SOCKEYE SALMON INVESTI(;ATIONS AND LIMNOL( )(;ICAL SAMPLIN(; ON

TUSTUMENA LAKE: 1995 ANNUAL REPORT

G. L. Todd G. B. Kyle

REGIONAL INFORMATION REPORT' NO. 5596-06

Alaska Department of Fish and (' ~ a m e Division of Commercial Fisheries Manzgernent 2nd De~e!opment

P.O. Bou 25526 Juneau, Alaska 99802-5526

February 1996

AUTHORS

Gary L. Todd is the project biologist for the Tustumena Lake sockeye salmon investigations for the Alaska Depart~nent of Fish and Game, Lirnnology Unit, Division of Commercial Fisheries Management and Development, 34828 Kalifomsky Beach Road, Suite B, Soldotna Ak 99669.

Gary B. Kyle is the south-central Regional Lirnnologist for the Alaska Department of Fish and Gane. Lirnnology Unit, Division of Commercial Fisheries Management and Development, 34828 Kalifomsky Beach Road, Suite B, Soldotna Ak 99669.

ACKNOWLEDGMENTS

We thank Ken Tarbox (ADF&G) and his staf€ for adult salmon fishery information, and Pat Shields, Al Helminski, Richard Dederick, and Shawn Rogers (ADF&G) for assisting in data collection. Fundine for this project was provided by the Exvon Valdez Oil Spill Trustee Council.

TABLE OF CONTENTS

Section

ABSTRACT ........................................................................................................................... 1

7 .................................................................................................................. INTRODUCTION ....- Study Site Description ........................................................................................................ 2

................................................................................................................................. METHODS 3 . .

Adult Hatchery Contrlbutlon .............................................................................................. 3 Hydroacoustic~Townet Surveys ......................................................................................... 3 S~nol t Enumeration and Sampling ...................................................................................... 4

........................................................................................................ Lirnnological Surveys 5

RESULTS AND DISCUSSION .................................................................................................. 5 . . ............................................................................................. Adult Hatchery Contr~but~on 5

Fall Fry Abundance. Size. and Age ................................................................................... 6 ...................................................................................... Srnolt Abundance. Size. and Age 6

....................................................................................................... Li~nnological Surveys 7

LITERATURE CITED ............................................................................................................... 8

LIST OF TABLES

Table

I . Estimated number and percentage of hatchery-produced sockeye salmon in the escapement and total return for Tustumena Lake in 1995, and comparison to the mean for 1984- 1994.

2. Densities and population estimates of juvenile fish rearing in Tustumena Lake by transect based on the 22 September 1995 hydroacoustic survey, and comparison to the mean estimate for 198 1 - 1994.

3. Mean size and age composition of juvenile sockeye salmon fry caught in surface net tows by basin during 1995 in Tustumena Lake, and comparison to the means for 1980- 1994.

4. Daily population estimates of sockeye salmon smolts emigrating Tusturnena Lake, 1995.

5. Size and age of Tustumena Lake sockeye salmon smolts by sample period, 1995.

6. Size and age of Tusturnena Lake sockeye salmon smolts that were dyed and recaptured by sample period, 1995.

7. Population estimate (in thousands) of sockeye salmon smolts emigrating Tustumena Lake by age and sample period, 1995.

X. Lirnnolo~ical characreristics within the epilinmion (1 rn) of Tustumena Lake during 1995 by basin, and comparison to the means for 1980-1994.

9. Summary of zooplankton seasonal mean density and biomass within Tustumena Lake during 1995, and comparison to the means for 1980- 1994.

Pace -

10

LIST OF FIGURES

Figure

1. hlol-pho~netric map of Tustumena Lake showing the location of lirnnolo~ical sample stations and the nine transects used during the 22 September 1995 hydroacoustic survey.

2. Map of Tustumena Lake showing the distribution (density) of juvenile sockeye salmon for the 22 September 1995 hydroacoustic survey.

3. Juvenile sockeye salmon distribution by depth stratum for the three basins in Tustumena Lake based on the 22 September 1995 hydroacoustic survey.

4. Migration. timing for sockeye salmon smolts emigrating Tustumena Lake, 1995.

5. Density of Diuptonzus by sample date and station in Tustumena Lake, 1995.

6. Density of Cyclops by sample date and station in Tustumena Lake, 1995.

ABSTRACT

The I995 adult sockeye salmon Oncorhynch~rs rlerka escapement into Tustumena Lake based on sonar counts in the Kasilof River was estimated at 205,903. The hatchery contribution of adult sockeye salmon in the escapement for 1995 was 10,256 (5%). An estimated 3 1,197 hatchery-produced fish contributed to the total Tustumena Lake sockeye salmon return in 1995.

