salmon populations (Trosper 2002; Berkes 2003). Climate change is also implicated in salmonid population declines, especially in small to medium rain- and snow-dominated watersheds, where increased water temperatures have been associated with higher rates of prespawning mortality of adult migrant salmon (Atlas et al.   2017). The large scale at which many Pacific salmon populations continue to be managed today far exceeds the local scale at which the critical life history events of individual populations operate. This is because the majority of small to medium salmon-bearing systems lack the on-the-ground data required for accurate stock assessment and effective management (Nehlsen et al. 1991; Quinn 2018).  Migratory Coupling Between Pacific Salmon (Oncorhynchus spp.) and Their Freshwater Iteroparous Salmonid Predators and Scavengers Depending on the species, Pacific salmon juveniles migrate out of their natal watersheds within a few weeks to multiple years after they emerge from their gravel nests to feed and grow in productive marine waters, often hundreds of kilometres away from their natal streams or lakes (Rounsefell 1958; Quinn 2018). Upon attaining peak growth over six months to several years, semelparous adults undertake a single spawning migration to return to their natal streams to reproduce and subsequently die. The seasonal pulse of resources provided by the eggs and flesh of both spawning and senescent semelparous Pacific salmon provides an important, but 6  ephemeral food supply––commonly termed a nutrient subsidy (Polis et al. 1997) ––for an array of predators and scavengers, including grizzly bears (Ursus artos horribilis), bald eagles (Haliaeetus leucocephalus), and freshwater fishes such as prickly sculpin (Cottus asper), coastal cutthroat trout, and Dolly Varden. The ability of piscivorous freshwater fishes to exploit salmon subsidies depends on the degree to which these predators are resident or transient within the watershed. Resident freshwater fishes that have restricted territorial home ranges can only exploit salmon subsidies if and when they are available in their range. Transient freshwater fishes can move to exploit resources such as spawning or foraging habitats as they become available throughout the watershed, and anadromous fishes extend this transience to include use of marine waters. The ability to move between alternative foraging habitats releases freshwater fishes from many of the constraints of restricted localized territoriality, but the spatial and temporal context of resource availability can still create trade-offs. For anadromous predatory and scavenger fishes in salmon-bearing watersheds, the temporally overlapping opportunities to feed in productive marine waters or exploit salmon subsidies in fresh waters creates a potential trade-off when these opportunities occur in separate locations (e.g. Bond and Quinn 2013). The timing and extent of movement between marine and freshwaters are traits that have likely been shaped in part by selection to find optimal solutions to this trade-off, and by environmental factors such as flow rates and water temperature (Banks 1969; Moore et al. 2012; Bond and Quinn 2013). Unlike the highly predictable timing and extent of Pacific salmon migrations, anadromous predator and scavenger migrations can be more variable, especially for long-lived potentially iteroparous anadromous fishes that spawn multiple times (Bond et al. 2015). These fishes can express varying degrees of anadromy throughout their life cycle, which suggests that 7  optimization of the marine-freshwater foraging trade-off may result in age-specific migratory strategies. In addition to foraging and spawning, iteroparous anadromous fishes use freshwater habitats to overwinter, which further complicates the ability to disentangle the factors that underlie the timing and extent of migration of iteroparous predatory and scavenger freshwater fishes. Nonetheless, the associated movements of spawning and senescent salmon and their migratory predators and scavengers represent an intriguing potential case of migratory coupling, wherein predators undertake extensive migrations beyond the scale of territorial home ranges in order to forage on migrant prey (Furey et al. 2018).  Migratory coupling can have widespread ecological consequences by varying the strength of trophic interactions within and across landscapes and waterways over time (Furey et al. 2018). Migration influences community structure by introducing transient members to local communities, and by increasing local population densities when migrants converge as they approach their destination. Concentrations of prey can lead to increased competitive interactions among local consumers or foraging opportunities for local predators. Concentrations of predators can similarly lead to increased competition and predation, and can also release local competitors and prey that reside in the sites recently vacated by migrant predators. In general, predation plays a key ecological role by regulating prey population growth, which can otherwise have destabilizing effects on an ecosystem, and by broadly distributing prey nutrients throughout the food web, which can otherwise be unavailable or can be rapidly shunted out of a system (Hairston et al.1960; Paine 1980; Levin 1999). In river systems, food web dynamics can be complicated by the temporal and spatial scales of nutrient availability throughout the heterogenous waterways that link headwaters to estuaries and by the degree to which predators and prey are resident or transient in freshwater communities. Due to this complexity, the 8  dynamics of river food webs are still poorly understood (Power and Dietrich 2002), but the identification of migratory coupling within food webs may advance the understanding of river food web dynamics (Furey et al. 2018; Kanigan 2019). This may be especially true in salmon-bearing systems wherein the migratory patterns of populations of these species are already documented.  Life History Diversity of Dolly Varden (Salvelinus malma) Populations in Western North America Char (Salvelinus spp.) are represented by a diverse array of iteroparous freshwater and anadromous salmonid fishes that are distributed throughout the temperate and polar regions of the Northern Hemisphere (reviewed in Kershner et al. 2019). Geological, evolutionary, and ecological processes have resulted in a mosaic of sympatric and allopatric taxa that collectively exhibit an array of life histories (Kershner et al. 2019). There are currently five recognized species of char in North America, most of which are further represented by populations of ecomorphs that express a wide range of ecological and phenotypic traits (Whiteley et al. 2019). The diversity of char life histories stems from variation in all traits classically considered in life history theory (Stearns 1992), including growth pattern, age and size at maturity, frequency and size of reproductive allocations, and lifespan (e.g. Armstrong 1974; Reist et al. 2013; Bond et al. 2015). Habitat preference and migratory life history also vary greatly across species and populations of char. Although the general char life cycle begins in cold-water streams or tributaries, it is rarely restricted to these waters. Movement within and among aquatic habitats varies from lacustrine or fluvial (lake or stream) residence to transience among lakes and streams, to facultative anadromy (e.g. Armstrong 1974; Armstrong 1984, Swanson et al. 2010, 9  Jonsson and Jonsson 2011). Juveniles of anadromous populations eventually migrate as so-called “smolts” to marine waters after as many as eight years in fresh water, and may return to fresh water after foraging for as few as four weeks in marine waters. Dolly Varden (Figure 1) express a wide range of the life history strategies exhibited across the entire genus, and lacustrine, fluvial resident, and anadromous populations have been recorded throughout the Arctic and the Northern Pacific Rim. Two subspecies of Dolly Varden are recognized in western North America (Phillips et al. 1999; Kowalchuk et al. 2010): Northern Dolly Varden (S. malma malma) occur throughout the western Arctic and south to the Alaskan Peninsula; and Southern Dolly Varden (S. malma lordi) occur from Washington to the Gulf of Alaska. Putative contact zones between the two subspecies have been identified in the Gulf of Alaska between the Alaskan Peninsula and Cook Inlet (Taylor and May-McNally 2015), but it remains unclear whether the observed population admixture is due to historical interbreeding or if the two forms currently interbreed where they come into contact.   Figure 1. Southern Dolly Varden (Salvelinus malma lordi), lateral view. 10   Northern Dolly Varden have been studied in numerous systems throughout their range (e.g. Armstrong 1984; Redenbach and Taylor 2002, 2003; Stewart et al. 2010; Bond and Quinn 2013, May-MacNally et al. 2014; Dennert et al. 2016), and Canadian populations of S. m. malma were assessed as “Special Concern” by The Committee on the Status of Endangered Wildlife in Canada (COSEWIC 2011). Conversely, Southern Dolly Varden have been relatively understudied, and population-level observations have been limited to only a few watersheds on Vancouver Island in British Columbia and the Olympic Peninsula in Washington (Brown 1992; Michalski 2006). Northern and Southern Dolly Varden are recognized as two subspecies owing to the strong molecular evidence that the two forms have distinct evolutionary histories (Phillips et al. 1999; Taylor et al. 2008), and that they may even represent distinct species (Kowalchuk et al. 2010). Their ranges span vastly different latitudes and ecological conditions, which raises questions about how local adaptation may have differentially shaped the ecological characteristics of populations of the two subspecies. The degree to which an understanding of the biology and conservation status of Northern Dolly Varden can be applied to Southern Dolly Varden remains unclear. Despite near ubiquity in watersheds along the entirety of the British Columbia coast, there have been no formal studies that identify critical habitat and migratory patterns of Southern Dolly Varden north of Campbell River, BC. Southern Dolly Varden have not been assessed as a designatable unit as defined by COSEWIC (see COSEWIC 2018), primarily because the distribution, habitat use, and migratory life histories of Southern Dolly Varden in watersheds throughout their range are poorly known. At the regional scale, coastal watersheds in Washington, British Columbia, and Alaska likely support hundreds of populations of Dolly 11  Varden, almost all of which are virtually unstudied relative to other co-occurring salmonids such as sockeye salmon (Oncorhynchus nerka). Over half of the 214 unique lake-rearing sockeye salmon populations that are recognized in British Columbia occur in low-elevation coastal watersheds (Holtby and Ciruna 2007), which also likely support a similar diversity of Southern Dolly Varden populations given the tendency for Dolly Varden to co-occur with Pacific salmon in other areas (Morrow 1980; Armstrong 1984; DeCicco and Reist 1999).  