Assessment of Trout Abundance and Distribution in the Waiparous Creek Drainage, Alberta, 2006

Transcription

1 Assessment of Trout Abundance and Distribution in the Waiparous Creek Drainage, Alberta, 26 CONSERVATION REPORT SERIES

2 The Alberta Conservation Association is a Delegated Administrative Organization under Alberta s Wildlife Act. CONSERVATION REPORT SERIES 25% Post Consumer Fibre When separated, both the binding and paper in this document are recyclable

3 Assessment of Trout Abundance and Distribution in the Waiparous Creek Drainage, Alberta, 26 Kevin Fitzsimmons Alberta Conservation Association P. O. Box 142 Cochrane, Alberta, Canada T4C 1B4

4 Report Editors PETER AKU KELLEY KISSNER Alberta Conservation Association 5 Tuscany Meadows Cres NW #11, 9 Chippewa Rd Calgary, AB T3L 2T9 Sherwood Park, AB T8A 6J7 Conservation Report Series Type Data ISBN printed: ISBN online: Publication No.: T/24 Disclaimer: This document is an independent report prepared by the Alberta Conservation Association. The authors are solely responsible for the interpretations of data and statements made within this report. Reproduction and Availability: This report and its contents may be reproduced in whole, or in part, provided that this title page is included with such reproduction and/or appropriate acknowledgements are provided to the authors and sponsors of this project. Suggested Citation: Fitzsimmons, K. 28. Assessment of trout abundance and distribution in the Waiparous Creek drainage, Alberta, 26. Data Report, D 28 11, produced by the Alberta Conservation Association, Cochrane, Alberta, Canada. 39 pp. + App. Cover photo credit: David Fairless Digital copies of conservation reports can be obtained from: Alberta Conservation Association #11, 9 Chippewa Rd Sherwood Park, AB T8A 6J7 Toll Free: Tel: (78) Fax: (78) info@ab conservation.com Website: conservation.com i

5 EXECUTIVE SUMMARY The primary objective of this study was to provide estimates of abundance and distribution of bull trout, cutthroat trout, and brook trout in the Waiparous Creek drainage to assist with management of trout populations. In 26, I backpack electrofished stream sites along Waiparous, Meadow, Johnson and Unnamed creeks to obtain monitoring data needed to estimate fish abundance, distribution and movement. I sampled 5 m sites in two periods; early season (late spring to early summer; n = 59) and late season (early fall; n = 32). During the early season, I marked 53 bull trout and 133 cutthroat trout (all > 15 mm in fork length, FL). I used capture mark recapture population estimation methods at a subset of sites to estimate electrofishing capture efficiency. Maximum likelihood estimates of capture efficiency varied among creeks, but were typically higher for cutthroat trout (.36 to.65), followed by bull trout (.2 to.54), and then brook trout (.38 to.46). Of the three study species, bull trout capture efficiencies exhibited the least precision. Next, I used a non parametric generalized additive model to estimate fish (> 7 mm FL) abundance in 5 m intervals along study streams. Estimates were replicated 1, times to develop 95% confidence intervals around abundance projections. Based on capture mark recapture methods and non parametric modeling, brook trout was the most abundant of the three species along 4 km of Waiparous Creek. Waiparous Creek supported the most bull trout and cutthroat trout compared to tributaries of Waiparous Creek. Surveys were only conducted along Meadow Creek in the early sampling period on 13.5 km of stream. In this stream, I estimated relatively few bull trout or cutthroat trout, whereas I estimated brook trout to be 33 times more abundant than bull trout. In Johnson Creek, early sampling indicated that cutthroat trout was approximately 2.2 times more abundant than bull trout, and that brook trout was the most abundant species in this creek at 14 times more abundant than bull trout. Fall sampling did not occur along the full length of Johnson Creek. I failed to capture brook trout in Unnamed Creek (3.5 km). Bull trout abundance in Unnamed Creek was relatively low; however, the majority of fish could be considered juveniles. I captured few cutthroat trout in Unnamed Creek in either the early or late season. ii

6 To identify fish movement patterns and life history strategies used by fish in the Waiparous Creek drainage, I marked 53 bull trout and 133 cutthroat trout. Of tagged fish, I recaptured five bull trout and 17 cutthroat trout at later dates. One bull trout moved a minimum of 23.3 km between captures, and all others were recaptured within 5 m of their initial capture location. Although there was an increase in bull trout abundance between early and late sampling periods, considerable variance exists around these estimates and no strong support for a fluvial or ad fluvial life history strategy was found. Further, the number of mature large bull trout in the drainage was relatively unchanged between sampling periods. iii

7 ACKNOWLEDGEMENTS I thank Alberta Conservation Association (ACA) employees Marco Fontana, Chad Judd, Adrian Meinke, Mike Rodtka, Jay Wielickzo, and Thomas Winter, and Alberta Sustainable Resource Development (ASRD) staff Brian Lajeunesse and John Bilas for their assistance in data collection. I also thank Mike Blackburn (ASRD) and Mike Rodtka for their direction on project design, and Andrew Paul (ASRD) for his considerable help with the software programs R and MARK. Mike Rodtka and Cam Stevens provided helpful reviews on drafts of this document. iv

