title
Stock assessment of arctic char in the Agulowak and Agulukpak rivers of the Wood River lake system, 1993
author
Array ( [0] => Minard, R. E. J. J. Hasbrouck )
abstract
Mark-recapture experiments conducted during the summer of 1993 were used to estimate abundance of Arctic char Salvelinus alpinus near the Agulowak and Agulukpak rivers within the Wood River lake system, Alaska. Estimated abundance of char at the Agulowak River was 5,441 fish (95% CI 4,993 to 5,950). This is substantially less than the estimated historic average of over 11,700 fish. Abundance estimates at the Agulukpak River ranged between 4,011 and 7,118 char, which brackets the estimated historic average of over 6,300 fish. Size composition at the Agulowak River was similar to that observed in 1976-1979, suggesting the mechanism causing the decline acted on all segments of the population. Size distribution of Arctic char at the Agulukpak River was slightly lower than those sampled in 1976-1978. Based on these results the Department of Fish and Game, Division of Sport Fish recommends reducing the bag limit for Arctic char at the Agulowak River and limiting terminal tackle to single-hook artificial lures only.
date
1994-11-01
organization
ADF&G Division of Sport Fish
species
Array ( [0] => Arctic Char )
file_path
https://grey-lit.s3.wasabisys.com/stock-assessment-of-arctic-char-in-the-agulowak-and-agulukpak-rivers-of-the-wood-river-lake-system-1.pdf
thumb
https://grey-lit.s3.wasabisys.com/stock-assessment-of-arctic-char-in-the-agulowak-and-agulukpak-rivers-of-the-wood-river-lake-system-1-pdf-2-793x1024.jpg
content
Fishery Data Series No. 94-42 Stock Assessment of Arctic Char in the Agulowak and Agulukpak Rivers of the Wood River Lake System, 1993 bY R. Eric Minard and James J. Hasbrouck November 1994 Alaska Department of Fish and Game Division of Sport Fish FISHERY DATA SERIES NO. 94-42 STOCK ASSESSMENT OF ARCTIC CHAR IN THE AGULOWAK AND AGULUKPAK RIVERS OF THE WOOD RIVER LAKE SYSTEM, 1993l BY R. Eric Minard and James J. Hasbrouck Alaska Department of Fish and Game Division of Sport Fish Anchorage, Alaska November 1994 1 This investigation was partially financed by the Federal Aid in Sport Fish Restoration Act (16 U.S.C. 777-777K) under Project F-10-9, Job No. R-2-9. The Fishery Data Series was established in 1987 for the publication of technically oriented results for a single project or group of closely related projects. Fishery Data Series reports are intended for fishery and other technical professionals. Distribution is to state and local publication distribution centers, libraries and individuals and, on request, to other libraries, agencies, and individuals. This publication has undergone editorial and peer review. The Alaska Department of Fish and Game administers all programs and activities free from discrimination on the basis of sex, color, race, religion, national origin, age, marital status, pregnancy, parenthood, or disability. For information on alternative formats available for this and other department publications, contact the department ADA Coordinator at (voice) 907-465-4120, or (TDD) 907-465-3646. Any person who believes s/he has been discriminated against should write to: ADFbG, P.O. Box 25526, Juneau, AK 99802-5526; or O.E.O., U.S. Department of the Interior, Washington, DC 20240. TABLE OF CONTENTS Page LIST OF TABLES.......................................................... ii LIST OF FIGURES......................................................... iii ABSTRACT................................................................ 1 INTRODUCTION............................................................ 2 STUDY OBJECTIVES........................................................ 2 METHODS................................................................. 2 Population Abundance .............................................. 2 Biological Sampling ............................................... 8 RESULTS................................................................. 9 Abundance Estimates ............................................... 9 Agulowak River ................................................ 9 Agulukpak River ............................................... 9 Biological Sampling ............................................... 14 DISCUSSION.............................................................. 18 ACKNOWLEDGMENTS......................................................... 22 LITERATURE CITED........................................................ 22 -i- LIST OF TABLES Table 1. 2. 3. 4. 5. 6. 7. 8. Historic estimates of abundance of Arctic char at the Agulowak River . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Historic estimates of abundance of Arctic char at the Agulukpak River . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Estimated sport harvest of Arctic char in the Wood River drainage from 1977-1992 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Summary of capture data of Arctic char during two sampling events at the Agulowak and Agulukpak rivers in 1993 . . . . . . . . . . . . . . . . . . . . . . . . 