BULL TROUT REDD COUNT SURVEYS IN SELECT KOOTENAY LAKE TRIBUTARIES (2011) AND RECOMMENDATIONS FOR FUTURE SURVEYS

February 2012

BULL TROUT REDD COUNT SURVEYS IN SELECT KOOTENAY LAKE TRIBUTARIES (2011) AND RECOMMENDATIONS FOR FUTURE SURVEYS

February 2012

Prepared for:

Fish and Wildlife Compensation Program – Columbia Basin 103-333 Victoria St. Nelson, BC

Prepared by:

Greg Andrusak, RPBio & Harvey Andrusak, RPBio

Redfish Consulting Ltd. 5244 Highway 3A Nelson, BC V1L 6N6

Cover Photo: ‘Canyon section of lower Woodbury Creek in mid-October.’ Photograph taken on the 19th of October 2011 by Greg Andrusak.

The Fish and Wildlife Compensation Program is a joint initiative between BC Hydro, the BC Ministry of Ministry of Forests, Lands and Natural Resource Operations (MFLNRO) and Fisheries & Oceans Canada (DFO) to conserve and enhance fish and wildlife populations affected by the construction of BC Hydro dams in Canada's portion of the Columbia Basin.

Suggested Citation: G.F. Andrusak and H. Andrusak. 2012. Bull trout (Salvelinus confluentus) redd count surveys in select Kootenay Lake tributaries (2011) and recommendations for future surveys.. Report prepared for Fish and Wildlife Compensation Program – Columbia Basin(Nelson, BC) by Redfish Consulting Ltd. (Nelson, BC). FWCP Report No. XXX. XX pp. + app.

Executive Summary

The Fish and Wildlife Compensation Program – Columbia Basin (FWCP) annually funds two major compensation projects on Kootenay Lake: a large scale nutrient restoration project, and Meadow Creek kokanee spawning channel. A primary rationale for these projects is restoration of bull trout and Gerrard rainbow trout populations; however, there has been no lake-wide measure of bull trout abundance to use as a performance measure. The purpose of this study was to establish a first lake-wide index of spawning bull trout distribution and abundance in Kootenay Lake tributaries using redd counts. These counts, if repeated over the long term, will provide a valuable performance measure for evaluating the response of bull trout to FWCP compensation efforts and management decisions on the lake.

During the fall of 2011, a comprehensive assessment of bull trout spawning was conducted in all Kootenay Lake tributaries below Duncan Dam with previously documented adfluvial bull trout presence. This report also includes results of a redd count on the Westfall River upstream of Duncan Dam, using data provided from a separate study. Approximately 260 km of stream habitat was surveyed to obtain spawner abundance and distribution in tributaries to the lake, slightly higher than the proposed 230 km. Including the Westfall River above the Duncan dam, another 21 km was surveyed, bringing the total to ~280 km. Surveys were conducted by separating the lake into three geographical groups, proximal areas included; Lardeau/Duncan tributaries (Group 1), Central Kootenay Lake tributaries (Group 2) and South Arm tributaries (Group 3). Despite the magnitude of the surveys, a number of potentially key bull trout tributaries were not assessed or identified, especially within the upper Duncan River, Lardeau River, Trout Lake basin and Kootenay River drainage.

Initial reconnaissance surveys were conducted in September on most systems to identify locations of barriers to fish passage and observe the spatial extent of bull trout spawning within each stream. These surveys indicated spawning bull trout were distributed over a large geographic area within Kootenay Lake. Much of their distribution is inter-connected by large sub-basins (Duncan Reservoir) and rivers (Lardeau and Kootenay rivers) and within tributaries to these systems. However, many of the tributaries offer limited access to much of the available habitat for spawning as a result of naturally occurring barriers that obstruct fish passage. Some of the systems were dominated by glacial inputs with sections of moderate to steep gradients. Habitat within these steep gradient sections consisted of step-pool or cascade morphology, often proving difficult to survey. Some of the lower sections of tributaries, especially Hamill and Cooper creeks, were incised in steep bedrock canyons up to 500 m deep with a gradient of up to 25%, inaccessible by survey crews for safety reasons. Habitat within

Redfish Consulting Ltd. Page i

the lower gradient sections consisted of riffle-pool and/or cascade morphology and were less difficult to survey.

Within a week of the reconnaissance surveys, redd surveys commenced at the beginning of October, and confirmed that bull trout spawning activity had peaked by mid- September with redd construction largely completed by mid-October. Most importantly, survey timing was ideal for most streams with good weather that provided excellent viewing conditions for conducting redd counts. Overall, a total of 1,711 redds were enumerated within the 19 tributaries below Duncan Dam, with an additional 114 redds counted in a tributary to the Duncan Reservoir, the Westfall River. Two prominent systems, namely the Kaslo River and Midge Creek, accounted for ~50% of all redds enumerated in all steams surveyed, with counts exceeding 400 redds each. A total of 512 redds were enumerated within the entire Kaslo River drainage, including 439 redds in the upper 21 km of the Kaslo River.

Independent estimates of spawner numbers from resistivity counters provide the opportunity to convert redd counts into indices of adult escapement on select streams. An electronic resistivity counter, installed on the upper Kaslo River, enumerated ~1,180 bull trout kelts descending from the upper portion of the river following spawning. Using the calibrated electronic counts and the redd numbers observed in 2011 on the upper Kaslo River, a bull trout per redd ratio of 2.7 was obtained. This derived expansion factor in 2011 was slightly higher than the average of 2.4 bull trout derived over the previous four years. Using the range in the expansion factor of 1.9-2.4 bull trout per redd from data collected on Crawford Creek and the Kaslo river, accounting for harvest and iteroparity, it is conceivable that the spawning population utilizing Kootenay Lake tributaries exceeds 7,000 bull trout annually (see Discussion).

In summary, redd counts provide a relatively cost effective monitoring tool to assess bull trout population trends in tributaries to Kootenay Lake. The success of FWCP initiatives, and future bull trout conservation and management decisions rely on the ability of biologists to accurately assess and monitor their status or abundance, particularly in response to management actions that are implemented. A series of recommendations are provided for establishment of appropriate, representative index streams for future monitoring of spawning bull trout within tributaries of Kootenay Lake.

Redfish Consulting Ltd. Page ii

Acknowledgements

Funding for this project was provided by the Fish and Wildlife Compensation Program – Columbia Basin (FWCP). The FWCP is a partnership of BC Hydro, the BC Ministry of Forests, Lands and Natural Resource Operations (MFLNRO), BC Ministry of Environment, and Canada Department of Fisheries and Oceans. In kind support was also provided by several MFLNRO employees. Bull trout, an endemic char, are highly valued in the Kootenay Region and many dedicated people are involved in their protection, management and conservation. Specifically, Jeff Burrows (MFLNRO) and James Baxter (FWCP) are acknowledged for their efforts in management and conservation of these unique char.

BC Hydro and Trevor Oussoren are acknowledged and thanked for their agreement to include the Westfall River data within this report.

Stefan Himmer (contractor), Gary Pavan (contractor), Jimmy Robbins (contractor), Clint Tarala (contractor), Louise Porto (AMEC), Crystal Lawrence (AMEC), Murray Pearson, Matt Neufeld (MFLNRO) Sue Pollard (MFLNRO) and Albert Chirico (Ministry of Environment) are acknowledged for their hard work during the Kootenay Lake redd surveys. Jeremy Baxter (contractor), Ico de Zwart (MEC) and Al Irvine (MEC) are acknowledged for their hard work during the Westfall River redd surveys

Special thanks to Dam Helicopters and owner/pilot Duncan Wassick who provided helicopter service for the reconnaissance and survey work. As well, Gary Pavan for organizing all GIS data and providing GIS maps.

Thanks to Steve Arndt (FWCP) and James Baxter (FWCP) for review of this report

Redfish Consulting Ltd. Page iii

Table of Contents

EXECUTIVE SUMMARY ...... I ACKNOWLEDGEMENTS ...... III TABLE OF CONTENTS ...... IV LIST OF TABLES ...... VI LIST OF FIGURES ...... VI INTRODUCTION ...... 1 RATIONALE AND OBJECTIVES ...... 2 BACKGROUND ...... 3 BULL TROUT LIFE HISTORY OVERVIEW ...... 3 BULL TROUT AND KOOTENAY LAKE OVERVIEW ...... 4 SITE LOCATION ...... 6 Kootenay Lake ...... 6 DUNCAN RIVER TRIBUTARIES- ABOVE DUNCAN DAM ...... 7 Westfall River ...... 7 GROUP 1 (DUNCAN/LARDEAU TRIBUTARIES) ...... 9 Healy Creek ...... 9 Poplar Creek ...... 9 Hamill Creek ...... 9 Cooper Creek ...... 10 GROUP 2 (CENTRAL KOOTENAY LAKE TRIBUTARIES)...... 13 Kaslo River ...... 13 Woodbury Creek ...... 13 Coffee Creek ...... 14 Crawford Creek ...... 14 GROUP 3 (SOUTH ARM KOOTENAY LAKE TRIBUTARIES) ...... 16 Midge Creek ...... 16 Cultus Creek ...... 16 Summit Creek ...... 17 METHODS ...... 19 STREAM SELECTION ...... 19 RECONNAISSANCE SURVEYS ...... 19 REDD IDENTIFICATION ...... 19 REDD COUNT STANDARDIZATION AND INTER-OBSERVER VARIATION ...... 20 REDD SURVEYS ...... 21 ELECTRONIC RESISTIVITY COUNTER...... 22 RESULTS ...... 23 INTER-OBSERVER VARIATION ...... 23

Redfish Consulting Ltd. Page iv

ELECTRONIC RESISTIVITY COUNTER...... 24 REDD COUNTS ...... 25 Westfall River ...... 25 GROUP 1 (DUNCAN/LARDEAU TRIBUTARIES) ...... 25 Healy Creek ...... 25 Poplar Creek ...... 26 Hamill Creek ...... 27 Cooper Creek ...... 27 GROUP 2 (CENTRAL KOOTENAY LAKE TRIBUTARIES)...... 29 Kaslo River ...... 29 Woodbury Creek ...... 30 Coffee Creek ...... 31 Crawford Creek ...... 31 GROUP 3 (SOUTH ARM KOOTENAY LAKE TRIBUTARIES) ...... 33 Midge Creek ...... 33 Cultus Creek ...... 34 Summit Creek ...... 35 DISCUSSION ...... 37 KOOTENAY LAKE BULL TROUT MONITORING-OVERVIEW ...... 37 REDD SURVEY INTER-OBSERVER VARIATION ...... 38 KOOTENAY LAKE BULL TROUT ABUNDANCE...... 39 KOOTENAY LAKE BULL TROUT DISTRIBUTION AND REDD DENSITY ...... 42 KOOTENAY LAKE BULL TROUT HARVEST IMPLICATIONS ...... 44 FUTURE MONITORING ...... 45 Group 1 (Duncan/Lardeau tributaries) a minimum of two tributaries ...... 47 Group 2 (Central Kootenay Lake tributaries) a minimum of two tributaries ...... 47 Group 3 (South Arm Kootenay Lake tributaries) a minimum of one tributaries ...... 47 Duncan River tributaries- (above Duncan Dam) a minimum of one tributaries...... 47 CONCLUSIONS ...... 48 RECOMMENDATIONS ...... 49 REFERENCES ...... 50 APPENDIX 1–TRIBUTARY INFORMATION...... 57 APPENDIX 2–FIELD DATA FORMS ...... 58 APPENDIX 3–PHOTOS ...... 59 APPENDIX 4–REDD SURVEY DISTRIBUTIONS IN SELECT TRIBUTARIES ...... 65 APPENDIX 5–FUTURE MONITORING-REDD SURVEY TIMING ...... 75

Redfish Consulting Ltd. Page v

List of Tables

TABLE 1. STREAM LENGTH AND LOCATION FOR GROUP 1 (DUNCAN/LARDEAU TRIBUTARIES) SYSTEMS OR BASINS, AND THE WESTFALL RIVER...... 9 TABLE 2. STREAM LENGTH AND LOCATION FOR GROUP 2 (CENTRAL KOOTENAY LAKE TRIBUTARIES) SYSTEMS OR BASINS...... 13 TABLE 3. STREAM LENGTH AND LOCATION FOR GROUP 3 (SOUTH ARM KOOTENAY LAKE TRIBUTARIES) SYSTEMS OR BASINS...... 16 TABLE 4. NUMBER OF BULL TROUT REDDS COUNTED BY FOUR TWO-MAN CREWS IN TWO SECTIONS OF THE KASLO RIVER...... 23 TABLE 5. GROUP 1 (LARDEAU/DUNCAN TRIBUTARIES) BARRIER LOCATION, CHARACTERISTICS AND ACCESSIBLE HABITAT TO OBSTRUCTION...... 29 TABLE 6. GROUP 1(LARDEAU/DUNCAN TRIBUTARIES) REDD SURVEY DATA ...... 29 TABLE 7. GROUP 2 (CENTRAL KOOTENAY LAKE TRIBUTARIES) BARRIER LOCATION, CHARACTERISTICS AND ACCESSIBLE HABITAT TO OBSTRUCTION...... 32 TABLE 8. GROUP 2 (CENTRAL KOOTENAY LAKE TRIBUTARIES) REDD SURVEY DATA...... 33 TABLE 9. GROUP 3 (SOUTH ARM KOOTENAY LAKE TRIBUTARIES) BARRIER LOCATION, CHARACTERISTICS AND ACCESSIBLE HABITAT TO OBSTRUCTION...... 36 TABLE 10. GROUP 3 (SOUTH ARM KOOTENAY LAKE TRIBUTARIES) REDD SURVEY DATA...... 36 TABLE 11. DERIVATION OF EXPANSION FACTOR FROM ESTIMATED REDDS COUNTS AND ELECTRONIC RESISTIVITY COUNTS FROM THE KASLO RIVER (2006-2011) AND CRAWFORD CREEK (2008- 2010)...... 40 TABLE 12. ESTIMATED NUMBER OF BULL TROUT TRANSFERRED AND NUMBER OF TRANSFERS AT DUNCAN DAM (BC HYDRO DATA ON FILE.)...... 41 TABLE 13. REDD DENSITY BASED ON ACCESSIBLE HABITAT FROM COMPLETE/PARTIAL BARRIERS ON SELECT SYSTEMS (INCLUDING TRIBUTARIES) IN FALL 2011...... 44

List of Figures

FIGURE 1. MAP OF KOOTENAY LAKE AND SELECT TRIBUTARIES FOR CONDUCTING REDD SURVEYS IN FALL OF 2011...... 8 FIGURE 2. LARDEAU/DUNCAN RIVER TRIBUTARIES COMPRISING GROUP 1 NORTH AND THE WESTFALL RIVER...... 11 FIGURE 3. LARDEAU/DUNCAN RIVER TRIBUTARIES COMPRISING GROUP 1 SOUTH...... 12 FIGURE 4. CENTRAL KOOTENAY LAKE TRIBUTARIES COMPRISING GROUP 2...... 15 FIGURE 5. SOUTH ARM KOOTENAY LAKE TRIBUTARIES COMPRISING GROUP 3...... 18 FIGURE 6. NUMBER OF BULL TROUT REDDS COUNTED AND COUNTING ERRORS ...... 24

Redfish Consulting Ltd. Page vi

FIGURE 7. DISTRIBUTION OF REDDS COUNTED IN THE WESTFALL RIVER IN FALL OF 2011 ...... 65 FIGURE 8. REDD SURVEY CONDUCTED IN HEALY CREEK IN FALL OF 2011 ...... 65 FIGURE 9. DISTRIBUTION OF REDDS COUNTED IN POPLAR CREEK IN FALL OF 2011 ...... 66 FIGURE 10. DISTRIBUTION OF REDDS COUNTED IN HAMILL CREEK AND TRIBUTARIES IN FALL OF 2011 .. 67 FIGURE 11. DISTRIBUTION OF REDDS COUNTED IN COOPER CREEK AND TRIBUTARIES IN FALL OF 2011 .. 68 FIGURE 12. DISTRIBUTION OF REDDS COUNTED IN THE KASLO RIVER AND TRIBUTARIES IN FALL OF 2011 69 FIGURE 13. DISTRIBUTION OF REDDS COUNTED IN COFFEE CREEK AND WOODBURY CREEK IN FALL OF 2011 ...... 70 FIGURE 14. DISTRIBUTION OF REDDS COUNTED IN CRAWFORD CREEK AND TRIBUTARIES IN FALL OF 2011 ...... 71 FIGURE 15. DISTRIBUTION OF REDDS COUNTED IN MIDGE CREEK AND TRIBUTARIES IN FALL OF 2011 ... 72 FIGURE 16. DISTRIBUTION OF REDDS COUNTED IN CULTUS CREEK IN FALL OF 2011 ...... 73 FIGURE 17. DISTRIBUTION OF REDDS COUNTED IN SUMMIT CREEK AND TRIBUTARIES IN FALL OF 2011 . 74

Redfish Consulting Ltd. Page vii

Introduction

Bull trout (Salvelinus confluentus) are a char species endemic to western North America. Within British Columbia they have been identified as a species of special concern by the Provincial Government as a result of their specific habitat requirements, and declining trends in population abundance and distribution (BC Environment 1994; Hagen and Decker 2011). Despite the impacts of stream habitat alteration and exotic species introduction, British Columbia prevails as a species stronghold for bull trout in North America (McPhail 2007).