8

9 TABLE OF CONTENTS EXECUTIVE SUMMARY...ii ACKNOWLEDGEMENTS...iv TABLE OF CONTENTS...v LIST OF FIGURES...vi LIST OF TABLES... viii LIST OF APPENDICES...ix 1. INTRODUCTION STUDY AREA MATERIALS AND METHODS Study design Data collection Population modeling RESULTS Capture efficiencies Waiparous Creek Meadow Creek Johnson Creek Unnamed Creek Fish maturity and movement Summary LITERATURE CITED APPENDICES...4 v

10 LIST OF FIGURES Figure 1. Figure 2. Figure 3. Figure 4. Location of the Waiparous Creek drainage, including Waiparous, Meadow, Johnson and Unnamed creeks, relative to the City of Calgary and other major streams...3 Location of inventory and capture mark recapture sites in the Waiparous Creek drainage...5 Bull trout abundance estimates in Waiparous Creek...12 Cutthroat trout abundance estimates in Waiparous Creek...13 Figure 5. Brook trout abundance estimates in Waiparous Creek...14 Figure 6. Figure 7. Figure 8. Fork length frequency distribution of bull trout captured in Waiparous Creek in early and late sampling periods Fork length frequency distribution of cutthroat trout captured in Waiparous Creek in early and late sampling periods Fork length frequency distribution of brook trout captured in Waiparous Creek in early and late sampling periods Figure 9. Bull, cutthroat and brook trout abundance estimates in Meadow Creek...19 Figure 1. Fork length frequency distribution of bull, cutthroat and brook trout captured in Meadow Creek in the early sampling period...2 Figure 11. Bull trout abundance estimates in Johnson Creek...22 Figure 12. Cutthroat trout abundance estimates in Johnson Creek...23 Figure 13. Brook trout abundance estimates in Johnson Creek...24 Figure 14. Fork length frequency distribution of bull trout captured in Johnson Creek in early and late sampling periods...25 Figure 15. Fork length frequency distribution of cutthroat trout captured in Johnson Creek in early and late sampling periods Figure 16. Fork length frequency distribution of brook trout captured in Johnson Creek in early and late sampling periods...27 Figure 17. Bull trout abundance estimates in Unnamed Creek...29 Figure 18. Cutthroat trout abundance estimates in Unnamed Creek...3 Figure 19. Fork length frequency distribution of cutthroat trout captured in Unnamed Creek in early and late sampling periods Figure 2. Fork length frequency distribution for bull trout captured in Unnamed Creek in early and late sampling periods vi

11 Figure 21. Mature bull, cutthroat and brook trout captured in the Waiparous Creek drainage in Figure 22. Estimated mean abundance and 95% confidence limits for bull, cutthroat and brook trout for early and late sampling periods for all study streams in the Waiparous Creek drainage where comparisons could be made...36 vii

12 LIST OF TABLES Table 1. Table 2. Table 3. Table 4. Table 5. Table 6. Abundance of bull trout, cutthroat trout and brook trout in study streams in the Waiparous Creek drainage...9 Capture efficiencies for bull, cutthroat, and brook trout captured with capture mark recapture population estimate techniques at four locations in Waiparous Creek drainage and one location in Canyon Creek drainage...1 Abundance estimates and 95% confidence limits for Waiparous Creek bull, cutthroat, and brook trout in early and late sampling periods...11 Abundance estimates and 95% confidence limits for Meadow Creek bull, cutthroat, and brook trout in the early sampling period...18 Abundance estimates and 95% confidence limits for Johnson Creek bull, cutthroat, and brook trout in early and late sampling periods...21 Abundance estimates and 95% confidence limits for Unnamed Creek bull and cutthroat trout in early and late sampling periods...28 viii

13 LIST OF APPENDICES Appendix 1. Appendix 2. Appendix 3. Fish captures by species for study sites in the Waiparous Creek drainage in early and late sampling periods, Summary of capture mark recapture of trout at population estimate sites...43 Project location identification, habitat assessment identification, geographic coordinates, mean wetted width, and mean stream depth for study sites in the Waiparous Creek drainage, ix

14

15 1. INTRODUCTION Bull trout (Salvelinus confluentus) and westslope cutthroat trout (Oncorhynchus clarkii) are native to the Bow River drainage (Scott and Crossman 1973; Haas and McPhail 1991; Behnke 1992) and have experienced declines in distribution and abundance over the past century (Behnke 1992; Post and Johnson 22). Declines have been attributed to overexploitation, competition with exotic fish species, and habitat alteration and fragmentation (Rhude and Stelfox 1997; Dunham and Rieman 1999). In Alberta, bull trout is currently listed as Sensitive and a province wide zero harvest limit has been in place since Cutthroat trout in the South Saskatchewan River drainage, its native range in Alberta, is listed as Threatened (Committee on the Status of Endangered Wildlife in Canada, 26) and a recovery plan for this species is in preparation (Alberta Sustainable Resource Development and Alberta Conservation Association 26). One of the first tasks in prescribing management plans for sport fish is an assessment of fish abundance. For abundance estimates to be useful, inventory data need to be collected in a manner that reduces biases whenever possible. For example, selecting fish inventory locations based on access considerations or collecting data at locations known to contain relatively high densities of fish (i.e., non random site selection) may bias abundance estimates. Of particular concern is monitoring of fish abundance at good fishing areas, which often leads to an overestimation of fish numbers and failure to detect a collapsed population (Hilborn and Walters 1992). Here, my primary objective was to describe abundance and spatial distribution of bull trout, cutthroat trout and brook trout (S. fontinalis) in Waiparous, Meadow, Johnson and Unnamed creeks in two open water time periods (early and late season). To reduce sampling biases, I adopted a systematic survey design with a random starting location (Morrison et al. 21). Estimates of fish abundance and distribution were adapted from Paul and Dormer (25) and were based on fish captures at inventory sites and capture efficiencies derived from capture mark recapture methods employed at a subset of sites. It is important to calculate capture efficiencies when estimating abundance as efficiency can vary with factors such as fish species and capture technique. It is then possible to describe the variability in capture efficiencies and apply this to catch at inventory sites to estimate fish abundance by species. 1