10 Number of Arctic char recaptured and not recaptured during the recapture event by period of the marking event at the Agulowak and Agulukpak rivers during 1993 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Number of Arctic char captured during the marking (M) and recapture (C) events, number of marked char recaptured (R) during the recapture event, and relevant statistics of the estimated abundance of Arctic char at the mouths of the Agulowak and Agulukpak rivers during 1993 estimated with an unstratified Petersen model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Estimate of condition index (K) and associated standard error for Arctic char at the mouths of the Agulowak and Agulukpak rivers during 1993 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Results of pairwise Kolmogorov-Smirnov tests comparing the cumulative length distribution of Arctic char between years at the Agulowak and Agulukpak rivers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 -ii- LIST OF FIGURES Figure Pane 1. The Wood River Lake system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2. Length distributions of Arctic char at the Agulowak River in 1993 . . . 11 3. Length distributions of Arctic char captured with hook and line and purse seine at the Agulukpak River in 1993 . . . . . . . . . . . . . . . . . . . . . . 15 4. Length distributions of Arctic char at the Agulukpak River in 1993 . . 16 5. Length distributions of Arctic char sampled at the Agulowak River 1976-1979 and 1993, and the Agulukpak River 1976-1978 and 1993 . . . . . . 20 -iii- -iv- ABSTRACT Mark-recapture experiments conducted during the summer of 1993 were used to estimate abundance of Arctic char Salvelinus alpinus near the Agulowak and Agulukpak rivers within the Wood River lake system, Alaska. Estimated abun- dance of char at the Agulowak River was 5,441 fish (95% CI 4,993 to 5,950). This is substantially less than the estimated historic average of over 11,700 fish. Abundance estimates at the Agulukpak River ranged between 4,011 and 7,118 char, which brackets the estimated historic average of over 6,300 fish. Size composition at the Agulowak River was similar to that observed in 1976- 1979, suggesting the mechanism causing the decline acted on all segments of the population. Size distribution of Arctic char at the Agulukpak River was slightly lower than those sampled in 1976-1978. Based on these results the Department of Fish and Game, Division of Sport Fish recommends reducing the bag limit for Arctic char at the Agulowak River and limiting terminal tackle to single-hook artificial lures only. KEY WORDS: Agulowak River, Agulukpak River, Arctic char, Salvelinus alpinus, Wood River lake system, mark-recapture experiment, size composi- tion. -l- INTRODUCTION Arctic char Salvelinus alpinus is an important gamefish to southwest Alaska's multimillion dollar sport fishing industry. Arctic char occur in most of the drainages of Bristol Bay but especially thrive in the Wood River system (Figure 1). The largest directed sport fishery for Arctic char in southwest Alaska occurs in the Wood River system at the mouths of the Agulowak and the Agulukpak rivers. Arctic char in the Wood River Lake system, particularly in the Agulowak and Agulukpak rivers, received considerable attention histori- cally because of their predation on sockeye salmon smolt. Each spring Arctic char concentrate at the mouths of these rivers to feed on outmigrating salmon smolt. From the 1920s through 1940 the federal and state governments sponsored lethal predator control programs that paid bounties on Arctic char tails. Predator control programs were terminated by 1941; however, a nonlethal program began in the mid 1970s. This program involved the temporary impound- ing of char from the mouths of the Agulowak and Agulukpak rivers to net pens during the smolt outmigration (Clark 1978). The program continued through 1980 with over 57,000 char sampled for length and, with additional studies, produced a number of population abundance estimates since 1954 (Tables 1 and 2). Recent estimates of sport harvest of char from the Wood River drainage have averaged 1,563 (Table 3). Most Arctic char taken in the Wood River Lake system are caught at the Agulowak and Agulukpak rivers (Minard 1989). Recent harvest estimates of Arctic char for the Agulowak River are over twice as large as estimates from 1977-1981 (Mills 1979-1982). To what extent the increased sport harvest affected the abundance and size composition of Arctic char in the Agulowak and Agulukpak rivers was unknown, prompting the Division of Sport Fish to conduct a mark-recapture stock assessment project. STUDY OBJECTIVES The objectives of this study were to: 1. estimate the abundance of Arctic char 2 250 mm in length feeding near the mouths of the Agulowak and Agulukpak rivers from 1 June to 8 July 1993; and 2. estimate the length distribution and the condition factor of Arctic char 2 250 mm in length near the mouths of the Agulowak and Agulukpak rivers each week from 1 June to 8 July 1993. METHODS Population Abundance Mark-recapture experiments were used to estimate the abundance of Arctic char 2 250 mm in length near the mouths of the Agulowak and Agulukpak rivers from 28 May to 8 July 1993. Char were captured with a 69 m by 11 m hand purse seine. Fishing was conducted during the twilight hours, with 8-20 seine sets -2- L - -L ynr Laka I I Lake Nerka I r-’ I WOOD-TIKCHIK -- STATE PARK GOUNDARY Y PACIFIC OCEAN t . \ WOOD RIVER Nunheask River; Figure 1. The Wood River Lake system. -3- Table 1. Historic estimates of abundance of Arctic char at the Agulowak River. Study Peak 95% Year Dates Estimate Confidence Interval Agency= Estimator 1954 6/11 - 9/8 11,297 1971 6/17 - 7/18 13,400 1972 6/22 - 7/2 12,402 1975 6/11 - 7/18 13,490 1976 5/26 - 7/26 11,062 1977 6/2 - 8/6 8,480 1978 6/l - 8/10 11,822 1979 5/22 - 7/27 15,627 1980 6/3 - 7/23 8,093 Average all years 11,741 not available 11,400 - 15,600 6,932 - 24,850 10,967 - 17,520 10,188 - 12,099 