Kootenay Lake and many of its tributaries provide a vast area of potentially suitable spawning and rearing habitat for bull trout. The lake itself provides a large lacustrine habitat in which a large portion of the adult population resides for the majority of their life history. However, very little is known about the status of the adfluvial bull trout population in Kootenay Lake including: location of major spawning systems and specific spawning sites within these systems, the relative importance of the different systems to overall bull trout production, and spawner escapement to any of these systems. This lack of information is considered problematic since bull trout are known to be highly vulnerable to fishing, and very sensitive to changes or alterations in their stream habitats (Dunham and Rieman 1999; Dunham and Rieman 2003).

Bull trout are the target of an intensive lake fishery with an estimated average of ~12,000 caught annually in Kootenay Lake (Redfish Consulting 2007). A recent, large scale exploitation study has also indicated that harvest rates for bull trout may be above sustainable levels on the lake (Andrusak and Thorley 2011). Needless to say, concern of over‐exploitation alone warrants continued monitoring of trend of Kootenay Lake’s bull trout populations. Expansion of an existing bull trout monitoring program on the Kaslo River and Crawford Creek systems (Andrusak 2010) to key tributaries within the Kootenay Lake would also be supportive in achieving two important program objectives (see Objectives and Rationale below); 1) obtain a better understanding of the health and status of bull trout populations in Kootenay Lake, and 2) assist with upper trophic level monitoring associated with the highly successful nutrient restoration program on Kootenay Lake (Schindler et al. 2011).

This report summarizes the first comprehensive assessment of adfluvial bull trout spawning in 19 tributaries to Kootenay Lake completed in the fall of 2011. Approximately 260 km of spawning habitat, not including the Duncan River tributaries, was surveyed in 19 select tributaries below Duncan Dam as part of the FWCP initiative, in partnership with the Ministry of Forests, Lands and Natural Resource Operations (MFLNRO). In addition, a primary tributary, the Westfall River, (~ 21 km) of the Duncan

Redfish Consulting Ltd. Page 1

Reservoir was also included in the 2011 results, assessed as part of a bull trout spawning assessment project of BC Hydro’s Water Use Plan (DDMMON#10) initiative. This report identifies the total accessible habitat from reconnaissance surveys and summarizes completed redd survey assessments of each of the select tributaries and by geographical groups (i.e. Group 1-Duncan/Lardeau tributaries; Group 2-Central Kootenay Lake tributaries; Group 3-South Arm Kootenay Lake tributaries). The report also summarizes the continuation of results (2006-2011) collected from the electronic resistivity counter deployed on the upper Kaslo River in the fall of 2011.

Rationale and Objectives The Fish and Wildlife Compensation Program – Columbia Basin (FWCP) annually funds a large scale nutrient restoration project and kokanee spawning channel on Kootenay Lake. Since bull trout are likely the most abundant top-piscivore in the lake, and a primary rationale for the compensation projects is restoration of bull trout and Gerrard rainbow trout populations, the FWCP has initiated a plan to provide bull trout monitoring for Kootenay Lake. This began in 2009 with a comprehensive review of existing information on adfluvial bull trout distribution in the lake’s tributaries, and a consideration of several population monitoring options (Hagen and Decker 2009). The priority recommendation from that report was to conduct redd counts in a selected suite of tributaries with documented evidence of adfluvial bull trout presence. Study objectives include;  Summarize results of redd surveys, reconnaissance surveys, barrier locations and accessible habitat on 19 select tributaries to Kootenay Lake, including adjoining tributaries to these systems.  Summarize data by geographical group (i.e. Group 1-Duncan/Lardeau tributaries; Group 2- Central Kootenay Lake tributaries; Group 3-South Arm Kootenay Lake tributaries)  Summarize complete redd surveys on 1 select tributaries (Westfall River) to Duncan Reservoir under a separate BC Hydro WUP study (DDMMON#10)  Summarize electronic resistivity counter results from 2011 on the Kaslo River out-migrating adfluvial bull trout  Discuss the efficacy of methodologies to establish an index of abundance for bull trout  Discuss the estimate of bull trout abundance from the resistivity counters  Provide estimates of bull trout escapement to the watersheds surveyed  Discuss uncertainty and variability inherent in redd surveys

Redfish Consulting Ltd. Page 2

Background

Bull Trout Life History Overview

Bull trout, native to western North America, are widely distributed throughout much of BC (McPhail 2007). They generally display three common life history patterns which consist of fluvial, adfluvial and resident populations distributed throughout their entire geographic range. Stream resident populations are typically separated from migratory populations (fluvial or adfluvial) by an obstacle or barrier to migration, either physical (e.g., waterfalls, dams; Latham 2002), physiological (e.g., unfavorably high temperatures; Rieman and McIntyre 1993; Dunham et al. 2003), or biological (e.g., presence of non-native competitor species; Paul and Post 2001). Bull trout spawn in the fall and depending on life history, reach sexual maturity between 5-6 years of age. This char species is considered slow growing and long lived, often exceeding 10 years of age. As well, depending on the life history form, size of older fish usually exceed 400 mm, with adfluvial forms attaining >600 mm at maturity.

Bull trout are exclusively adapted to cold water environments, a requirement in much of their early life history stages. Egg incubation and development are specifically related to water temperature, with optimal development and survival for bull trout occurring at 2- 4C (McPhail and Murray 1979). As well, water temperature is highly important in the growth and survival of bull trout older juveniles and adults (Selong et al. 2001; Dunham et al. 2003). Rearing juveniles also have specific habitat requirements, typically utilizing shallow areas with low current velocities along channel margins with un-embedded substrate (McPhail and Baxter 1996). Density dependent survival during the earliest juvenile stages (egg to age 1) has been found to regulate recruitment into the adult population for most bull trout populations (Johnston et al. 2007). In general, stream resident populations spend their entire life cycle within individual streams or stream reaches, fluvial and adfluvial bull trout rear in natal tributaries for 1-4 years before undergoing migrations downstream to larger rivers and lakes, respectively, with migration at age-2+ being the most common (Fraley and Sheppard 1989; Downs et al. 2006).

Fluvial and adfluvial populations usually spawn between mid-August and mid-October within the southern portion of BC, with many northern populations spawning earlier (McPhail and Murray 1979; Andrusak et al. 2011). Spawning redds range in size from 0.5-3.0 m2 (McPhail and Murray 1979), depending on the size of the female and the nature of the substrate being utilized. The characteristic form of bright, clean appearance of redds, and the low water conditions generally present during the early fall, allow for counts of redds to be utilized as an index of population abundance. However, a single female may construct more than one redd (Leggett 1980), and the

Redfish Consulting Ltd. Page 3

average number of redds per spawner needs to be assessed if redd counts are to be used to estimate the size of the spawning population (Dunham et al. 2001; Al- Chokhachy et al. 2009). Electronic resistivity counters have been reliably utilized in defining the number of adults to redds and the data also reduces some of the uncertainty and variability associated with redd counts (Andrusak 2010).

Bull Trout and Kootenay Lake Overview

Kootenay Lake has endured significant ecological changes during the last half century. Hydroelectric developments upstream of the lake during the 1960s and 1970s have been responsible for most of the alterations, although unregulated discharge of phosphorus and other mining developments and discharge into the headwaters also had a significant influence on lake productivity during the 1960s and 1970s (Northcote 1973; Daley et al. 1981; Hirst 1991; Ashley et al. 1997; Moody et al. 2007). Collectively these impacts caused major changes to the primary sport fish populations that have been well documented in a series of publications (Northcote 1973; Daley et al. 1981; Ashley et al. 1997; Moody et al. 2007; Schindler et al. 2011). Until 1992, when lake fertilization commenced Kootenay Lake was in a state of trophic depression (Ney 1996) as a result of upstream reservoirs that retained nutrients which adversely impacted lake productivity that in turn impacted the many of its fish populations (Ashley et al. 1997; Schindler et al. 2011).

Adfluvial populations of bull trout, primarily the adult and sub-adult life stage rely upon the large lacustrine habitat that Kootenay Lake supports. As with many piscivorous species of fish in Kootenay Lake, bull trout are highly dependent on kokanee (Oncorhynchus nerka) as their primary food source. The ecological footprint impacts from hydro-electric impoundment (Moody et al. 2007;) have had profound negative consequences to kokanee stocks on Kootenay Lake over the past few decades (Ashley et al. 1997;Schindler et al. 2011) resulting in a cascading effect on the piscivorous populations. With the recovery of the lakes' kokanee stocks, as a result of the large scale nutrient restoration program (Schindler et al. 2011), bull trout populations have likely benefited from increased in-lake survival and growth conditions, similar to that reported on the Arrow Lakes Reservoir (ALR; Arndt 2004a,b).

While Kootenay Lakes’ bull trout populations have likely recovered to some extent, they are still highly vulnerable to over-fishing due to the intense recreational fishery. The lack of more comprehensive information is problematic since bull trout populations appear to be highly susceptible to over-harvest, with other studied populations having exploitation rates exceeding sustainable levels (Post et al. 2003; Johnston et al. 2007; Andrusak and Thorley 2011). Thus, potential overfishing of bull trout and their sensitivity to habitat perturbations in their natal spawning and rearing tributaries

Redfish Consulting Ltd. Page 4

(Dunham and Rieman 1999; Dunham and Rieman 2003) led the provincial government to list bull trout as a species of concern throughout BC.

Obtaining accurate estimates of bull trout escapement are imperative in monitoring population trends but often difficult to obtain due to the large spatial scale in their distribution (Rieman and Myers 1997; Dunham et al. 2001; Al‐Chokhachy et al. 2005; Muhlfeld et al. 2006; Andrusak et al. 2011). While reliable annual escapement estimates provide the key data required for managing and sustaining bull trout, it is also known that utilization of informative tools such as a power analysis often suggest a minimum of 15 years of counts are required in order to detect a 50% change in the population based on the known variability (Maxell 1999; Al-Chokhachy et al. 2009; Andrusak et al. 2011). With the exception of the Kaslo River watershed and Crawford Creek (Andrusak 2010) and some earlier work on the Duncan Dam (O’Brien 1999, Olmsted et al. 2001), little information exists on the status of bull trout spawner numbers in most of Kootenay Lakes’ tributaries. Fortunately the Kaslo and Crawford data sets provide establishment of long-term indices for two bull trout streams on Kootenay Lake, with a continuation of data since 2006.

Kootenay Lake bull trout spawn in the headwaters of numerous tributaries that are often inaccessible. Frequently redd surveys are confounded by variable life history characteristics, behavior, and spatial and temporal distributions that collectively make spawning bull trout difficult to assess (Rieman and McIntyre 1996; Rieman and Myers 1997; Dunham et al. 2001). Uncertainty also exists in expanding redd counts to estimate population size due to the unknown relationship between the number of redds created per fish which has been demonstrated to change through the wide geographic distribution bull trout inhabit, and in the same system over different years. For example, Al-Chokhachy et al. (2005) suggested an average of 2.68 bull trout/redd but indicated ranges between 1.2 to 4.3 bull trout per redd depending on the various life history forms being monitored. It is also suggested in the literature that there is strong correlation on a logarithmic scale between escapement estimates and redd counts, but observer errors and the spatial and temporal variability in bull trout life history can invite considerable uncertainty (Muhlfeld et al. 2006).

Despite the above mentioned uncertainties, complete census redd surveys provide a relatively inexpensive method of monitoring trends in populations over time (Dunham et al. 2001; Andrusak et al. 2011). Accordingly, the literature does have a number of studies that have evaluated the validity of this method for detecting trends in population size (Rieman and Myers 1997; Dunham et al. 2001; Al‐Chokhachy et al. 2005; Muhlfeld et al. 2006).

Redfish Consulting Ltd. Page 5

The primary objective of this monitoring program was to assess bull trout spawning abundance and distribution in select tributaries to Kootenay Lake to better understand the total population size, as detailed in Hagen and Decker (2009). However, due to the difficulty in access, extent of their spawning distribution and size of the basin, only a select number of streams could be assessed from the total available. Similar concerns of spatial scale for monitoring bull trout population trends on the Williston Reservoir were also discussed by Andrusak et al. (2011). While the plan by Hagen and Decker (2009) recommended a number of tributaries to the lake, there are a number of other potentially important bull trout systems, especially in the Duncan‐Lardeau group, which could also contribute substantively to Kootenay Lake. Site Location

Kootenay Lake

Kootenay Lake, located in the upper Columbia River drainage of Southeast British Columbia, lies between the Selkirk and Purcell Mountain ranges (Figure 1). The main lake is 107 km long, approximately 4 km wide with a mean depth of 94 m and a maximum of 154 m (Daley et al. 1981). The lake is fed by two major river systems: the Lardeau/Duncan system at the north end (North Arm) and the Kootenay River that flows into the south end (South Arm). The outlet of the main lake, at Balfour, British Columbia, forms the upper end of the West Arm. At this outlet, a sill lies at a depth of approximately 8 m producing a distinct boundary between the main lake and the West Arm. The West Arm is about 40 km long with a mean depth of only 13 m. It is physically and limnologically different from the main lake, comprised of a series of shallow basins interconnected by narrow riverine sections. The West Arm of Kootenay Lake flows in a westerly direction becoming the lower Kootenay River, which flows into the Columbia River at Castlegar, BC.

The North Arm of the lake receives 21% of the entire inflow to the lake via Lardeau/Duncan drainage (Binsted and Ashley 2006). Virtually all tributaries of the Lardeau River likely support adfluvial bull trout spawning and rearing given their cold glacial inputs (H. Andrusak pers. comm.). Not including the Lardeau tributaries, the notable streams flowing into the east side and west side of the North Arm and the lower Duncan River include; Hamill, Creek, Cooper Creek, Schroeder Creek, Campbell, Powder Creek, Woodbury Creek, Coffee Creek and the Kaslo River watershed. Tributaries of the Lardeau/Duncan geographical constitute Group 1 north and south (Figure 2; Figure 3). Streams tributary to the north arm and central Kootenay Lake constitute Group 2 (Figure 4) candidate streams for this study.

Redfish Consulting Ltd. Page 6

The South Arm of the lake receives 56% of the entire inflow to the lake via the Kootenay River drainage and represents about two thirds of the entire lake surface and volume (Daley et al. 1981, Binsted and Ashley 2006). The Kootenay River drainage originates on the western slopes of the Rocky Mountains in eastern BC and flows southwest to Canal Flats, BC where it enters the Rocky Mountain trench and flows south into Montana. Downstream of the Libby Dam in Montana there is a natural waterfall (Kootenai Falls) that represents a barrier to all upstream fish movement. Below the falls the river flows west through Northern Idaho to Bonners Ferry where it shortly thereafter swings north to flow into the South Arm of the lake near Creston, BC. The primary streams flowing into the east side of the South Arm include the Goat River, Boulder Creek, Akokli Creek, Sanca Creek, Lockhart Creek, Grey Creek, and Crawford Creek. A major bull trout system on the East side of the South Arm is Crawford Creek where bull trout spawning has been well documented by Andrusak (2011). Boundary, Summit, Midgely, Next, Cultus, and Midge creeks flow into the west side of the Kootenay River and Kootenay Lake. Midge, Cultus, and Boundary creeks within this group are the largest systems that support adfluvial bull trout. Tributaries of the South Arm of Kootenay Lake constitute geographical Group 3 (Figure 5). Midge and Summit creeks are by far the largest streams in this group.

Duncan River tributaries- above Duncan Dam

Westfall River

The Westfall River, a tributary to the upper Duncan River and the Duncan Reservoir, was also included in this study despite being conducted under a separate BC Hydro WUP study (DDMMON#10)

The Westfall River is a 5th order watershed that is located approximately 25 km upstream of the north end of Duncan Reservoir (Figure 1). This river’s watershed covers a gross drainage of ~230 km2. The river flows easterly over ~ 29 km before draining into the upper Duncan River (Figure 2; MOE data on file). The lower 2 km section is incised in a steep bedrock canyon up to 200 m deep with a gradient of up to 10%. Upstream of the canyon area the watershed is mostly high-elevation, with moderate gradients ranging from 3-10%. The higher elevation areas are dominated by glacial inputs with a base elevation of 1200 m near the creeks origin, and peaks towering up to 2700 m.

Small but notable tributaries to the Westfall River include; Marsh Adams Creek and Silvertip Creek (Appendix 1; MOE data on file). Marsh Adams Creek is the largest tributary which extends ~12 km north beyond the confluence with the Westfall River.

Redfish Consulting Ltd. Page 7

Figure 1. Map of Kootenay Lake and select tributaries for conducting redd surveys in fall of 2011.

Redfish Consulting Ltd. Page 8

Group 1 (Duncan/Lardeau tributaries)

Group 1 includes four drainages emptying into the Duncan and Lardeau rivers below Duncan Dam (Table 1).

Table 1. Stream length and location for Group 1 (Duncan/Lardeau tributaries) systems or basins, and the Westfall River.

Stream Length Watershed Name Watershed Code Waterbody ID UTM Location (km) Area (km2) Hamill Creek 340-218400-02900 00000DUNC 11 503535 5561424 37 300 Cooper Creek 340-218400-02000 00000DUNC 11 503265 5560637 17 257 Poplar Creek 340-218400-07200-39200 00000DUNC 11 491324 5584889 27 156 Healy Creek 340-218400-07200-56600 00000DUNC 11 484255 5592405 23 152 Westfall River 340-218400-67700 00000DUNC 11 486084 5625459 27 230

Healy Creek

Healy Creek is a 4th order watershed that is located approximately 45 km north of the Meadow Creek town-site (Figure 1). The watershed covers a gross drainage of ~152 km2. The stream flows west over ~23 km in length before draining into the Lardeau River (Figure 2; MOE data on file). Upstream at the 3 km section, Healy Creek is incised in a steep bedrock canyon up to 200 m deep with a gradient of up to 10%. The remainder of the watershed is very rugged, with gradients ranging from 3-15%.