16 Through examination of temporal data of trout abundance estimates for early and late sampling periods, as well as movement data obtained from recaptures of tagged fish, managers may better understand the life history strategies of trout in Waiparous Creek drainage. For example, a significant increase in mature bull trout abundance in this drainage in the late sampling period would be indicative of a fluvial (versus resident) life history strategy and that significant movements were being made between Waiparous Creek and the Ghost River of Ghost Reservoir (Hogen and Scanecchia 26). 2. STUDY AREA Waiparous Creek is a fifth order tributary to the Ghost River (Bow River drainage) within the Rocky Mountain and Foothills Natural Regions of Alberta (Natural Regions Committee 26). Located approximately 5 km northwest of Calgary, Alberta, Waiparous Creek is 51 km in length and drains an area of approximately 332 km 2 (Figure 1). Waiparous Creek supports populations of bull trout, cutthroat trout, brook trout and mountain whitefish (Prosopium williamsoni) (Fitzsimmons and Fontana 23). Waiparous Creek and its major tributaries, Meadow Creek, Johnson Creek and Unnamed Creek were my study streams (Figure 1). 2

17 Unnamed Creek Johnson Creek Meadow Creek Waiparous Creek ô ôó4 Ghost River N Kilometers ôó1a Calgary Bow River River Kananaskis ô ôó4 ôó1 ô Jumpingpound Creek Elbow River Figure 1. Location of the Waiparous Creek drainage, including Waiparous, Meadow, Johnson and Unnamed creeks, relative to the City of Calgary and other major streams. 3

18 3. MATERIALS AND METHODS 3.1 Study design I geo referenced all inventory and capture mark recapture population estimate sites in the field (Universal Transverse Mercator projection, NAD 83) using a hand held global positioning system unit. I collected both fish abundance and stream habitat data at inventory sites. The locations of these sites were determined through a systematic study design with a random starting location. All inventory sites were 5 m in length and inter site distance was 1, m on Waiparous Creek and 5 m on Johnson, Meadow and Unnamed creeks (Figure 2). I conduced sampling at inventory sites in two time periods, early and late in the season. Early season sampling was conducted from 3 May to 7 July 26 (n = 59, Figure 2) and late season sampling from 5 to 21 September 26 (n = 32, Figure 2). Late season sampling was conducted at the same locations as early sampling. Within a study stream, sampling commenced at the furthest downstream site and progressed to upstream sites. If bull or cutthroat trout were not captured at two consecutive sample sites, I did not complete additional upstream sampling. This situation was encountered in the headwaters of Waiparous and Unnamed creeks during early season sampling (Figure 2). Waterfall barriers likely prevented fish from reaching their upper reaches. Early season sampling on Johnson Creek continued systematically upstream until headwaters were reached. I found the site on Johnson Creek immediately downstream of site 62 (Figure 2) to be dry, and no data were collected. Sampling on Meadow Creek continued systematically upstream until a large beaver dam prevented sampling at the site between location 72 and 82 (Figure 2). I sampled three more inventory sites upstream of this skipped site until another large beaver dam at the headwaters of the creek prevented further sampling. Late season sampling on Waiparous and Unnamed creeks did not include the sites on each creek in the headwaters where no fish were captured during the early sampling period. Because of time and budgetary constraints, I only conducted late season sampling along the first 6.5 km (lower section) of Johnson Creek, and I did not conduct late season sampling on Meadow Creek. 4

19 Figure 2. Location of inventory and capture mark recapture sites in the Waiparous Creek drainage. Sites have been labelled with the last two digits of the Alberta Sustainable Resource Development, Fisheries Management Information System project location identification. 5

20 I collected data at capture mark recapture sites from 1 to 28 July 26 by backpack electrofishing one site in each of the four study streams, following methods of Otis et al. (1978) (Figure 2). The location of capture mark recapture sites was based on suitable locations to set blocking nets at these sites, as well as access considerations. Capturemark recapture sites on Unnamed Creek and Waiparous Creek were 818 m and 66 m in length, respectively. Capture mark recapture sites were 1, m in length on Meadow Creek and Johnson Creek to allow for sufficient numbers of fish to be marked for analysis. 3.2 Data collection I sampled fish populations using a Smith Root model 12 B backpack electrofisher outputting pulsed current (voltage 4 7 V, frequency 25 5 Hz, and duration 1 4 ms). Sites were electrofished with a single pass in a downstream to upstream direction. Electrofishing was not conducted when water clarity was reduced from what would normally be expected for an individual stream. At all sites, I recorded electrofishing effort (seconds) and fish species, fork length (mm) and weight (g) every 5 m. When discernible, I also recorded fish sex and life stage. To aid in discerning movement patterns, I tagged bull trout and cutthroat trout 15 mm with Floy T bar anchor tags with unique colour and number combinations to identify individual fish. The number of fish captured by species for locations shown in Figure 2 is presented in Appendix 1. At capture mark recapture sites, I marked fish electrofished on the single marking run by clipping their adipose fin. Each site was electrofished 48 h later with a single pass recapture run and I recorded the number of previously marked fish to unmarked fish. Fish emigration and immigration were assumed to be eliminated at capture markrecapture sites by placing block nets at the top and bottom of the site for the duration of the estimate. I revisited each site 24 h after the marking run to clean and inspect block nets. A summary of fish captures at capture mark recapture sites is presented in Appendix 2. 6