not available not available not available not available FRI FRI FRI ADFCG ADFdG ADF&G ADF&G ADFdG ADF6G Schnabelb Petersen' Petersend Petersene Schnabel= R/C Ratiof R/C Ratio' Petersenh Petersenh a FRI = Fisheries Research Institute, University of Washington ADFdG = Alaska Department of Fish and Game b Thompson et al. (1971). c Rogers et al. (1972). d Rogers (1972). e Meacham (1977a). f Meacham (1977b). ' Clark (1978). h Fried and Laner (1980). -f4- Table 2. Historic estimates of abundance of Arctic char at the Agulukpak River. Year Study Peak 95% Confidence Dates Estimate Interval Agencya Estimator 1972 6/19 - 7/25 7,262 5,847 - 9,628 FRI Petersenb 1976 6/16 - 7/9 7,824 5,976 - 12,484 ADF&G Petersen' 1977 6/16 - 7/10 5,836 not available ADFdG R/C Ratiod 1978 6/5 - 7/13 4,357 not available ADFdG R/C Ratio= Average all years 6,320 a FRI = Fisheries Research Institute, University of Washington ADF&G = Alaska Department of Fish and Game b Rogers (1972). c Meacham (1977a). d Meacham (1977b). e Clark (1978). -5- Table 3. Estimated sport harvest of Arctic char in the Wood River drainage from 1977- 1992.a Year Harvest Estimate 1977 435 1978 905 1979 685 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 Average 1988-1992 646 529 1,048 2,108 1,559 882 526 2,335 564 2,348 1,362 1,724 1,818 1,563 a Mills (1979-1993). -6- made on any given evening. Arctic char were also captured at the Agulukpak River using hook and line gear to supplement low seine catches. Capture and sampling of fish occurred at the Agulowak River from 28 May through 14 June (marking event) and from 28 June through 30 June (recapture event). Capture and sampling of fish occurred at the Agulukpak River from 17 June through 24 June (marking event) and from 6 July through 8 July (recapture event). Fish 2 250 mm were marked with an individually numbered Floy T-Anchor tag and the adipose fin clipped to assess tag loss. Fork length (snout to fork-of- tail) was measured to the nearest millimeter on all fish prior to release. Tag number was recorded for all recaptured char and fork length measured prior to release. Recaptured char with a finclip but with no tag present were recorded as a tag loss and marked with a new tag. All captured char were released near the point of capture. All tagging and fork length data were recorded on a Sport Fish Division Revised Tagging Mark-Sense Form Version 1.0 (Heineman 1991). We used the Chapman modified Petersen two-sample mark-recapture model (Seber 1982, page 59) to estimate abundance. Assumptions of the Petersen estimator are: 1. the population is closed, with no additions (recruitment or immigration) or losses (mortality or emigration) between sampling events; 2. marking does not affect capture probability during the recapture event; 3. all Arctic char 2 250 mm in length have an equal probability of capture during the marking event or the recapture event; or marked fish mix completely with unmarked fish prior to the recapture event; 4. 5. marks (tags) are not lost between sampling events; and all marked fish captured during the recapture event are identified and recorded. correctly The firs t assumption addresses the notion of closure. Though the two rivers were not closed to migration, and mortality or removal was possible, the assumption of closure was met by simply limiting the duration of the study period. It was also assumed that growth related recruitment did not occur. If char emigrate but no immigration occurs between sampling events, we assume marked and unmarked individuals left at the same rate. No correction is necessary and the estimate is of abundance during the first event. If char immigrate but no emigration occurs between sampling events, no correction is needed but the population estimate is for the second sampling event (Bernard and Hansen 1992). If both immigration and emigration occur between sampling events, no correction is possible and the population estimate is biased high. To determine if migration was a problem, we tested the null hypothesis that the probability of capture during the recapture event (i.e., recapture rate) -7- was equal among all time periods of the marking event. Time periods were composed of 2-3 sequential capture dates. The alternative hypothesis was that the probability of capture during the recapture event was different for at least one time period of the marking event. While both harvest and hook-and-release mortality may have occurred at both rivers, we assumed this mortality would equally affect marked and unmarked individuals. Additionally, the fish captured during the study were handled with utmost care to avoid injury or stress related mortality. The second assumption, addressing whether capture and handling during the marking event affects the probability of capture during the recapture event, could not be tested directly. However, results of tests examining violation of the third assumption provided indirect evidence of whether the second assumption was violated. Careful handling of the fish and fishing throughout the study area should have minimized problems of violating this assumption. The third assumption addresses the issue that all fish have an equal probability of capture during the study period. To determine if the probabil- ity of capture differed due to size (i.e., size selective bias of the capture gear), Kolmogorov-Smirnov (KS) tests were applied to length data collected from the marking and recapture events at each river. In one KS test, the null hypothesis was that the cumulative length distribution of all