A total of 10.5 million + 2.0 million sockeye salrnon juveniles were estimated in Tustu~nena Lake based on a hydroacoustic/townet survey conducted in September of 1995. The age composition of sockeye salmon fiy collected was 84.6% age-0 and 15.4% age- 1. The mean length of age-0 fiy was 64 mrq which was larger than all previous years since 1980, and the mean weight was 2.8 g, which was larger than all previous years except for 1993.

The estimated number of sockeye salrnon smolts that emigrated Tustumena Lake in 1995 was 7.3 million + 2.2 million. The dady peak emigration occurred on 8 June when an estimated 1.6 million srnolts emigrated. The smolt population was comprised of 4.8 million (66.1%) age-1 smolts and 2.5 million (33.9%) age-2 smolts. The mean size of age-1 and age-2 smolts was 71.5 rnm and 2.8 g, and 84.2 mrn and 4.3 g, respectively.

L

Limnological surveys in 1995 indicated a higher seasonal mean water temperature, lower concentrations of key nutrients (phosphorus and nitrogen), and lower algal biomass (chlorophyll U J

than the means for 1980-1994. The euphotic zone depth was greater at all three stations in 1995 than the 1980-1994 means. The density of Diaptom~ls in 1995 was 16% geater than the mean during 1980- 1994, while Cyclops density was 40% lower compared to the 1980- 1994 means.

KEY WORDS: sockeye, f?y stocking, smolt and adult production, glacial lake, limnology

INTRODUCTION

The Tusturnena Lake s0ckej.e sahnon Otzcot-hytzch~~s tzerlicl investigation project was initiated in 198 1 to assess juvenile and adult sockeye salmon populations relative to the efFects of stocking hatchery fry. The primary goals of this project are to determine in-lake factors affecting productivity and to assess the s t o c h g program relative to averting negative impacts to the natural production of sockeye salmon.

htially, the working hypothesis of sockeye salmon production in Tustumena Lake was that this system operated in a typical density-dependent fashion, where the variability in smolt production (juvenile- rearing success) is related to fluctuations in freshwater trophic levels (Foerster 1944; Johnson 1965: Brocksen et al. 1970; Koenings and Burkett 19x7; Kyle et al. 1988). However, if juvenile fish density is not sufficient to challenge the lirnnetic forage base, and trophic-level responses become uncoupled from fLsh density, observed trophic changes may be correlated to environmental variables (density- independent). In Tustumena Lake, Kyle (1992) reported that trophlc-level responses were inconsktenr with unconditional density-dependent sockeye salrnon production. A prelirmnary analysis in Tustu~nena Lake suggests that sockeye sahnon production is density-independent, as environmentally dependent variables were responsible for a majority of the annual variation in the production of ~5 ild age- 1 smolts (Koenings et al. 1988).

This report summarizes hheries and limnological data collected during 1995. This infor~nation B used to assess juvenile and adult sockeye salmon production and changes in lirnnolo~ical parameters in Tustumena Lake for comparison to Skilak and Kenai lakes ( S c h d t et al. 1995). Sockeye salrnon

"0s were investigations in Tustumena Lake have been on-going since 1980. In 1974 sockeye sahnon e,, first taken from Glacier Flats Creek. Since then, sockeye salmon eggs have been taken each year f om Glacier Flats Creek, Bear Creek, or both, and fry have been stocked in Tustumena Lake (with the exception of 1994. when the fry were infected with infectious hernatopoietic necrosis (THN virus and destroyed) at levels ranging from 6-17 million. Previous years of information on thks project are reported in Ragg et al. (1984; 1987), federal-aid (Anadro~nous Fish Conservation) annual reports ( 1987- 1992) by various authors, a summary report for years 1980- 199 1 by Kyle (1 992), and annual reports by Todd and Kyle (1994, 1995).

Study Site Description

Tustumena Lake (60' 10' N, 150" 55' W) is located on the Kenai Peninsula, approximately 25 lun south of Soldotna, Alaska. This lake has a surface area of 294.5 km2 (73,942 acres), a mean and muirnum depth of 24 m and 320 m, and is approximately 40-km long and 8-km wide (Figure 1). Tustu~nena Lake is fed by clearwater creeks as well as glacial creeks that originate in the Harding Icefield. The lake outlet (Kasilof River) drains into Cook Inlet over a watershed area of 1,376 km'. Tustumena Lake lies within the Kenai National Wildlife Refuge and project activities are regulated under a special use pennit issued each year to the Alaska Department of Fish and Game (ADF&G) by the United States Fish and Wildlife Service (FWS). The Cook Inlet Aquaculture Association (CIAA) collects eggs at Bear Creek on Tusturnena Lake for the fry stocking project in Tustumena Lake and for stocking fiy in other lakes in Cook Inlet.