21 At all sites, I measured water temperature (.1 C) and conductivity (.1 μohm 1 ) immediately prior to electrofishing. I measured stream wetted width, rooted width and maximum depth at transects spaced 5 m apart within a study site. Stream wetted width (.1 m) was measured as the width of the water surface at a right angle to the direction of flow and stream rooted width (.1 m) was measured as the distance at a right angle to the direction of flow from woody rooted vegetation to woody rooted vegetation. Stream depth (.1 m) was measured at the place of maximum depth at each wetted/rooted width transect. I estimated the percentages of stream pool, riffle and run for each 5 stream sections between transects. Mean wetted width and maximum stream depth for each location are provided in Appendix 3. At each survey site, I recorded any significant observations relating to fish or fish habitat that were not addressed on inventory forms. I also noted barriers and potential barriers to fish migration. Types of barriers included waterfalls, culverts, beaver dams, subterranean flows, low water, water velocity or other. I took representative photos at each sample site using a digital camera. All fish and habitat data collected were entered into the Alberta Sustainable Resources Development, Fisheries Management Information System (FMIS) database under Project Inventory ID Population modeling I followed methods by Paul and Dormer (25) when estimating fish capture efficiencies, population sizes and uncertainty in population sizes of trout in the Waiparous Creek drainage. I corrected species catch data at systematic inventory sites using electrofishing capture efficiencies picked at random from a modeled distribution of capture efficiency per species. Fish abundance at each inventory site was expressed as the observed catch plus the number of fish expected to have been missed. I estimated abundance and spatial distribution for fish 7 mm FL only (Peterson et al. 24). Fish abundance was modeled along the entirety of each study stream. Modeling incorporated two levels of uncertainty. First, I used multiple capture markrecapture estimates of fish capture efficiencies to project a beta distribution of capture 7

22 efficiencies per species. The beta distribution was used to model capture efficiency as it ranges in values from to 1, which lends itself to describing proportions, and it s two shape parameters (α and β) are defined by the mean and the variance of the multiple capture efficiency estimates. where, x and υ are the mean and the variance, respectively, of the capture efficiency estimates. Second, I addressed uncertainty in fish abundance per species and location using a constant capture efficiency. A negative binomial distribution of possible fish missed at each site was generated with a capture efficiency drawn at random from the beta distribution and the number of fished observed at each site as parameters. Fish abundance at each location was then expressed as the number of fish observed at the site, plus a random value from the negative binomial distribution of fish missed at the site. I then estimated fish spatial distribution and abundance with a generalized additive model (GAM) in consecutive 5 m increments for each stream. This exercise was repeated 1, times to calculate 95% confidence limits around means. I estimated electrofishing capture efficiency using Program MARK software (Cooch and White 1998) with closed captures. Population modeling was completed using the R software program (R Development Core Team 26). 8

23 4. RESULTS Table 1 presents abundance (fish/km) of bull trout, cutthroat trout, and brook trout in early and late sampling periods. Values presented are based on the actual number of fish captured in study streams (all fish > 7 mm FL) and the length (km) of the stream electrofished. In all streams where brook trout was captured, its abundance was higher than that of any other species. In Waiparous Creek, brook trout was three times more abundant than bull trout, and in Meadow Creek it was 4 times more abundant than bull trout. Cutthroat trout was the second most abundant species in Meadow and Johnson creeks. Bull trout was the most abundant species in Unnamed Creek. Table 1. Abundance (fish/km) of bull trout, cutthroat trout and brook trout in study streams in the Waiparous Creek drainage. Bull trout/km Cutthroat trout/km Brook trout/km Early Late Early Late Early Late Waiparous Meadow 5.4 NA NA NA 1 Johnson Unnamed NA 2 NA 2 1 Estimate not available because late season sampling was not conducted. 2 Estimate not available because no brook trout were captured. 4.1 Capture efficiencies I modeled bull trout, cutthroat trout and brook trout electrofishing capture efficiencies and associated uncertainties for fish captured with capture mark recapture techniques (Table 2). In general, maximum likelihood estimates of capture probabilities were greatest for cutthroat trout, followed by brook trout and then bull trout. However, there was considerable overlap in the modeled capture probabilities with brook trout having the narrowest distribution. 9