fish released during the marking event was equal to the cumulative length distribution of all fish captured during the recapture event. A second test evaluated the null hypothesis that the cumulative length distribution of all fish released during the marking event was equal to the length distribution of fish recap- tured during the recapture event. The fourth and fifth assumptions, dealing with lost tags and identification of marked fish, were addressed by employing a secondary mark: an adipose finclip. Each fish that was captured was carefully inspected for a tag and finclip. This process ensured that fish which lost tags were identified, and that all fish were adequately inspected to ensure their capture histories were accurately recorded. Biological Sampling To attain the desired precision for estimating the length distributions, a minimum of 130 char were sampled (Thompson 1987) from each river during each event. All captured char were measured for fork length and the first 26 unmarked char sampled each evening were weighed to the nearest 10 g. All length and weight data were recorded on a Sport Fish Division Revised Tagging Mark-Sense Form Version 1.0 (Heineman 1991). Mean length, weight, and their associated variances were calculated using normal procedures. The relationship of the natural log of weight as a func- tion of the natural log of length (condition) was estimated using standard regression procedures (Draper and Smith 1981). Analysis of covariance was used to test whether the regression slope (i.e., estimates of the condition index) and the intercept estimates differed among weeks at each river. Although Arctic char at the Agulowak and Agulukpak rivers were measured in previous studies (Baker 1988, Clark 1978, Fried and Laner 1980, Meacham 1977a -8- and 1977b), none of these studies determined if changes in size composition occurred among years. Since effort, capture methodology, and time of capture were similar among these studies, we tested whether changes in size composition occurred in these two populations among years. Length data were collected from Arctic char at the Agulukpak River during the summers of 1976- 1978 and at the Agulowak River during the summers of 1976-1979. The length distribution data were combined into 10 mm length categories. An Anderson- Darling K-sample test (Scholz and Stephens 1987) was used to test the hypothe- sis that the cumulative length distribution of char at each river is equal among years. If differences existed, a KS test was performed on each pairwise comparison to determine which years were different. RESULTS Abundance Estimates Agulowak River: During the marking event 1,545 unique live Arctic char 2 250 mm were captured and released (Table 4). Tag loss during the marking event was 3%. During the recapture event 919 unique char were captured. Of these fish 260 were recap- tures from the marking event: 228 with tags and 32 (12%) that lost their tags. We assumed tag loss occurred between events on char originally captured during the marking event. The cumulative length distribution of char captured and released during the marking event was significantly greater (ni = 1,543; n2 = 919; D = 0.0063; P = 0.02) than that of char captured during the recapture event (Figure 2). The cumulative length distribution of char recaptured during the recapture event was significantly greater (nl = 1,543; n2 = 260; D = 0.1354; P < 0.001) than that of char captured and released during the marking event. These results indicate size selective sampling, but generally when size selective gear bias occurs the cumulative length distributions cross each other. We concluded that large sample sizes and growth of recaptured fish caused these results and that the probability of capture was not affected by size selective sampling. There was no significant difference (x2 = 2.95, df = 3, P = 0.40) in recapture rate among four periods of the marking event (Table 5). borderline significant difference (x2 = Although there was a 15.34, df = 8, P = 0.05) in recapture rate among capture dates, there was no consistent pattern or trend. Based on these results, an unstratified Chapman modified Petersen model was used to estimate abundance of Arctic char at the Agulowak River. The estimated abundance was 5,441 (SE = 259) char (Table 6). Agulukpak River: During the marking event 1,020 unique char 2 250 mm were captured and released with tags (Table 4). Five fish of 126 recaptures (4%) lost their tag during the marking event. During the recapture event 250 unique char were captured. Of these fish, one fish (3%) lost its tag and 34 additional fish were recaptured with tags for a total of 35 recaptures. -9- Table 4. Summary of capture data of Arctic char during two sampling events at the Agulowak and Agulukpak rivers in 1993. Event Total Duplicate Unique Lost Tags Unique Records Tag Nos Captures Number Percent Recaps Agulowak River Marking Event 5/28 - 6/14 1,869 Recapture Event 6/28 - 6/30 935 324 16 1,545 919 9 3 32 12 228 Apulukpak River Marking Event 6/17 - 6/24 1,146 126 1,020 5 4 Recapture Event 7/06 - 7/08 254 4 250 1 3 34 -lO- 1 0.9 1 f 0.8 f 0.7 g 0.6 .- p 0.5 ; 0.4 'E + 0.3 5 0.2 0 0.1 0 ;:/ ‘/:h3elease (n = 1543) ,i’ ,./’ ,/’ 200 .,/ . .._... ....I. 1 300 400 500 600 700 Length 0.6 - 01 . . . . . -..