Poplar Creek

Poplar Creek, considered a 4th order tributary of the Lardeau River, is a large system that flows east from the Goat Range Provincial Park and the Selkirk Mountain range, covering 156 km2 (Figure 1; MOE data on file). The stream is ~27 km in length and drains into the Lardeau River near the historic community of Poplar Creek BC (Figure 3). Much of the habitat consists of high gradient, large boulder-cobble substrates and a high procession of step pool morphology. Similar to Cooper and Hamill creeks, the lower 5 km section is incised in a steep bedrock canyon up to 300 m deep with a gradient of up to 15%. Base elevation at the start of the surveys is ~900 m, with gradients ranging from 2-23%.

Hamill Creek

Hamill Creek is considered a 5th order tributary of the Duncan River (Figure 1). The watershed covers ~300 km2, draining from the Purcell Mountains at the north end of Kootenay Lake (Figure 3; MOE data on file). The stream flows west for ~37 km in length before joining with the Duncan River. The lower 5 km section is incised in a steep

Redfish Consulting Ltd. Page 9

bedrock canyon up to 300 m deep with a gradient of up to 17%. The remainder of the watershed is mostly high-elevation and very rugged, with gradients ranging from 3-10%. The higher elevation areas are dominated by glacial inputs with a base elevation of 1570 m near the creeks origin, with peaks towering up to 3200 m. Clint Creek is the primary tributary to Hamill Creek (Appendix 1). Much of the creeks entire length is adjacent to the historic Earl Grey Pass trail, most of which is included in the Purcell Wilderness Conservancy.

Hamill Creek sustained an extensive wild fire in the fall of 2007, which covered > 1500 ha within the watershed. Following the event in the drainage, significant peak flow increases and debris transport were observed (Jordan 2008).

Cooper Creek

Cooper Creek, considered a 5th order tributary of the Duncan River, is a large system that flows east from the Goat Range Provincial Park and the Selkirk Mountain range, covering 257 km2 (Figure 1). The stream is ~17 km in length and drains into the Duncan River near the community of Cooper Creek BC (Figure 3; MOE data on file). Similar to Hamill Creek, the lower 2-7 km section is incised in a steep bedrock canyon up to 1000 m deep with a gradient of up to 23%. Base elevation at the start of the surveys is ~1200 m, with gradients ranging from 2-23%. Cooper Creek has limited road access via a series of logging roads, using a Forest Service (FSR) road on the south side of Meadow Mountain. It is known that bull trout spawn in McKian Creek and South Cooper Creek (Chirico 1993).

South Cooper is a primary tributary which extends ~12 km south beyond the confluence with Cooper Creek (Appendix 1; MOE data on file). The higher elevation areas are dominated by high gradients, large boulder-cobble substrates and a high procession of step pool morphology. Moreover, the higher elevation areas are dominated by glacial inputs with a base elevation of 1100 m near the start of the survey and gradients ranging from 3-30%.

McKian Creek is the largest tributary which extends ~16 km north beyond the confluence with Cooper Creek (Appendix 1; MOE data on file). Similar to South Cooper, habitat consists of high gradient, large boulder-cobble substrates and a high procession of step pool morphology. The stream is dominated by glacial inputs with a base elevation of 1300 m at the start of the survey and lower gradients ranging from 4-10%.

Redfish Consulting Ltd. Page 10

Figure 2. Lardeau/Duncan river tributaries comprising Group 1 north and the Westfall River. Creeks included in the survey and their total lengths are indicated in bold.

Redfish Consulting Ltd. Page 11

Figure 3. Lardeau/Duncan river tributaries comprising Group 1 south. Creeks included in the survey and their total lengths are indicated in bold.

Redfish Consulting Ltd. Page 12 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Group 2 (Central Kootenay Lake tributaries)

Group 2 includes four systems emptying into the north arm and central portions of Kootenay Lake (Table 2).

Table 2. Stream length and location for Group 2 (Central Kootenay Lake tributaries) systems or basins.

Stream Length Watershed Name Watershed Code Waterbody ID UTM Location (km) Area (km2) Kaslo River 340-215300 00000KOTL 11 507136 5528029 33 435 Woodbury Creek 340-213900 00000KOTL 11 506944 5513517 21 131 Coffee Creek 340-213100 00000KOTL 11 506890 5505032 21 95 Crawford Creek 340-243500 00000KOTL 11 513507 5501475 24 187

Kaslo River

The Kaslo River is considered a 4th order tributary of Kootenay Lake (Figure 1). The watershed covers a gross drainage area of 435 km2, and is one of the larger tributaries that flow into the north arm of the lake near Kaslo BC (Figure 4; MOE data on file). With its origin at Fish Lake, the river is paralleled by Highway 31A for much of the 33 km length before converging with Kootenay Lake. The upper Kaslo River is a highly complex mixture of heterogeneous habitats characterized by cobble-gravel substrates, large woody debris (LWD) accumulations and moderate gradients varying from 1 to 10%. Small notable tributaries to the upper Kaslo River include; Twelve Mile Creek and Rossiter Creek. Keen Creek is the primary tributary to the Kaslo River, located ~8 km upstream of the lake (Appendix 1; MOE data on file). From the confluence Keen Creek extends ~29 km to its headwater origin in Kokanee Glacier Provincial Park and covers a gross drainage area of 92.2 km2. Habitat on Keen Creek is more homogeneous, consisting of mostly high gradient, large boulder-cobble substrates and a high procession of step pool morphology (Andrusak 2010).

Woodbury Creek

Woodbury Creek is considered a 4th order tributary of Kootenay Lake (Figure 1). The watershed covers a gross drainage of 131 km2 and drains a steep sided glacial valley that extends for approximately 21 km down to the lake (Figure 4; MOE data on file). Its headwaters originate in Kokanee Glacier Provincial Park and the Selkirk Mountain range. Woodbury Creek can be characterized as a cascade-pool morphology system with gradients ranging from 4% to 15%. A 1.5m waterfall is located at a small hydro structure 800m upstream of the Highway 31 bridge. This small waterfall is a seasonal migration barrier. Small notable tributaries to Woodbury Creek include; Nelles Creek, Pontiac Creek, Silver Creek and Spray Creek. However, these tributaries do not support

Redfish Consulting Ltd. Page 13 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011 spawning adfluvial bull trout due to accessibility and the steep gradients which pose natural barriers to migration.

Coffee Creek

Coffee Creek is considered a 3rd order tributary of Kootenay Lake (Figure 1). The watershed covers gross drainage of 95 km2 and drains a steep sided glacial valley that extends for approximately 21 km down to the lake (Figure 4; MOE data on file). Similar to Woodbury Creek, its headwaters originate in Kokanee Glacier Provincial Park and the Selkirk Mountain range. Coffee Creek can be characterized as a cascade-pool morphology system with gradients ranging from 4% to 15%. Coffee Creek channel morphology has significantly altered following a large a flood/slide event in November of 1999.

Crawford Creek

Crawford Creek, considered a 4th order tributary of Kootenay Lake, flows into Crawford Bay on the east side of the lake (Figure 1 MOE data on file). The watershed covers a gross drainage area of 187 km2. Crawford Creek extends ~24 km upstream of its confluence with Kootenay Lake draining from the west slopes of the Purcell mountain range (Figure 4). A steep sided canyon area with some large pools formed by bedrock exists approximately 10 km upstream of the lake but this site is not a barrier to large migratory fish. Most of the stream is dominated by riffle habitat dominated by boulders and cobble with little LWD accumulation. However, the upper 4 km of Crawford Creek does support some lower gradient highly complex heterogeneous habitat with large LWD accumulations. This habitat then gives way to 1.7 km of high gradient step-pool morphology where the creek origins begin at the base of the Purcell Mountains. A 1:20,000 fish and fish habitat survey was conducted in 1997 which identified large adfluvial bull trout a distance upstream (Timberland Consultants 1997).

Canyon Creek, Hooker Creek and Houghton Creek are three major tributaries to Crawford Creek that are known to support spawning adfluvial bull trout (Appendix 1; MOE data on file). Canyon Creek, Hooker Creek extend ~6-8 km upstream of its confluence with Crawford Creek, while Houghton Creek extends ~10 km upstream.

Redfish Consulting Ltd. Page 14 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Figure 4. Central Kootenay Lake tributaries comprising Group 2. Creeks included in the survey and their total lengths are indicated in bold.

Redfish Consulting Ltd. Page 15 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Group 3 (South Arm Kootenay Lake tributaries)

Group 3 includes the tributaries found in 3 drainages entering the south arm of the lake and Kootenay River (Table 3)

Table 3. Stream length and location for Group 3 (South Arm Kootenay Lake tributaries) systems or basins.

Stream Length Watershed Name Watershed Code Waterbody ID UTM Location (km) Area (km2) Midge Creek 340-311100 00000KOTL 11 514321 5469359 27 263 Cultus Creek 340-326800 00000KOTL 11 515507 5463722 26 112 Summit Creek 340-418600 00000KOTL 11 528063 5442997 38 300

Midge Creek

Midge Creek, a 6th order tributary of Kootenay Lake, is located approximately 35 kilometres northwest of Creston, B.C (Figure 1). The watershed covers a gross drainage area of 263 km2. Midge Creek is a large tributary that flows ~ 27 km southeast from its headwaters in West Arm Provincial Park, draining the eastern slopes of the Nelson Range of the Selkirk Mountains into the south arm of Kootenay Lake (Figure 5; MOE data on file). Midge Creek can be characterized as a cascade-pool morphology system with gradients ranging between 4% to 14%.

Kutetl Creek, Conway Creek, Württemberg Creek and Seeman Creek are the major tributaries to Midge Creek that are known to support spawning adfluvial bull trout (Appendix 1; MOE data on file). Seeman Creek is the largest of these tributaries, with a gross drainage of 69 km2, that flow into Midge Creek. Seeman Creek is ~15 km in length with varying gradients from 3-10%. Kutetl Creek is a moderate sized tributary that flows east ~11 km before draining into upper Midge Creek. Surprisingly, Kutetl Creek has low to moderate gradients ranging from 3-7%. Württemberg Creek, Conway and Hughes are relatively small systems with < 10 km in total length (Appendix 1; MOE data on file).

Cultus Creek

Cultus Creek, a 4th order tributary of Kootenay Lake, is located approximately 30 kilometres northwest of Creston, B.C (Figure 1). The watershed covers a gross drainage area of 112 km2. Cultus Creek is moderate to small tributary that flows ~26 km southeast from its headwaters, draining the eastern slopes of the Nelson Range of the Selkirk Mountains into the south arm of Kootenay Lake (Figure 5; MOE data on file).

Redfish Consulting Ltd. Page 16 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Cultus Creek can be characterized as a cascade-pool morphology system with gradients ranging between 3% and 10%.

Laib Creek and Kloosh Creek are the two major tributaries to Cultus Creek. However, these tributaries are known not to support spawning adfluvial bull trout (G. Nellestijn pers. comm.).

Summit Creek

Summit Creek, a 4th order tributary of Kootenay Lake, is located approximately 10 kilometres west of Creston (Figure 1). The watershed covers a gross drainage area of 300 km2. Its headwaters originate in Stagleap Provincial Park and parallel Highway 3 for much of the ~38 km length before converging with the Kootenay River in the Creston Wildlife Management Area (CWMA; Figure 5). Summit Creek can be characterized as a primarily large boulder-cobble riffle system with moderate gradients ranging between 2% to 8%.

Small but notable tributaries to the Summit Creek include; Char Creek, Bayonne Creek, Blazed Creek and Maryland Creek (Appendix 1; MOE data on file). Blazed Creek is the largest of these tributaries, with a gross drainage of 54 km2. Blazed Creek is ~10 km in length with varying gradients from 3-10%. Char Creek, Bayonne Creek, and Maryland Creek are relatively small tributaries with < 10 km in total length (Appendix 1; MOE data on file).

Redfish Consulting Ltd. Page 17 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Figure 5. South Arm Kootenay Lake tributaries comprising Group 3. Creeks included in the survey and their total lengths are indicated in bold.

Redfish Consulting Ltd. Page 18 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Methods

Stream Selection

This survey included all tributaries and sub-tributaries below Duncan Dam with previous information documenting a historic presence of adfluvial bull trout (see Hagen and Decker 2009 for details). As previously mentioned, there may be other tributaries that support adfluvial bull trout, particularly those influenced by cold glacial flows in the Lardeau River, Kootenay River, Trout Lake and Duncan River drainages.

Data from the Westfall River above Duncan Dam is reported herein as well, inclusion of this data has been collected under a separate BC Hydro study (DDMMON#10)

Reconnaissance surveys

Initial reconnaissance was undertaken on most systems to identify locations of barriers to fish passage and observe the spatial extent of bull trout spawning within the system. Systems with barriers previously documented or that were established were re-visited to re-confirm the barriers existence. The reconnaissance surveys provided certainty of the extent of spawning and accessible length of stream to be surveyed when passage barriers were well known. As well, these surveys also provided important information on spawn timing and determination on when to conduct primary redd surveys for each system, thus reducing temporal variability often associated with redd surveys (Rieman and McIntyre 1996; Dunham et al. 2001).

Available information prior to conducting the reconnaissance surveys, from individual systems, was also obtained from the Ministry of Environment using the Fisheries Inventory Data Queries (FIDQ) website (http://a100.gov.bc.ca/pub/fidq/main.do).

Redd identification

Redds were identified as approximately dish-shaped excavations in the bed material, often of brighter appearance than surrounding substrates, accompanied by a deposit beginning in the excavated pit and spilling out of it in a downstream direction. A bull trout redd can be defined as the entire area of gravel excavated by the female, the size of which can range from 0.5 m2 to 3.0 m2 (McPhail and Murray 1979) depending on the size of the female and the nature of the substrate being utilized. It also appears that a single female can spawn in more than one redd if gravel accumulations at the first location are of limited size (Leggett 1980). Disturbances in the bed material caused by fish were discriminated from natural scour by: i) the presence of tail stroke marks; ii) an over-steepened (as opposed to smooth) pit wall often accompanied by perched substrate that could be easily dislodged down into the pit, and often demarcated by

Redfish Consulting Ltd. Page 19 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011 sand deposited in the velocity break caused by the front wall; iii) excavation marks alongside the front portion of the deposit demarcating the pit associated with earlier egg laying events (bull trout will deposit eggs in several nests as the redd is built in an upstream direction); and iv) a highly characteristic overall shape that included a ‘backstop’ of gravel deposited onto the unexcavated substrates, a deposit made up of gravels continuous with this backstop and continuing upstream into the pit, and a pit typically broader than the deposit and of a circular shape resulting from the sweeping of gravels from all sides to cover the eggs (in a portion of redds gravels are swept into the pit from only one side, often a shallow gravel bar on the shore side).

A second important determination was whether fish had actually spawned at a location where an excavation had been started. ‘Test digs’ were considered to be pits, often small, accompanied by substrate mounded up on the unexcavated bed material downstream but with no substrate swept into the pit itself, which would denote at least one egg deposition event. In the case of a ‘test dig’ determination the mound of gravels would typically be short and narrow around the downstream side of a relatively small pit.

In areas of limited gravel or high redd abundance, or where spawning site selection is highly specific, superimposition of redds upon one another can occur (McPhail and Baxter 1999). For this study, the redd count was based on an evaluation, with the most recent complete redd(s) counted and the disturbed remains of prior redds being included and considered a redd when the aforementioned criteria was met (see above). A greatly extended deposit length (subjectively evaluated to be at least twice the length of a ‘typical’ deposit length) was grounds to consider whether a second female had made use of the pit created by a first female to construct a separate redd.

Redd count standardization and inter-observer variation

Understanding the variability of counts when conducting redd surveys is critical in obtaining reliable estimates for monitoring population trends. There are many potential sources of uncertainty when undertaking redd survey assessments (Dunham et al. 2001; Muhlfeld et al. 2006).

A study protocol was designed to assess and minimize the effects of observer crew differences (inter-observer variability) on precision of redd counts. Crew members with less than one year of redd counting were partnered with a more experienced observer during a one day standardization survey on the Kaslo River. The two-person crews surveyed two pre-determined reaches (< 1 km) of the upper Kaslo River at the beginning of the study period, providing a comparison of replicate surveys (by different crews) in a portion of the study reaches, similar to that detailed by (Dunham et al. 2001;

Redfish Consulting Ltd. Page 20 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Muhlfeld 2006). None of these observers had prior knowledge of redd locations within study reaches. Prior to the redd counts all observers examined an existing bull trout redd to be standardized on redd characteristics and attributes (i.e. size and shape).

Prior to conducting the standardization of all other crews, redd surveys were conducted on two sections of the upper Kaslo River by the most experienced observer. Total redd count (baseline number) obtained by the experienced observer in the two sections was compared to numbers (observed numbers) obtained by all other crews. Scoring of redd counts was occasionally complicated by close proximity or superimposition of redds. Due to the occasional difficulty of identifying individual redds, data was summarized only by each test reach.

