I began my assignment on 1 April 1991 to compile, analyze, and report or edit back-year accumulations of file data. These data were collected during studies to determine if alternative hatchery practices could potentially increase survival to adulthood of salmon reared at Trinity River Hatchery (TRH). Reports on the first two Job 8 task studies have been completed, and summaries follow. I have begun analysis on the third and fourth studies, which will be combined into one report.
The goal of the first study was to determine if survival of fall-run chinook salmon could be improved by releasing them into the Trinity River at downstream sites. Results showed survival of fingerlings was increased by releasing them into the lower Trinity River (at Willow Creek and Hoopa), but homing tendency to return to TRH as spawners was reduced. Results for yearlings did not clearly show survival was increased by releasing them into the upper Trinity River (at Junction City and Lime Point).
The goal of the second study was to compare survival of fall-run chinook salmon based on age at release. Three age-at-release types were examined: fingerlings (reared 6 to 7 months), yearlings (reared 10 to 13 months), and yearlings plus (reared 13 to 15 months). Results showed older fish survived better and contributed more to fisheries and spawner escapements as adults. Younger fish were always larger (FL) at age 2 and generally at age 3; all release types were of similar size at age 4.
1. To provide for the compilation, analysis, write-up or editing of multi-year accumulations of previously collected file data on Trinity River basin salmon and steelhead that are beyond the scope of current Project activities.
2. To provide timely, as-needed technical support to the Project Supervisor in responding to unprogrammed information and data analysis requests regarding Trinity River basin salmon and steelhead stocks.
The goal of the study was to determine if survival of hatchery-reared fall-run chinook salmon could be increased by releasing them at downstream sites. This determination was based on survival comparisons between groups of tagged fish released near Trinity River Hatchery (TRH) and at various downstream sites. The comparisons were made for two age-at-release types: fingerlings released in June, and yearlings released in October or November.
This study was carried out by taking a portion of the annual hatchery production of TRH; dividing that portion into groups to represent a release at a particular site; tagging the groups then releasing them into the Trinity River; collecting recovery data obtained from surveys of the fisheries and at TRH; calculating ocean catch rates and using these rates to infer relative survival; comparing survival of groups from the same brood year (BY) and age-at-release; and estimating the contributions to the fisheries for all groups in the study.
For each BY (1977 through 1979) of the initial phase of the study, two groups of fingerlings were released into the Trinity River: one near TRH and the other at a downstream site. The downstream sites were Tish Tang Campground (1977 BY) located 153 river kilometers (RKM) below TRH, Hoopa (1978 BY) located 159 RKM below TRH, and Willow Creek (1979 BY) located 138 RKM below TRH. For each BY (1982 through 1984) of the second phase of the study, groups of both fingerlings and yearlings were released at TRH and three downstream sites: Steelbridge, located 15 RKM below TRH; Junction City, located 50 RKM below TRH; and Lime Point, located 61 RKM below TRH.
We used the following criteria to compare performance of groups from the same BY and age-at-release:
1. Hypothesis testing using the sign test (Siegel 1956) at the 0.05 level of significance to compare survival of fingerlings released at TRH with those released at downstream sites. We assumed equal mortality in the ocean for groups from the same BY and of the same age, and differences in ocean catch rates are due to mortality that occurs during the post-release freshwater emigration phase. There were only three BYs of results for yearlings, which were not enough for hypothesis testing.
2. Mortality-rate estimates experienced by hatchery-released fish during emigration. We calculated mortality rates for the sections of the Trinity River between TRH and the downstream release sites using the following formula from Hallock and Reisenbichler (1979):
1-(RTRH/Rds)
where,
RTRH = ocean catch rate of adults for tag-code group released near TRH and
Rds = ocean catch rate of adults for group released at downstream site.
3. Homing tendency rates for groups released at downstream sites. Homing tendency rate is a measurement of the behavioral tendency of fish released downstream to return to TRH, compared to fish released at TRH. The homing tendency rate was calculated using a formula provided by Hallock and Reisenbichler (1979). The homing tendency rate is the quotient of two ratios:
(Hds/Ods)/(HTRH/OTRH)
where,
Estimates of contributions to the ocean fisheries were obtained from the Pacific States Marine Fisheries Commission (1991). Estimates of contributions to the inland gill-net fisheries were obtained from the U S Fish and Wildlife Service (Arcata, CA) and the Hoopa Valley Tribal Council's Fishery Department. Estimates of contributions to the Klamath River sport fishery were expansions from data provided by the Department of Fish and Game's (DFG's) Klamath River Project. Estimates of contributions to the Trinity River sport fishery and natural spawner escapement were obtained from DFG's Trinity River Project. Finally, counts were made of all fish entering TRH.