-- 250 300 350 400 450 500 550 600 650 Length Figure 2. Length distributions of Arctic char at the Agulowak River in 1993. -ll- Table 5. Number of Arctic char recaptured and not recaptured during the recapture event by period of the marking event at the Agulowak and Agulukpak rivers during 1993. Marking Not Recapture Period Recaptured= Recaptured Total Rate Agulowak River 28 May - 2 June 27 151 178 0.152 3 June - 4 June 57 372 429 0.133 7 June - 9 June 68 424 492 0.138 11 June - 14 June 76 370 446 0.170 Agulukpak River 17 June - 19 June 7 179 186 0.038 20 June - 22 June 8 501 509 0.016 23 June - 24 June 19 306 325 0.058 a Recapture event occurred from 28-30 June at the Agulowak River and 6-8 July at the Agulukpak River. -12- Table 6. Number of Arctic char captured during the marking (M) and recapture (C) events, number of marked char recaptured (R) during the recapture event, and relevant statistics of the estimated abundance of Arctic char at the mouths of the Agulowak and Agulukpak rivers during 1993 estimated with an unstratified Petersen model. Statistic Agulowak Agulukpak River River M 1,543 1,030 C 919 250 R 260 35 N 5,441 7,118a Var(N) 67,292 1,131,794 SE(N) 259 1,064 95% Confidence Interval 4,993 - 5,950 5,032 - 9,203 95% Relative Precision 9% 29% a This estimate was modified to a range of 4,011-7,118 fish after additional testing and analyses. -13- Since fish were caught using both hook and line gear and the hand purse seine, a KS test was used to determine if the cumulative length distribution was different between the gear types. If no difference was detected then the data could be combined. There was no significant difference (nl = 1,067; n2 = 195; D = 0.0777; P = 0.26) in the cumulative length distribution between gear types (Figure 3), so we combined the data. The cumulative length distribution of char captured and released during the marking event was not significantly different (ni = 1,015; n2 = 248; D = 0.0603; P = 0.44) from that of all char captured during the recapture event (Figure 4). The cumulative length distri- bution of char recaptured during the recapture event was also not signifi- cantly different (ni = 1,015; n2 = 34; D = 0.1642; P = 0.32) than that of all char captured and released during the marking event. We concluded there was no size selective sampling at the Agulukpak River. Based on the results of the KS tests an unstratified Chapman modified Petersen model was used to estimate abundance. Abundance was estimated at 7,118 (SE = 1,064) char (Table 6). However, at the Agulukpak River there was a significant difference (x2 = 11.38, df = 2, P = 0.003) in recapture rate, primarily due to the low recapture rate of char marked during the middle period of the marking event (Table 5). Further analysis indicated the recapture rate was not different (x2 = 3.10, df = 1, P = 0.08) between the first two marking periods and that char captured during the first two marking periods had a significantly lower (x2 = 9.35, df = 1, P = 0.002) recapture rate than those captured during the last marking period. We concluded from these results and discussion with the field crew (Lew Coggins, School of Fisheries and Ocean Sciences, University of Alaska- Southeast, Juneau, personal communication), that both immigration and emigration occurred between sampling events at the Agulukpak River. This makes the Petersen estimate biased high (overestimate abundance) with an unknown degree of bias. We decided to estimate abundance with mark-recapture data collected only during the marking event using program CAPTURE (Otis et. al. 1978). This program fits different closed population models to mark- recapture data and, through a series of likelihood ratio tests, determines which model best fits the data. The models differ by relaxing in various ways the assumption that each individual has an equal probability of capture. The model that best fit the Arctic char data from the Agulukpak River allowed the probability of capture to differ over time (Chao 1989) and gave an abundance estimate of 4,011 (SE = 346) char. Biological Sampling Length and weight data of each river were divided into three groups (Table 7). These groups corresponded to the first half and the second half of the marking event, and the recapture event. Analysis of covariance detected no signifi- cant difference in the condition indices (i.e., estimated slope of natural log of body weight as a linear function of natural log of fork length) among the three groups at either the Agulowak (F = 0.80; df = 2, 330; P = 0.45) or the Agulukpak (F = 0.66; df = 2, 286; P = 0.52) rivers. Therefore, we combined the data of each river and tested the null hypothesis that condition was equal between the two rivers. Condition index at the Agulukpak River [K = 2.92; -14- 2 0.6 b 5 0.5 IL‘ 0.4 ? -g 0.3 i 0.2 s 0.1 0 Length Figure 3. Length distributions of Arctic char captured with hook and line and purse seine at the Agulukpak River in 1993. -15- 1 u E 0.9 $j 0.8 E E 0.7 2 0.6 Ii 9 0.5 ‘E m 5 0.4 E 3 0.3 0.2 0.1 0 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 . . ..N ,,,y .;;’ ,.;A ,’ : ;y ,/,,’ ‘I ;,,j; ;,g- Release (n = 1 ,015) ,.,‘,! l&74 ,/ Capture (n=248) / 2 2 /B j/’ ‘.’ / y’ // .y .--_ ._... ,-> ~_.