To compare the precision of redd counts, absolute residual deviations from mean observed redd counts in each section were generated. Relative bias in relation to baseline redd counts was calculated as the difference between observed and baseline redd counts converted to a percentage of the baseline redd count. Observer error was defined in two ways: (1) apparent error or the absolute value of relative bias (percent) in redd counts, and (2) total error or the sum of omissions and false identifications, standardized as a percentage of the actual redd count.

Redd surveys

Experienced two man crews were used for all redd surveys to reduce observer error or variability in counts due to multiple surveyors (Muhlfeld et al. 2006). Often experienced crew members with counting experience were paired with less experienced crew members with less than 1 year experience. Candidate streams were surveyed with crew members walking in a downstream direction on opposing banks when possible. Surveys started from the known barrier location to fish passage based on assessments from the reconnaissance surveys conducted during the peak of spawning.

Complete redds were enumerated and UTMs recorded by geo-referenced time and waypoints using a handheld 62s Garmin GPS. Data for each redd was recorded on waterproof paper (see Appendix 2). GPS track logs were also initiated at the start of the survey and used as an overlay in GIS mapping for assessing spatial distribution of spawning within each system.

Incomplete redds were also identified where the observed redd was associated with a spawning adult or pair of spawning adults. Similar to complete redds, incomplete redds were enumerated and UTMs recorded by geo-referenced time and waypoints using a handheld 62s Garmin GPS. Incomplete redds were used as a valuable tool in assessing

Redfish Consulting Ltd. Page 21 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011 temporal variability in survey timing where no previous information existed (Rieman and McIntyre 1996; Dunham et al. 2001).

Electronic Resistivity Counter

A resistivity counter was installed on the Kaslo River, similar to previous years of work on this system (Andrusak 2010). The resistivity counter detects the passage of fish across an array of three electrodes (positive, ground and negative), placed submerged across the river bottom, or in a channel, in an insulated base (Aprahamian et al. 1996). The counter electronics continually monitor the bulk resistance of the water above the counting array and calibrates the hardware for changes in this resistance every 30 minutes. When a fish passes over the three electrodes, a change in resistance occurs, as a fish is more conductive than the water it displaces. This change of resistance is recorded and analyzed by the counter using a firmware algorithm to determine if it fits a typical fish pattern. If the counter assessed that a fish passed over the array the time, direction of travel and peak signal size (maximum change of resistance measurement) of the fish event, is recorded and stored for later downloading and analysis (see Aprahamian et al. 1996 for more details of counter design and operation). More detail on the counter methodology and count interpretation of down streaming bull trout spawners (kelts) can be found in Andrusak (2010).

The design of the flat pad sensor units in this project were similar to that used on McKinley Creek in the Cariboo Region, with an outer aluminum frame and an inner vinyl sleeve into, which were set 3 stainless steel electrodes (McCubbing et al. 1999; Galesloot and McCubbing 2003; Peard et al. 2005). The flat pads were held in place and weighted down by biodegradable burlap sand bags. The sand bags were placed on the periphery of each pad (6-8 per pad) to ensure they did not interfere with fish passage over the electrodes.

The sensor units installed (4 in total on upper Kaslo River) covered a width of 12 m, with channel 1 on the near bank and channel 4 on the far bank (Photo 1). The remaining margin areas, in shallow water (< 20 cm), were blocked off by natural rock weirs to discourage fish passage. At the Kaslo River site, the Logie 2100C counter was hard wired into the electrodes and positioned on the bank along with video validation equipment stored in a metal work box (Photo 2). Validation of counts is conducted by video (Photo 3) and assessment of all directional counts from graphics (Photo 4) and counts by Logie 2100C following the spawning period. The counter was installed on September 25 and operated continuously until it was removed on October 23.

Redfish Consulting Ltd. Page 22 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Results

Inter-observer variation

Redd counts were conducted within 2 sections of the upper Kaslo River, one section ~426 m in length and the other section ~ 1,100 m in length. A total of 4 two-man crews conducted the standardization sections of the Kaslo River on October 3.

Overall, variation in counts among observers was considered to be low and well within the range reported in the literature of 25%-254% and 78%-130% cited in Dunham et al. (2001) and Muhlfeld (2006), respectively. Initial counts were conducted by an experienced observer and formed the basis for comparison (i.e. baseline to observed). Counts in the upper section of the river ranged between 13 and 22, or 81%–138% of the baseline (actual) estimate of 16 redds. (Table 4; Figure 6). Similarly, counts in the lower section of the river ranged between 25 and 55, or 63%–125% of the baseline (actual) estimate of 44 redds (Table 4; Figure 6). Crews frequently committed both omissions and false identifications compared to the baseline count. False identifications were two times higher in the lower section compared to the upper section. Despite having a higher number of redd omissions in the lower section, redd densities were markedly similar over the lengths surveyed at 40 redds per km. Much of the variability associated within this section likely was a result of superimposition (Albert Chirico, MOE biologist, pers. comm.).

Table 4. Number of bull trout redds counted by four two-man crews in two sections of the Kaslo River. Sections were used to conduct an assessment of observer efficiency. Percent error is evaluated as the % difference between baseline counts conducted by an experienced crew and observed counts conducted by subsequent crews for the upper and lower sections of the Kaslo River.

Upper Section Lower Section Combined Sections Crew Redd % of Redd % of Redd % of baseline (2 man) Count baseline Count baseline Count total 1 17 106 54 123% 71 118% 2 21 131 49 111% 70 117% 3 22 138 55 125% 77 128% 4 13 81 25 57% 38 63% Baseline redd count (16) 100% (44) 100% (60) 100%

Redfish Consulting Ltd. Page 23 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Figure 6. Number of bull trout redds counted and counting errors (omissions or ‘‘omitted’’ and false identifications or added as “false’’) by four two person crews in two sections of the upper Kaslo River in fall of 2011. Dashed horizontal lines refer to best estimates (actual) of redd number of 44 and 16 for the lower and upper section, respectively.

Electronic Resistivity Counter

The electronic resistivity counter, installed on the upper Kaslo River, enumerated ~1,180 bull trout kelts descending from the upper portion of the river following spawning. Using the calibrated electronic counts and the redd numbers observed in 2011 on the upper Kaslo River, a bull trout per redd ratio of 2.7 was obtained. The derived expansion factor for 2011 was slightly higher than the average of 2.4 bull trout derived from 2006, 2008- 2009 and 2011. No data was collected from the counter in 2007 and data collected in 2010 was considered to be biased as a result of the poor visibility during the redd surveys.

Redfish Consulting Ltd. Page 24 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Redd Counts

Westfall River

A reconnaissance survey under a separate BC Hydro contract was conducted on the Westfall River on September 7 using access along the Westfall FSR to identify any accessibility issues or safety concerns. As well, on the morning of October 17th, prior to conducting the redd surveys, a helicopter was utilized to conduct the final assessment of barriers on the Westfall River and its tributaries.

Redd surveys commenced on October 17, utilizing a helicopter to deploy two experienced two man crew into upper portion of the Westfall River. Crew deployment was based on the aerial reconnaissance of the system. While much of the higher elevation (1200 m) is dominated by glacial inputs, most of this habitat was suitable for spawning, having low to moderate gradients of 3-10%. Nevertheless, the conditions during the survey on the Westfall River were ideal with excellent visibility over the entire 21 km surveyed (Table 5) No ground redd survey on 8 km of Marsh Adam Creek was conducted. However, an aerial redd survey was conducted of the entire Marsh Adam system during the reconnaissance. The two experienced two man crew surveyed the entire system in 2.5 days.

A complete barrier to adfluvial bull trout migrants was identified at ~21.2 km upstream on the Westfall River (Table 5; Photo 5). The barrier was a large bedrock waterfall reaching 5 m in height. However, the first redd was not observed until ~298 m downstream. As well, a complete barrier to adfluvial bull trout migrants was identified at ~8 km upstream on the Marsh Adam Creek, similar to that detailed in O'Brien (1999). The barrier was a series of bedrock/ice chutes and falls that reached 5 m in height. Silvertip Creek was not surveyed since it was considered not accessible for bull trout migrants during the over-flight. In total, the Westfall River drainage provides ~30 km of accessible habitat for spawning.

A total of 114 redds were enumerated in the Westfall River drainage (Table 6; Figure 7). Only 2 redds from the total were observed in the tributary Marsh Adam Creek at ~4.5 km from the aerial over-flight. Redd survey timing was considered optimal with only one spawning (male) bull trout observed over the 30 km surveyed.

Group 1 (Duncan/Lardeau tributaries)

Healy Creek

The Healy Creek system had been monitored for bull trout spawning for the past two years by the author prior to conducting these assessments with no spawning adfluvial

Redfish Consulting Ltd. Page 25 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011 bull trout observed over the entire 23 km available. This led to the conclusion that a barrier exists at ~2km upstream of the confluence with the Lardeau River and this has been confirmed during a kokanee aerial reconnaissance of the Lardeau River (E. Schindler MFLNRO Nelson BC pers. comm.).

A ground reconnaissance/redd survey was conducted on October 12 on Healy Creek. The crew walked upstream from the confluence with the Lardeau River. The lower section below the canyon provides ~500 m of meandering low gradient (< 3%) habitat that was utilized by spawning kokanee. Thereafter, the 1.5 km canyon section of Healy Creek begins and extends from 0.5 km to ~2 km. Due to the difficulty in access and safety concerns, no observation of the barrier could be obtained. Similar to Cooper Creek, much of the canyon section is incised in a steep bedrock canyon up to 200 m deep with a gradient of up to 20%. However, ground crews indicated the potential for a drop (falls) or velocity (chute) barrier within the 1.5 km canyon section was quite high.

No redds were observed in suitable habitat above the canyon section over a 5 km ground survey section, indicating adfluvial bull trout do not currently utilize or access this system above the canyon (Table 5; Figure 8). Chirico (1993) also confirmed absence of adfluvial bull trout when captured bull trout above the canyon section were all considered to be the resident life history form. As well, spawning habitat was rated low as a result of the high percentage of fines and sediments in the system (Chirico 1993).

Poplar Creek

A reconnaissance surveys was conducted on Poplar Creek in September 10, prior to conducting redd surveys to identify any accessibility issues or safety concerns. Once again a steep canyon section from 3.5 to 2.5 km on Polar Creek was not surveyed due to safety issues. The canyon was incised in steep bedrock 200 m deep with a gradient of up to 10%.

Redd surveys commenced on October 5 & 12 utilizing a ground crew from AMEC. However, due to the difficult conditions for surveying only the lower 0-2.5 km section of Poplar Creek was completed. A subsequent crew was deployed to complete the upper 3.4-4.4 km portion of the creek. As with many of the other systems, most of the habitat consists of high gradient, large boulder-cobble substrates and a high procession of step pool morphology. Difficulty in terrain severely limited the crews’ ability to assess this system (Louise Porto, AMEC, pers. comm.). With the exception of 1.0 km of canyon, a two man crew surveyed the entire system in 1.5 days.

A complete barrier to adfluvial bull trout migrants was identified at ~4.4 km upstream on Poplar Creek (Table 5; Photo 6). The barrier was a bedrock chute that exceeded 5

Redfish Consulting Ltd. Page 26 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011 meters in height. Nevertheless, Poplar Creek provided 4.5 km of accessible spawning habitat to adfluvial bull trout.

A total of 53 redds were enumerated in the Poplar Creek drainage (Table 6;Figure 9). Only 1 adult spawner was observed during the survey and redd survey timing was considered optimal.

Hamill Creek

An initial reconnaissance survey was conducted on Hamill Creek on September 28, one day prior to conducting redd surveys, to identify a possible barrier within the lower 2-4 km section of the creek. This lower section is incised in a steep bedrock canyon up to 1000 m deep with a gradient of up to 17% and was deemed to be unsafe for conducting redd surveys. Nevertheless, both ground and helicopter reconnaissance indicated that no barrier exists within the lower canyon portion of Hamill Creek.

Redd surveys commenced on September 29, utilizing a helicopter to deploy an experienced two man crew into Hamill Creek. The crew initiated the redd survey ~35.4 km upstream on Hamill Creek with a starting elevation of 1570 m (Table 5). With the exception of 2.5 km of canyon, the crew surveyed the entire system over the following 5 days.

Two complete barriers (< 10 m apart) to adfluvial bull trout migrants were identified at ~35.4 km upstream on Hamill Creek (Table 5). Both barriers were comprised of a series of bedrock chutes and a series of steps that reached > 5 meters in height (Photo 7). While the surveys commenced at ~35 km upstream, redds were not observed for 15 km below this barrier suggesting another complete or partial barrier above where redds were observed. Survey crews identified a large cascade section that was believed to be a velocity barrier under higher flows (Photo 8). Nonetheless, Hamill Creek during this survey provided a minimum of ~23.8 km of accessible spawning habitat to adfluvial bull trout.

A total of 216 redds were enumerated in the Hamill Creek drainage (Table 6; Figure 10). Another 7 redds were observed but incomplete and associated with either an adult or an adult pair within Hamill Creek. Only 3 redds from the total were observed in the lower end or alluvial area (~500 m) of Clint Creek. Redd survey timing was considered optimal with a total of 16 spawning bull trout observed over the 35 km surveyed.

Cooper Creek

Several reconnaissance surveys were conducted on Cooper Creek in September 10, prior to conducting redd surveys, to identify a possible barrier within the lower 2-7 km section

Redfish Consulting Ltd. Page 27 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011 of the creek. Similar to Hamill Creek, this lower section is incised in a steep bedrock canyon up to 500 m deep with a gradient of up to 23%. Local white-water kayak enthusiasts considered this high gradient section to be a class five rapid (Jimmy Robbins pers. comm.). As a result, no redd surveys were conducted within the 2-7 km canyon due to safety concerns. Both ground and helicopter, reconnaissance indicated that no barrier exists within the lower canyon portion of Cooper Creek.

Redd surveys commenced on October 13, utilizing a helicopter to deploy two experienced two man crews into the Cooper Creek system. One crew was deployed ~14 km upstream in McKian Creek based on accessibility and observation of possible barriers from reconnaissance using the aerial overview. Similarly, the other crew was deployed ~12 km upstream in South Cooper Creek based on accessibility and observation of barriers from the over-flight. During the downstream walks on McKian Creek, crews also conducted a survey of Spokane Creek. After completing the South Cooper Creek surveys, the two man crew was re-deployed onto the mainstem of Cooper Creek ~16 km from the confluence with the Duncan River. The entire watershed was completed in 3 days with the four member crew. Of importance, most of the 1st and 2nd order tributaries to these sub-basins are non-fish bearing and mainly ephemeral with gradients greater than 30% (Timberland Consultants 2001).

A complete barrier to adfluvial bull trout migrants was identified at ~13 km upstream on McKian Creek (Table 5; Photo 9). This barrier was a combination of large wood debris (LWD) and large boulder accumulations reaching 3 m in height. However, the first redd was not observed until ~1.5 km downstream. The survey crew identified two large logjam structures that were considered partial barriers to fish access at ~11.4 km, just above the first redd observation. A complete barrier to adfluvial bull trout migrants was identified at ~1.3 km upstream on South Cooper Creek (Table 5; Photo 10). This barrier comprised a large wood jam structure with a 3 m drop, with little or no plunge pool below. The first redd observation was ~50 m below this structure. Lastly, a complete 10 m barrier was re-confirmed (FIDQ data) on Cooper Creek ~16.1 km upstream from the confluence with the Duncan River (Table 5). Although no redds were observed until 340 m below this structure, indicating a partial barrier may exist downstream. The survey crew identified a 2 m partial barrier downstream which may cause access issues. Only three redds were observed above this partial barrier. Nonetheless, Cooper Creek drainage provided about 30 km of accessible spawning habitat to adfluvial bull trout.

A total of 134 redds were enumerated in the Cooper Creek drainage (Table 6;Figure 11). Of the total, McKian Creek and South Cooper Creek had 53 and 17 redds, respectively. The remaining 64 redds were observed within the main channel of Cooper Creek. Redd survey timing was considered optimal with no spawning bull trout observed over the 30 km surveyed.

Redfish Consulting Ltd. Page 28 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Table 5. Group 1 (Lardeau/Duncan tributaries) barrier location, characteristics and accessible habitat to obstruction. Italicized systems represent streams that had partial barriers located below the complete barrier

Accessible Name Complete Height Habitat Group Name (Tributary) (System) UTM or Partial Type (m) (km) Group 1 Healy Creek Healy Creek 11 484492 5593233 complete chute unk 1.1 Group 1 Poplar Creek Poplar Creek 11 488309 5582262 complete chute 5 m 4.4 Group 1 Hamill Creek Hamill Creek 11 529610 5567587 complete chute 5 m 35.4 Group 1 Hamill Creek Hamill Creek 11 511785 5565679 partial cascade 3 m 23.8 Group 1 Clint Creek Hamill Creek 11 506158 5562651 complete cascade 2 m 0.1 Group 1 Cooper Creek Cooper Creek 11 489507 5556503 complete falls 10 m 16.1 Group 1 Cooper Creek Cooper Creek 11 489802 5556423 partial logjam 1 m 16 Group 1 McKian Creek Cooper Creek 11 485687 5564001 complete chute 3 m 13 Group 1 McKian Creek Cooper Creek 11 486882 5564811 partial logjam 1 m 11.4 Group 1 Spokane Creek Cooper Creek 11 490135 5561787 complete chute 1 m 0.8 Group 1 South Cooper Creek Cooper Creek 11 492337 5556395 complete falls 3 m 1.3 Duncan Westfall River Westfall River 11 472181 5632938 complete falls 5m 21.2

Table 6. Group 1(Lardeau/Duncan tributaries) redd survey data

Complete Incomplete Fish Group Name (Tributary) Name (System) Date Surveyed redds redds Count Group 1 Clint Creek Hamill Creek October 4 3 0 0 Group 1 Hamill Creek Hamill Creek September 29 213 7 16 Group 1 Cooper Creek Cooper Creek October 13 64 0 0 Group 1 McKian Creek Cooper Creek October 13 53 0 0 Group 1 South Cooper Creek Cooper Creek October 13 17 0 0 Group 1 Poplar Creek Poplar Creek October 5 53 0 1 Duncan Westfall River Westfall River October 17 114 0 0 Group 1 Healy Creek Healy Creek October 12 0 0 0

Group 2 (Central Kootenay Lake tributaries)

Kaslo River

No reconnaissance surveys were conducted on the Kaslo River drainage in 2011 since the Kaslo River has been part of an ongoing bull trout monitoring index since 2006 (Andrusak 2010).