24 Table 2. Capture efficiencies for bull, cutthroat, and brook trout captured with capture mark recapture population estimate techniques at four locations in Waiparous Creek drainage and one location in Canyon Creek drainage. Capture efficiency (q) Population estimate site Bull trout Cutthroat trout Brook trout Waiparous Creek NA Meadow Creek NA Johnson Creek NA Unnamed Creek.5366 NA NA Canyon Creek NA NA Unnamed Creek NA NA Mean q Beta distribution 95% CI Population estimated in 25 (Fitzsimmons 27) in Canyon Creek drainage (Elbow River). NA indicates that no estimate was obtained. 4.2 Waiparous Creek In Waiparous Creek, cutthroat trout abundance was similar between early and late season sampling periods. Bull trout was almost two times more abundant in the late sampling period and brook trout was 1.4 times more abundant in the late sampling period (Table 3). However, abundance estimates for trout species in Waiparous Creek were statistically similar based on overlapping confidence intervals (Table 3; Figures 3 5). Frequency histograms of FL for bull, cutthroat and brook trout indicated that size of fish captured did not vary greatly between early and late sampling periods, suggesting a resident life history strategy for trout species in Waiparous Creek (Figures 6 8). Furthermore, although bull trout abundance estimates increased almost two fold from early to late sampling, confidence limits for these estimates were large and overlapping (Figure 3). Mature bull trout of sizes characteristic of a fluvial life history (fish > 4 mm, Post and Johnson 22) decreased from only five to four individuals (5% to 2%) between early and late sampling periods. 1

25 Table 3. Abundance estimates and 95% confidence limits for Waiparous Creek bull, cutthroat, and brook trout in early and late sampling periods using catch data and capture efficiencies from capture mark recapture population estimates and the nonparametric model. Species Time Period Mean abundance (4 km of stream) Lower 95% CL (4 km of stream) Upper 95% CL (4 km of stream) Bull trout Early 1, ,45 Late 2, ,86 Cutthroat trout Early ,523 Late ,799 Brook trout Early 1,952 1,552 2,485 Late 2,824 2,253 3,667 11

26 Bull trout early Bull trout late 2 Abundance per 5 m Distance upstream (m) Figure 3. Bull trout abundance estimates per 5 m in Waiparous Creek made using the nonparametric model and fish captures at inventory sites corrected with capture efficiencies from capture mark recapture techniques. Shown is the mean (1, model runs) of the 5 m sections over which estimates were made for both the early and late sampling periods. Banding the mean is the upper and lower 95% confidence limits. 12

27 5 Cutthroat trout early Cutthroat trout late Abundance per 5 m Distance upstream (m) Figure 4. Cutthroat trout abundance estimates per 5 m in Waiparous Creek made using the nonparametric model and fish captures at inventory sites corrected with capture efficiencies from capture mark recapture techniques. Shown is the mean (1, model runs) of the 5 m sections over which estimates were made for both the early and late sampling periods. Banding the mean is the upper and lower 95% confidence limits. 13

28 125 1 Brook trout early Brook trout late Abundance per 5 m Distance upstream (m) Figure 5. Brook trout abundance estimates per 5 m in Waiparous Creek made using the nonparametric model and fish captures at inventory sites corrected with capture efficiencies from capture mark recapture techniques. Shown is the mean (1, model runs) of the 5 m sections over which estimates were made for both the early and late sampling periods. Banding the mean is the upper and lower 95% confidence limits. 14

29 Frequency Early Late Fork Length (cm) Figure 6. Fork length frequency distribution of bull trout captured in Waiparous Creek in early and late sampling periods. 15

30 Early Frequency Late Figure Fork Length (cm) Fork length frequency distribution of cutthroat trout captured in Waiparous Creek in early and late sampling periods. 16

31 1 9 8 Early Frequency Late Figure Fork Length (cm) Fork length frequency distribution of brook trout captured in Waiparous Creek in early and late sampling periods. 17

32 4.3 Meadow Creek I conducted electrofishing surveys in Meadow Creek in the early sampling period only (Table 4 and Figure 9). Throughout Meadow Creek, brook trout was consistently more abundant than bull trout and cutthroat trout (approximately 34 times more abundant than bull trout, Table 4 and Figure 9) and represented nearly 92% of all fish captured in Meadow Creek. Based on sizes of cutthroat and brook trout captured in Meadow Creek, it is likely that juveniles to adult fish were present (Figure 1). The smallest bull trout captured was 125 mm FL indicating that Meadow Creek may not provide significant spawning or rearing habitat for bull trout. Table 4. Abundance estimates and 95% confidence limits for Meadow Creek bull, cutthroat, and brook trout in the early sampling period using catch data and capture efficiencies from capture mark recapture population estimates and the nonparametric model. Species Time period Mean abundance (13.5 km of stream) Lower 95% CL (13.5 km of stream) Upper 95% CL (13.5 km of stream) Bull trout Early Cutthroat trout Early Brook trout Early 6,535 5,267 8,152 18

33 45 4 Brook trout Bull trout Cutthroat trout 35 3 Abundance per 5 m Distance upstream (m) Figure 9. Bull, cutthroat and brook trout abundance estimates per 5 m in Meadow Creek made using the nonparametric model and fish captures at inventory sites corrected with capture efficiencies from capture mark recapture techniques. Shown is the mean (1, model runs) of the 5 m sections over which estimates were made for the early sampling period. Banding the mean is the upper and lower 95% confidence limits. 19

34 1 8 Bull trout early Cutthroat trout early 3 Frequency Brook trout early Fork Length Figure 1. Fork length frequency distribution of bull, cutthroat and brook trout captured in Meadow Creek in the early sampling period. 2