Throughout the year in an average coastal watershed, optimal Dolly Varden habitat may alternate between productive marine waters for feeding in the summer, rivers or tributaries for feeding on eggs and flesh from spawning and senescent salmon, respectively, cold-water tributaries for spawning, and lakes for overwintering. Summer droughts may reduce flow rates and increase temperatures in the rivers and streams that connect these habitats, which can create suboptimal or lethal abiotic conditions and a prevent access to optimal habitat. Migration provides an opportunity to remove constraints on growth or survival imposed by declining physiological and ecological conditions such as temperature, oxygen level, and food availability within single habitats that vary spatially and seasonally. Although facultative anadromy appears to be common in Dolly Varden in salmon-bearing watersheds, the frequency and extent of their marine migrations is variable, and the factors underlying this variation have only been studied in a few populations of Northern Dolly Varden in Alaska (Bond and Quinn 2013). The combination of anadromy and seasonal exploitation of salmon subsidies (e.g. Denton et al. 2009) suggests a case for migratory coupling between Dolly Varden and Pacific salmon (Furey et al. 2018), but uncertainty remains regarding the extent of individual Dolly Varden movements beyond territorial home ranges to exploit migrant prey subsidies, and the relative importance of foraging, 12  spawning, and overwintering, which are all potential causes of migration into Pacific salmon spawning sites.  The Koeye River Watershed on the Central Coast of British Columbia One watershed where spawning Pacific salmon and other salmonids interact is the Koeye River, situated on the remote central coast of British Columbia, about 100 km due north of the northeastern tip of Vancouver Island. The Koeye River drains a 186 km2 mainland watershed as it flows approximately 35 km westward from headwaters in the Pacific Range to its mouth. The watershed contains one of the region’s largest complexes of intact primary forest, characterised by high densities of giant specimens of Sitka spruce (Picea sitchensis), Western redcedar (Thuja plicata), and Western hemlock (Pseudotsuga menziesii). The area supports a large grizzly bear population and is an important migratory stopover site and breeding area for numerous species of land and sea birds, including the marbled murrelet (Brachyramphus marmoratus), the northern goshawk (Accipiter gentilis), the sandhill crane (Antigone candensis), and numerous species of shorebirds (Calidridae) and waterfowl (Anatidae). The watershed and its coastal margins support all five species of Pacific salmon and is among the most productive sockeye salmon systems in the region.  The rain- and snow-dominated watershed is headed by two medium-sized lakes (Figure 2). Upper Koeye Lake (227 ha) is located 17 km from the river mouth at 140 m elevation. Lower Koeye Lake (450 ha) is located approximately 6 km downstream at 53 m elevation. A steep canyon below the outlet of Upper Koeye Lake is a putative barrier to salmonid movement between lakes, and therefore all Pacific salmon spawning and rearing in the watershed are concentrated in or below Lower Koeye Lake and its tributaries. The Koeye River experiences 13  peak flows after heavy rains in the late fall and winter, and high flows during a powerful freshet in late spring. The lower section of the river flows westward from Lower Koeye Lake for 11 km directly into outer-coastal waters, a feature that is globally rare in glacially-impacted regions where rivers typically flow into inner-coastal waters of inlets and fjords. A short, narrow canyon upstream of the exposed mouth shelters a series of forested river islands, low-lying meadows, and braided brackish tidal streams. The estuary extends for 5 km from the mouth throughout the majority of the tidal influence, and encompasses a broad gradient of habitat characteristics ranging from large sandy beaches at the mouth, to highly saline beds of eelgrass (Zostera marina), to variably saline muddy patches of rockweed (Fucus spp.), to salt marsh, and to primarily freshwater stream channels with pool-riffle sequences typical of gentle gradient first-order coastal streams. The estuary supports a diverse community of fishes and provides important rearing habitat for coho salmon (Oncorhynchus kisutch) and various marine fishes, including notably large populations of Pacific sand lance (Ammodytes hexapterus) in sandy substrates near the mouth (Seitz et al. 2020). Local land users and biologists have long documented large seasonal aggregations of Southern Dolly Varden during the summer in the lower reaches of the estuary (William Housty, Heiltsuk Integrated Resource Management Department, 2018, pers. comm.).  14   Figure 2. Location of the Koeye River watershed on the Central Coast of British Columbia.  The Koeye River is situated in Heiltsuk traditional territory and has been a site of great cultural significance for the Heiltsuk for thousands of years. Numerous village sites, stone fish traps, culturally modified trees, and deep shell middens throughout the watershed trace thousands of years of continuous occupation and resource management by the Heiltsuk through to the present day. Since the 1990s, the area has been the site of the community-led Koeye Camp program which has a mandate “to open the eyes of our children to their responsibility as stewards of our land, culture, and resources.” Koeye Camp integrates scientific and cultural learning into 15  week-long summer camps for youth from the nearby communities. All scientific researchers who are conducting fieldwork as part of the ongoing long-term ecological monitoring of the area also participate in the camp program, and research on Pacific salmon and Dolly Varden in the estuary has greatly contributed to the Koeye Camp program mandate by creating opportunities for youth campers to practice hands-on field techniques used by conservation biologists and fisheries managers.  The Koeye River watershed is the site of a Heiltsuk-led salmon monitoring project. Researchers use traditional fishing technology (e.g. a redcedar weir deployed in the lower river to capture and tag adults during the spawning migration) and tracking methods (e.g. an array of passive sensors deployed throughout the watershed, and visual counts of colour tags in spawning channels) to monitor sockeye salmon throughout the watershed in order to ultimately develop a sockeye salmon management plan (Atlas et al. 2017). The project has collected information on the movements and habitat use of other species in the community, including Southern Dolly Varden. The ongoing deployment of monitoring and tracking infrastructure and the means to access remote portions of the watershed throughout the year present a rare opportunity not only to collect distributional and habitat use information on a mid-coast population of Southern Dolly Varden in British Columbia, but also to study the migratory life histories of Pacific salmon and Dolly Varden in the same watershed in the context of migratory coupling.    16  Thesis Objectives The broad objective of my research was to inform conservation ecology of Southern Dolly Varden on the mid-coast of British Columbia, which has rarely been examined despite the importance of this large region in the overall geographic distribution of this fish. To achieve this goal, I conducted ecological surveys of a single population of Southern Dolly Varden in a mainland coastal watershed in order to: (i) identify spawning and foraging locations, (ii) characterise the spatial and seasonal variation in abundance and size of individuals of various size classes, (iii) characterise population-scale seasonal movement patterns, and (iv) assess the extent of migratory coupling with migratory Pacific salmon. The results of this study should contribute to the conservation of Southern Dolly Varden by providing a rare population-level assessment of the locations of potentially critical habitats and the seasonal movements that link them, which are both aspects that would be considered in an eventual COSEWIC conservation status assessment of Southern Dolly Varden. The results also contribute to the growing body of research surrounding the importance of migratory coupling in food web dynamics, which has a promising theoretical basis, but still lacks many thoroughly tested examples in nature. The results of this research will also contribute to the ongoing Heiltsuk-led Pacific salmon study in the Koeye River watershed, which will ultimately be used in the design of a Pacific salmon management plan for Koeye and other important salmon systems throughout Heiltsuk Territory.   17  Materials and Methods  Stream and Tributary Surveys In order to identify locations of Dolly Varden spawning and rearing sites in the Koeye River watershed, I surveyed nine streams across a broad longitudinal extent of the watershed for presence of juvenile Dolly Varden (fry, i.e. immature fish that emerged recently, and parr, i.e., older immature fish with fork length greater than 80 mm that continue to display juvenile markings) in June and July 2018 (Figure 3). I surveyed each stream at least once, and surveyed streams twice if no Dolly Varden were captured during the first survey. I surveyed streams that appeared to contain suitable spawning and juvenile rearing habitat (i.e. low gradient pool-riffle sequences that contained sand, gravel, or cobble streambed substrates and cover such as large woody debris in-stream or along the shoreline; see Armstrong and Morrow 1980). Each stream survey consisted of deploying at least six minnow traps in suitable salmonid spawning habitat throughout the stream. I put a perforated bait cup filled with sockeye salmon eggs in each trap and set the traps at the heads of pools (where juvenile salmonids often feed on drift) at intervals extending as far upstream as possible. I checked the traps after 12-24 hours and identified, enumerated, and measured the fork length of individuals of all species. During one particular survey of a tributary of Lower Koeye Lake (Figure 3, “Lake Trib Right”), I tagged and released 36 Dolly Varden parr following the tagging protocol outlined below to study the movement of juveniles above the lake. I did not tag individuals during any of the other stream surveys. 18   Figure 3. Location of streams surveyed throughout the Koeye River watershed. Pink lines indicate the surveyed extent of each stream.  Downriver Movement Monitoring After occupying upstream sections of a watershed for a period of time, either as overwintering mature individuals or as developing immature individuals, anadromous salmonids move downriver to access estuarine or marine habitats. For Pacific salmon, this downriver movement represents a concerted migration of juveniles, or so-called “smolts”, that takes place in the spring and early summer in coastal watersheds of western North America. The Koeye River salmon monitoring project deployed a rotating screw trap (hereafter, smolt trap) to monitor Pacific salmon smolts during the smolt migration season from late April to late June each year from 2015 to 2019. The smolt trap was deployed at a single site mid-channel in the mainstem of the lower Koeye River (Figure 4). The open end of the trap was oriented to target individuals that were specifically moving downstream. In addition to capturing salmon smolts, the smolt trap 19  captured other salmonids that were moving downstream, including Dolly Varden. All salmonid individuals were enumerated, measured, and tagged with a Passive Integrated Transponder (PIT) tag following the tagging protocol outline below. Throughout the duration of the smolt trap sampling period, 431 Dolly Varden were tagged and released.   Figure 4. Location of Dolly Varden sampling sites throughout the Koeye River watershed. Black points and the black square represent beach seine sites throughout the estuary and the location of the smolt trap, respectively. Black bars represent the location of the four Radio Frequency Identification (RFID) antennae deployed throughout the watershed.   