All five species of Pacific salrnon are found in the Tusturnena Lake system though sockeye sahnon predo~ninate. Sockeye ssalrnon escapements have been estimated in the Kasilof River by sonar since 1968. Annual escapements have ranged from a low of 40,000 in 1973 to a high of 505,000 in 1985 (Ruesch and Fox 1995). The estimated commercial fishing exploitation rate of sockeye salmon bound for the Kasilof River has ranzed from 50% to 85% (Ruesch and Fox 1995). Sockeye salmon returning to the Tustu~nena Lake system are a major contributor to the total Cook Met sockeye salmon harvest.

METHODS

Adult Hatchery Contribution

The contribution of hatchery-produced fish to the Tusturnena Lake sockeye salmon escapement in 1995 was estimated from respective hatchery smolt data and the corresponding return of adults by age class. The adult sockeye salmon a,oe composition is determined fiom data collected at the Kasilof River sonar/fishwheel site, operated each year by ADF&G. As not all hatchery-released fingerlings were marked, the total number of hatchery-produced sockeye salmon in the escapement was computed from: 1) the corresponding years of release, 2) the number of fingerlings released, and 3) the number of fingerlings marked. The contribution rate of hatchery-produced sockeye salmon in the total return (escapement and harvest) was determined by making the assumption that the same ratio of hatchery-to-wild fish estimated in the escapement also occurred in the commercial, personal-use, and sport catches of Tusturnena Lake sockeye salmon. No spawning escapement surveys were conducted in 1395, but CIA4 conducted an escapement count during the egg take at Bear Creek (Fandrei 1995).

HydroacousticlTo wnet Surveys

A hydroacoustic survey was conducted on 22 September 1995 to estimate the number and distribution of sockeye saIrnon fingerlings rearing in Tusturnena Lake. The survey c o q n s e d of recording data along nine transects perpendicular to the longitudinal axis of the lake (Figure 1). The lake was divided into three areas, and three transects per area were selected randomly. Recording of down-looking acoustic data along the transects was done at night because juvenile sockeye salmon in this lake are Inore dispersed during darkness. A ~ a ~ e l l a n " NAV 5000A global positioning system (GPS) receiver was used to maintain transect course. A BioSonics model-105 echosounder system with a 6/15" dual- bean transducer was used for the survey. Fish signals were recorded electronically using a Sony digital audio tape recording system and on paper using a BioSonics model- 115 chart recorder.

Analysis of the recorded hydroacoustic tapes was conducted by Dr. Richard Thorne of BioSonics, Inc., under a State of Alaska contract. Fish densities were low enough to allow the use of echo-counting techniques (Thorne 1983) for the population estimate. The number of echoes from fish targets were counted in l0-minute increments along the nine transects and in five depth intervals. Sampling volumes were estimated by the duration-in-beam technique (Nunnallee and Mathisen 1972; Nunnallee 1980; Thorne 1988). For each depth interval and 10-min increment, fish densities (no./m3) were summed to determine the total fish density (no./m2) for each transect. Mean fkh densities were weighted by time, since end-of-transect increments were usually less than 10 minutes.

"Mention of commercial products 'and trade names does not constitute endorsement by the ADF&G, Commercial Fisheries Management ,and Development (CFND) Division.

A weighted mean fish density and an associated variance was co~nputed from the three transects per area. and a weighted fish population estimate for each of the three areas was computed. The total lake population estimate and variance was obtained by adding the weighted h h population estimates and variances for the three areas. The 9 5 4 confidence interval is reported as fi k 2.45(S. E. ) , where :'? is the fish population estimate, and 2.45 is the t-value for p = 0.05 for eight degrees of freedom.

Townetting was conducted in conjunction with the hydroacoustic surveys to determine species of acoustically-counted fish, and to determine age and size of juvenile sockeye salmon. Townetting procedures consisted of using a 3 x 3-m net pulled at a speed of approximately 1 .O- 1.25 d s e c . As the fish in Tustumena Lake are located relatively near the surface during both night and day, surface net tows were conducted during the day on 26 September 1995. AU captured juvenile sockeye salmon were preserved in 10% buffered formalin, and later sampled for individual snout-to-fork lengths (nearest 1.0 m n ) and weights (nearest 0.1 g). A scale smear was taken from each fish, and ages were determined with the use of a microfiche reader.

Smolt Enumeration and Sampling

One inclined-phne smolt trap (Todd 1994) was placed in ;he Kisiiof River on 8 May and operated through 30 June 1995. The trap measured 1.5 m wide by 1 m in height, and was 3.8 m long. Each day, captured smolts were either individually enumerated and released, or when the daily srnoit numbers exceeded 7,000, the number of smolts were estimated using a biomass technique. T L procedure entailed weishing 50 individual sockeye salmon smolts every day to obtain a mean srnolt weight, which was divided by the total weight of the sockeye salmon smolts cauzht to estllnate the number of smolts.