-- 300 350 400 450 500 550 600 650 Length Recapture (n=34) 300 350 400 4ko 500 550 600 Sk0 Length Figure 4. Length distributions of Arctic char at the Agulukpak River in 1993. -16- Table 7. Estimate of condition index (K) and associated standard error for Arctic char at the mouths of the Agulowak and Agulukpak rivers during 1993. Time Period n K SE(K) Agulowak River 28 May - 4 June 130 2.60 0.13 7 June - 14 June 128 2.79 0.18 28 June - 30 June 78 2.87 0.14 Total 336 2.69 0.09 Agulukpak River 17 June - 20 June 104 3.03 0.13 22 June - 24 June 104 2.82 0.11 6 July - 8 July 84 2.87 0.17 Total 292 2.92 0.08 -17- SE(K) = 0.081 was nearly significantly greater (F = 3.61; df = 1, 624; P = 0.06) than that at the Agulowak River [K = 2.69; SE(K) = 0.091. When comparing the cumulative length distribution of each river with data collected in previous studies, the sample size at the Agulowak River exceeded the maximum limits of the software to conduct the Anderson-Darling tests. It was assumed that with 19,691 observations there would be a significant differ- ence among years and we conducted the pairwise comparisons between years. The length distributions fell into three groups: 1978, 1979 < 1993 < 1976, 1977 (Table 8). The cumulative length distributions appear to show no major difference among years (Figure SA). The length distribution of char at the Agulukpak River was significantly different (Th = 44.64, P < 0.001) among years. Based on the test results (Table 8) and a plot of the distributions (Figure 5B), the major conclusion is that char sampled in 1993 were smaller than those sampled in 1976-1978. DISCUSSION The abundance estimates from the Chapman modified Petersen model are biased if both immigration and emigration to these rivers occurred during the mark- recapture experiments. From observations made by the field crew (Lew Coggins, School of Fisheries and Ocean Sciences, University of Alaska-Southeast, Juneau, personal communication), it is likely immigration occurred at the Agulowak River during the marking event and perhaps between events. However, it is doubtful that emigration occurred to any great degree because: (a) The recapture rate of char marked the first day of the marking event (19%) was high relative to the overall recapture rate (15%). (b) Daily catches during the recapture event were as high as those during the latter portion of the marking event. Assuming equal effort was expended during these times, which is likely because the crew was familiar with the gear and the area, this implies the number of char did not decline between events. However, note that emigration could occur and abundance decline, but not to the point where it reduced catches. (c) If both immigration and emigration occurred, we would expect that char present and captured early during the marking event would emigrate sooner or at a greater rate, and thus have a different recapture rate, than char marked towards the end of the marking event (some of which, presumably, immigrated to the Agulowak River during the marking event). Although these arguments are weak, we concluded immigration occurred but emigration was minimal at worst. In this case the population estimate of 5,441 is unbiased but reflects the abundance during the second event. This is the lowest estimate of abundance of Arctic char at the Agulowak River (Table 1). Both immigration and emigration likely occurred at the Agulukpak River. It is likely immigration occurred based on discussion with the field crew and -18- Table 8. Results of pairwise Kolmogorov-Smirnov tests comparing the cumulative length distribution of Arctic char between years at the Agulowak and Agulukpak rivers. Year nb 1977 Year Tested Against= 1978 1979= 1993 Apulowak 1976 2,608 NSD 78 << 76 79 -c-c 76 93 < 76 1977 5,710 78 << 77 79 << 77 93 << 77 1978 7,061 NSD 78 << 93 1979 1,510 79 < 93 1993 2,802 Aaulukpak 1976 2,816 NSD NSD 93 << 76 1977 2,429 78 < 77 93 << 77 1978 1,890 93 << 78 1993 1,197 a NSD = No significant difference at P > 0.01; Year X < Year Y = Length distribution of Year X significantly smaller than that of Year Y at 0.01 2 P > 0.001; Year X << Year Y = Length distribution of Year X significantly smaller than that of Year Y at P 5 0.001. b Number of char sampled for length. c Only the Agulowak River was sampled in 1979. -19- s s 100 1 A. AGULOWi 80 t 60 +I978 40 -8-1979 -1993 20 0 250 300 350 400 450 500 550 600 650 700 LENGTH (MM) 100 B. AGULUKPAK 80 60 O+-- I t- +I976 a977 +I978 f 1993 250 300 350 400 450 500 550 600 650 700 LENGTH(MM) Figure 5. Length distributions of Arctic char sampled at the Agulowak River 1976-1979 and 1993, and the Agulukpak River 1976-1978 and 1993. -2o- because all of the abundance estimates from CAPTURE, based solely on data from the marking event, were at least 3,000 char lower than the Petersen estimate using all of the data. It also appeared that emigration occurred because recapture rates differed among char marked "early" vs. "late" of the marking event. It is possible that char captured early during the marking event were emigrating at a greater rate or began emigrating earlier than those captured later in the marking event. Daily catches were also lower during the recapture event than during the latter portion of the marking event, but the field crew spent much time sorting adult sockeye salmon from the seine net during the recapture event. The crew believed the large numbers of adult sockeye salmon perhaps caused Arctic char to leave the area. Because it seems both immigration and emigration occurred, the Petersen estimate of 7,118 is biased high with an unknown degree of bias. The population estimate of 4,011 from CAPTURE is of limited value. The estimate is valid only for the 8-day period of the marking event (17-24 June), and if immigration actually occurred this estimate is biased low with an unknown degree of bias. Although bias makes both estimates at the Agulukpak River unsatisfactory, the population was likely between 4,000-7,000 char. This range of values is in line with previous estimates. Results of the condition factor analysis were a bit surprising. Initially it