Estuary Monitoring Estuaries provide critical foraging and sheltering habitat to multiple life stages of anadromous salmonids and are the transition point between freshwater and marine portions of the life cycle. The Koeye River salmon monitoring project used beach seine surveys to monitor fish community structure at 19 sites throughout the estuary (Figure 4) every 10-14 days from April to 20  September each year from 2015 to 2018. The most extensive sampling took place in 2017 and 2018, which represented 500 beach seine sets over 28 rounds of sampling. Each seine set used one of three different seine nets depending on the depth and spatial features of the sampling site: a 22 m x 3.1 m net was used in the deeper sites at the river mouth; a 30 m x 1.8 m net was used nearly everywhere else throughout the estuary; and a 13.7 m x 1.2 m pole seine was used in several shallow and narrow sites. Captured individuals were identified to species, enumerated, measured, and quickly released alive. Dolly Varden individuals were separated from the rest of the sampled individuals and were processed and tagged following the tagging protocol outlined below.   PIT tagging protocol for stream, river, and estuary surveys Anadromous salmonids seasonally occupy habitats throughout coastal watersheds and their associated marine waters. Individuals move among habitats seasonally and at various life stages, and this movement can represent concerted population-scale migrations if it is a regular characteristic of the life cycle. To document the movement patterns of Dolly Varden in the Koeye River watershed, I monitored the detection timing and location of individuals that had been previously tagged with a unique PIT tag and released. The PIT tags were implanted in Dolly Varden sampled at three locations throughout the Koeye River watershed: at the lower river in a smolt trap, throughout the estuary in beach seine nets, and in a tributary of Lower Koeye Lake in minnow traps (Figure 4). Each sampled Dolly Varden was measured (fork length) and enumerated, and a handheld Radio Frequency Identification (RFID) reader was used to determine if the individual had been previously tagged with a PIT tag, in which case it was recorded as a recapture and released immediately at the capture site. Newly-captured individuals 21  (i.e. not previously tagged) that had a fork length (FL) < 80 mm, or that had visible injuries and/or scale loss, were also released immediately in order to minimize post-tagging mortality. Newly-captured individuals that were suitable for PIT tagging were placed in a container filled with an aqueous solution of buffered MS-222 (5 mg/L) until they lost equilibrium. A lock needle was used to make a small incision in either the peritoneal cavity or below the dorsal fin depending on the size of the individual, and a sterile PIT tag was implanted into the incision. A 12 mm tag was implanted in the peritoneal cavity of individuals with FL < 250 mm, and a 12, 23, or 32 mm tag was implanted in the dorsal muscle of individuals with FL > 250mm. Dolly Varden may spawn more than once in their life time (they are iteroparous) and are not reliably identified by sex outside of the breeding season. Consequently, PIT tags were implanted in the dorsal muscle whenever possible to avoid potential tag loss from the peritoneal cavity of females while releasing eggs during spawning. Tagged individuals were then placed into a 68 L container of aerated water sourced from the capture site to recover from the anesthetic, and released at the capture site once they regained equilibrium and began swimming normally.  Tracking Tagged Individuals To better understand the extent of movements and use of habitat by individual Dolly Varden through time in the Koeye River watershed, a permanent network of RFID antennae was used to passively track the movement of tagged Dolly Varden (Figure 4). The RFID network was in annual operation beginning in 2016 as part of the ongoing Koeye River salmon monitoring project. Each antenna was anchored into the river bed and spanned the width of the river. Three antennae were in operation in 2016: one in the lower river near the upper limit of the tidal influence, one at the outlet of Lower Koeye Lake, and one in the lower reaches of the Upper 22  Koeye River (above Lower Koeye Lake). Beginning in 2017, a fourth antenna was in seasonal operation in a tributary of Lower Koeye Lake. The RFID network was operated principally to monitor the spawning migration of sockeye salmon and antennae were only active between spring and fall each year, when adults migrate upriver and into spawning tributaries of Lower Koeye Lake. The two antennae in the lower river were activated each spring, before the returning adult sockeye salmon entered the river mouth, and the two antennae above Lower Koeye Lake were activated in late summer, before adults moved into spawning grounds from the lake. All antennae were active until late October, after sockeye salmon had completed spawning. There was occasional variation in RFID operation over the years of the study. For example, in September 2017 the antenna in the Upper Koeye River was damaged by a grizzly bear and was inactive for eight days. High river flow rates periodically displaced some antennae, which likely reduced the detection range for several days until they were repositioned and reconfigured.  Statistical Methods To examine size-class structure of the populations sampled during stream, river, and estuary surveys of the Koeye River watershed, I performed a mixture analysis to determine the number of modes in the size distributions of individuals sampled with each survey technique. Mixture analysis uses maximum-likelihood and Akaike Information Criteria (AIC) to select the best model from a set of models that each fit some number (k) of normal distributions to multimodal size distribution data (Akaike 1973; Macdonald and Pitcher 1979). I calculated the relative difference in AIC values from the k different model comparisons by rescaling AIC to  Δi = AICi – AICmin  23  where AICmin is the minimum of the k different AICi values. The model that fit the best value of k to the size distribution data therefore has Δi = 0, and all remaining values of k have positive Δi values. I used the following rules of thumb outlined in Burnham and Anderson (2004) to determine which value(s) of k received substantial support: Values of k associated with Δi ≤ 2 have substantial support, those with 4 ≤ Δi ≤ 7 have less support, and those with Δi > 10 have little to no support. I interpreted the value(s) of k with substantial support as the number of distinct size-classes present in the sampled population. I conducted mixture analyses of the size distributions of Dolly Varden sampled in the estuary within and across each year of sampling to determine if size-class structure was stable across years. To assess the presence, abundance, or size of juvenile individuals in the sampled streams along an upstream gradient in the watershed, I grouped streams into three regions: estuary, river, and lake, and tallied the total number of juveniles that were sampled in each region. I also calculated the total number of individuals in each size class, using size thresholds based on a combination of the results of the mixture analyses and previously published data on age-class size thresholds for a different population of Southern Dolly Varden studied by Michalski (2006). To determine if there was a relationship between body size and timing of downriver passage of Dolly Varden captured in the smolt trap, I fit a simple linear model of length ~ date pooled across all years for all potential smolts (i.e. individuals undertaking their first passage to estuarine or marine waters), which I conservatively defined as individuals with length < 160 mm. To examine the diversity of individual-level movement of Dolly Varden throughout the watershed, I produced a capture history for each of the 1,185 PIT-tagged individuals to summarize the timing and location of detections throughout the watershed over the duration of the study. I then classified subsets of capture histories that represented common detection 24  patterns based on total number detections, time between detections, and spatial history of detections– which I interpreted as movement throughout the watershed– as well as for several record-setting or otherwise interesting individuals.  To determine if the timing of Dolly Varden aggregate movement was associated with Pacific salmon migrations in the watershed, I plotted the date of all detections of PIT-tagged individuals pooled across sites throughout the entire watershed (i.e. the estuary, lower river, lake outlet, and lake tributaries) from early spring until late fall over the five years of the study. To determine if the variance in detection dates (i.e. the breadth of run timing) were different between Dolly Varden and adult migrant sockeye salmon at the lower, lake outlet, and above-lake sites, I used a Fligner-Killeen test of homogeneity of variance, which is similar to a Levene’s test, but is more robust for non-normally distributed data that include outliers.   I produced all analyses and data figures using the R statistical environment version 3.6.2 (R Development Core Team 2020) using the additional ‘mixtools’ package for the mixture analysis and numerous packages within the ‘tidyverse’ for data wrangling and visualization.   25  Results  Stream Surveys Dolly Varden juveniles (fry and parr) were sampled using minnow traps deployed from late June to July 2018 in nine streams of the Koeye River. Juveniles were present in all three regions of the Koeye River watershed (estuary, river, and lake tributaries; Figure 2), and juvenile presence, abundance, body size generally increased upstream from estuary streams to lake streams (Table 1). Mixture analysis and AIC of the size distribution of sampled individuals pooled across all regions in the watershed provided substantial support for three size classes of juveniles in streams (k = 3, AIC = 786.5; Ntotal = 117, group means = 64.7 mm [±5.15mm SD], 87.23 mm [±9.49mm SD], and 121.1 mm [±19.94 mm SD] respectively; Figure 5; Appendix Table A1).  Table 1. Summary of variation in presence, abundance, and size classes of Dolly Varden in streams throughout the Koeye River watershed. Stream Name Region Total Length > 75 mm Length > 120 mm J Creek Estuary 0 0 0 Camp Creek 0 0 0 Pole Seine Creek 0 0 0 R5S3 Creek 7 4 0 Drinkwater Creek River 4 4 0 Canyon Creek 25 11 0 Lake Narrows Creek Lake 1 1 0 Lake Tributary Left 8 4 4 Lake Tributary Right 72 63 9 26    Figure 5. Size distribution of juvenile Dolly Varden sampled in nine streams in the Koeye River watershed (N = 117). Vertical dashed lines represent size thresholds of age-classes (in years) documented for Southern Dolly Varden in the Campbell Lake watershed, British Columbia (Michalski 2006).   Downriver Movement Monitoring Dolly Varden smolts (smaller individuals presumed to be initiating their first anadromous migration) and other larger anadromous migrants were sampled using a rotating screw trap deployed in the lower Koeye River from late April to late June each year from 2015 to 2019. Mixture analysis and AIC of sampled individuals pooled across all five years of monitoring provided substantial support for three size classes of anadromous migrants (k = 3, AIC = 3317; Ntotal = 431, group means = 119.7 mm [±16.9 mm SD], 177.8 mm [±25.9 mm SD], and 258.1 mm [±32.1 mm SD] respectively; Figure 6; Appendix Table A1). Fish length increased slightly as the monitoring season progressed (Figure 7), and this trend was strongest among fish presumed to be smolts (i.e. under 160 mm; N = 364, model = lm(length ~ date), R2 = 0.058, p = 0.000001919; Figure 8). This size threshold represents the transition point between the first and second size classes of individuals measured at the smolt trap (Figure 6).  