The population estimate of sockeye salmon smolts migrating frorn Tustumena Lake was made using weekly trap efficiencies determined throush a mark-and-recapture procedure (Rawson 19x3). Durins each week in May and June 900-2,000 sockeye salmon smolts were dyed and released upstream of the traps to estimate trap efficiency. The enumerated smolts were placed in an aerated transport tank containing a dye solution (1 g of Bismarck Brown Y dye to 30 L of water), and remained in the dye for 30 minutes during transport to the release location, approximately 2 lun upstream. The dyed smolts were placed in a live box in the river to recover and assess handling mortality prior to release. Dyed smolts recaptured in the trap were enumerated, sampled and released. Weekly migrations were estimated from the recovery of dyed smolts to determine trap catch efficiency. The percentage of the total migration comprising age-1 and age-2 smolts was estimated for each weekly period using scales obtained from samples collected that week. Ths percentage was then applied to the estimated total smolt migration for the weekly period to obtain estimates of the number of migrating smolts for each age class (Flagg et al. 1984).

Every other day through the peak of the smolt emigration, a random sample of 50 sockeye salmon smolts was sampled for age and size (AWL sample). After the emigration peaked, the smolts were sampled twice weekly. The smolts were anesthetized in a solution of MS-222 and measured for snout- to-fork length (nearest millimeter) and weight (nearest 0.1 g). A scale smear was taken from each of the sampled smolts and read with the use of a microfiche reader to determine age. The size and age of recaptured dyed smolts were recorded and compared to the size of smolts that were randomly sampled (not dyed and recaptured) to assess m y bias in these data.

In 1994 no hatchery reared fry were released into Tustumena Lake because fry at Crooked Creek hatchery were diagnosed with the EN virus and were destroyed. Consequently, the smolts in 1995 were not exarnined for hatchery marks; some age-2 smolts were of hatchery origin, but because of the anticipated anall number of recovered ~narked fish based on previous years, it was not cost effective to examine the population.

Limnological Surveys

Lunnology sampling was conducted once every three weeks during the ice-free period at stations representing each basin of the lake (Figure 1). Chemical and biological samples were analyzed according to standard limnology procedures (Koenings et aL 1987). Dissolved oxygen (D.O.) and temperature profiles (taken at 1 m intervals from the surface to 5 m, then at 5-m intervals to 50 m or the bottom of station A) were recorded at each station using a Yellow Springs Inc. model-57 temperatureD.0. meter. Water samples for nutrient and general water-quality analysis were taken from depths of 1 m and 50 m at stations B and C, and 1 m and 25 m at station A. The 1% incident li,oht level (euphotic zone depth) was measured using an International Light submersible photometer. Zooplankton were collected with a 0.5-m diameter zooplankton net of 153 Jtm mesh down to 50 m at stations B and C and -30 m at station A. Zooplankton from each station were preserved in a 10%- buffered formalin solution for species identzcation, enumeration, and sizing (Koenings et al. 1987). Two additional zooplankton stations were sampled on each sarnple trip; one midway between stations '4 and B (station D), and one midway between stations B and C (station E), down to 50 m each.

RESULTS AND DISCUSSION

Adult Hatchery Contribution

In 1995, the estimated number (and percentage) of hatchery fish in the sockeye salmon escapement was 10,256 (5%), which was lower than the mean of 60,328 (25%) for 1984-1994 (Table 1). The estimated commercial harvest of Tustumena Lake sockeye sdmon was 393,644 in 1995 (Randall Davis. ADF&G, Soldotna, personal communication). Using a hatchery contribution rate of 5%, an estimated 19,607 hatchery-produced sockeye sahnon were harvested in the commercial fishery (Table 1). In addition, an estimated 1,333 hatchery-produced sockeye salmon were caught in the personal- use, subsistence, and sport fisheries in 1995 (Jeff Fox, ADF&G, Soldotna, personal communication). Thus, of the total return of sockeye salmon to Tustumena Lake in 1995, an estimated 3 1,197 fish were hatchery-produced, which is similar to 1994 when 3 1,236 hatchery-produced fish were estimated in the total return.