was thought that the condition index would increase over time as the char gained weight while eating sockeye smolt. In retrospect, char likely also increase in length, so that although char grow and improve their "condition," the relationship between natural log length and natural log weight did not change significantly (at least statistically). It is also possible that these results provide another indication that immigration occurred during the mark- recapture experiments. The cumulative length distribution of Arctic char in 1993 at the Agulowak River is within those distributions seen historically, but char at the Agulukpak River are smaller on average than those in the late 1970s. There are several possible explanations for these results, especially at the Agulukpak River. It is possible that large char were overexploited or more susceptible to hook and line mortality, or perhaps a large cohort of younger, smaller fish is dominating the population and causing the size distribution to shift. Given these results, it appears that the population of Arctic char feeding at the Agulukpak River is stable relative to historic levels. The minor shift in size composition is not alarming at this point. Regulations governing this fishery appear to provide adequate protection and no changes are recommended. Abundance of Arctic char feeding at the Agulowak River has dropped by half of that previously reported. The fact that size composition has not changed from the historic levels suggests the decline has been across all segments of the population. Sport harvest of Arctic char in the Wood River lakes system averaged 1,500 fish annually since 1988 and was as high as 2,348 fish in 1988 (Mills 1989). The vast majority of this harvest comes from the Agulowak River (Minard 1989). Given an abundance of 5,441 char at the Agulowak and a likely average harvest of 1,350 char per year, it is conceivable that exploitation on this stock exceeds 20% annually. In addition to a relatively high level of exploitation, there is a very high degree of catch and release fishing -21- directed toward Arctic char in the Agulowak River. Catch and release mortality could also be contributing to the decline. During September 1993, we conducted mark-recapture experiments on prespawning populations of Arctic char at Ice, Youth, and Sunshine creeks. These creeks are major spawning areas for Arctic char (McBride 1979) and, like the Agulowak River, all three creeks flow into Lake Aleknagik. We attempted to estimate abundance of Arctic char at these three creeks to determine if the decline in abundance at the Agulowak River was caused by a decline in the spawning component of the Arctic char population in Lake Aleknagik, or by a decline in Arctic char throughout the rest of the lake. Unfortunately, poor weather and high water did not allow us to capture and mark sufficient numbers of fish to get an abundance estimate at any of the creeks. Given the apparent decline of Arctic char at the Agulowak River, the Division of Sport Fish proposed that the Alaska Board of Fisheries take action to allow this stock to recover. Regulatory changes that reduce harvest and the potential effects of catch-and-release mortality were recommended. Specifi- cally, bag and possession limits should be reduced from ten fish to two fish per day, and terminal tackle limited to single hook artificial lures, to reverse the apparent decline in char abundance. Periodic monitoring of the Arctic char populations of both the Agulowak and Agulukpak rivers should be planned to evaluate the effect of these regulation changes. ACKNOWLEDGMENTS We wish to thank Lew Coggins, who supervised the field crew that collected the data, and field crew members: David Wightman, Arron Rogers, Brandon Cherry, and Carl Rutz for their dedication and skill which made this project possible. We also appreciate the efforts of Sandy Sonnichsen in editing the data and performing preliminary analyses, and the support and advice of Doug McBride. LITERATURE CITED Baker, T. T. 1988. Age and growth of Arctic char in the Wood River system, Alaska. Master's thesis, University of Alaska-Southeast, Juneau. Bernard, D. R. and P. A. Hansen. 1992. Mark-recapture experiments to estimate the abundance of fish. Alaska Department of Fish and Game, Division of Sport Fish, Special Publication No. 92-4, Anchorage. Clark, J. H. 1978. Agulowak and Agulukpak Arctic char studies-1978. Wood River char predation control memo series. Alaska Department of Fish and Game, Division of Commercial Fisheries. Division of Sport Fish Archive, Dillingham. Chao, A. 1989. Estimating population size for sparse data in capture- recapture experiments. Biometrics 45:427-438. Draper, N. R. and H. Smith. 1981. Applied regression analysis, second edition. John Wiley and Sons, New York. -22- LITERATURE CITED (Continued) Fried, S. M. and J. J. Laner. 1980. 1979 and 1980 Wood River system char impoundment studies. Alaska Department of Fish and Game, Division of Fisheries Rehabilitation, Enhancement and Development. Division of Sport Fish Archive, Dillingham. Heineman, G. H. 1991. Instructions for using sport fish creel survey and biological mark-sense forms, 1991. Alaska Department of Fish and Game, Sport Fish Division, Research and Technical Services Section, Anchorage. Meacham, C. P. 1977a. Arctic char predation assessment and control investigations within the Wood River system, Alaska, 1975 and 1976. Alaska Department of Fish and Game, Division of Commercial Fisheries, Research Section, Anchorage. ---.