1+ 2+024640 80 120 160Fork Length (mm)CountSize distribution of juvenile fish captured in streams in summer 2018 (N = 117)27    Figure 6. Size distribution of Dolly Varden sampled at the smolt trap (N = 431). Horizontal lines represent size ranges of age-classes (in years) documented for southern Dolly Varden in the Campbell Lake watershed, British Columbia (Michalski 2006).   2 y3 y4 y010203040100 200 300Fork Length (mm)CountSize distribution of fish captured in smolt trap in spring of 2015−2019 (N = 431)28   Figure 7. Body length of Dolly Varden sampled at the smolt trap as the sampling season progressed (N = 431, pooled across 2015-2019 seasons). Horizontal line represents the presumed fork length threshold for smolts which are the focus of the analysis presented in Figure 8 below.   ●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●100150200250300May 1 May 15 June 1 June 15Fork Length (mm)Size of fish over the 2015−2019 smolt trap seasons (N = 431)−−− −− − −−− −29   Figure 8. Body length of Dolly Varden smolts (i.e. below 160 mm fork length) sampled at the smolt trap throughout the season (N = 364, pooled across 2015-2019 seasons).   Estuary Monitoring Dolly Varden in the estuary were sampled using beach seines from April to September each year from 2015 to 2018. Mixture analysis and AIC of sampled individuals pooled across all four years of monitoring provided substantial support for four size classes of individuals in the estuary aggregations (k = 4, AIC = 6507; Ntotal = 680, size class means = 136.7 mm [±19.0 mm SD], ●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●R2 = 0.058p = 1.919e−0675100125150175May 1 May 15 June 1Fork Length (mm)Variation in size of smolts over the 2015−2019 smolt trap seasons (N = 364)30  183.0 mm [±14.6 mm SD], 254.4 mm [±36.0 mm SD], and 357.5 mm [±39.9 mm SD] respectively; Figure 9; Appendix Table A1). These size classes of Dolly Varden from the Koeye River estuary were stable across years of monitoring, and mixture analysis of sampled individuals within individual years supported size classes approximately corresponding to those of individuals pooled across all years (Figure 10 a-d; Table A1): Mixture analysis and AIC of sampled individuals across individual years from 2016 to 2018 provided substantial support for both four and five size classes of individuals in 2015 (AICk=4 = 1829, AICk=5 = 1831, N = 194; Table A1); four size classes in 2016 (AICk=4 = 1274, N = 134; Table A1); two, three, and four size classes in 2017 (AICk=2 = 1817, AICk=3 = 1818, AICk=4 = 1819, N = 202; Table A1); and four size classes in 2018 (AICk=4 = 1374, N = 150; Table A1).   Figure 9. Size distribution of Dolly Varden sampled in the Koeye River estuary, pooled across all years (N = 680). Vertical dashed lines represent mean length of each size class. Horizontal lines represent size ranges of age-classes (in years) documented for southern Dolly Varden in the Campbell Lake watershed, British Columbia (Michalski 2006).   2345+01020100 200 300 400Length (mm)CountSize distribution of fish captured in estuary in summer of 2015−2018 (N = 680)31   Figure 10. Size distribution of Dolly Varden sampled in the Koeye River estuary for each year. Vertical dashed lines represent mean body length of each size-class identified by mixture analysis of individuals pooled across all years (Figure 9).  Tagging and Detection Summary A total of 1,185 Dolly Varden were tagged with a passive integrated transponder (PIT) tag between 9 May 2015 and 23 July 2019 at various sites throughout the Koeye River watershed. Between 28 May 2015 and 20 October 2019, 372 Dolly Varden (31.3% of all tagged fish) were 2015(N = 194)A05102016(N = 134)B05102017(N = 202)C05102018(N = 150)D0510100 200 300 400CountSize distribution of fish captured in estuary in summerLength (mm)32  detected 1,106 times either passively by the RFID antennae, or during smolt trap and estuary monitoring (Table 2). Of the 372 individuals that were redetected, the number of detections per individual ranged from 1 to 26 detections (mean = 2.97 [±2.98 SD]; Table 2). Of the 1,149 Dolly Varden that were tagged below the lake (i.e. at the smolt trap or in the estuary), 100 were detected by a RFID antenna above the lake at some point during the study period. The maximum number of times that an individual Dolly Varden moved from below the lake to above the lake over the duration of the study was four times. The time interval between the date of the initial tagging and the date of the most recent detection ranged from 0 (i.e. subsequently detected in the same year) to 3.25 years (mean = 0.67 years [±0.70 years SD]).   A total of 715 Dolly Varden were tagged in the estuary, and 251 (35.1%) of these individuals were subsequently detected at least once (Table 2). The majority of detections occurred within the same year across all detection locations, and the number of detections gradually decreased during each subsequent year. No individuals that had been tagged in the estuary were detected in the estuary three years after tagging, but individuals were detected everywhere else in the watershed, including above the lake, after as many as three years since being tagged (Table 3). A total of 434 Dolly Varden were tagged at the smolt trap, and 117 (26.9%) of these individuals were eventually detected again at least once (Table 2). The majority of detections occurred within the same year across all detection locations, and the number of detections gradually decreased during each subsequent year. Only two individuals were detected after three years since being tagged, and these were in the lower river (Table 4). A total of 36 Dolly Varden parr were tagged in the upper tributary of the lake in 2018, and four (11.1%) of these individuals were eventually detected again at least once (Table 2). A 33  single individual was detected the same year that it was tagged, and three individuals were detected one year after being tagged.   The most commonly observed capture history of Dolly Varden tagged either in the estuary or at the smolt trap is detection within the same year after being tagged, with no subsequent detections (Table 5). Several individuals that were tagged in the estuary were subsequently detected above the lake multiple times throughout the duration of the study, including one individual that was detected above and below the lake each year of the study, including the year that it was tagged (Table 5). Several individuals that were tagged at the smolt trap were also subsequently detected above and below the lake, but none were detected above the lake three years after being tagged. One individual was detected below the lake in each year of the study but was never detected above the lake (Table 5).  34  Initial Tagging Location Individuals Tagged Individuals Detected (All Sites) Individuals Detected Below Lake Individuals Detected Above Lake Max. Trips Above Lake Per Individual Total Detections Max. Detections  Per Individual Mean Detections  Per Individual Estuary 715 251 235 69 4 750 26 2.99 [±3.15 SD] Smolt Trap 434 117 111 31 3 346 12 2.96 [±2.63 SD] Lake Tributary 36 4 0 4 - 10 4 2.4 [±1.73 SD] Total 1185 372 346 104 - 1106 26 2.97 [±2.98 SD]  Table 2. Tagging and Detection summary of Dolly Varden in the Koeye River watershed.       35    Time interval between initial tagging in the estuary and subsequent detections (Excluding individuals that were never detected) Detection Location Same Year 1 Year 2 Years 3 Years No. Total Detections No. Unique Individuals No. Total Detections No. Unique Individuals No. Total Detections No. Unique Individuals No. Total Detections No. Unique Individuals Estuary 15 14 11 10 2 2 0 0 Lower River 167 124 107 59 31 13 3 3 Lake Outlet 108 73 47 32 7 5 3 3 Above Lake 116 49 108 29 23 5 2 2  Table 3. Detection summary over the duration of the study for Dolly Varden initially tagged in the estuary.      36    Time interval between initial tagging at the smolt trap and subsequent detections (Excluding individuals that were never detected) Detection Location Same Year 1 Year 2 Years 3 Years No. Total Detections No. Unique Individuals No. Total Detections No. Unique Individuals No. Total Detections No. Unique Individuals No. Total Detections No. Unique Individuals Estuary 5 5 1 1 0 0 0 0 Lower River 114 72 48 22 7 4 2 2 Lake Outlet 49 38 14 13 5 5 0 0 Above Lake 53 20 32 13 16 4 0 0  Table 4. Detection summary over the duration of the study for Dolly Varden initially tagged at the smolt trap.      37  Initial Tagging Location Time interval between tagging date and subsequent detections above or below the lake Same Year 1 Year 2 Years 3 Years No. Individuals Below Lake Above Lake Below Lake Above Lake Below Lake Above Lake Below Lake Above Lake Estuary  251 individuals redetected 1 0 0 0 0 0 0 0 113 1 1 0 0 0 0 0 0 24 0 0 0 1 0 0 0 0 6 0 0 0 0 0 0 1 0 2 1 1 0 1 1 1 0 0 1 0 1 1 1 1 1 0 1 1 1 0 1 0 1 0 1 1 1 0 0 1 0 0 0 1 0 1 Smolt Trap  117 individuals redetected 1 0 0 0 0 0 0 0 64 0 0 1 0 0 0 0 0 10 1 1 1 1 0 0 0 0 2 1 0 0 1 0 0 0 0 2 1 0 1 0 1 1 0 0 2 1 1 0 1 0 1 0 0 1 0 0 1 0 1 1 0 0 1 0 0 1 0 1 0 1 0 1 0 0 0 0 1 0 1 0 1  Table 5. Select capture histories for individual Dolly Varden in the Koeye River watershed.38  Timing of Population-Scale Movement Throughout the Watershed Dolly Varden individuals that had been previously tagged with a unique PIT tag were detected using RFID antennae and traps at various sites throughout the Koeye River watershed between 9 May 2015 and 20 October 2019 (see Figure 4). A distinct pattern of seasonal detections emerged at the aggregate scale of all Dolly Varden detected throughout the watershed (Figure 11). The distribution of detection dates for Dolly Varden captured moving downstream in the spring at the smolt trap and throughout the summer in the estuary broadly overlapped the sockeye salmon and coho salmon smolt migrations (Figure 11). The distribution of detection dates for Dolly Varden captured moving upstream at the lower river RFID and the lake outlet RFID broadly overlapped the adult sockeye salmon upriver spawning migration, and the variance in Dolly Varden detection dates was significantly greater than the variance in sockeye salmon dates (lower river: Fligner-Killeen chi-squared = 610.62, p < 0.000001, lake outlet: Fligner-Killeen chi-squared = 288.95, p < 0.000001; see Figure 11 and Table 6). The distribution of detection dates for Dolly Varden captured moving into streams above the lake at the upper river RFID and lake tributary RFID more closely overlapped the detection dates of adult sockeye salmon moving into these streams, but the variance in Dolly Varden date of detection was also significantly different from that of sockeye salmon (Fligner-Killeen chi-squared = 63.725, p < 0.000001; see Figure 11 and Table 6).  39   Figure 11. Date of passage of Dolly Varden and Pacific salmon (sockeye salmon and coho salmon) at detection sites throughout the Koeye River watershed. The upper panel represents downriver movement in lower river (A) and occupation of the estuary (B). The middle panel represents upriver movement in the lower river (C) and at the lake outlet (D). The lower panel represents movement into tributaries above the lake (E).     