Redd surveys commenced on October 4, utilizing two experienced two man crews (MFLNRO staff) on the Kaslo River. The crew initiated the redd survey ~29 km upstream

Redfish Consulting Ltd. Page 29 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011 on the Kaslo River with a starting elevation of 994 m (Table 7). As well, a second experienced two man crew was deployed on Keen Creek on October 5, a large tributary to the Kaslo River. This crew initiated the redd survey ~6 km upstream of the confluence with the Kaslo River with a starting elevation of 956 m (Table 7). The two crews surveyed the entire system over the following 4 days.

A complete barrier to adfluvial bull trout migrants was identified at ~28.8 km upstream on the upper Kaslo River (Table 7; Photo 11). The barrier is a 3 m waterfall near the historic town-site of Retallic BC, adjacent to Highway 31A. As well, a complete barrier to adfluvial bull trout migrants was identified at ~6.4 km upstream on Keen Creek. Similar to the upper Kaslo river barrier, this barrier is a 4-5 m waterfall and is clearly marked on the FSR as the boundary for angling (Table 7; Photo 12). Small tributaries to the Kaslo River including; Rossiter Creek and Twelve Mile Creek each have established barriers ~400m upstream from the confluence. Both barriers are bedrock chutes that provide a series of steps and drops ~2-3 m high. Since the removal of the old Kaslo hydroelectric dam in the early-1970s (MOE on file), the Kaslo River drainage provides ~35 km of accessible spawning habitat to adfluvial bull trout.

A total of 512 complete redds were enumerated in the Kaslo River drainage (Table 8; Figure 12). Another 44 redds were observed but incomplete and associated with either an adult or an adult pair in the Kaslo River or Keen Creek. The upper Kaslo River had the highest redd count of 422, most of which were distributed in the upper 20 km of the system. Keen Creek had a total of 73 redds distributed in the upper 3 km of the watershed. Small tributaries including; Rossiter Creek and Twelve Mile Creek contributed 3 each to the total. Once again, redd survey timing was considered good although some spawning was observed during the time of the survey (a total of 72 spawning bull trout observed over the 35 km surveyed).

Woodbury Creek

Several reconnaissance surveys were conducted on Woodbury Creek in September 7, 12, 25, prior to conducting redd surveys, to identify a possible barrier within the lower section of the creek where the existing hydro-electric dam is located. As well, a reconnaissance of the canyon section at 1.0-3.7 km was conducted to ensure the safety of crews assessing the creek. Similar to other systems, the canyon section is incised in a steep bedrock canyon up to 100 m deep with gradients varying from 8-15%.

Redd surveys commenced on October 19, utilizing an experienced two man crew on Woodbury Creek. Although based on accessibility, only the lower portion of the creek, below the dam, was assessed during the actual redd survey. A partial barrier to adfluvial bull trout migrants was identified at the hydro-electric dam ~600 m upstream from

Redfish Consulting Ltd. Page 30 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Kootenay Lake. The partial barrier is a 2 m waterfall as a result of concrete sill used in dam construction (Table 7; Photo 13). Although in some years, based on flows, adfluvial bull trout have the potential to access ~10 km of suitable habitat for spawning above the obstruction (J. Bell pers. comm.).

A total of 31 redds were observed below the obstruction (Table 8; Figure 13). No redds or fish were observed above the dam section over 10 km of suitable spawning habitat. However, upstream of the dam resident and adfluvial population of bull trout have been confirmed (FISS data). Redd survey timing was considered good and only 4 spawning bull trout were observed.

Coffee Creek

A ground reconnaissance survey was conducted on Coffee Creek on September 12, prior to conducting redd surveys. Coffee Creek drainage is accessible utilizing a FSR that ends at the bridge where the barrier is located (FISS data).

Redd surveys commenced on October 5, utilizing an experienced two man crew on Coffee Creek. The crew initiated the redd survey ~6.7 km upstream from Kootenay Lake (Table 7). Despite the challenging survey as a result of major changes in channel morphology, the crews surveyed the entire system over 1 day.

A complete barrier to adfluvial bull trout migrants was identified at ~6.7 km upstream from the confluence with Kootenay Lake (Table 7; Photo 14). The barrier is a series of 4 m waterfalls at the end of the old Coffee Creek FSR. Coffee Creek provided 6.7 km of accessible spawning habitat for adfluvial bull trout.

A total of 77 redds were enumerated in the Coffee Creek drainage (Table 8; Figure 13). Another 15 redds were observed but incomplete and associated with either an adult or an adult pair. Redd survey timing was likely too early for this system, with a total of 32 spawning bull trout observed over the 6.7 km surveyed.

Crawford Creek

A ground reconnaissance survey was conducted on Crawford Creek in September 8, prior to conducting redd surveys despite being part of the ongoing bull trout monitoring index since 2008 (Andrusak 2010).

Redd surveys commenced on October 16, utilizing two experienced two man crews. The crew initiated the redd survey ~24 km upstream on Crawford Creek with a starting elevation of ~1500 m (Table 7). The conditions during the survey were ideal with

Redfish Consulting Ltd. Page 31 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

excellent visibility over the entire 24 km surveyed on Crawford Creek. The crews surveyed the entire system over the following 3 days.

A complete barrier to adfluvial bull trout migrants was identified at 24.2 km upstream on from the confluence with Kootenay Lake (Table 7; Photo 15). The barrier is a series of boulder cascades reaching 5 m in height with a gradient of up to 25%. Small tributaries to Crawford Creek including; Canyon Creek, Hooker Creek and Houghton Creek have barriers less than 600 m upstream from the confluence with Crawford Creek. Canyon Creek and Hooker Creek barriers are partial obstructions comprised of a series of logjams with drops of 1 m high. Both of these systems likely have complete barriers within the lower sections since the gradients exceed 20%. Houghton Creek has a complete barrier comprised of a bedrock chute ~2m in height. Moreover, Crawford Creek has two moderate canyon sections at 3 km and 4 km that are considered barriers to other migratory fish species, especially kokanee. In total, Crawford Creek provided >25 km of accessible spawning habitat to adfluvial bull trout.

A total of 209 complete redds were enumerated in the Crawford Creek drainage (Table 8; Figure 14). Only 1 adult spawner was observed during the survey and redd survey timing was considered optimal.

Table 7. Group 2 (Central Kootenay Lake tributaries) barrier location, characteristics and accessible habitat to obstruction. .

Accessible Name Complete Height Habitat Group (Tributary) Name (System) UTM or Partial Type (m) (km) Group 2 Kaslo River Kaslo River 11 490153 5543094 complete falls 3 m 28.8 Group 2 Keen Creek Kaslo River 11 495966 5529375 complete falls 4 m 6.4 Group 2 Twelve Mile Kaslo River 11 495805 5538918 complete chute 2 m 0.3 Group 2 Rossiter Creek Kaslo River 11 492677 5541725 complete chute 2 m 0.4 Group 2 Woodbury Creek Woodbury Creek 11 506505 5513847 partial falls 2 m 0.6 Group 2 Coffee Creek Coffee Creek 11 500890 5505774 complete falls 4 m 6.7 Group 2 Crawford Creek Crawford Creek 11 522634 5514246 complete cascade 5 m 24.2 Group 2 Canyon Creek Crawford Creek 11 523277 5508389 partial logjam 1 m 0.3 Group 2 Hooker Creek Crawford Creek 11 520365 5506278 partial logjam 1 m 0.1 Group 2 Houghton Creek Crawford Creek 11 515856 5505809 complete chute 2m 0.6

Redfish Consulting Ltd. Page 32 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Table 8. Group 2 (Central Kootenay Lake tributaries) redd survey data.

Date(s) Complete Incomplete Fish Group Name (Tributary) Name (System) Surveyed redds redds Count Group 2 Kaslo River Kaslo River October 4 422 19 33 Group 2 Keen Creek Kaslo River October 5 73 19 34 Group 2 Rossiter Creek Kaslo River October 4 14 3 4 Group 2 Twelve Mile Creek Kaslo River October 4 3 3 1 Group 2 Coffee Creek Coffee Creek October 5 77 15 32 Group 2 Woodbury Creek Woodbury Creek October 19 31 0 4 Group 2 Crawford Creek Crawford Creek October 16 209 1 1

Group 3 (South Arm Kootenay Lake tributaries)

Midge Creek

A ground reconnaissance survey was conducted on Midge Creek on September 20, prior to conducting redd surveys, to identify a possible barriers or obstructions to fish passage. Ground crews had good access to the upper portion of the watershed and tributaries utilizing ATV (all-terrain vehicle) along the Porcupine Creek FSR located in the Nature Conservancy of Canada (NCC) property, formerly known as Darkwoods property.

Redd surveys commenced on October 7, utilizing two experienced two man crews. The crew initiated the redd survey ~30 km upstream from the confluence with Kootenay Lake (Table 9). Kutetl Creek, Württemberg Creek and Seeman Creek were initially surveyed during the reconnaissance to identify the upstream extent of fish distribution (Table 9). Helicopter was used to survey Conway Creek and Hughes Creek prior to conducting the redd surveys. The crews surveyed the entire system over the following 4 days.

A complete barrier to adfluvial bull trout migrants was identified at ~20.3 km upstream on Midge Creek (Table 9; Photo 16). The barrier was a series of four waterfalls and/or bedrock chutes varying from 2-4 m in height. Fish were observed immediately downstream of these obstructions. Two partial barriers to adfluvial bull trout migrants were identified at ~12.6 km upstream on Seeman Creek (Table 9). These barriers comprised a large wood jam structure with a 1 m drop, with little or no plunge pool below. As well, low water limited the accessibility into reaches beyond these obstructions. Wurttemberg Creek, a tributary of Seeman Creek, had a complete bedrock chute barrier at ~800 m upstream from the confluence (Table 9; Photo 17). A complete barrier was identified at ~9.9 km upstream on Kutetl Creek (Photo 18). However, the

Redfish Consulting Ltd. Page 33 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011 first redd was not observed until ~600 m downstream, suggesting a partial barrier downstream. The survey crew identified a large logjam structures were considered partial barriers to fish access at ~ 9.3 km, just above the first redd observation. A complete bedrock chute barrier was identified at ~7.5 km upstream during the aerial over-flight on Conway Creek, re-confirming observations by Purcell Resources (2002). Hughes Creek was observed to be inaccessible to migrating fish at ~800 m upstream from its confluence with Midge Creek due to low flow. In an overview, Midge Creek drainage provided >50 km of accessible spawning habitat to adfluvial bull trout (Figure 15)

A total of 400 redds were enumerated in the Midge Creek drainage (Table 10; Figure 15). Of the total, Seeman Creek and Wurttemberg Creek had 139 and 30 redds, respectively. Kutetl Creek and Conway Creek had 55 and 16 redds, respectively. The remaining 160 redds were observed within the main channel of Midge Creek. Only 1 incomplete redd was observed and associated with either an adult or an adult pair in Midge Creek. Redd survey timing was considered optimal with only 5 spawning bull trout observed over the 50 km surveyed.

Cultus Creek

Similar to Midge Creek, a ground reconnaissance survey was conducted on Cultus Creek on September 12, 21, prior to conducting redd surveys, to identify possible barriers or obstructions to fish passage. Ground crews had good access to the upper portion of the watershed and tributaries utilizing ATV along FSRs located in the Nature Conservancy of Canada (NCC) property. The reconnaissance was conducted over ~20 km of the watershed, adjacent to the FSR.

Based on the reconnaissance and the upstream distribution of fish, redd surveys commenced on October 8. The crew initiated the redd survey ~3.5 km upstream from the confluence with Kootenay Lake. With the limited accessible habitat, the crew conducted the survey in 1 day (Table 9).

A series of barriers to adfluvial bull trout migrants were identified at ~3.3 km upstream on Cultus Creek (Table 9; Photo 19). The barriers consisted of a series of four waterfalls and/or bedrock chutes varying from 2-3 m in height. Large aggregations (>20 bull trout) were observed immediately downstream of these obstructions and no evidence of spawning was observed over ~ >15 km during the reconnaissance survey. Despite being a complete barrier during the 2011 surveys, it is believed that these barriers may pose only a partial obstruction under higher flows during some years, allowing access to > 15 km in habitat. Nonetheless, Cultus Creek provided 3.3 km of accessible spawning habitat to adfluvial bull trout in 2011.

Redfish Consulting Ltd. Page 34 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Despite having only 3.3 km of accessible habitat, a total of 50 redds were enumerated in Cultus Creek (Table 9; Figure 16). Redd survey timing was considered optimal with no spawning bull trout observed over the length surveyed.

Summit Creek

Ground reconnaissance survey were conducted on Summit Creek on September 16 and 25, prior to conducting redd surveys, to identify possible barriers or obstructions to fish passage. Ground crews had good access to the watershed and tributaries utilizing FSRs and Highway 3 for much of the ~40 km length. The reconnaissance commenced near the headwaters which originate in Stagleap Provincial Park located at the summit of the Salmo / Creston highway.

Redd surveys commenced on October 8-10 & 17-18, utilizing an experienced crew. The crews initiated the redd survey ~35 km upstream from the confluence with the Kootenay River Table 9). All tributaries including Char, Bayonne, Maryland and Blazed creeks were assessed. Due to the vast accessible habitat, crews surveyed the entire system over a 4 day period.

A complete barrier to adfluvial bull trout migrants was identified at ~33 km upstream on Summit Creek (Table 9; Photo 20). The barrier was a 3 m waterfall, confirming observations by Purcell Resources Inc. (2001). However, the first redd was not observed until ~3 km downstream, suggesting a partial barrier further downstream. The survey crew did not observe any partial barrier downstream or just above the first redd observed. A complete barrier to adfluvial bull trout migrants was identified at ~1.2 km upstream on Bayonne Creek (Table 9; Photo 21). The barrier was a 5m waterfall impassable by adfluvial bull trout. Similarly, Char and Maryland creeks had partial barriers impassable to migrating bull trout less than 300m upstream of the confluence with Summit Creek. Although Char Creek barrier was considered to be a partial barrier under the observed flows (Photo 22), a complete barrier was identified 2.9 km upstream by Purcell Resources Inc. (2001). Maryland Creek had a 20 m high bedrock chute barrier, impassable to fish (Photo 23). A complete barrier to adfluvial bull trout migrants was identified at 0.7 km upstream on Blazed Creek (Table 9; Photo 24). The barrier was a bedrock chute reaching in a 2 m in height. In an overview, Summit Creek drainage provided more than 35 km of accessible spawning habitat to adfluvial bull trout.

Despite having over 35 km of accessible habitat, only 29 redds were enumerated in the entire Summit drainage (Figure 17). The low redd count is also substantiated by the limited presence of spawners during the reconnaissance survey. Including all tributaries, only one redd was observed in Blazed Creek. Redd survey timing was considered good with only 1 spawning bull trout observed over the length surveyed.

Redfish Consulting Ltd. Page 35 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Table 9. Group 3 (South Arm Kootenay Lake tributaries) barrier location, characteristics and accessible habitat to obstruction. Italicized systems represent streams that had partial barriers located below the complete barrier.

Accessible Complete Height Habitat Group Name (Tributary) Name (System) UTM or Partial Type (m) (km) Group 3 Midge Creek Midge Creek 11 501842 5479245 complete falls 2 m 20.3 Group 3 Kutetl Creek Midge Creek 11 493686 5475913 complete chute 2 m 9.9 Group 3 Wurttemberg Creek Midge Creek 11 499032 5472341 complete chute 3 m 0.8 Group 3 Conway Creek Midge Creek 11 507398 5479650 complete chute 3 m 7.5 Group 3 Conway Creek Midge Creek 11 507327 5479716 partial logjam 3 m 5.3 Group 3 Conway Creek Midge Creek 11 508156 5477957 partial logjam 1 m 4.8 Group 3 Seeman Creek Midge Creek 11 494823 5471633 partial logjam 1 m 12.6 Group 3 Seeman Creek Midge Creek 11 495215 5471693 partial logjam 1 m 12.5 Group 3 Cultus Creek Cultus Creek 11 512711 5462726 partial chute 2 m 3.3 Group 3 Summit Creek Summit Creek 11 500059 5438012 complete falls 3 m 33.1 Group 3 Bayonne Creek Summit Creek 11 504525 5439942 complete falls 5 m 1.3 Group 3 Blazed Creek Summit Creek 11 514066 5442811 complete chute 2 m 0.7 Group 3 Char Creek Summit Creek 11 502245 5435250 complete chute 1 m 2.9 Group 3 Char Creek Summit Creek 11 503010 5437650 partial chute 1 m 0.2 Group 3 Maryland Creek Summit Creek 11 510310 5439073 complete chute 20 m 0.5

Table 10. Group 3 (South Arm Kootenay Lake tributaries) redd survey data.