35 4.4 Johnson Creek I completed inventory sampling on 15.5 km (i.e., the entire length) of Johnson Creek in the early time period and on the first 6.5 km of the stream in the late time period (Table 5, Figures 11 13). In the first 6.5 km of Johnson Creek, abundance of bull trout did not vary from early to late sampling periods (Figure 11), whereas abundances of cutthroat and brook trout were higher in the early sampling period (Figures 12 and 13). Fork length frequency histograms of bull, cutthroat and brook trout in Johnson Creek (Figures 14 16) did not indicate a shift in fish size distribution between early and late sampling periods for any species. Based on a size of 15 mm for the lower size limit for mature resident bull trout populations (Post and Johnson 22), catch decreased from 43 bull trout in the early sampling period to 14 bull trout in the late sampling period. The portion of the total catch in each sampling period that could be considered mature resident bull trout increased from 8% to 88% from the early to late sampling period. The greater numbers of fish captured in the early sampling period was attributed to greater effort in the early sampling period needed to conduct a population estimate that was not conducted in the late sampling period. Considering fluvial bull trout life histories (fish > 4 mm, Post and Johnson 22), mature bull trout increased from to 2 (13%) from the early to the late sampling period. Table 5. Abundance estimates and 95% confidence limits for Johnson Creek bull, cutthroat, and brook trout in early and late sampling periods using capture efficiencies from mark recapture population estimates and the nonparametric model. Species Time period Mean abundance Lower 95% CL Upper 95% CL Bull trout Early 15.5 km Late 6.5 km Cutthroat trout Early 15.5 km ,297 Late 6.5 km Brook trout Early 15.5 km 3,88 3,136 4,676 Late 6.5 km ,1 21

36 3 Bull trout early Bull trout late Abundance per 5 m 1 Distance upstream (m) Figure 11. Bull trout abundance estimates per 5 m in Johnson Creek made using the nonparametric model and fish captures at inventory sites corrected with capture efficiencies from capture mark recapture techniques. Shown is the mean (1, model runs) of the 5 m sections over which estimates were made for both the early and late sampling periods. Banding the mean is the upper and lower 95% confidence limits. 22

37 6 5 Cutthroat trout early Cutthroat trout late 4 Abundance per 5 m Distance upstream (m) Figure 12. Cutthroat trout abundance estimates per 5 m in Johnson Creek made using the nonparametric model and fish captures at inventory sites corrected with capture efficiencies from capture mark recapture techniques. Shown is the mean (1, model runs) of the 5 m sections over which estimates were made for both the early and late sampling periods. Banding the mean is the upper and lower 95% confidence limits. 23

38 4 35 Brook trout early Brook trout late 3 Abundance per 5 m Distance upstream (m) Figure 13. Brook trout abundance estimates per 5 m in Johnson Creek made using the nonparametric model and fish captures at inventory sites corrected with capture efficiencies from capture mark recapture techniques. Shown is the mean (1, model runs) of the 5 m sections over which estimates were made for both the early and late sampling periods. Banding the mean is the upper and lower 95% confidence limits. 24

39 2 15 Early 1 5 Frequency 2 15 Late Fork Length (cm) Figure 14. Fork length frequency distribution of bull trout captured in Johnson Creek in early and late sampling periods. Early sampling occurred on 15.5 km of stream, whereas late sampling occurred only on the first 6.5 km from the mouth of Johnson Creek upstream. 25

40 4 Early Frequency Late Fork Length (cm) Figure 15. Fork length frequency distribution of cutthroat trout captured in Johnson Creek in early and late sampling periods. Early sampling occurred on 15.5 km of stream, whereas late sampling occurred only on the first 6.5 km from the mouth of Johnson Creek upstream. 26

41 Early Frequency 2 5 Late Fork Length (cm) Figure 16. Fork length frequency distribution of brook trout captured in Johnson Creek in early and late sampling periods. Early sampling occurred on 15.5 km of stream, whereas late sampling occurred only on the first 6.5 km from the mouth of Johnson Creek upstream. 27

42 4.5 Unnamed Creek Bull trout abundance in Unnamed Creek was greater in the early sampling period than the late sampling period (Table 6 and Figure 17). I estimated cutthroat trout distribution and abundance for the early sampling period (Table 6 and Figure 18), but not for the late sampling period because too few fish were captured to allow estimates to be made. Cutthroat trout abundance in Unnamed Creek was very low with all captured fish > 14 mm FL (Figure 19). The frequency distribution of FL for bull trout in Unnamed Creek suggested that fish size distribution did not vary greatly between sampling periods (Figure 2). Greater numbers of fish captured in the early sampling period than the late sampling period was due, in part, to the early sampling period including data from the capture mark recapture population estimate that was not repeated in the late sampling period. Table 6. Abundance estimates and 95% confidence limits for Unnamed Creek bull and cutthroat trout in early and late sampling periods using capture efficiencies from mark recapture population estimates and the nonparametric model. No estimates have been made for cutthroat trout in the late sampling period as too few fish were captured. Species Time period Mean abundance (3.5 km of stream) Lower 95% CL (3.5 km of stream) Upper 95% CL (3.5 km of stream) Bull trout Early Late Cutthroat trout Early Late NA NA NA NA = Not available due to insufficient captures to allow for population modeling. 28

43 1 75 Bull trout early Bull trout late Abundance per 5 m Distance upstream (m) Figure 17. Bull trout abundance estimates per 5 m in Unnamed Creek made using the nonparametric model and fish captures at inventory sites corrected with capture efficiencies from capture mark recapture techniques. Shown is the mean (1, model runs) of the 5 m sections over which estimates were made for both the early and late sampling periods. Banding the mean is the upper and lower 95% confidence limits. 29