40  Detection Location Dolly Varden Sockeye Salmon Coho Salmon Smolt Trap May 16 ± 23.3 days (N = 434) May 6 ± 9.68 days (N = 6523) May 9 ± 12.1 days (N = 5722) Estuary June 24 ± 23.5 days (N = 749) May 18 ± 4.83 days (N = 220) May 31 ± 40.8 days (N = 1274) Lower River August 4 ± 36.2 days (N = 479) July 2 ± 14.0 days (N = 3988) - Lake Outlet August 17 ± 40.8 days (N = 233) July 7 ± 14.1 days (N = 2353 - Above Lake September 23 ± 23.4 days (N = 396) September 27 ± 12.3 days (N = 2591) -  Table 6. Summary of date of passage of Dolly Varden, sockeye salmon, and coho salmon at detection sites throughout the Koeye River watershed (Mean date with associated standard deviation in days).   41  Discussion  Dolly Varden life history  The broad goal of this research was to inform the conservation biology of Southern Dolly Varden on the mid-coast of British Columbia by closely studying Dolly Varden within the Koeye River watershed and summarizing data collected over a period of five years. This watershed is representative of numerous small- to medium-sized Pacific salmon-bearing watersheds on the British Columbia coast, but is also an uncommon example of an intact primary forest basin free of anthropogenic disturbance from its upper headwaters in the Coast Mountains to its estuary on the outer coast of the Pacific Ocean. This aspect of the Koeye River watershed –– along with its geographic position in the middle of the Great Bear Rainforest, one of the most intact regions remaining in North America outside of the Arctic Circle –– makes it an ideal system for studying the population biology of freshwater fishes in a relatively undisturbed state. In Canada, freshwater fish conservation assessments (e.g. COSEWIC assessments) and management plans rely heavily on an understanding of the geographic distribution of taxonomic groups, both at regional scales (e.g. the number and locations of populations throughout the range of a species) and at local scales (e.g. the number of locations containing critical habitat). For example, COSEWIC assessed the conservation status of the Northern Dolly Varden throughout its Canadian range as Special Concern primarily due to the relative scarcity of locations that supported key spawning and overwintering habitat and the vulnerability of these areas to stochastic events and climate change (COSEWIC 2011). Conservation assessments can also benefit from an understanding of the life history of freshwater fishes in undisturbed systems, because it provides a baseline when comparing populations from more disturbed locations. 42  Unlike the northern subspecies, Southern Dolly Varden have not received a conservation status assessment by COSEWIC, and this is due in part to a lack of local-scale information about the distribution and abundance of populations throughout the majority of the range, specifically on the mid-coast of British Columbia. My research therefore sought to provide local-scale information about the life history of Southern Dolly Varden and the location and seasonal usage of important habitats, and this was specifically carried out in an intact watershed on the mid-coast of British Columbia that could serve as a baseline for future studies of other populations and systems. The Koeye River watershed also supports several populations of Pacific salmon that spawn throughout the entire watershed. My work also sought to assess the potential for migratory coupling between Dolly Varden and Pacific salmon by examining the timing of seasonal movements of Dolly Varden and their Pacific salmon prey throughout the watershed. Freshwater food web ecology is challenged by the complexity of freshwater habitats and community dynamics. Habitats may undergo seasonal hydrological changes, and populations of organisms may move in and out of habitats, or seasonally fluctuate in size. Modelling the population dynamics of just a single species of interest is often a difficult task for resource managers due to field data collection limitations, and broader community-level interactions such as competition and predation may be largely unknown. Migratory coupling theory has the potential to simplify these systems by identifying and modelling relationships between species of interest (e.g. species of Pacific salmon) and other members of the freshwater fish community in order to make stronger ecological predictions about community dynamics. However, migratory coupling between freshwater predators and prey has not been adequately assessed in many freshwater systems, even in Pacific salmon systems where the seasonal resource subsidies 43  provided by spawning adults and out-migrating juveniles are thought to be a critical aspect of many coastal food webs (Naiman et al. 2002; Shindler et al. 2003).  My results provide evidence that Dolly Varden use a broad extent of the Koeye River watershed, from the river mouth to tributaries of the lake. I identified juvenile rearing habitat in tributary streams of the lower river and above the lake, and detected fry in these areas, which suggests that Dolly Varden spawn throughout the watershed both above and below the lake. I found that upper sections of the watershed (i.e. above the lake) supported more spawning and rearing habitat and contained more and larger juveniles, which is consistent with patterns observed in populations of Northern Dolly Varden in southwest Alaska (e.g. Bond et al. 2014) suggesting that it is a species-wide trait. This pattern may be driven by differences in habitat characteristics such as flow rates, productivity, and water temperature along the elevational gradient of these watersheds. Upper portions of these watersheds are fed by larger rainfall catchments and contain greater stream habitat complexity (e.g. large woody debris provides cover and stabilizes gravel to form deep persistent pools) than lower sections, which likely creates suitable habitat conditions throughout the year. Lower portions of the watershed contain streams that offer more ephemeral habitats, and are not suitable for rearing throughout the entire year. It is possible that these lower streams support Dolly Varden spawning in late fall when water levels are high, incubation over the winter, and a short juvenile rearing period following emergence in the early spring, but that low flow rates and high temperatures in the summer force fry to descend into the main river or estuary, resulting in a pattern of low or no habitat occupancy during summer surveys.  I also found evidence that larger parr were able to move among tributaries of the lake by descending into the lake and then ascending neighbouring streams. This behaviour may have 44  developed to allow access seasonal resources such as salmon eggs that were not readily available in their natal stream (e.g. Armstrong 1984). I was not able to determine the temporal scale of these movements, but it is possible that more mature parr are even able to move back and forth between streams on the scale of days to access quality foraging habitat when it becomes available due to changes in flow rates or the onset of salmon spawning, or to access thermal conditions that optimize foraging and metabolic assimilation. Similar foraging behaviours have been observed in juvenile coho salmon, where individuals that underwent weekly and even daily migrations to exploit thermal heterogeneity at the scale of hundreds of meters within and among streams matured at significantly higher rates than individuals that remained in their natal sites (Armstrong et al. 2013). I observed size classes of juveniles in the Koeye River watershed that approximately corresponded to multiple age classes of Southern Dolly Varden rearing in streams in the Campbell Lake watershed on Vancouver Island, BC (Michalski 2006). My results support the idea that juveniles feed in or near their natal streams for multiple years before descending to the lake or further down river. The largest size classes I observed in natal and rearing streams were comparable to the lower size range of individuals captured and measured at the smolt trap moving downriver towards the estuary. This suggests that Dolly Varden parr undergo smoltification when they attain a fork length of approximately 120mm, although there is considerable variation in this estimate. This estimate of smolt size is slightly lower than the mean length observed in other systems of Southern Dolly Varden. For example, Dolly Varden smolts captured moving downriver in the Keogh River on Vancouver Island had a mean fork length of 156mm (Ward and McCubbing 1998), and smolts in Hood Bay Creek in Southeastern Alaska had a mean fork length of 135mm (Armstrong 1970). In both of these other cases, smolts 45  consisted of individuals from a variety of age classes (2-4 years). I observed an increase in size of smolts sampled at the smolt trap throughout the season, which is can most easily attributed to individual growth over time (i.e. individuals are on average larger towards the end of the smolt season because they have had more time to grow). I also observed smolt-sized fish in the estuary throughout the summer, along with fish of multiple larger size classes.  I identified patterns in seasonal movements that were characterized by downriver movement into the estuary in late spring and summer, and upriver movement into and above the lake in the fall. The estuary supports large aggregations of mature Dolly Varden throughout the summer and appears to be an important foraging area for multiple size classes, including putative smolts. Dolly Varden individuals that were tagged either in the spring during the Pacific salmon smolt downriver migration season or in the summer in the estuary were subsequently detected throughout the watershed using passive RFID sensors. There was considerable variation in the extent and timing of movements throughout the watershed: some individuals that were tagged as putative smolts during the spring (i.e., moving downriver for the first time) were subsequently detected moving up river and above the lake during the same year; some individuals were not detected upriver until as many as three years after being tagged; some individuals appeared to only make one upriver journey and never being detected again; and some individuals made multiple journeys above and below the lake throughout the duration of the study. I observed similar variation in the extent and timing of movement of the individuals that were tagged in the estuary as mature fish. This variation in the duration of marine or estuarine residence prior to upriver migration appears to be characteristic of anadromous char (e.g. Armstrong 1984), and even in the population of sockeye salmon in the Koeye River, wherein a spawning run in a given year may be comprised of multiple cohorts of mature individuals that resided in marine areas for 46  one, two, or three years prior to moving upriver. While it is clear that single-winter residents account for a minority of spawning sockeye salmon in a given run, and that two- and three-year ocean residents dominate the spawning runs, the proportional representation of various cohorts of Dolly Varden in each spawning season remains unknown. This is due in part to the inability of passive monitoring techniques to discern spawning migrations from overwintering migrations, and the fact that individual Dolly Varden may undertake numerous migrations annually or throughout a watershed before ever spawning.   