Date(s) Complete Incomplete Fish Group Name (Tributary) Name (System) Surveyed redds redds Count Group 3 Conway Creek Midge Creek October 7 16 0 0 Group 3 Hughes Creek Midge Creek October 8 0 0 0 Group 3 Kutetl Creek Midge Creek October 7 55 0 1 Group 3 Midge Creek Midge Creek October 7-9 160 1 5 Group 3 Seeman Creek Midge Creek October 7 139 0 1 Group 3 Wurttemberg Creek Midge Creek October 7 30 0 0 Group 3 Blazed Creek Summit Creek October 10 1 0 0 Group 3 Char Creek Summit Creek October 8 0 0 0 Group 3 Maryland Creek Summit Creek October 10 2 0 0 Group 3 Summit Creek Summit Creek October 10 26 0 1 Group 3 Cultus Creek Cultus Creek October 8 50 0 0

Redfish Consulting Ltd. Page 36 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Discussion

Kootenay Lake bull trout monitoring-overview

The primary purposes of the 2011 redd surveys were to assess the distribution and relative abundance of adfluvial bull trout spawning in 19 select tributaries to Kootenay Lake. The lake and most of its tributaries provide a vast area of potential habitat for adfluvial bull trout populations. However, many of these tributaries to the lake offer limited access to much of the available habitat for spawning as a result of naturally occurring barriers that obstruct fish passage. Conversely, some of the most prominent Kootenay Lake bull trout spawning and rearing systems have been unnaturally impacted by the construction and formation of hydro-electric dams (Northcote 1973; Hirst 1991). Nevertheless, completion of this first large scale monitoring program for bull trout has greatly improved the knowledge base of their spawning distribution and abundance within the Kootenay Lake system.

The 2011 survey was the initial and logical action item emanating from the bull trout plan for Kootenay Lake described by Hagen and Decker (2009). The rationale of the entire plan included assessing: 1) the response of bull trout to nutrient restoration and kokanee enhancement in Kootenay Lake, 2) determining the current conservation status of populations in individual spawning tributaries, and 3) the identification of tributaries and reaches that currently support adfluvial populations, and the contribution of each to overall production in Kootenay Lake. The 2011 survey was aimed at surveying a total of 230 km of stream in all tributaries to Kootenay Lake (below Duncan Dam) with confirmed historic adfluvial populations (Hagen and Decker 2009). Inclusion of the Westfall River, a tributary of the Duncan Reservoir, resulted in survey of slightly over 280 km of accessible stream habitat surveyed for spawning bull trout in Kootenay Lake tributaries. As noted earlier, the Westfall River counts would include bull trout from Duncan Reservoir as well as Kootenay Lake fish that transferred through the dam. Despite the magnitude of the surveys, a number of potentially key tributaries for bull trout were not assessed or identified in the plan by Hagen and Decker (2009). Undoubtedly, the information collected from this survey will be the basis for establishing a future baseline index of abundance for bull trout on Kootenay Lake. An added benefit of conducting the spawning study in 2011 is that it coincided with a creel survey that will provide bull trout harvest estimates for Kootenay Lake (report in progress), and an exploitation study that will provide estimates of the natural and fishing mortality rates for the same year.

Understandably, obtaining accurate estimates of escapements is essential for managing exploited fish populations and imperative in monitoring population trends over time (Rieman and Myers 1997; Dunham and Davis 2001; Walters and Martell 2004).

Redfish Consulting Ltd. Page 37 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Establishment of a reliable index to ascertain trends in abundance and population status relies on and includes; 1) the repeatability of the survey 2) reducing uncertainty and variability around estimates 3) the ability to provide a cost effective index over time. While redd counts often provide a cost effective method of obtaining an index of adult escapement, their precision and accuracy are often compromised as a result of their uncertainty in obtaining reliable estimates and ability to detect sensitive changes in population status (Maxwell 1999; Dunham et al. 2001; Al-Chokhachy et al. 2005; Andrusak et al. 2011). Moreover, bull trout spawn in the headwaters of numerous tributaries that are often difficult to access and expensive to establish repeatable indices on. Assessments are further confounded by variable spatial and temporal distributions that collectively make spawning bull trout difficult to assess (Rieman and McIntyre 1996; Rieman and Myers 1997; Dunham et al. 2001). Yet, the success of future bull trout conservation and management decisions will be dependent on the ability of biologists to accurately assess and monitor their status or abundance, particularly in response to management and restoration actions that are implemented.

Redd survey inter-observer variation

Much of the variability in counts between observers can be substantially reduced when using experienced crew members (Dunham et al. 2001; Muhlfeld 2006). Despite this, life history (resident/adfluvial), redd density, habitat complexity, redd survey timing, water visibility and flow, redd size, redd age and weather conditions are also key factors affecting inter-observer efficiency when conducting redd counts.

In the replicated counts on the Kaslo River, much of the crew variability and detection error commonly occurred in areas of the Kaslo River where superimposition was highly prevalent, similar to that observed in Dunham et al. (2001) and Muhlfeld (2006). Given the differences in total counts between crews, the Kaslo River test counts indicate two sources of observer error in redd counts which include 1) omission, and 2) false identification. Omissions are considered to be related to redd densities because they are proportional to true redd numbers, while false identifications appear to be independent of redd frequency (Muhlfeld 2006). Moreover, false identification are uniformly distributed and related to the length of stream surveyed (i.e. longer the section the more false identifications). These characteristics were both observed on the Kaslo River redd count and standardization exercise. Nevertheless, while false identifications are independent of redd frequency, in systems with low redd densities, redd counts have been observed to overestimate the true redd number due to false positives (Muhlfeld 2006). The latter scenario is most concerning when populations are at low abundances. While estimates of observer efficiency are not available for each system and despite the large differences observed between most streams, observability should be obtained for each system in attempt to reduce uncertainty in estimates. It is also important to

Redfish Consulting Ltd. Page 38 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011 acknowledge also that the spawning habitat in the two Kaslo River test reaches were more conducive to redd identification (lower gradient with more extensive gravel deposits) than the majority of reaches in other tributaries (higher gradient with smaller pockets of gravel); hence the degree of potential variation among crews in other tributaries may have been underestimated. The test reaches were chosen because they were logistically feasible and were the only sections that could be surveyed by multiple crews with the available funding and resources.

Concern over the spatial and temporal variability in conducting redd surveys on such a large scale system, similar to Kootenay Lake, likely imposes more uncertainty than derived expansion factors for extrapolating a lake wide estimate of abundance. Decker and Hagen (2007) detailed their concerns of high flows on observability and temporal variability in spawn timing as contributing factors affecting the ability to conduct redd surveys in Arrow Lakes tributaries in 2005 and 2006. A similar circumstance was experienced on the Kaslo River in 2010, when less than optimal visibility obscured redds, resulting in poor observer efficiency (Matt Neufeld pers. comm.; MFLNRO). In that year, the counter estimate on the Kaslo River did not suggest a major decline in bull trout numbers as the redd survey indicated, suggesting the redd counts were underestimating under these conditions (Andrusak 2010).

While experienced crews are important in reducing uncertainty, survey timing is likely more critical in obtaining accurate redd counts (Dunham et al. 2001). It was observed that bull trout spawning activity peaked by mid- September and that redd construction ended by mid-October. This is a very narrow window for conducting redd counts on a large number of systems therefore, the timing of redd counts should be determined for each population of interest through periodical surveys to ensure that spawn timing is essentially complete. Resistivity counters offer an excellent understanding of the run timing and ideal time to implement redd surveys (Andrusak 2010). Importantly, the 2011 surveys indicated that survey was excellent for select streams surveyed.

Kootenay Lake bull trout abundance

A total of 1711 redds were enumerated within tributaries to Kootenay Lake, not including tributaries above Duncan Dam. This estimate does not represent a complete escapement for the entire lake population since a number of key systems with potential to support adfluvial bull trout were not surveyed. These include several tributaries to the Duncan and Lardeau rivers (notably Mobbs, Tenderfoot, Hope, Cascade, Lake and Rapid creeks and Trout Lake tributaries), Meadow Creek system (Legget 1980), as well as several tributaries to the South Arm (e.g. Sanca, Boundary) and Kootenay River tributaries. Despite these shortcomings of the Kootenay Lake bull trout plan, the 2011 estimates likely represent the majority of the total escapement. The most prominent (>

Redfish Consulting Ltd. Page 39 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

400 redds enumerated) bull trout systems included the Kaslo River and Midge Creek, together accounting for about 50% of all redds enumerated in the study. Likewise, Crawford Creek and Hamill Creek were relatively similar in their contribution, with > 200 redds enumerated on each system. Not surprisingly, the Kaslo River and Crawford Creek were amongst the highest redd counts of any system, affirming their importance as potential index systems and use of resistivity counts for bull trout on Kootenay Lake (Andrusak 2010).

While the large scale monitoring failed to obtain a total escapement for the lake, the magnitude of the known escapement indicates the bull trout population on the lake is quite robust. A crude estimate of escapement for the surveyed tributaries can be made using resistivity counter data from the Kaslo River, Keen and Crawford Creeks generated over the last few years using data from the calibrated resistivity counters. Despite some concerns using redd counts as an estimate of escapement (Rieman and Myers 1997; Muhlfeld 2006), combining the redd count data with the electronic counter data provides the opportunity to calibrate direct spawner counts with redd counts. Despite the acceptance of resistivity counter data by Hagen and Decker (2009), and without a total count on all available bull trout tributaries, the use of an expansion factor derived from the calibrated resistivity counter is currently the best method of deriving a lake wide escapement estimate. An average expansion factor of 2.4 (SD ± 0.21) bull trout per redd has been derived from the Kaslo River over a four year period (Table 11). Likewise, an average expansion factor of 1.9 (SD ± 0.17) bull trout per redd was derived from Crawford Creek over a three year period (Table 11). Despite the differences in the expansion factors between the two systems, likely attributable to the quality of habitat and accessible length, the estimates are similar to the average of 2.68 bull trout per redd (range of 1.2 to 4.3 bull trout per redd) reported by Al-Chokhachy et al. (2005). Based on the low variability of the estimates and the small range observed across two distinctively different systems, the use of 1.9-2.4 bull trout per redd can be applied to all systems to estimate escapement for the surveyed streams. Table 11. Derivation of expansion factor from estimated redds counts and electronic resistivity counts from the Kaslo River (2006-2011) and Crawford Creek (2008-2010). Kaslo River (upper) Crawford Creek Year # Redds Electronic Count Expansion Factor # Redds Electronic Count Expansion Factor 2006 321 716 2.2 n/a n/a n/a 2007 458 n/a n/a n/a n/a n/a 2008 471 1,197 2.5 188 336 1.8 2009 542 1,219 2.2 268 486 1.8 2010 302* 1,170 3.8* 182 389 2.1 2011 439 1,180 2.7 n/a n/a n/a Avg. 446.2 1096 2.4 213 404 1.9 Note*- 2010 Kaslo River redd count considered an unreliable estimate due to poor visibility

Redfish Consulting Ltd. Page 40 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Concomitant with the 2011 bull trout monitoring survey, a comprehensive lake wide creel census was also conducted. The redd counts (i.e. escapement), combined with harvest data can provides some measure of total potential spawner abundance in the lake. Utilizing the expansion factor range of 1.9-2.4 bull trout per redd provides an estimate of 3,251-4,106 spawners to the Kootenay Lake tributaries surveyed in 2011, excluding the Duncan Reservoir and tributaries. Inclusion of the Duncan Dam transfers, which account for ~300-1000 spawners annually (Table 12; BC Hydro 2010), the range estimate increases to 3,765-4,621 spawners. These estimates are considered conservative since a number of important tributaries were not surveyed, especially the tributaries to the Lardeau River and Kootenay River. Taking into account harvest data, based on the KLRT mail out survey (Redfish Consulting Ltd. 2007) and recent data from the Kootenay Lake Creel (in prep), an average of 5,000 bull trout (> 2 kg) are also harvested annually. Much of this harvest is assumed to be from the adult population based on age structure data from the fishery (Golder Associates Ltd. 2010). Andrusak and Thorley (2011) have reported that spawning repeatability is quite high in bull trout (> 60%), likely a result of good in-lake feeding conditions since nutrient addition, similar to that reported on the ALR by Arndt (2004). Accounting for 50% of the harvest, of which, the majority likely consists of maturing adults (~2,500 adults) and information on repeatability (Andrusak and Thorley 2011), the relative lake wide annual spawner abundance could potentially exceed 7,000.

The 2011 results are considered highly informative, and provide a baseline measurement that can be used for future lake-wide monitoring. The future success of bull trout conservation and management decisions on Kootenay Lake will be dependent on the ability of biologists to accurately assess and monitor their status, particularly in response to restoration and management actions that are implemented.

Table 12. Estimated number of bull trout transferred and number of transfers at Duncan Dam (BC Hydro data on file.). Year # of transfers Estimated # of bull trout 2007 9 371 2008 9 553 *2009 9 725 2010 10 971 2011 8 515 Avg 9 627 Note*-italicized year represents enumerated number

Redfish Consulting Ltd. Page 41 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Kootenay Lake bull trout distribution and redd density

Distribution of spawning bull trout within tributaries to Kootenay Lakes' covers a large geographic area, inter-connected by large sub-basins (Duncan Reservoir) and rivers (Lardeau and Kootenay rivers). Despite the construction of hydro-electric dams, the remaining inter-connectivity has likely maintained the genetic diversity within the Kootenay Lake bull trout population. Evidence and importance of connectivity can be observed after the removal of the Kaslo community hydro-electric dam in the mid-1970s (H. Andrusak pers. comm.), re-establishing the Kaslo River escapements of over 1,000 spawners annually. Annual transfers of bull trout at the Duncan Dam have been vital in maintaining the health and status of bull trout in Kootenay Lake (BC Hydro 2010; DDMMON#6). Importantly, Golder Associates Ltd. (2010; DDMMON#5) indicated greater than 50% of adults sampled in all three areas (Duncan, Duncan Dam and Kootenay Lake) were from tributaries above Duncan Dam. Proposed micro-hydro developments on select tributaries pose the most imminent threat to bull trout and their connectivity to Kootenay Lake. Maintaining this connectivity and diversity is vital in ensuring the future conservation of bull trout in the region.

Due to natural obstructions to fish passage, many systems provided only a fraction of the available habitat (total length) within the system to spawning bull trout. Most barriers posed a complete obstruction (falls or bedrock chutes) to migrant fish passage and surveys were conducted on accessible habitat downstream from the location of the barrier. However, some systems were observed to have partial barriers, mainly large wood jams that limited access seasonally or under lower flows. Environment Canada data indicates the fall of 2011 had unseasonable low precipitation compared to previous years (http://climate.weatheroffice.gc.ca), therefore it is possible that the extent of migration was more limited in 2011 than it would be in a year of higher flows. A good example of a partial barrier was the circumstance in 2011 that left only 0.8 km of accessible habitat in Woodbury Creek due to the small hydro dam that impeded fish passage. In other years this site has not been a barrier and bull trout have been observed well beyond the dam (interview with local anglers; H. Andrusak pers. comm.). There were also several systems at the higher elevations where low flows restricted further fish. At these sites the main stream consisted of small was divided into several small channels with gradients exceeding 15% and little available habitat beyond these flow related obstructions.

Within all systems that did not have barriers near the stream mouth, the spatial distribution of redds indicate that the majority of redds were found in the upper portions of watersheds, similar to that observed in previous years on the Kaslo River and Crawford Creek drainages (Andrusak 2010). These spatial patterns have also observed in other systems in British Columbia where bull trout populations are closely monitored

Redfish Consulting Ltd. Page 42 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

(Decker and Hagen 2008; Hagen and Decker 2009). As a result, redd density (redds/km) as a biostandard over the entire accessible length can be somewhat misleading, since most of the lower portions of many of these systems do not support spawning similar to the Kaslo River and Crawford Creek (Table 13; Andrusak 2010). Nevertheless for the sake of comparison the mean redd density in surveyed tributaries (total number divided by total accessible length, summed over all tributaries) was 7.9 redds/km (Table 13), resulting in a biostandard of 15.8 spawners/km based on an expansion factor of 2.0 spawners per redd cited in Hagen and Decker (2008). Interestingly, this density is slightly higher than 12.6 spawners/km identified by Hagen and Decker (2008) and 13 spawners/km presented in Lindsay (1977a,b) on Arrow Lakes tributaries. It is suggested that the difference in densities is a reflection of differences in lake productivity between the two systems. Nevertheless, using the biostandard to extrapolate over the accessible habitat surveyed in 2011 on Kootenay Lake, an estimate of 3,700 spawning bull trout is derived. Once again, this does not account for ~ 70 km of stream habitat not surveyed, which could bring the estimate to ~4, 800 spawners. Nonetheless, this estimate based on accessible habitat would be considered conservative in comparison to estimates derived from the redd count-resistivity counter expansion factor (section above).