The relatively low return of hatchery-produced sockeye salrnon in 1995 in part can be attributed to a lowered stocking level (beginning in 1988); however, apparently other factors are involved. That is, using the hstorical average fry-to-smolt and smolt-to-adult survivals suggests that from a 6 million fry stocking level about 135,000 adults should be produced. Other factors affecting hatchery adult production include excessive fry transport mortality and/or underestimation of hatchery production. A dramatic decline in the contribution of smolts produced from the f fy stocking %st occurred in 1990. In the fry transports of 1989 and 1990, the f?y appeared to be stressed due to warm ambient water temperature during transport. Thus, some of the hatchery-produced adults (age class 2.3) that returned in 1995 were exposed to stress conditions during the 1990 f fy transport. The procedure for

estimating hatchery smolt production has been followed since the inception of the smolt ~nonitoring progan, so no obvioub procedure errors are evident that could have affected the recent srnolt estimates. Since 1990, hatchery adult production as been based on the contribution percentage of hatchery smolts and the corresponding escapement and harvest. This was different from previous years ( 1984- 1989) when weirs and personnel were in place to observe marked hatchery fish. In addition, thr, change in computing hatchery contribution was initiated because of the change in the fry transport t echque (starting in 1986) from ground stocking of fry in the creeks to aerid s tochg in the lake. We believe aerial stocking increases the chance of weak imprinting to release sites and has resulted in a decline in hatchery srnolt survival (Kyle 1992).

Fall Fry Abundance, Size, and Age

An estimated 10.5 + 2.0 million juvenile sockeye salmon iYy were rearing in Tusturnena Lake during late September 1995 (Table 2). The highest fLFh density (83/103 rn? was observed near the middle of transect 6 (Figure 2). Transects 1-3 located in the eastern portion of the lake where there is a major ~Iacial input usually have the lowest fish densities in the fall, and t h was the case in 1995 in which fish 0

densities were generally < 15 per lo3 m'. The relative percent of fish density by stratum was highest near the surface (1.5-5 m) for stations B and C (Figure 3). At station A a slightly higher percentage of fish density was at the 5-9 m stratum than near the surface (1.5-5 m).

Townetting in the fall of 1995 resulted in a catch of 319 juvenile sockeye salmon, which comprised 98 % of the total catch; 7 threespine sticklebacks Gasterostcus awleatus were also caught. Of the 3 1 Y sockeye salmon fry caught. 84.6% were age-0 and 15.4% were age-1 which were similar to the 1980- 1994 means of 82.3% and 17.7%, respectively (Table 3). Age-0 fish averaged 64 mrn and 2.8 _g and the age- 1 fish averaged 78 m and 5.2 g (Table 3). The mean size of age-0 fry in 1995 were considerably larger (0.8 g) than the 1980-1994 mean but the mean size of age-1 fry in 1995 was similar to the 1980- 1994 mean.

Slnolt Abrrndance, Size, and .Age

An estimated 7,302,605 f 2,180,180 sockeye salmon smolts emigrated Tusturnena Lake during 8 May-30 June 1995 (Table 4). During the course of the rnigration one large peak occurred; 7-8 June when 2.7 million smolts emipated (Figure 4). Peak emigration timing (when 50% of the emigration was reached) in 1995 was 8 June, which was one day later than the 198 1- 1994 mean of 7 June. The initiation of migration (when 1,000 smolt are captured in the trap) was 22 May, which was five days later than the 1981-1994 mean of 17 May. On 8 June 1995 the highest daily number (55,800) of smolts emigrated, and based on a 3.6% capture rate the daily emigration was estimated at 1.6 million (Table 4).

Ln 1995 the srnolt migration consisted of 66% age- 1 and 34% age-2 smolts that averaged 72 rnm and 2.8 g, and 84 rnrn and 4.3 g, respectively (Table 5). The mean size and age composition of age-1 and age-2 smolts in 1995 were similar to the 198 1 - 1994 means (Table 5). The size of age- 1 and age-2 smolts that were dyed and recaptured in 1995 (Table 6) were similar to the randornly-sampled smolts (Table 5), and the mean age composition was slightly different (e.g., 72% age-1 for the recaptured sample compared to 66% age-1 for the random sample). Based on the sampled age compositions, an estimated 4,826,100 + 500,000 age-1 smolts and 2,476,500 + 500,000 age-2 smolts comprised the

total 1995 smolt outmi~ration (Table 7). No hatchery contribution rate was calculated for the 1995 srnolt out~~igration as there were no age-1 hatchery smolts produced (the frjr were destroyed after findill2 that they were infected with the EN-virus).