---- 1977b. Agulowak River Arctic char studies-1977. Wood River char predation control memo series. Alaska Department of Fish and Game, Division of Commercial Fisheries. Division of Sport Fish Archive, Dillingham. McBride, D. N. 1979. Homing of Arctic char to feeding and spawning sites . In Master's thesis, University of Alaska-Southeast, Juneau. Mills, M. J. 1979. Alaska statewide sport fish harvest studies. Alaska Department of Fish and Game, Federal Aid in Fish Restoration, Annual Report of Progress, 1978-1979, Project F-9-11, 20 (SW-I-A), Juneau. -* 1980. Alaska statewide sport fish harvest studies. Alaska Department of Fish and Game, Federal Aid in Fish Restoration, Annual Report of Progress, 1979-1980, Project F-9-12, 21 (SW-I-A), Juneau. -- 1981a. Alaska statewide sport fish harvest studies (1979). Alaska Department of Fish and Game, Federal Aid in Fish Restoration, Annual Report of Progress, 1980-1981, Project F-9-13, 22 (SW-I-A), Juneau. -' 1981b. Alaska statewide sport fish harvest studies (1980). Alaska Department of Fish and Game, Federal Aid in Fish Restoration, Annual Report of Progress, 1980-1981, Project F-9-13, 23 (SW-I-A), Juneau. -* 1982. Alaska statewide sport fish harvest studies (1981). Alaska Department of Fish and Game, Federal Aid in Fish Restoration, Annual Report of Progress, 1981-1982, Project F-9-14, 24 (SW-I-A), Juneau. -* 1983. Alaska statewide sport fish harvest studies (1982). Alaska Department of Fish and Game, Federal Aid in Fish Restoration, Annual Report of Progress, 1982-1983, Project F-9-15, 25 (SW-I-A), Juneau. -* 1984. Alaska statewide sport fish harvest studies (1983). Alaska Department of Fish and Game, Federal Aid in Fish Restoration, Annual Report of Progress, 1983-1984, Project F-9-16, 26 (SW-I-A), Juneau. -23- LITERATURE CITED (Continued) -* 1985. Alaska statewide sport fish harvest studies (1984). Alaska Department of Fish and Game, Federal Aid in Fish Restoration, Annual Report of Progress, 1984-1985, Project F-9-17, 26 (SW-I-A), Juneau. -* 1986. Alaska statewide sport fish harvest studies (1985). Alaska Department of Fish and Game, Federal Aid in Fish Restoration, Annual Report of Progress, 1985-1986, Project F-10-1, 27 CRT-21, Juneau. -* 1987. Alaska statewide sport fisheries harvest report. 1986 data. Alaska Department of Fish and Game, Fishery Data Series No. 2, Juneau. -- 1988. Alaska statewide sport fisheries harvest report. 1987 data. Alaska Department of Fish and Game, Fishery Data Series No. 52, Juneau. -* 1989. Alaska statewide sport fisheries harvest report. 1988 data. Alaska Department of Fish and Game, Fishery Data Series No. 122, Juneau. -- 1990. Harvest and participation in Alaska sport fisheries during 1989. Alaska Department of Fish and Game, Fishery Data Series No. 90- 44, Anchorage. -* 1991. Harvest, catch, and participation in Alaska sport fisheries during 1990. Alaska Department of Fish and Game, Fishery Data Series No. 91-58, Anchorage. -* 1992. Harvest, catch, and participation in Alaska sport fisheries during 1991. Alaska Department of Fish and Game, Fishery Data Series No. 92-40, Anchorage. -* 1993. Harvest, catch and participation in Alaska sport fisheries during 1992. Alaska Department of Fish and Game, Fishery Data Series No. 93-42, Anchorage. Minard, R. E. 1989. Effort, catch, and harvest statistics, for the sport fisheries on the Agulukpak and Agulowak Rivers, Wood River Lakes system, Alaska, 1986-1988. Alaska Department of Fish and Game, Fishery Data Series No. 90, Juneau. Otis, D. L., K. P. Burnham, G. C. White, and D. R. Anderson. 1978. Statistical inference from capture data on closed animal populations. Wildlife Monograph No. 62. Rogers, D. E. 1972. Alaskan salmon studies: Part A. Study of red salmon in the Nushagak district. Fisheries Research Institute, University of Washington. NOAA-NMFS, Periodic Report No. 1, Seattle. Rogers, D. E., L. Gilbertson, and D. Eggers. 1972. Predator-prey relationship between Arctic char and sockeye salmon smolts at the Agulowak River, Lake Aleknagik, in 1971. Fisheries Research Institute, University of Washington, Circular No. 72-7, Seattle. -24- LITERATURE CITED (Continued) Scholz, F. W. and M. A. Stephens. 1987. K-sample Anderson-Darling test. Journal of the American Statistical Association 82:918-924. Seber, G. A. F. 1982. The estimation of animal abundance, second edition. MacMillan Publishing Company, New York. Thompson, R. B., C. R. Weaver, and W. D. Gronlund. 1971. A study of the role of Arctic char, Salvelinus alpinus, as a salmon predator in the Wood River lakes, Alaska, with notes on its life history. National Marine Fisheries Service Manuscript Report File No. 96, Auke Bay, Alaska. Thompson, S. K. 1987. Sample size for estimating multinomial proportions. The American Statistician 41:42-46. -25- . TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES ABSTRACT INTRODUCTION STUDY OBJECTIVES METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS LITERATURE CITED
geography
Array ( [0] => Array ( [state] => Alaska [province] => ) )
batch_date
Wed, 06 May 2020 18:53:35 +0000
original_file_path
http://www.adfg.alaska.gov/FedAidPDFs/fds94-42.pdf
group__id
5eb3087fb0cb20.66466719
document_format
1.1.7
disabled
-1
processing_date
1609350751
year
1994
hash
Array ( [document] => e0caeee1e2d45c93f59aeb3cd0367f417ac68d3c [content] => f1bdd440d0d092b121a78c1a0e97ad4de5a1daf0 [dedupe] => 9b7ed6ea553135ab88b1dd2ec5498da2a5f1230d )
id
doc-5eb3087f82ecd3be87026338
__hash__
Content: f1bdd440d0d092b121a78c1a0e97ad4de5a1daf0 | Abstract: 3bf503d10bf61fdd5bd069e463bc3b066af62224