Despite the successful recovery of kokanee on Kootenay Lake (Schindler et al. 2011), lack of pre-nutrient data on bull trout escapement makes the ability to assess their response to nutrient restoration difficult. Assessment of the benefit of nutrient addition to bull trout, as an objective of this study, will require a minimum of 10-15 years of annual escapement data, similar to that collected on Gerrard rainbow trout. Even then, without pre-fertilization escapement information, annual bull trout escapement data alone will not provide a definitive measure of the success of nutrient addition on Kootenay Lake. While the bull trout are considered one component of the aquatic ecosystem, assessing the entire trophic level response to nutrient addition will provide the best measure of success of the program, not outlined by Hagen and Decker (2009). Another potentially productive approach to assessing bull trout response to fertilization would be to analyze pre-fertilization growth (MOE data on file) in the 1980s vs. fertilization era. Nonetheless, the data provided by this study show that the 2011 population is quite robust, and that establishing a repeatable index for bull trout population monitoring is not only feasible but warranted.

Redfish Consulting Ltd. Page 43 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Table 13. Redd density based on accessible habitat from complete/partial barriers on select systems (including tributaries) in fall 2011.

Group Name (System) # complete redds Accessible Habitat (km) Redds/km Group 1 Hamill Creek 216 23.8 9 Group 1 Cooper Creek 134 29.6 4.4 Group 1 Poplar Creek 53 4.4 12 Above Duncan Dam Westfall River 114 21.2 5.4 Group 2 Kaslo River 512 35.9 14.3 Group 2 Coffee Creek 77 6.7 11.5 Group 2 Woodbury Creek 31 0.6 52 Group 2 Crawford Creek 209 25.2 8.4 Group 3 Midge Creek 400 49.7 8.0 Group 3 Cultus Creek 50 3.3 15.5 Group 3 Summit Creek 29 35.7 0.8 Total 1825 237

Kootenay Lake bull trout harvest implications

While harvest levels are a critical factor influencing adult bull trout survival rates in Kootenay Lake, use of the historic or current fishery data alone cannot provide the quantitative review needed to assess the benefits of fertilization to the upper trophic levels. Even the combination of harvest levels and escapement can only provide inferences following nutrient addition, without pre-nutrient data. However, obtaining estimates of harvest rates (exploitation), in combination with escapement and harvest data, can provide the necessary information to ensure the future of conservation of bull trout within Kootenay Lake (Andrusak and Thorley 2011). Moreover, adult and juvenile bull trout age data and micro-chemistry information is available for Kootenay Lake (Golder Associates Ltd. 2010). Assessment of bull trout productivity (compensatory capacity), utilizing juvenile assessments in combination with adult escapement, can also provide reference points or thresholds for regulating the intensive fishery which is considered the most imminent threat to the stocks on the lake (Andrusak and Thorley 2011). Bison et al. (2003) estimated bull trout fishing mortality was between 12-24% before implementing more conservative fishing regulations to avoid overfishing on Adams Lake. Similarly, Johnston et al. (2007) recommended more conservative fishing regulations on Kananaskis Lake following a retention fishery that nearly collapsed the population, resulting from 50% mortality rate for adult size bull trout. Kootenay Lake is far more productive than Adams or Kananaskis Lake yet preliminary data indicates exploitation rates of 20% are well above optimal levels of 13% for bull trout on Kootenay Lake (Andrusak and Thorley 2011). The age‐structured population model also suggests

Redfish Consulting Ltd. Page 44 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011 that an exploitation rate of 30% could potentially lead to a stock collapse (Andrusak and Thorley 2011).

Future monitoring

Depending on the priority of the monitoring program, reducing the uncertainty in redd counts can be accomplished by conducting replicate counts within an index site, measuring observer efficiency, measuring sources of observer efficiency and having a better understanding of the spatial and temporal variation in spawning activity (Andrusak et al. 2011). Isaak and Thurow (2006) suggest that conducting redd counts with a spatially continuous, temporally replicated sampling design will reduce errors associated with simple random designs and provide more accurate ecosystem views. In comparison, Jacobs et al. (2009) evaluated the utility of the generalized random tessellation stratified (GRTS) sampling design to determine bull trout population status through redd counts. Gallagher et al. (2007) recommended conducting redd counts over pre-determined 3-5 km stream reaches that are randomly selected with an annual selection of at least 10% of all reaches. However, these designs can often be more time consuming and rely heavily on statistical inference that often cannot capture the spatial and temporal variation unless substantial effort is appropriated through replication. Lack of randomization in site selection and shifts in fish distribution may bias results from index site monitoring and mask population trends (Isaak and Thurow 2006). Invariably, there is a trade-off between coverage (more sites) that satisfies population monitoring versus the ability to reduce the uncertainty or inter-annual variability within each index. Moreover, index areas may not represent population dynamics of salmonids at a regional scale (Rieman and McIntire 1996) and may miss redds due to annual variation in spatial distributions or spawning activity (Maxell 1999; Dunham et al. 2001).

While there is much debate over appropriate sampling designs to determine bull trout population status through redd counts, complete surveys of representative index systems are the most appropriate and practical means of monitoring bull trout on Kootenay Lake, alleviating the confounding issues of sub-sampling from stratified random designs.

Furthermore, establishing annual monitoring on select tributaries also requires an index that can be repeated within the same year and between years. Systems that can be cost effectively replicated within the same year can be valuable in understanding uncertainty associated with redd survey methodology in an attempt to improve the accuracy and precision of the index. Often, some systems cannot be repeated annually because of the difficultly in access, inclement weather, changes in seasonal flows that can have a profound effect on counting conditions and observability, similar to that reported on the Illecillewaet and Incomappleux Rivers (Decker and Hagen 2007). Notably, the 2011

Redfish Consulting Ltd. Page 45 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011 surveys were conducted with ideal counting conditions and survey timing, improving the certainty of the results. Such conditions cannot be expected every year, thus the emphasis of utilizing resistivity counters in addition to redd counts that are confounded by the vagaries of weather and variable flow conditions. Lastly, the ability to provide a cost effective index over time precludes the ability to monitor certain systems that require substantial financial and logistical support. For example, Cooper Creek and Hamill Creek support good numbers of bull trout but may be too cost prohibitive to repeat, notwithstanding the additional concerns of annual variation in environmental conditions and safety considerations in walking canyon sections of this system.

While it is acknowledged that bull trout populations may also have local and regional scale sub-populations, monitoring large spatial systems, such as Kootenay Lake, often precludes the ability to measure these local or geographic sub-populations. In order to gain some insight into the lake wide bull trout population(s) and owing to its large expanse, understanding the spatial and temporal variability in distribution within the watershed and basin should be the priority for their conservation and population management. Therefore, tributaries selected for annual monitoring should ideally be selected from the core range (Hagen and Decker 2011) and/or within the geographical sub-groups in this study (i.e. Group 1-3), also including tributaries above the Duncan Dam and follow survey timing schedule similar to that outlined in Appendix 5. Understandably, the importance of this emanates from the concern that population trends and habitat associations observed may not be representative of production dynamics for Kootenay Lake as a whole and may result in misleading conclusions, despite the population having numbers above the effective population size (McElhany et al. 2000; Rieman and Allendorf 2001).

Following the data and information collected from the 2011 surveys, it is recommended that Cooper and Hamill creeks not be continued as part of a future monitoring program due to the difficultly in access and owing to dangerous conditions at some locations, not to mention the prohibitive cost to deploy crews into the remote areas surveyed. Similarly, Poplar and Woodbury creeks pose similar concerns and the inability to complete full census due to the terrain.

Redfish Consulting Ltd. Page 46 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Therefore future monitoring should conduct complete census surveys during late- September-October (see Appendix 5) on:

Group 1 (Duncan/Lardeau tributaries) a minimum of two tributaries

1. Meadow Creek 2. Mobbs Creek 3. Poplar Creek as an alternate (despite the difficulty in assessing)

Group 2 (Central Kootenay Lake tributaries) a minimum of two tributaries

1. Kaslo River 2. Crawford Creek 3. Coffee 4. Woodbury Creek as an alternates (dependent on year)

Group 3 (South Arm Kootenay Lake tributaries) a minimum of one tributaries

1. Midge Creek 2. Cultus Creek as alternates

Duncan River tributaries- (above Duncan Dam) a minimum of one tributaries

1. Westfall River 2. Houston Creek (only in years where turbidity is good)

Reconnaissance of other systems should include;

1. Duncan River tributaries- (above Duncan Dam) including; Giegerich Creek 2. Group 1 (Lardeau tributaries) including; Tenderfoot Creek, Rapid Creek and Lardeau Creek 3. Group 3 (South Arm tributaries) including; Sanca Creek, Next Creek and Boundary Creek

Redfish Consulting Ltd. Page 47 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Conclusions

The 2011 survey of Kootenay Lake bull trout spawning in 19 tributaries and sub- tributaries revealed important abundance and distribution data, previously not available. Some systems provided spawning migrations more than 20 km in distance along the mainstem stream while others were less than 5 km. The longest migrations as evidenced by observable redds were on the Kaslo River and Midge Creek, in addition, to, supporting ~50% of all redds in the survey. The annual transfer of 400-1000 migrant bull trout per year at the Duncan Dam indicates significant numbers continue to contribute to the total population in Kootenay Lake. The only system surveyed that did not support adfluvial bull trout was Healy Creek which has a barrier < 1-2km from the confluence with the Lardeau River. Woodbury Creek has a partial barrier located very close to the lake that in some years has been successfully ascended by bull trout.

A total of 1,711 redds were counted in the 19 systems to Kootenay Lake, with an additional 114 redds counted in a tributary to the Duncan Reservoir, the Westfall River. Using a biostandard determined from Kaslo River and Crawford creek resistivity counter data the estimated number of spawners based on the 1711 redds ranged from 3,251- 4,106. Taking into account annual harvest, a crude estimate of total spawner abundance for Kootenay Lake likely exceeds 7,000.

Independent estimates of spawner numbers from the resistivity counters provides the opportunity to conduct future indices of adult escapement estimates on select streams using redd count methodology. Preliminary data from the 2011 sport fishery suggests bull trout harvest is fairly high and this factor alone warrants continued monitoring to ascertain if escapements to index streams are adequate to sustain the population. Notwithstanding the sport fishery harvest issue, bull trout abundance in Kootenay Lake appears to be very high.

As previously mentioned, the primary rationale for this study was to establish a first lake-wide index of spawning bull trout distribution and abundance in Kootenay Lake tributaries using redd counts. Undoubtedly, these counts, if repeated over the long term, will provide a valuable index for monitoring trends in bull trout populations while providing some performance measure for evaluating FWCP compensation efforts on the lake. However, any direct assessment of a response in the bull trout population (increased or decreased abundance) following nutrient addition will prove difficult and cumbersome since little or no information exists prior to the nutrient addition era (i.e. no measure of escapement).

Redfish Consulting Ltd. Page 48 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Recommendations

The following recommendations are made with the knowledge that other monitoring projects such as KLRT creel census, kokanee surveys and fishery exploitation are on- going:

 Conduct annual, complete census redd surveys on candidate systems in each of the three core groups surveyed in 2011. These should include Meadow Creek or at least two Lardeau tributaries and Westfall River under Group 1, Kaslo River and Crawford Creek under Group 2 and Midge Creek under Group 3.

 To have better understanding of the basin wide contribution of spawning bull trout, Lardeau River tributaries (Mobbs Creek and/or Tenderfoot Creek), Kootenay River tributaries (i.e. Boundary Creek) and select tributaries to Trout Lake (i.e. Lardeau Creek) should be assessed.

 Assess observer error and variability within a number of the select monitoring streams.

 Experienced survey crews in 2011 should continue future survey work based on survey timing window outlined in Appendix 5. Redd surveys should be conducted by experienced ground crews to reduce observer error in estimates.

 Surveys should be conducted from late-September to mid-October (Appendix 5).

 Resistivity counters should continue to be installed on Kaslo River and Crawford Creek. A counter should also be considered for Meadow Creek.

 The suggested index streams should be monitored for temperature and flows (if available) over the monitoring period, but more specifically applies to juvenile assessment work.

 Woodbury Creek obstruction should be assessed for potential improvement to fish passage, similar to what occurred in the early 1970s for the Kaslo River.

 Assessment and improvement of fish passage on Woodbury Creek should be conducted to access > 10 km of available habitat

 Genetic analysis of Kootenay Lake bull trout is not required owing to large spawner numbers and continued connectivity within the basin.

 Redd counts should be standardized and entered into an Access database held by the Ministry of Forests, Lands, and Natural Resource Operations.

Redfish Consulting Ltd. Page 49 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

References

Al-Chokhachy, R., P. Budy, and H Schaller. 2005. Understanding the significance of redd counts: a comparison between two methods for estimating the abundance of and monitoring bull trout populations. North American Journal of Fisheries Management 25:1505-1512.

Al-Chokhachy, R., P. Budy and M. Conner 2009. Detecting declines in the Abundance of a bull trout (Salvelinus confluentus) population: understanding the accuracy, precision and costs of our efforts. Can. J. Fish. Aquat. Sci. 66: 649-658

Andrusak, H. 1987. Kootenay Lake Sport Fishery 1984-86 unpublished MS, Fisheries Branch, Nelson, BC Report No. KO-19 31 p.

Andrusak, G. and D.J.F. McCubbing. 2006. Kaslo River Adult Bull Trout Pilot Spawner Assessment. Prepared for the Habitat conservation Trust Fund of BC and the Ministry of Environment, Nelson BC.

Andrusak, G.F. 2010. Kaslo River and Crawford Creek Adult Bull Trout Spawner Assessment. Prepared for the Fish and Wildlife Compensation Program, Columbia Basin, Nelson BC, the Habitat Conservation Trust Foundation and the Ministry of Environment, Nelson, BC. 40 pp.

Andrusak, G.F. H. Andrusak, and A.R. Langston. 2011. Bull trout (Salvelinus confluentus) redd count surveys in select Williston Reservoir tributaries (2001 – 2010) and recommendations for future surveys. Fish and Wildlife Compensation Program – Peace Region Report No. XXX. XX pp. + app.

Andrusak G.F. and Thorley J.L. (2011) Kootenay Lake Exploitation Study: Fishing and Natural Mortality of Large Rainbow Trout and Bull Trout. A Poisson Consulting Ltd. and Redfish Consulting Ltd. Report prepared for the Habitat Conservation Trust Foundation

Aprahamian, M.W, Nicholson S.M, McCubbing, D.J.F and Davidson I. 1996 The use of resistivity fish counters in fish stock assessment. In Stock Assessment in Inland Waters ed I.Cowx Chapter 3, 27-43.

Arndt, S. 2004a. Arrow Lakes Reservoir Creel Survey 2000-2002. Columbia Basin Fish and Wildlife Compensation Program, Nelson BC.

Redfish Consulting Ltd. Page 50 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Arndt, S., 2004b. Post-Fertilization Diet, Condition, and Growth of Bull Trout and Rainbow Trout in Arrow Lakes Reservoir Fish and Wildlife Compensation Program – Columbia Basin, Nelson, BC.

Ashley, K., L. C. Thompson, D. C. Lasenby, L. McEachern, K. E. Smokorowski and D. Sebastian (1997). Restoration of an Interior Lake Ecosystem: The Kootenay Lake Fertilization Experiment. Water Quality Resources Journal of Canada 32: 295‐323.

BC Environment. 1994. A Strategic Plan for the Conservation and Management of Char in British Columbia. BC Ministry of Environment, Fisheries Program, Victoria, BC.

BC Hydro. 2010. Duncan Dam Bull Trout Passage Monitoring (Year 2) Data Report. BC Hydro, Water License Requirements. Study No. DDMMON-6.

Binsted, G.A., and K.I. Ashley. 2006. Phosphorus Loading to Kootenay Lake from the Kootenay and Duncan rivers and Experimental Fertilization Program. Prepared for the British Columbia Conservation Foundation.

Chirico, A. 1993. Kootenay Lake tributary fisheries reconnaissance survey. Prepared for the Ministry of Environment, Nelson BC

Daley, R.J., E.C. Carmack, C.B.J. Gray, C.H. Pharo, S. Jasper, and R.C. Wiegand. 1981. The Effects of Upstream Impoundments on Kootenay Lake, BC Canada Inland Waters Directorate, Research Institute, Scientific Series, West Vancouver, British Columbia.

Decker, A. S., and J. Hagen. 2007. Distribution of adfluvial bull trout in tributaries of the Arrow Lakes Reservoir (2004-2006) and the feasibility of monitoring juvenile and adult abundance. Final report prepared for the Fish and Wildlife Compensation Program, Nelson, BC, March 2007. 129 p.

Decker, A. S., and J. Hagen. 2008. Adfluvial bull trout spawner abundance in tributaries to the Arrow Lakes Reservoir (2004-2007). Technical report prepared for the Fish and Wildlife Compensation Program – Columbia Basin, Nelson, BC.

Downs, C., D. Horan, E. Morgan-Harris, and R. Jakubowski. 2006. Spawning demographics and juvenile dispersal of an adfluvial bull trout population in Trestle Creek, Idaho. North American Journal of Fisheries Management 26:190- 200.

Redfish Consulting Ltd. Page 51 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Dunham, J., and B. Rieman. 1999. Meta-population structure of bull trout: influences of physical, biotic, and geometrical landscape characteristics. Ecological Applications 9:642-650

Dunham, J., B. Rieman and K. Davis. 2001. Sources and Magnitude of Sampling Error in Redd Counts for Bull Trout. North American Journal of Fisheries Management 21:343-352.