By analyzing detection history of tagged individuals, I determined that the timing of Dolly Varden seasonal movements broadly overlapped that of Pacific salmon migrations in the Koeye River watershed. Downriver movement below the lake and into the estuary coincided with the tail end of the smolt migrations of sockeye and coho salmon in the spring. Dolly Varden are known to prey upon juvenile salmon (e.g. Armstrong 1970), and the overlap in timing of downriver migration supports the potential for predation of juvenile sockeye and coho salmon by Dolly Varden during their downriver migrations. A similar overlap in timing between Dolly Varden and coho salmon smolts was observed in the Keogh River on Vancouver Island, which suggests that this behavior may be common trait in populations in salmon-bearing watersheds (Ward and McCubbing 1998). Dolly Varden remained abundant in the estuary throughout the summer, after sockeye and coho salmon smolt abundance had decreased. Upriver Dolly Varden movement above the estuary and into the upper reaches of the lower Koeye River broadly lagged behind adult sockeye salmon moving into the lake throughout the late summer. Movement of Dolly Varden above the lake and into tributaries was more tightly associated with the timing of sockeye salmon spawning migration. This latter tighter overlap is consistent with the idea that Dolly Varden were keying in on the salmon egg foraging opportunity which of course does not 47  exist until adult salmon arrive at their spawning locations. The increased association in timing of upriver movement as sockeye salmon entered their spawning streams and began spawning supports the existence of migratory coupling between ovivorous Dolly Varden and their sockeye salmon egg prey. This foraging behavior is likely a species-wide trait that is expressed in populations in salmon-bearing watersheds (e.g. Armstrong 1970; Bond and Quinn 2013) and even in other species of char such as Arctic char (Salvelinus alpinus) and bull trout. A similar migratory coupling has been documented between bull trout (a congener of Dolly Varden) and sockeye salmon in the Chilko Lake system, where adult bull trout undertook migrations away from their home ranges to aggregate and forage on juvenile sockeye salmon at the lake outlet at the onset of the smolt migration (Kanigan 2019).  Beyond taxonomy, one major difference between Northern and Southern Dolly Varden is the range of latitudes their distributions span. Northern Dolly Varden occupy northern regions of Canada and Alaska as far south as the Gulf of Alaska (approximately 55ºN). They are replaced to the south by populations of Southern Dolly Varden which occur in coastal watersheds as far south as Oregon (approximately 46ºN). The major climate, landscape, and ecological differences between these regions have likely driven differences in life history throughout the range of the species in northwestern North America. Northern Dolly Varden occupy watersheds that experience freezing winter temperatures and narrow periods of time when suitable water levels exist in Pacific salmon spawning tributaries which result in similarly narrow Pacific salmon spawning seasons (Quinn 2018). This results in narrower opportunities for Northern Dolly Varden to take advantage of spawning salmon prior to the onset of freezing and lowered water levels in the head waters. Anadromous Northern Dolly Varden that have spent the summer foraging in marine and estuarine habitats therefore face a tradeoff between remaining in 48  productive marine waters until moving upriver in time to overwinter and moving upriver earlier in time for the salmon subsidy (Bond and Quinn 2013). Southern Dolly Varden are unlikely to face similar tradeoffs throughout their more temperate range because southern watersheds rarely experience these freezing temperatures. Pacific salmon have broader spawning seasons and individuals in certain populations spawn well into the wither months, long after marine waters have passed peak productivity (Quinn 2018). This may allow Southern Dolly Varden to remain in marine and estuarine habitats when foraging opportunities are still optimal, while still allowing time to move upriver to take advantage of the salmon subsidies. My results from studying the timing of movement of Koeye River Southern Dolly Varden indicate that many individuals move upriver much later into the fall than those from populations in the north (Bond and Quinn 2013), which suggests these fish are less constrained in the timing of their movement throughout the watershed.  Migratory coupling The classic migratory coupling framework (Furey et al. 2018) considers coupling to exist when predators move beyond their home range to actively track prey during the prey’s seasonal movement or at their final destination. Dolly Varden have long been acknowledged as predators and scavengers of the eggs and flesh of Pacific salmon (Armstrong 1970), and were even the target of an aggressive bounty hunting program that operated throughout Western Alaska from 1920 until 1941 on the premise that Dolly Varden exerted significant predation pressure on salmon smolts (this was later determined to be inaccurate; see Morton 1982). Dolly Varden are also known to spawn in similar habitats and at similar times as Pacific salmon like sockeye and coho salmon, and to overwinter in or near these locations (Armstrong and Morrow 1980). 49  Individuals from anadromous populations also forage and grow in marine or estuarine habitats, much like Pacific salmon, and must move among these habitats as they become seasonally optimal. These aspects of Dolly Varden biology alone could result in the movement patterns I observed in the Koeye River population, without implicating any migratory coupling with Pacific salmon. Two of my observations, however, provide compelling evidence that Dolly Varden are at least taking advantage of the salmon subsidy in the watershed. First, the much closer associations in timing above the lake in the fall, specifically during the sockeye salmon spawning season, is consistent with a prediction of migratory coupling. Second, I made many first-hand observations of Dolly Varden consuming eggs during on-the-ground salmon counts in the fall which provide direct evidence of the benefits of migratory coupling. The migratory coupling framework distinguishes migratory coupling from other forms of predation on migratory prey by requiring the predator to migrate beyond their home range in order to track their prey. Without considering the influence of salmon subsidies on Dolly Varden movement, it is likely that the home range of Dolly Varden would expand or shift throughout the year due to changes in productivity in marine foraging areas and hydrological conditions of spawning and overwintering habitats. A strong argument for the existence of a migratory coupling between Dolly Varden and Pacific salmon could be made if there was evidence that Dolly Varden detour away from the migratory route they would otherwise take between marine foraging grounds and spawning and overwintering grounds in order to exploit salmon resources. Multiple confounding factors have likely driven Dolly Varden life history evolution along a similar migratory trajectory in the Koeye River watershed and other coastal watersheds, including the seasonal salmon subsidy, and it is not possible to quantitatively assess their individual contributions with my data set. It is also possible that Dolly Varden primarily use upper portions of the watershed 50  for their own spawning and overwintering, but that the sockeye salmon subsidy is what permits individuals in lower regions of the watershed to access these upper portions that would otherwise be too metabolically costly to access on an estuarine and freshwater invertebrate diet alone.  Conservation implications My research informs the conservation biology of Dolly Varden, specifically for the population in the Koeye River watershed and more broadly for populations throughout the mid-coast of British Columbia. I found that Dolly Varden in the Koeye River watershed make use of habitats that collectively span a great extent of the watershed, from the river mouth to the tributaries in the headwaters above the lake. Individual fish make use of these habitats within a single year, and in many cases, over multiple years, which suggests that habitat connectivity is a critical aspect in the full expression of Dolly Varden life history in an intact watershed. Habitat connectivity is perhaps the most critical ecological requirement of migratory freshwater fishes in general, as individual fitness and population persistence relies on the ability to move among habitats that are used for shelter, foraging, and reproduction. Although there appears to be flexibility in the timing of movement throughout a watershed at least at the population level, Dolly Varden life history still requires many of the complex habitats that exist throughout an intact coastal watershed like Koeye. Spawning occurs in the fall, in cold streams and lake tributaries, but fry do not emerge until the following spring, and may rear in natal streams for multiple years before descending into the lake or downriver (Reed 1967; Griffith 1979). This means that these streams must consistently harbor suitable conditions (i.e. clean water with consistent flow and temperatures below 15°C) year-round if they are to support Dolly Varden spawning and rearing–– in Koeye and likely in other coastal watersheds throughout the Southern Dolly Varden range. It is also 51  evident that Dolly Varden undertake large-scale movements throughout the watershed between the headwaters and the estuary. These movements require connectivity throughout the watershed in order to allow individuals to access critical habitat throughout the year and throughout their life cycles. The life history and habitat requirements of Dolly Varden partly mirror that of Pacific salmon and other anadromous salmonids in coastal British Columbia, and while Dolly Varden are not exploited to the same extent as are many populations of Pacific salmon, Dolly Varden and Pacific salmon likely share other major threats pertaining to habitat loss, connectivity loss, and climate change. The conservation status of many populations of Pacific salmon in Canada therefore suggests that range-wide assessments of Dolly Varden are needed in order to mitigate the factors that are already known to threaten other species of anadromous freshwater fishes.  Limitations of my research There were several limitations to this research. My stream surveys took place over a single summer when stream water levels were relatively low. This reduced the amount of available habitat that would otherwise be available in the fall and winter when Dolly Varden spawning occurs. As a result, I was not able to confidently identify actual spawning sites, but rather the areas in the streams that were suitable for juvenile rearing during the summer. I conducted the stream surveys by deploying minnow traps, and this may not have been as effective as other detection methods such as electrofishing. A reduction in detection efficiency could lead to underestimating or even failure to detect individuals that were truly present in the stream. I observed very few (and often zero) juveniles in streams in the lower sections of the watershed (i.e. draining into the estuary or lower river). These streams may contain little to no suitable rearing habitat year-round, but may still be used for spawning. In this case, fry would descend 52  their natal stream into the estuary or river before summer temperatures created unsuitable conditions, and before I had the opportunity to sample these individuals in my surveys. This pattern has been observed in populations of Northern Dolly Varden (Bond et al. 2014), and is therefore a potential aspect of Dolly Varden spawning activity in the Koeye River watershed.  