Lirnrzological Surveys

Eight lirnnological surveys were conducted on Tustu~nena Lake during 1995. In 1995, the seasonal mean epilirnnetic ( 1 m) concentrations of corrected total phosphorous ( U P ) and ammonia were lower at all three stations than for the station means during 1980-1994 (Table 8). Total nitrogen (Kjeldahl nitrogen plus nitrate + nitrite) was lower (3-7%) at stations A and C co~npared to the 1980-1991 rneans, but because of lower CTP values in 1995, the aro~nic ratio of total nitrozen (TN) to CTP ii.as higher (27%) than the means at all stations. In addition, seasonal mean chlorophyll a concentrations (algal biornass) were lower at stations A and C but higher at station B than the historical averages (1980- 1994); peak concentrations were also lower at station A and C and 35% higher at station B in 1995. General observations indicate that the weather during the spring was normal but the summer was drier than normal and the fall was much wetter than normal. The seasonal Inean water temperature was higher at station A, and the date of heat maximum was earlier (by 2-3 weeks) for all three stations than the means for 1980- 1994.

Ln 1995. the seasonal mean Diaprom~!,~ biomass was about 25% higher at stations A and B and the ssasond mean density 17,,as also hgher compared to that for 1980- 1994: at station C the biomasc and density were both lower in 1995 than the mean (Table 9). For stations D and E, the seasonal mean Diaptomus biomass was 20% higher in 1995 than the 1990- 1994 mean, and the density for station D in 1995 was similar to the 1990-1994 mean while at station E the density was 25% higher. The seasonal mean Diuptomm biomass and density for stations A, B, and C in 1995 were 15% and 25% higher, respectively than the 1980-1994 rneans. In 1995, the Diaptomus density peaked on 2 August for stations A, B, and E, on 25 August at station C, and on 13 September at station D (Figure 5).

Cyclops seasonal mean biomass in 1995 was about 13% lower and density was about 30% lower at stations A and B compared to the means for 1980-1994; however, at station C both biomass and density were proportionately less (about 58%) compared to that for 1980- 1994 (Table 9). The seasonal Inean Cyclops biomass and density for stations A, B, and C were 25% and 40% lower, respectively in 1995 than the 1980-1994 means. For stations D and E, the Cyclops biornass and density were both about 30% and 40% lower, respectively than the 1990-1994 means. During 1995 Cyclops density varied considerably at each station, and except for station A and B no obvious peaks in abundance occurred (Figure 6).

LITERATURE CITED

Brocksen. R. W., G. E. Davis, and C. E. Warren. 1970. Analysis of trophlc processes on the basis of density-dependent functions. Pages 186-199 iiz J. A. Steel (editor) Marine Food Chains. Oliver and Boyd, Edinburgh.

Fandrei, G. 1995. Bear Creek (Tustumena Lake) sockeye salmon adult count and gamete collection progress report, 1995. Cook Inlet Aquaculture Association. 12 p.

Flazg, L. B., M. J. Owecke, and D. C. Waite. 1984. Sockeye salmon smolt studies, Kasilof River. Alaska, 1982. Alaska Department of Fish and Game, FRED Division Report Series No. 27. Juneau, Alaska.

Flagg, L. B., G. A. Gray, J. P. Koenings. and K. E. Tarbox. 1987. Sumnmary of sockeye sahnon investigations in Tustu~nena Lake, 1921 1- 1986. Progress Report No. 5. Alaska Department of Fish and Game. Soldotna, Alaska.

Foerster. R. E. 1944. The relationship of lake population density to size of young sockeye salmon (Oncorhytzchra tzerka). J. Fish. Res. Bd. Can. 6267-280.

Johnson, W. E. 1965. On the mechanism of self-regulation of population abundance in (Otzcorhynch~~s ~erl ia) . Mitt. Intern. Verein. Limnol. 3 1 :66-87.

Koenings, J. P. and R. D. Burkett. 1987. The production patterns of sockeye salmon (Oncorhynclzus nerku) smolts relative to temperature regimes, euphotic volume, fry density, and forage basz within Alaska lakes. Pages 419-434 in H. D. Smith, L. Margolis, and C. C. Woods (editors) Sockeye Salmon (Oncorhytzchrrs izerka) Population Biology and Future Management. Can. Spec. Pubi Fish Aquat. Sci. 96.

Koenings, J. P., J. A. Edrnundson, G. B. Kyle, and J. M. Edmundson. 1987. Lirnnological field and laboratory manual: methods for assessing aquatic production. Alaska Department of Fish and Garne, FRED Division Report Series No. 7 1. Juneau, Alaska.

Koenings, J. P., J. A. Edmundson, and G. B. Kyle. 1988. Density-independent environmental innfluences on juvenile sockeye salmon (Oncorhynchw nerku) recruitment and smolt production. Alaska Department of Fish and Game. Soldotna, Alaska. (Itz revision).

Kyle, G. B. 1992. S umrnary of sockeye salmon (Otzcorhynchus izerka) investigations in Tustu~nena Lake, 198 1-199 1. Alaska Department of Fish and Game, FRED Division Report Series No. 122. Soldotna, Alaska.