Dunham, J., B. Rieman and Gwynne Chandler. 2003. Influences of Temperature and Environmental Variables on the Distribution of Bull Trout within Streams at the Southern Margin of Its Range. North American Journal of Fisheries Management: 894–904.

Fraley, J.J. and B.B. Shepard. 1989. Life history, ecology, and population status of migratory bull trout (Salvelinus confluentus) in the Flathead Lake and River system, Montana. Northwest Science 63: 133-143.

Gallagher, S. P., P. K. J. Hahn, and D. H. Johnson. 2007. Redd counts. Pages 197–234 in D. H. Johnson, B. M. Shrier, J. S. O’Neal, J. A. Knutzen, X. Augerot, T. A. O’Neil, and T. N. Pearsons. Salmonid field protocols handbook: techniques for assessing status and trends in salmon and trout populations. American Fisheries Society, Bethesda, Maryland..

Gallagher, S.P., P.B. Adams, D.W. Wright and B.W. Collins 2010. Performance of Spawner Survey Techniques at Low Abundance Levels. North American Journal of Fisheries Management 30: 1086-1097, 2010.

Galesloot, M and D.J.F McCubbing. 2003. Chinook Escapement into the Bonaparte River, Summer 2003. Secwepemc Fisheries Commission Report. 51pp

Golder Associates Ltd. 2010. Upper Duncan Bull Trout Migration Monitoring – Final Report March 2010. Report prepared for BC Hydro, Castlegar, BC Golder Report No. 09-1480-0051: 49 p. + 8 app.

Hagen, J., and S. Decker 2009. Bull Trout Monitoring Plan for Kootenay Lake. Contract report prepared for the Fish and Wildlife Compensation Program-Columbia Basin, Nelson BC.

Hagen, J., and S. Decker 2011. Bull Trout Population ‘Health’ in British Columbia: Distribution, Abundance, Trends, and Threats. Prepared for: Ministry of Environment, Ecosystems Protection & Sustainability Branch, Aquatic Conservation Science Section Victoria, British Columbia

Redfish Consulting Ltd. Page 52 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Hirst, S. M. 1991. Impacts of the operations of existing hydroelectric developments on fishery resources in British Columbia, Volume II: inland fisheries. 91. Canada. Department of Fisheries and Oceans. Canadian Manuscript Report of Fisheries and Aquatic Sciences.

Isaak, D. J.; Thurow, R. F. 2006. Network-scale spatial and temporal variation in Chinook salmon redd distributions: patterns inferred from spatially continuous replicate surveys. Canadian Journal of Fisheries and Aquatic Sciences. 63: 285-296.

Jacobs, S. E., W. Gaeumand, M. A. Weeber, S. L. Gunckel, and S. J. Starcevich. 2009. Utility of a probabilistic sampling design to determine bull trout population status using redd counts in basins of the Columbia River plateau. North American Journal of Fisheries Management29:1590–1604.

Johnston, F. D., J. R. Post, C. J. Mushens, J. D. Stelfox, A. J. Paul and B. Lajeunesse. 2007. The demography of recovery of an overexploited bull trout, Salvelinus confluentus, population. Canadian Journal of Fisheries and Aquatic Sciences 64:113-126.

Jordan, P. 2008. Hamill Creek Fire, 2007, N70139. Post-Wildfire Risk Analysis. Forest Sciences Section, Southern Interior Forest Region, BC Ministry of Forests and Range, Nelson, BC. Prepared for BC Ministry of Forests and Range: Southeast Fire Centre, Southern Interior Forest Region, and Kootenay Lake Forest District

Latham, S. J. 2002. Historical and anthropogenic influences on genetic variation in bull trout (Salvelinus confluentus) in the Arrow Lakes, British Columbia. M.Sc. thesis, University of British Columbia, Vancouver BC.

Leggett, J. W. 1980. Reproductive ecology and behaviour of Dolly Varden charr in British Columbia. Pages 721-737 in Balon, E. K. editor. Charrs, salmonid fishes of the genus Salvelinus. W. Junk, The Hague, Netherlands.

Lindsay, R. A. 1977a. Investigations on fish populations that will be affected by the Revelstoke 1880 Dam, Volume II. Report prepared by BC Fish and Wildlife Branch for BC Hydro and Power Authority.

Lindsay, R. A. 1977b. Revelstoke Dam fish compensation on the Upper Arrow Lake 1977. Report prepared for Kootenay Region Fish and Wildlife Branch, Nelson, and BC Hydro and Power Authority, Vancouver.

Maxell, B. A. 1999. A prospective power analysis on the monitoring of bull trout stocks using redd counts. North American Journal of Fisheries Management 19:860–866

Redfish Consulting Ltd. Page 53 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

McCubbing, D.J.F 1999. Steelhead and Rainbow Trout Migration in the Deadman River in 1999, based on fish counter, trap and radio-tracking observations. MELP Project Report. McCubbing, D.J.F and G.F. Andrusak. 2006. Lardeau River Fish Counter: Gerrard Rainbow Trout Enumeration Report. Prepared for the Habitat conservation Trust Fund of BC and the Ministry of Environment, Nelson BC.

McCubbing, D.J.F and L. Burroughs. 2002. Application of flat pad counting technology in the enumeration of Quesnel Rainbow Trout on McKinley Creek Prepared for the Habitat conservation Trust Fund of BC and the Ministry of Environment .2002.

McPhail, J. D. 2007. The Freshwater Fishes of British Columbia. University of Alberta Press, Edmonton Alberta. 620p.

McPhail, J. D., and C. B. Murray. 1979. The early life history and ecology of Dolly Varden (Salvelinus malma) in the upper Arrow Lakes. Report to BC Hydro and British Columbia Ministry of Environment, Fisheries Branch, Nelson BC.

McPhail, J. D., and J. S. Baxter. 1996. A review of bull trout (Salvelinus confluentus) life history and habitat use in relation to compensation and improvement opportunities. Province of British Columbia Fisheries Management Report No. 104.

McElhany, P., M.H. Ruckelhaus, M.J. Ford, T.C. Wainwright, and E.P. Bjorkstedt. 2000. Viable salmonid populations and the recovery of evolutionarily significant units. U.S. Dept. Commer., NOAA Tech. Memo. NMFS-NWFSC-42.

Moody, A., P. Slaney, and J. Stockner. 2007. Footprint impact of BC Hydro dams on aquatic and wetland productivity in the Columbia Basin. Report prepared by AIM Ecological Consultants Ltd. In association with Eco-Logic Ltd. and PSlaney Aquatic Science Ltd. for Fish and Wildlife Compensation Program, Nelson, BC.

Muhlfeld C.C., Shepard, B.B, Taper, M.L and D.F. Staples. 2006. Observer Error Structure in Bull Trout Redd Counts in Montana Streams: Implications for Inference on True Redd Numbers. Transactions of the American Fisheries Society 135:643- 654.

Ney, J. 1996. Oligotrophication and its discontents: effects of reduced nutrient loading on reservoir fisheries. American Fisheries Society Symposium 16:285-295.

Northcote, T.G. 1973. Some Impacts of Man on Kootenay Lake and Its Salmonids. Great Lakes Fisheries Commission Tech. Rep. 25.

Redfish Consulting Ltd. Page 54 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

O’Brien, D.S. 1999. The Duncan bull trout radio telemetry program. Prepared for Columbia Basin Fish & Wildlife Compensation Program, Nelson BC.

Paragamian, V, Gidley, K. and J. Walters. 2010. Kootenai River Fisheries Investigations: Salmonid Studies. Bull Trout Studies Summary Report. Prepared by Idaho Department of Fish and Game for U.S. Department of Energy Bonneville Power Administration Division of Fish and Wildlife

Paul, A. J., and J. R. Post. 2001. Spatial distribution of native and nonnative salmonids in streams of the eastern slopes of the Canadian Rocky Mountains. Transactions of the American Fisheries Society 130:417-430.

Post, J. R., C. Mushens, A. Paul, and M. Sullivan. 2003. Assessment of alternative harvest regulations for sustaining recreational fisheries: model development and application for bull trout. North American Journal of Fisheries Management 23:22-34.

Purcell Resources. 2001. Reconnaissance (1:20,000) Fish and Fish Habitat Inventory of Summit Creek Watershed (WSC: 340-418600). Prepared for J.J. Huscroft Ltd.

Purcell Resources. 2002. Reconnaissance (1:20,000) Fish and Fish Habitat Inventory of Midge Creek Watershed (WSC: 340-31110). Prepared for J.J. Huscroft Ltd.

Redfish Consulting Ltd. 2007. Kootenay Lake Rainbow Trout Survey Questionnaire Results 2003-2006. Contract report prepared for the Ministry of Environment, Fisheries and the Habitat Conservation Trust Fund of BC.

Rieman, B. E., and J. D. McIntyre. 1993. Demographic and habitat requirements for the conservation of bull trout Salvelinus confluentus. USDA, Forest Service, Intermountain Research Station, General Technical Report INT-302, Ogden, UT.

Rieman, B.E. and McIntyre, J.D. 1996. Spatial and Temporal Variability in Bull Trout Redd Counts. North American Journal of Fisheries Management 16:132-141, 1996.

Rieman, B.E. and L. Myers. 1997. Use of Redd Counts to Detect Trends in Bull Trout (Salvelinus confluentus) Populations. Conservation Biology 1015-1018 Volume 11. No. 4,1997.

Rieman, B.E. and F.W. Allendorf. 2001. Effective population size and genetic conservation criteria for bull trout. North American Journal of Fisheries Management. 21:330-338.

Redfish Consulting Ltd. Page 55 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Schindler, E.U., D. Sebastian, H. Andrusak, L. Vidmanic, G.F. Andrusak, M. Bassett, F., T. Weir and K.I. Ashley. 2011. Kootenay Lake Fertilization Experiment, Year 17 (North Arm) and Year 5 (South Arm) (2008) Report. Fisheries Project Report No. RD 131, Ministry of Environment, Province of British Columbia.

Selong, J.H., T.E. McMahon, A.V. Zale, and F.T. Barrows. 2001. Effect of Temperature on Growth and Survival of Bull Trout, with Application of an Improved Method for Determining Thermal Tolerance in Fishes. Transactions of the American Fisheries Society Vol 130, Issue 6. pp. 1026–1037, 2001.

Timberland Consultants Ltd. 1997. Crawford Creek Interior Watershed Procedure (IWAP). Contract report prepared for the Ministry of Environment, Lands and Parks (MELP) and the Ministry of Forests, Kootenay Lake Forest District. 1996

Timberland Consultants Ltd. 2001. Reconnaissance (1:20,000) Fish and Fish Habitat Inventory of Cooper Creek/McKian Creek Watershed (WSC: 340-218400-02000). Prepared for Meadow Creek Cedar and Slocan Forest Products Ltd.

Walters, C.J. and S.J.D. Martell. 2004. Fisheries Ecology and Management. Princeton University Press, Princeton, New Jersey.

Redfish Consulting Ltd. Page 56 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Appendix 1–Tributary information Stream Waterbody UTM Length Watershed Name (Tributary) Name (System) Watershed Code ID Location (km) Area (km2) 340-218400-02900- 11 506093 Clint Creek Hamill Creek 09800 00000DUNC 5562685 15 71 340-218400-02000- 11 494596 McKian Creek Cooper Creek 43700 00000DUNC 5558908 14 97 340-218400-02000- 11 490665 Spokane Creek Cooper Creek 43700-3510 00000DUNC 5562257 5 14 340-218400-02000- 11 489766 South Cooper Creek Cooper Creek 68600 00000DUNC 5555666 16 41 340-218400-67700- 11 479617 Marsh Adams Creek Westfall River 29600 00000DUNC 5626236 12 47 340-218400-67700- 11 473524 Silvertip Creek Westfall River 60000 00000DUNC 5630351 7 14 11 501151 Keen Creek Kaslo River 340-215300-23800 00000KOTL 5529801 29 203 11 492460 Rossiter Creek Kaslo River 340-215300-74400 00000KOTL 5541472 7 14 11 495714 Twelve Mile Creek Kaslo River 340-215300-61700 00000KOTL 5539257 10 25 11 523070 Canyon Creek Crawford Creek 340-243500-58100 00000KOTL 5508508 7 22 11 520337 Hooker Creek Crawford Creek 340-243500-44000 00000KOTL 5506384 7.5 20 11 515414 Houghton Creek Crawford Creek 340-243500-23600 00000KOTL 5506146 12 27 11 501738 Kutetl Creek Midge Creek 340-311100-69100 00000KOTL 5479175 11 53 11 504900 Seeman Creek Midge Creek 340-311100-49100 00000KOTL 5474970 15 69 340-311100-49100- 11 499072 Wurttemberg Creek Midge Creek 49700 00000KOTL 5472961 3.5 23 11 505732 Conway Creek Midge Creek 340-311100-45100 00000KOTL 5474428 10 30 11 505538 Hughes Creek Midge Creek 340-311100-45900 00000KOTL 5474420 8 15 11 504978 Bayonne Creek Summit Creek 340-418600-73700 00000KOTL 5438991 9 45 11 514453 Blazed Creek Summit Creek 340-418600-44000 00000KOTL 5442264 15.5 54 11 502924 Char Creek Summit Creek 340-418600-80100 00000KOTL 5437858 5 13 11 510328 Maryland Creek Summit Creek 340-418600-58000 00000KOTL 5439534 8 24

Redfish Consulting Ltd. Page 57 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Appendix 2–Field data forms

2011 Kootenay Lake Bull Trout Survey Location: Date Start Time: Crew: End Time: Temp: Visibility: Description/Comment *GPS Tracking every second, Camera, GPS, Watch all the same time *Take a picture of your GPS with Time to the second

REDD / Fish Observations Location Incomplete Dist (km) / Complete Time Male Female REDD# WayPnt # Redds

Redfish Consulting Ltd. Page 58 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Appendix 3–Photos

Photo 1. The upper Kaslo River counter in 2011 Photo 2..Work box housing counter and video equipment

Photo 3. Video record of an up-streaming bull trout Photo 4. Example of a typical graphical fish trace of an on the upper Kaslo River upstream bull trout on the upper Kaslo River

Redfish Consulting Ltd. Page 59 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Photo 5. Westfall River complete barrier fall 2011 Photo 6. Poplar Creek complete barrier fall 2011

Photo 7. Hamill Creek complete bedrock barrier fall Photo 8. Hamill Creek partial cascade barrier fall 2011 2011

Redfish Consulting Ltd. Page 60 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Photo 9. McKian Creek complete barrier fall 2011 Photo 10. South Cooper Creek complete barrier fall 2011

Photo 11. Kaslo River complete barrier fall 2011. Photo 12. Keen Creek complete barrier fall 2011

Redfish Consulting Ltd. Page 61 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Photo 13. Woodbury Creek partial barrier fall 2011 Photo 14. Coffee Creek complete barrier fall 2011.

Photo 15. Crawford Creek complete barrier fall 2011 Photo 16. Midge Creek Complete barrier fall 2011

Redfish Consulting Ltd. Page 62 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Photo 17. Württemberg Creek complete barrier fall Photo 18. Kutetl Creek complete barrier fall 2011 2011

Photo 19. Cultus Creek complete barrier fall 2011 Photo 20. Summit Creek complete barrier fall 2011

Redfish Consulting Ltd. Page 63 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Photo 21. Bayonne Creek complete barrier fall 2011 Photo 22. Char Creek complete barrier fall 2011

Photo 23. Maryland Creek complete barrier fall 2011 Photo 24. Blazed Creek complete barrier fall 2011

Redfish Consulting Ltd. Page 64 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Appendix 4–Redd survey distributions in select tributaries

Figure 7. Distribution of redds counted in the Westfall River in fall of 2011

Figure 8. Redd survey conducted in Healy Creek in fall of 2011

Redfish Consulting Ltd. Page 65 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Figure 9. Distribution of redds counted in Poplar Creek in fall of 2011

Redfish Consulting Ltd. Page 66 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Figure 10. Distribution of redds counted in Hamill Creek and tributaries in fall of 2011

Redfish Consulting Ltd. Page 67 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Figure 11. Distribution of redds counted in Cooper Creek and tributaries in fall of 2011

Redfish Consulting Ltd. Page 68 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Figure 12. Distribution of redds counted in the Kaslo River and tributaries in fall of 2011

Redfish Consulting Ltd. Page 69 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Figure 13. Distribution of redds counted in Coffee Creek and Woodbury Creek in fall of 2011

Redfish Consulting Ltd. Page 70 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Figure 14. Distribution of redds counted in Crawford Creek and tributaries in fall of 2011

Redfish Consulting Ltd. Page 71 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Figure 15. Distribution of redds counted in Midge Creek and tributaries in fall of 2011

Redfish Consulting Ltd. Page 72 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Figure 16. Distribution of redds counted in Cultus Creek in fall of 2011

Redfish Consulting Ltd. Page 73 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Figure 17. Distribution of redds counted in Summit Creek and tributaries in fall of 2011

Redfish Consulting Ltd. Page 74 Bull Trout Redd Count Surveys in Select Kootenay Lake Tributaries-2011

Appendix 5–Future monitoring-redd survey timing

October 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Group Stream/River Redd Surveys Timing Meadow Creek Group 1 (Lardeau tributaries) Mobbs Creek Kaslo River Group 2 (Central KL tributaries) Crawford Creek Coffee Creek Group 3 (South KL tributaries) Midge Creek Group 4 (above Duncan Dam) Westfall River

Redfish Consulting Ltd. Page 75