44 25 2 Cutthroat trout early Abundance per 5 m Distance upstream (m) Figure 18. Cutthroat trout abundance estimates per 5 m in Unnamed Creek made using the nonparametric model and fish captures at inventory sites corrected with capture efficiencies from capture mark recapture techniques. Shown is the mean (1, model runs) of the 5 m sections over which estimates were made. Banding the mean is the upper and lower 95% confidence limits. No estimates have been made for the late sampling period as too few fish were captured. 3

45 1 8 Early Frequency 1 8 Late Fork Length (cm) Figure 19. Fork length frequency distribution of cutthroat trout captured in Unnamed Creek in early and late sampling periods. 31

46 5 4 Early Frequency 5 4 Late Fork Length (cm) Figure 2. Fork length frequency distribution for bull trout captured in Unnamed Creek in early and late sampling periods. 32

47 4.6 Fish maturity and movement I described seasonal changes in abundance of mature trout to aid with understanding fish movement patterns. During the early sampling period, cutthroat trout (n = 33) were the only sport fish captured in the Waiparous Creek drainage that were mature. These fish were captured in approximately the middle three quarters of Waiparous Creek, the lower portion of Meadow Creek, and the lower two thirds of Johnson Creek. During late season sampling, mature bull trout (n = 7) and brook trout (n = 68) were captured (Figure 21). Mature bull trout were captured in Waiparous Creek from the confluence of Meadow Creek to the confluence of Johnson Creek, in the lower portion of Johnson Creek, and at the lower inventory site on Unnamed Creek. During late season sampling, mature brook trout were captured in the lower three quarters of Waiparous Creek and in the first 6.5 km of Johnson Creek. As late season sampling was not conducted on Meadow Creek, no maturity information was available for this creek in this time period. During early inventory sampling in the Waiparous Creek drainage, 53 bull trout (Waiparous Creek n = 15, Unnamed Creek n = 2, Johnson Creek n = 15, Meadow Creek n = 21) and 133 cutthroat trout (Waiparous Creek n = 5, Unnamed Creek n = 2, Johnson Creek n = 2, Meadow Creek n = 61) were tagged with Floy T bar anchor tags. Of tagged fish, five bull trout and 17 cutthroat trout were recaptured at later dates. Four of these bull trout recaptures were within 5 m of their initial capture (on Waiparous and Johnson creeks) and one moved from project location 9969 (Meadow Creek) on 5 July 26 to project location 9944 (Unnamed Creek) on 13 September 26 (Figure 2). If traveling the most parsimonious route, this fish moved 23.3 km during this time period. Sex and reproductive status were not known when recaptured. Of 17 recaptured cutthroat trout, 13 were within 5 m of their initial capture and tagging location (on Waiparous, Johnson and Meadow creeks), two fish moved 6 m downstream on Meadow Creek between 7 and 28 July 26, one moved 4.94 km upstream on Meadow Creek between 21 June and 6 July 26, and one moved 1.7 km upstream on Waiparous Creek between 1 and 19 May

48 \ \ \ \ $T$T$T$T$T$T$T \ S $T $T$T$T$T \ \ $T$T$T$T$T$T$T$T$T $T$T$T S \ S $S$T $T $T$T$T$T$T $T $T$T$T$T S$T$T$T$T$T$T$T$T $T$T$T \\ $T$T$T$TS \ $T$T \ $T$T$T$T$T Legend N \ Mature cutthroat trout $T Mature brook trout S Mature bull trout Kilometers $T$T$T$T \ $T$T Figure 21. Mature bull, cutthroat and brook trout captured in the Waiparous Creek drainage in 26. Mature cutthroat trout captures were in the early sampling period and mature bull and brook trout captures were in the late sampling period. 34

49 4.7 Summary I estimated abundance and 95 percent confidence limits for bull, cutthroat and brook trout during early and late season sampling in the Waiparous Creek drainage. These estimates compared early and late season abundance for streams where early and late season sampling was conducted. These estimates took into account unequal sampling effort between early and late seasons. There was not strong support for a greater abundance of any one species during either early or late season sampling in Waiparous Creek. Overall, bull trout abundance estimates were greater for late season than early season sampling (Figure 22), which may be indicative of a fluvial life history in this population. However, the confidence limits around bull trout estimates for late season sampling were quite large. This substantial variability in late season estimates for bull trout was due to a highly variable capture efficiency, and failure of the spatial model to fit tight limits around the patchy distribution of bull trout observed in Waiparous Creek in the late season. 35

50 Abundance Early mean and 95% CI Late mean and 95% CI Bull trout Cutthroat trout Brook trout Figure 22. Estimated mean abundance and 95% confidence limits for bull, cutthroat and brook trout for early and late sampling periods for all study streams in the Waiparous Creek drainage where comparisons could be made. 36