Kyle, G. B., J. P. Koenings, and B. M. Barrett. 1988. Density-dependent, trophic-level responses to an introduced run of sockeye salmon (Oncoi-hynchm nerka) at Frazer Lake, Kodiak Island, Alaska. Can. J. Fish. Aquat. Sci 45:856-867.

Numdee, E. P. 1980. Application of an empirically scaled digital echo integrator for assessment of juvenile sockeye salmon populations. Ph.D. Thesis. University of Washington. Seattle, Washington. 15 1 p.

Numallee, E. P. and 0. A. Mathisen. 1972. An acoustic survey of Lake Wenatchee, Washington. University of Washington, College of Fisheries, Fisheries Research Institute. Circular No. 72- 13.

Schmidt, D.C., K. E. Tarbox, G. B. Kyle, and S. R. Carlson. 1995. Sockeye salmon overescapelnents: 1993 annual report on Kenai River and Kodiak investigations. Alaska Department of Fish and Game, Regional Information Report No. 5J95- 15. Juneau, Alaska.

Rawson, K. 1984. An estimate of the size of a migrating population of juvenile salmon using an index of trap efficiency obtained by dye marking. Alaska Department of Fish and Game, FRED Division Report Series No. 28. Juneau, Alaska.

Ruesch, P. H., and J. Fox. 1995. Upper Cook Inlet cornmerical fisheries annual rnanagernent report, 1994. Alaska Department of Fish and Game, Regional Information Report No. 2A95-26. Juneau, Alaska.

Thome. R. E. 19S2. Hvciroacoustics. Pages 239-253 it7 L. Nielson and D. Johnson (editors) Fisineries Techntques. Amer. Fish. Soc. Bethesda, Maryland.

Thorne, R. E. 1988. An empirical evaluation of the duration-in-beam technique for hydroacoustic estimation. Can. J. Fish. Aquat. Sci. 45: 1244- 1248.

Todd, G. L. 1994. A lightweight. inclined-plane trap for sampling salmon smolts in rivers. Alaska Fishery Research Bulletin l(2): 179- 186.

Todd, G. L. and G. B. Kyle. 1994. Tustumena Lake project report: sockeye salmon investigations (fall 1992- 1993). .4laska Depm~nent of Fish and Garne, Commercial Fisheries Management and Development Division. Regional Information Report Series No. 5394- 1 1. Soldotna, Alaska.

Todd, G. L. and G. B. Kyle. 1995. Tustumena Lake project report: sockeye salmon investigations 1994. Alaska Department of Fish and Game, Commercial Fisheries Management and Development Division. Regional Information Report Series No. 5J95-12. Soldotna, Alaska.

Table I. Estimated number and percentage of hatchery-produced sockeye salmon in the escapement and total return for Tustumena Lake in 1995, and comparison to the mean for 1984-1994.

Kasilof River sonar count (escapement) Estimated number and (%) of hatchery-produced fish in the escapement Estimated commercial fishing exploitation rate a

Estimated commercial harvest " Estimated number of hatchery-produced fisl I commercially caught Estimated number of fish caught in personal- use (dipnet and gillnet) and sport fisheries\b Estimated number of hatchery-produced fish in personal-use fisheries

Estimated number of hatchery-produced fish in the total return

Mean 1995 1984-1 994 205,903 231,372

10,256 60,328 5.0% 24.6%

" Exploitation rate and commercial harvest are preliminary.

cu-

m

E

!2 "0

<

5

Table 3. Mean size and age composition of juvenile sockeye salmon fry caught in surface net tows by basin during 1995 in Tustumena Lake, and comparison to the means for 1980-1994.

Mean Mean Mean Mean Sample Age length weiglit Sample Age length weight

Basin size comp. (mm) S.D. (g) S.D. size comp. (mm) S.D. (g) S.D.

Mean 1995 270 84.6% 64 5.93 2.8 0.72 49 15.4% 78 3.34 5.2 0.71

Mean 1980-1 994 256 82.3% 57 2.0

Table 4. Daily population estimates of sockeye salmon smolts emigrating Tustumena Lake, 1995.

Date

09-May 10-May 11-May 12-May 13-May 14-May 15-May 1 6-May 1 7-May 1 8-May 19-May 20-May 21-May 22-May 23-May 24-May 25-May 26-May 27-May 28-May 29-May 30-May 31-May 01 -Jun 02-Jun 03-Jun 04-Jun 05-Jun 06-Jun 07-Jun 08-Jun 09-Jun 10-Jun 1 1-Jun 12-Jun 13-Jun 14-Jun 15-Jun 16-Jun 17-Jun 18-Jun 19-Jun 20-Jun 21-Jun 22-Jun 23-Jun 24-Jun 25-Jun 26-Jun 27-Jun 28-Jun 29-Jun 30-Jun