Without otolith analysis to confirm the age and habitat use history of individuals, it was not possible to determine the exact age with complete certainty. This was a particular limitation during tagging individuals at the smolt trap, as it was not possible to determine if these individuals were in fact smolts undertaking their first downriver migration to the estuary or marine habitats, or if they had previously completed migrations down and upriver throughout the watershed. A disadvantage of otolith analysis, however, is that it requires lethal sampling, in which case it would not be possible to continue observing individuals for multiple years in order to study aspects of population-level life history. Dolly Varden life history spans many years, and it is unlikely that a five-year study has fully captured the complexity of their biology. Even fish tagged in 2015 may not have yet initiated complete movements above the lake, and fish tagged in 2018 have not been present in the study system for long enough to contribute to our understanding of long-term movement patterns.   It was also not possible to determine the watershed-level natal origin of individuals. This particular limitation applies most strongly to individuals in the estuary, because it is possible that Dolly Varden from stream and river systems adjacent to Koeye regularly move into the Koeye estuary to forage–– this has been observed in Dolly Varden systems in south-eastern Alaska (e.g. Armstrong 1984). The estuary is uncommonly large and productive in the region, and may provide critical habitat to fish that originate from outside the Koeye system. This therefore reduces the overall probability of subsequently detecting individuals upriver after tagging them 53  in the estuary, as there may be a proportion of tagged fish that will never move upriver in the Koeye watershed. Conversely, the upper portions of the Koeye watershed may be used for overwintering, but not spawning, by individuals that originated from outside the watershed. These individuals may move downriver in the spring alongside individuals of Koeye origin, even though they are not members of the same local-scale population. I was not able to address either of these aspects in my study.  Because the primary purpose of the RFID sensor network was to study the spawning migrations of adult sockeye salmon during the summer and fall, and because winter storms often create extremely high flow rates in the river, the sensors were not operated throughout the winter. Therefore, Dolly Varden movement could not be observed throughout the winter, and fish could have potentially moved upriver late into the winter, or moved downriver early in the spring without being detected. This observation bias in RFID operation dates could contribute to PIT tag detection error. Detection error was an inherent aspect of the RFID sensors even while operating under ideal conditions. The antennae had limited detection range (typically within 40cm), and are only able to detect a single PIT tag at any given time (if two fish were to move past the antennae at the same time, only one, or possibly none, of them would be recorded). The probably of false positives is negligible (i.e. detections of fish that were not actually present), but the probability of false negatives is likely considerable (i.e. not detecting a fish in a given year or at a given place even though the fish was still alive and present in the system that year and maybe have even moved past the sensors).  54  Future research There is great potential for future research to be carried out on the Dolly Varden in the Koeye River. The long-term goals of the Pacific salmon monitoring work will continue to foster the opportunity to study the long-lived Dolly Varden that co-occur with salmon in the watershed. With more years of tagging and detecting individuals, it will be possible to study population demography using statistical modelling approaches like spatial capture-recapture models to estimate population size and inter-annual survival rates. The sockeye salmon study has been operated for long enough to begin to estimate marine and in-river survival of individual cohorts, as well as the effects of climate conditions on salmon survival. Although the goals of this work are primarily to inform the sustainable management of a local sockeye salmon fishery, similar approaches can be taken with the ever-growing Dolly Varden PIT tag detection history dataset, which would give more insight into the complexity of iteroparous anadromy in char. Anadromous Dolly Varden are known to exhibit flexibility in their migratory life histories throughout their lifetime (e.g. timing and extent of migration) in order to optimize foraging, sheltering, and reproductive tradeoffs that change with age and size (e.g. Bond et al. 2015), and the long-term Dolly Varden PIT tagging study in Koeye will offer an opportunity to study how common this life history flexibility is at the population level. Dolly Varden use the full extent of the watershed for spawning and rearing habitat, and it is likely that different spawning regions of the watershed select for different migratory life histories. This could be studied in greater detail by expanding the PIT tagging efforts throughout the watershed to target juveniles in or near their natal streams in order to observe any differences in timing and extent of movement throughout the watershed. My stream surveys identified differences in habitat use by juveniles across the lower, middle, and upper portions of the watershed, and passive observation of movement of 55  individuals from natal streams throughout the watershed would be a natural extension to my work.  The continuation of a long-term PIT tagging study would also give insight into the apparent migratory coupling that exists between Dolly Varden and sockeye salmon. Although the opportunity to exploit seasonal resources in sockeye spawning streams is likely a major factor that drives Dolly Varden movement into these parts of the watershed, this is confounded by the need to access overwintering and spawning habitats which could drive a similar pattern in timing and location of Dolly Varden movement. One way to begin to disentangle these confounding factors (and to therefore determine the extent to which Dolly Varden and sockeye salmon fit into the traditional migratory coupling framework) is to study the variability of Dolly Varden movement across years to see how variability in sockeye salmon abundance and other ecological conditions may be influencing Dolly Varden movement.       56  References Akaike, H. (1973). Maximum likelihood identification of Gaussian autoregressive moving average models. Biometrika, 60(2), 255-265.  Armstrong, J. B., Schindler, D. E., Ruff, C. P., Brooks, G. T., Bentley, K. E., & Torgersen, C. E. (2013). Diel horizontal migration in streams: juvenile fish exploit spatial heterogeneity in thermal and trophic resources. Ecology, 94(9), 2066-2075.  Armstrong, R. H. (1970). Age, food, and migration of Dolly Varden smolts in southeastern Alaska. Journal of the Fisheries Board of Canada, 27(6), 991-1004.  Armstrong, R. H. (1974). Migration of anadromous Dolly Varden (Salvelinus malma) in southeastern Alaska. Journal of the Fisheries Board of Canada, 31(4), 435-444.  Armstrong, R. H. (1984). Migration of anadromous Dolly Varden charr in southeastern Alaska-a manager’s nightmare. In Biology of the Arctic charr, Proceedings of the International Symposium on Arctic Charr, Winnipeg, Manitoba, May (Vol. 198, No. 1, p. 5599570).  Armstrong, R. H., & Morrow, J. E. (1980). The dolly varden. Charrs: salmonid fishes of the genus Salvelinus, 99-140.  Atlas, W. I., Housty, W. G., Béliveau, A., DeRoy, B., Callegari, G., Reid, M., & Moore, J. W. (2017). Ancient fish weir technology for modern stewardship: lessons from community-based salmon monitoring. Ecosystem Health and Sustainability, 3(6), 1341284.  Banks, J. W. (1969). A review of the literature on the upstream migration of adult salmonids. Journal of Fish Biology, 1(2), 85-136.  Baras, E., & Lucas, M. C. (2001). Impacts of man's modifications of river hydrology on the migration of freshwater fishes: a mechanistic perspective. International Journal of Ecohydrology & Hydrobiology, 1(3), 291-304.  Barraclough, T. G., Vogler, A. P., & Harvey, P. H. (1998). Revealing the factors that promote speciation. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences, 353(1366), 241-249.  Beck, M. W., Heck, K. L., Able, K. W., Childers, D. L., Eggleston, D. B., Gillanders, B. M., Halpern, B., Hays, C. G., Hoshino, K., Minello, T. J., Orth, R. J., Sheridan, P. F., & Weinstein, M. P. (2001). The identification, conservation, and management of estuarine and marine nurseries for fish and invertebrates: a better understanding of the habitats that serve as nurseries for marine species and the factors that create site-specific variability in nursery quality will improve conservation and management of these areas. Bioscience, 51(8), 633-641.  57  Berkes, F. (2003). Alternatives to conventional management: lessons from small-scale fisheries. Environments, 31(1), 5-20.  Bloom, D. D., Weir, J. T., Piller, K. R., & Lovejoy, N. R. (2013). Do freshwater fishes diversify faster than marine fishes? A test using state-dependent diversification analyses and molecular phylogenetics of New World silversides (Atherinopsidae). Evolution, 67(7), 2040-2057.  Bond, M. H., Crane, P. A., Larson, W. A., & Quinn, T. P. (2014). Is isolation by adaptation driving genetic divergence among proximate Dolly Varden char populations? Ecology and Evolution, 4(12), 2515-2532.  Bond, M. H., Miller, J. A., & Quinn, T. P. (2015). Beyond dichotomous life histories in partially migrating populations: cessation of anadromy in a long-lived fish. Ecology, 96(7), 1899-1910.  Bond, M. H., & Quinn, T. P. (2013). Patterns and influences on Dolly Varden migratory timing in the Chignik Lakes, Alaska, and comparison of populations throughout the northeastern Pacific and Arctic oceans. Canadian Journal of Fisheries and Aquatic Sciences, 70(5), 655-665.  Brown, L. G. (1992). On the zoogeography and life history of Washington native char Dolly Varden (Salvelinus malma) and bull trout (Salvelinus confluentus). Washington Department of Wildlife. Fisheries Management Division Report, Olympia, Washington.  Burham, K. P., & Anderson, D. R. (2002). Model selection and multimodel inference: A practical information-theoretic approach. Springer. New York.  Burnham, K. P., & Anderson, D. R. (2004). Multimodel inference: understanding AIC and BIC in model selection. Sociological methods & research, 33(2), 261-304.  Carrete Vega, G., & Wiens, J. J. (2012). Why are there so few fish in the sea? Proceedings of the Royal Society B: Biological Sciences, 279(1737), 2323-2329.  Connors, K., Jones, E., Kellock, K., Hertz, E., Honka, L., & Belzile, J. (2018). BC Central Coast: a snapshot of salmon populations and their habitats. Pac. Salmon Fdn. Tech. Rep. https://salmonwatersheds. ca/library/lib_442. COSEWIC (2011). COSEWIC assessment and status report on the Dolly Varden Salvelinus malma malma (Western Arctic populations) in Canada. Committee on the Status of Endangered wildlife in Canada. Canadian Wildlife Service, Ottawa, Ont. COSEWIC (2018). COSEWIC guidelines for recognizing designatable units. Canadian Wildlife Service, Ottawa, Ont.  Dawson, M. N. (2012). Species richness, habitable volume, and species densities in freshwater, the sea, and on land. Frontiers of Biogeography, 4(3).  58  DeCicco, F., & Reist, J. (1999). Distribution of Dolly Varden (Salvelinus malma) in North-East Asia. In Proc. 8th and 9th ISACF Workshops on Arctic Char, Maine (Vol. 7, pp. 13-18).  Dennert, A. M., May-McNally, S. L., Bond, M. H., Quinn, T. P., & Taylor, E. B. (2016). Trophic biology and migratory patterns of sympatric Dolly Varden (Salvelinus malma) and Arctic char (Salvelinus alpinus). Canadian Journal of Zoology, 94(8), 529-539.  