51 5. LITERATURE CITED Alberta Sustainable Resource Development and Alberta Conservation Association 26. Status of the westslope cutthroat trout (Oncorhynchus clarkii) in Alberta. Alberta Sustainable Resource Development, Wildlife Status Report No. 61, Edmonton, Alberta. 34 pp. Behnke, R.J Native trout of western North America. American Fisheries Society Monograph 6. Committee on the Status of Endangered Wildlife in Canada (COSEWIC). 26. Assessment and update status report on westslope cutthroat trout Oncorhynchus clarkii lewisi (British Columbia population and Alberta population) in Canada. Committee on the Status of Endangered Wildlife in Canada. Ottawa, ON. Vii + 67 pp. Available online at: Cooch E., and G. White Program MARK: a gentle introduction. Dunham, J.B., and B. E. Rieman Metapopulation structure of bull trout: influences of physical, biotic, and geometrical landscape characteristics. Ecological Applications 9: Fitzsimmons, K. 28. Monitoring bull trout and cutthroat trout populations in Canyon and Prairie Creek drainages, Elbow River, Alberta, 25. Data report, D 28 1, produced by the Alberta Conservation Association, Cochrane, Alberta, Canada. 27 pp. + App. Fitzsimmons, K., and M. Fontana. 23. Cochrane and area upper Bow River cooperative Fisheries inventory project 23/24 annual report. Alberta Conservation Association, Cochrane, Alberta. 192 pp. 37

52 Haas, G.R., and J.D. McPhail Systematics and distributions of dolly varden (Salvelinus malma) and bull trout (Salvelinus confluentus) in North America. Canadian Journal of Fisheries and Aquatic Sciences 48: Hilborn, R., and C. Walters Quantitative fisheries stock assessment: choice, dynamics, and uncertainty. Chapman and Hall Inc., New York, New York. 57 pp. Hogen, D.M., and D.L. Scanecchia. 26. Distinct fluvial and adfluvial migration patterns of a relict charr, Salvelinus confluentus, stock in a mountainous watershed, Idaho, USA. Ecology of Freshwater Fish 15: Morrison, M.L., W.M. Block, M.D. Strickland, and W.L. Kendall. 21. Wildlife study design. Springer Verlag, New York, New York. 21 pp. Natural Regions Committee 26. Natural regions and subregions of Alberta. Compiled by D.J. Downing and W.W. Pettapiece. Government of Alberta, Publication No. I/5, Edmonton, Alberta. 254 pp. Otis, D.L., K.P. Burnham, G.C. White and, D.R. Anderson Statistical inference from capture data on closed animal populations. Wildlife monographs No 62. Paul, A.J., and C.G. Dormer. 25. Effect of a server flood on the cutthroat trout population of Silvester Creek, Alberta. G8 Legacy Chair in Wildlife Ecology, Department of Biological Sciences, University of Calgary, Calgary, Alberta. 49 pp. Peterson, J.T., R.F. Thurow, and J.W. Guzevich. 24. An evaluation of multipass electrofishing for estimating the abundance of stream dwelling salmonids. Transaction of the American Fisheries Society 133: Post, J.R., and F.D. Johnston. 22. Status of the bull trout (Salvelinus confluentus) in Alberta. Alberta Sustainable Resource Development, Fish and Wildlife 38

53 Division, and Alberta Conservation Association, Wildlife Status Report No. 39, Edmonton, Alberta. 4 pp. R Development Core Team. 25. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL project.org. Rhude, L.A., and J.D. Stelfox Status of bull trout in Alberta s fisheries management area three. Pages In: W.C. Mackay, M.K. Brewin and M. Monita, editors. Friends of the bull trout conference proceedings. Bull trout Task Force (Alberta), c/o Trout Unlimited Canada, Calgary. Scott, W.B., and E.J. Crossman Freshwater fishes of Canada. Fisheries Research Board of Canada Bulletin

54 6. APPENDICES Appendix 1. Fish captures by species for study sites in the Waiparous Creek drainage in early and late sampling periods, 26. Species codes: BLTR = bull trout; CTTR = cutthroat trout; BKTR = brook trout; BURB = burbot; MNWH = mountain whitefish; LNDC = longnose dace. Early Sampling Period Late Sampling Period Location BLTR CTTR BKTR BURB MNWH LNDC Location BLTR CTTR BKTR LNDC

55 Appendix 1. Continued. Early Sampling Period Late Sampling Period Location BLTR CTTR BKTR BURB MNWH LNDC Location BLTR CTTR BKTR LNDC NA NA 41

56 Appendix 1. Continued. Early Sampling Period Late Sampling Period Location BLTR CTTR BKTR BURB MNWH LNDC Location BLTR CTTR BKTR LNDC

57 Appendix 2. Summary of capture mark recapture of trout at population estimate sites. Species codes: BLTR = bull trout; CTTR = cutthroat trout; BKTR = brook trout. Waiparous Meadow Johnson Unnamed BLTR CTTR BKTR BLTR CTTR BKTR BLTR CTTR BKTR BLTR CTTR BKTR Marked at time Captured at time Captured at time 2 and marked at time

58 Appendix 3. Project location identification (ID), habitat assessment identification, geographic coordinates (UTM, NAD 83, Zone 11), mean wettedwidth, and mean stream depth for study sites in the Waiparous Creek drainage, 26. The habitat assessment ID uniquely identifies the early and late sampling inventory sites within each project location ID with habitat assessment ID between 1 68 denoting early sampling and denoting late sampling. Project location ID Activity date Habitat assessment ID Easting Northing Mean wettedwidth (m) Mean depth (m) May Sep May Sep May Sep May Sep May Sep May Sep May Sep May Sep Sep May Sep May Sep May Sep May Sep May Sep May May Sep May Sep