Totals

Unmarked No No. fish caught

released recovered (1 000s)

479 13 0.026 479 13 0.01 8 479 13 0.078 479 13 0.081 479 13 0.082 479 13 0.169 479 13 0.144 479 13 0.699 479 13 0.480 479 13 0.583 479 13 0.768 479 13 0.893 900 68 0.91 9 900 68 2.699 900 68 2.287 900 68 1.869 900 68 1.401 900 68 1.382 900 68 1.526 900 35 1.357 900 35 5.573 900 35 11.286 900 35 8.1 77 900 35 1.826 900 35 0.856 900 35 1.720 900 32 4.223 900 32 15.083 900 32 4.655 900 32 38.704 900 32 55.791 900 32 24.912 900 32 8.462

2,000 132 14.051 2,000 132 8.841 2,000 132 14.117 2,000 132 11.581 2,000 132 4.836 2.000 132 4.813 2,000 132 10.703

914 50 9.991 914 50 8.230 914 50 2.178 914 50 1.023 914 50 1.364 914 50 1.360 914 50 0.781

1,594 94 1.1 57 1,594 94 1.562 1,594 94 6.462 1.594 94 2.192 1,594 94 1.607 1,594 94 1.252

Estimated migration

1,030 71 3

3.089 3,208 3.247 6.693 5,703

27,683 27.683 23.089 30,416 35,366 12,329 36.208 30.681 25.073 18,795 18.540 20,472 35 852

147,241 235,234 21 6,040 48,244 22,616 45,443

122,352 436,995 134,868

1,121,358 1.61 6,413

721,767 245,167 214,400 134,902 215,407 176.71 1 73,791 73,440

163.314 186,088 153,289 40,567 19,054 25.405 25,331 14,547 19,816 26,753

110,676 37,543 27,523 21,443

95% Confidence interval lower upper

Table 5. Size and age of Tustumena Lake sockeye salmon smolts by sample period, 1995.

Mean Mean Sample Age Age length weight

Sample period size class comp. (mm) S.D. (9) S.D.

Weighted 762 mean (1 995) 337

Mean (1 980-1 994)

a The AWL samples for this period was estimated from adjacent periods (4-1 0 June and 18-24 June), as the AWL samples were taken incorrectly.

Table 6. Size and age of Tusturnena Lake sockeye salmon smolts that were dyed and recaptured by sample period, 1995.

Mean Mean Sample Age Age length weight

Sample period size class comp. (mm) S.D. (9) S.D.

Weighted mean (1 995)

Table 7. Population estimate (in thousands) of sockeye salmon smolts emigrating Tustumena Lake by age and sample period, 1995.

Age 1 Age 2 Migration 95% Confidence interval Migration 95% Confidence interval

Sample period estimate lower upper estimate lower upper

Total 4,826.1 Age Composition 66.1 %

Table 8. Limnological characteristics within the epilimnion (1 m) of Tustumer~~i Lnke during 1995 by basin, and comparison to the means for 1980-1 994.

Parameter

Station A Station B Station C All stations - Mean Mean Mean

1995 1980-94 1995 1980-94 1995 1980-94 1995 1980-94

Turbidity (NTU) a

Seasonal mean water temp. ("C)

Date of 1 m water at 4' C

Date of heat max. r C )

Euphotic zone depth (m)

Total phosphorus (pg/L)

NTP 01911)

CTP (pglL)

Total nitrogen (pg/L) Total Kjeldahl nitrogen (pg/L)

Ammonia (pglL) Nitrate + Nitrite (pg/L)

Atomic ratio (TN:CTP)

Silicon (pg lL)

Alkalinity (mglL CaCO,)

pH units

Chlorophyll a (seasonal mean) Chlorophyll a (seasonal peak)

a Mean turbidity is for 1984-1994.

Total phosphorus corrected for turbidity (NTP) = -0.224 + 0.688 TP (ug/L).

" Total phosphorus corrected for turbidity and inorganic particulate phosphorus (CTP) = 0.0701 + 0.249 NTP.

0' -0

C

m m a m

63 7

a

m

7

E 2

- h

m (I) C m E m

(I)

2 0

.- m

m

mm

w

ui

o

o

a'w

m

w

a- ui

m- d

0

- C 0

(I) .- L

m F 0

0

OL

D rr-

o

o

(00 o

m

o

o

ma

' m

m

rg

b

*m

c

qw

5.0 - 9.0 9.0 - 15 15-27

Depth Stratum (m)

Figure 3. Juvenile sockeye salmon distribution by depth stratum for the three basins in Tustumena Lake based on the 22 September 1995 hydroacoustic survey.

Figure 4. Migration timing for sockeye salmon smolts emigrating Tustumena Lake, 1995.

.- (I) IZ a, n