Denton, K. P., Rich Jr, H. B., & Quinn, T. P. (2009). Diet, movement, and growth of Dolly Varden in response to sockeye salmon subsidies. Transactions of the American Fisheries Society, 138(6), 1207-1219.   Furey, N. B., Armstrong, J. B., Beauchamp, D. A., & Hinch, S. G. (2018). Migratory coupling between predators and prey. Nature Ecology & Evolution, 2(12), 1846-1853.  Gende, S. M., & Quinn, T. P. (2004). The relative importance of prey density and social dominance in determining energy intake by bears feeding on Pacific salmon. Canadian Journal of Zoology, 82(1), 75-85.  Griffith, R. P. (1979). The spawning and rearing habitat of Dolly Varden char and Yellowstone cutthroat trout in allopatry and in sympatry with selected salmonids. Victoria: British Columbia Ministry of Environment, Fish and Wildlife Branch.  Hairston, N. G., Smith, F. E., & Slobodkin, L. B. (1960). Community structure, population control, and competition. The American Naturalist, 94(879), 421-425.  Hatfield, T. (2001). Status of the stickleback species pair, Gasterosteus spp., in Hadley Lake, Lasqueti Island, British Columbia. Canadian Field-Naturalist, 115(4), 579-583.  Hocking, M. D., & Reimchen, T. E. (2009). Salmon species, density and watershed size predict magnitude of marine enrichment in riparian food webs. Oikos, 118(9), 1307-1318.  Holtby, L. B., Ciruna, K. A., & Department of Fisheries and Oceans, Ottawa, ON (Canada); Canadian Science Advisory Secretariat, Ottawa, ON (Canada). (2008). Conservation units for Pacific salmon under the Wild Salmon Policy (No. 2007/070). DFO, Ottawa, ON (Canada).  Jonsson, B., & Jonsson, N. (2011). Habitats as Template for Life Histories. In Ecology of Atlantic Salmon and Brown Trout (pp. 1-21). Springer, Dordrecht.  Kanigan, A. M. (2019). The movements and distribution of bull trout (Salvelinus confluentus) in response to sockeye salmon (Oncorhynchus nerka) migrations in the Chilko Lake system, British Columbia (MSc Thesis, University of British Columbia).  59  Kershner, J. L., Williams, J. E., Gresswell, R. E., & Lobón-Cerviá, J. (2019). Global Status of Trout and Char: Conservation Challenges in the Twenty-First Century. In Trout and char of the world (pp. 717-760). American Fisheries Society.  Kowalchuk, M. W., Sawatzky, C. D., & Reist, J. D. (2010). A Review of the taxonomic structure within Dolly Varden, Salvelinus malma (Walbaum 1792), of North America. Fisheries and Oceans Canada, Science.  Kundzewicz, Z. W., Mata, L. J., Arnell, N. W., Döll, P., Jimenez, B., Miller, K., Oki, T., Sen, Z., & Shiklomanov, I. (2008). The implications of projected climate change for freshwater resources and their management. Hydrological sciences journal, 53(1), 3-10.  Levin, S. A. (1999). Towards a science of ecological management. Conservation Ecology, 3(2).  Levy, D. A., & Slaney, T. L. (1993). A review of habitat capacity for salmon spawning and rearing. Resources Inventory Committee.  Macdonald, P. D. M., & Pitcher, T. J. (1979). Age-groups from size-frequency data: a versatile and efficient method of analyzing distribution mixtures. Journal of the Fisheries Board of Canada, 36(8), 987-1001.  May, R. M. (1994). Biological diversity: differences between land and sea. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences, 343(1303), 105-111.  May-McNally, S. L., Quinn, T. P., Woods, P. J., & Taylor, E. B. (2015). Evidence for genetic distinction among sympatric ecotypes of Arctic char (Salvelinus alpinus) in south-western Alaskan lakes. Ecology of Freshwater Fish, 24(4), 562-574.  Michalski, T. (2006). Dolly Varden Stock Status and Habitat Preferences in the Lower Campbell Lake Watershed, Vancouver Island, British Columbia.  Moore, A., Bendall, B., Barry, J., Waring, C., Crooks, N., & Crooks, L. (2012). River temperature and adult anadromous Atlantic salmon, Salmo salar, and brown trout, Salmo trutta. Fisheries Management and Ecology, 19(6), 518-526.  Moore, J. S., Bajno, R., Reist, J. D., & Taylor, E. B. (2015). Post-glacial recolonization of the North American Arctic by Arctic char (Salvelinus alpinus): genetic evidence of multiple northern refugia and hybridization between glacial lineages. Journal of Biogeography, 42(11), 2089-2100.  Morton, W. M. (1982). Comparative catches and food habits of Dolly Varden and Arctic charrs, Salvelinus malma and S. alpinus, at Karluk, Alaska, in 1939–1941. Environmental Biology of Fishes, 7(1), 7-28.  60  Morrow, J. E. (1980). Analysis of the Dolly Varden charr, Salvelinus malma, of northwestern North America and northeastern Siberia. Charrs: salmonid fishes of the genus Salvelinus. Dr. W. Junk, the Hague, Netherlands, 323-338.  Naiman, R. J., Alldredge, J. R., Beauchamp, D. A., Bisson, P. A., Congleton, J., Henny, C. J., Huntly, N., Lamberson, R., Levings, C., Merrill, E. N., Pearcy, W. G., Rieman, B. E., Ruggerone, G. T., Scarnecchia, D., Smouse, P. E., & Wood, C. C. (2012). Developing a broader scientific foundation for river restoration: Columbia River food webs. Proceedings of the National Academy of Sciences, 109(52), 21201-21207.  Nehlsen, W., Williams, J. E., & Lichatowich, J. A. (1991). Pacific salmon at the crossroads: stocks at risk from California, Oregon, Idaho, and Washington. Fisheries, 16(2), 4-21.  Paine, R. T. (1980). Food webs: linkage, interaction strength and community infrastructure. Journal of animal ecology, 49(3), 667-685.  Pearson, M. P. (1999). The biology and management of the Salish sucker and Nooksack dace. In Proceedings of a Conference on the Biology and Management of Species and Habitats at Risk, Kamloops, BC (pp. 15-19).  Phillips, R. B., Gudex, L. I., Westrich, K. M., & DeCicco, A. (1999). Combined phylogenetic analysis of ribosomal ITS1 sequences and new chromosome data supports three subgroups of Dolly Varden char (Salvelinus malma). Canadian Journal of Fisheries and Aquatic Sciences, 56(8), 1504-1511.  Polis, G. A., W. B. Anderson, and R. D. Holt. (1997). Toward an integration of landscape ecology and food web ecology: the dynamics of spatially subsidized food webs. Annual Review of Ecology and Systematics 28: 289-316.  Power, M. E., & Dietrich, W. E. (2002). Food webs in river networks. Ecological Research, 17(4), 451-471.  Quinn, T. P. (2018). The behavior and ecology of Pacific salmon and trout. University of Washington press.  Quinn, T. P., & Adams, D. J. (1996). Environmental changes affecting the migratory timing of American shad and sockeye salmon. Ecology, 77(4), 1151-1162.  Quinn, T. P., Hodgson, S., & Peven, C. (1997). Temperature, flow, and the migration of adult sockeye salmon (Oncorhynchus nerka) in the Columbia River. Canadian Journal of Fisheries and Aquatic Sciences, 54(6), 1349-1360.  R Development Core Team. (2020). R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/   61  Redenbach, Z., & Taylor, E. B. (2002). Evidence for historical introgression along a contact zone between two species of char (Pisces: Salmonidae) in northwestern North America. Evolution, 56(5), 1021-1035.  Redenbach, Z., & Taylor, E. B. (2003). Evidence for bimodal hybrid zones between two species of char (Pisces: Salvelinus) in northwestern North America. Journal of Evolutionary Biology, 16(6), 1135-1148.  Reed, R. J. (1967). Observation of fishes associated with spawning salmon. Transactions of the American Fisheries Society, 96(1), 62-67.  Reist, J. D., Power, M., & Dempson, J. B. (2013). Arctic charr (Salvelinus alpinus): a case study of the importance of understanding biodiversity and taxonomic issues in northern fishes. Biodiversity, 14(1), 45-56.  Rounsefell, G.A. (1958) Anadromy in North American Salmonidae. Fish. Bull. 131, 171–185.   Schindler, D. E., Scheuerell, M. D., Moore, J. W., Gende, S. M., Francis, T. B., & Palen, W. J. (2003). Pacific salmon and the ecology of coastal ecosystems. Frontiers in Ecology and the Environment, 1(1), 31-37.  Seitz, K., Atlas, W., Millard-Martin, B., Reid, J., Heavyside, J., Hunt, B., & Moore, J. (2020). Size-spectra analysis in the estuary: assessing fish nursery function across a habitat mosaic. Ecosphere, 11(11)  Species at Risk Act, SC (2002), c 29, <http://canlii.ca/t/54tst> retrieved on 2020-11-02  Spence, B. C., & Hall, J. D. (2010). Spatiotemporal patterns in migration timing of coho salmon (Oncorhynchus kisutch) smolts in North America. Canadian Journal of Fisheries and Aquatic Sciences, 67(8), 1316-1334.  Stearns, S. C. (1992) The Evolution of Life Histories. Oxford University Press, Oxford  Stewart, D. B., Mochnacz, N. J., Reist, J. D., Carmichael, T. J., & Sawatzky, C. D. (2010). Fish life history and habitat use in the Northwest Territories: Dolly Varden (Salvelinus malma). Canadian Manuscript Report of Fisheries and Aquatic Sciences, 2915.  Stockner, J. G., & MacIsaac, E. A. (1996). British Columbia lake enrichment programme: two decades of habitat enhancement for sockeye salmon. Regulated Rivers: Research & Management, 12(4-5), 547-561.  Swanson, H. K., Kidd, K. A., & Reist, J. D. (2010). Effects of partially anadromous Arctic charr (Salvelinus alpinus) populations on ecology of coastal Arctic lakes. Ecosystems, 13(2), 261-274. Taylor, E. B. (1991). A review of local adaptation in Salmonidae, with particular reference to Pacific and Atlantic salmon. Aquaculture, 98(1-3), 185-207. 62   Taylor, E. B., Lowery, E., Lilliestråle, A., Elz, A., & Quinn, T. P. (2008). Genetic analysis of sympatric char populations in western Alaska: Arctic char (Salvelinus alpinus) and Dolly Varden (Salvelinus malma) are not two sides of the same coin. Journal of Evolutionary Biology, 21(6), 1609-1625.  Taylor, E. B., & May-McNally, S. L. (2015). Genetic analysis of Dolly Varden (Salvelinus malma) across its North American range: evidence for a contact zone in southcentral Alaska. Canadian Journal of Fisheries and Aquatic Sciences, 72(7), 1048-1057.  Trosper, R. L. (2002). Northwest coast indigenous institutions that supported resilience and sustainability. Ecological Economics, 41(2), 329-344.  Ward, B. R., & McCubbing, D. J. F. (1998). Adult steelhead and salmonid smolts at the Keogh River during spring 1998 in comparison to the historic record. Prov. BC Fish. Tech. Circ, (102), 84.  Waycott, M., Duarte, C. M., Carruthers, T. J., Orth, R. J., Dennison, W. C., Olyarnik, S., & Kendrick, G. A. (2009). Accelerating loss of seagrasses across the globe threatens coastal ecosystems. Proceedings of the national academy of sciences, 106(30), 12377-12381.  Whiteley, A. R., Penaluna, B. E., Taylor, E. B., Weiss, S., Abadia-Cardoso, A., & Gomez-Uchida, D. (2019). Trout and char: taxonomy, systematics and biogeography. Trout and Charr of the World. eds JL Kershner, NE Williams, RE Gresswell, and H. Lobón-Cerviá (Bethesda, MD: American Fisheries Society), 95-140.  63  Appendix                                       Table A1. AIC scores from mixture analysis of size distributions for each survey type of Dolly Varden where k represents the number of potential size classes in the mixture model. The most highly supported values of k (i.e. Δi ≤ 2) for each comparison are shaded in grey. Location k AIC Δi Streams 1 810.2 23.7 2 788.6 2.1 3 786.5 0 4 789.4 2.9 5 790.9 4.4 Smolt Trap 1 3562 245 2 3329 12 3 3317 0 4 3321 4 5 3324 7 Estuary (all years) 1 6647 140 2 6536 29 3 6523 16 4 6507 0 5 6512 5 Estuary (2015) 1 1868 39 2 1851 22 3 1845 16 4 1829 0 5 1831 2 Estuary  (2016) 1 1291 17 2 1282 8 3 1278 4 4 1274 0 5 1278 4 Estuary (2017) 1 1895 78 2 1817 0 3 1818 1 4 1819 2 5 1821 4 Estuary (2018) 1 1430 56 2 1394 20 3 1381 7 4 1374 0 5 1379 5 

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}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            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:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0397105/manifest

Comment

Related Items