The results of our evaluation of hatchery versus downstream releases were as follows:
1. The survival of fingerlings was increased by releasing them downstream (sign test for x = 0, p < 0.05, and n = 5); mortalities observed when fingerlings from the 1977 BY were released downstream precluded the inclusion of that sample in the test. Survival, as indicated by ocean catch rates, was nearly 10 times greater for fingerlings from the 1979 BY released at Willow Creek than for those released near TRH. For the five BYs for which comparisons are available, the average survival rate was nearly four times greater for groups released at the downstream-most sites than for those released near TRH.
2. Estimated mortality rates for fingerlings were calculated for five sections of stream. The mortality rates of these sections were as follows: TRH to Steelbridge - 0.13, TRH to Junction City - 0.43, TRH to Lime Point - 0.49, TRH to Willow Creek - 0.90, and TRH to Hoopa - 0.70. Estimated mortality rates of yearlings were calculated for three sections of stream. These rates were as follows: TRH to Steelbridge - 0.08, TRH to Junction City - 0.15, and TRH to Lime Point - 0.07.
3. Homing tendency rates for the groups of fingerlings released at downstream sites, compared with groups of the same BY released at TRH were as follows: Steelbridge - 0.37, Junction City - 0.48, Lime Point - 0.56, Willow Creek - 0.21, and Hoopa - 0.07. The rates for yearlings were as follows: Steelbridge - 0.74, Junction City - 0.48, and Lime Point - 0.52.
The BY contribution rates (% of original release) to the fisheries and spawner escapements ranged from 0.28% to 2.52% for fingerlings released at TRH and from 0.34% to 7.22% for fingerlings released downstream. The contribution rates for yearlings ranged from 2.00% to 13.03% for fish released at TRH and 2.92% to 7.95% for fish released downstream.
Survival of fall-run chinook salmon fingerlings was increased by releasing them at downstream sites. Results for fish released at Hoopa and Willow Creek suggest survival can potentially be increased from four to ten fold. It appears that survival is related to the distance downstream from TRH that fingerlings are released.
Selection of a planting location for fingerlings will involve a tradeoff between survival and homing tendencies. A greater fraction of the fish released downstream will survive, but a lesser fraction of these survivors will return to TRH. However, the results show total hatchery return rates of fingerlings can be greater for those fish released downstream, when homing tendencies are offset by improved survival.
Relative survival was consistent in the ocean, gill-net, and inland sport fisheries, and spawner escapements. That is, when ocean catch was high, so was inland gill-net catch, inland sport catch, and spawner escapement; and, when ocean catch was low, so was gill-net, inland sport catch, and spawner escapement.
Hallock, R.J., and R.R. Reisenbichler. 1979. Evaluation of returns from chinook salmon, Oncorhynchus tshawytscha, released as fingerlings from Coleman and Nimbus hatcheries and in the Sacramento River estuary. Unpublished office report available from Dept. of Fish and Game, Inland Fisheries Div., 1416 Ninth St., Sacramento, CA 95814
Pacific States Marine Fisheries Commission. 1991. Coast wide Coded-wire tag database. Portland, Oregon
Siegel, S. 1956. Nonparametric Statistics for the Behavioral Sciences. McGraw-Hill Book Co., Inc. 1956
This study was conducted to compare the relationship between age at release and survival to adulthood (three-, four-, and five-year-old fish) for fall-run chinook salmon reared at Trinity River Hatchery (TRH). Three age-at-release types selected for this study were: fingerlings - released at 6 to 7 months; yearlings - released at 10 to 12 months; and yearlings-plus - released at 15 to 17 months. Relative survival within each brood year (BY) was inferred from recovery rates of coded-wire tags (CWT) collected during surveys of ocean commercial and sport fisheries, inland gill-net and sport fisheries, and at TRH.
This study was carried out by taking a sample of the annual hatchery production from the 1977, 1978, 1979, 1983, 1984, and 1986 BYs. Each sample was then divided into three groups representing fingerlings, yearlings, and yearlings-plus. The groups were then tagged with CWTs, marked with an adipose-fin clip (Ad), and released into the Trinity River near the hatchery. Tags were later recovered during surveys of the fisheries and at the hatchery.
We used the following criteria to compare performance within BYs of fingerlings, yearlings, and yearlings-plus:
1. Hypothesis testing using the sign test (Siegel 1956) at the 0.05 level of significance to compare survival rates of groups representing different ages at release. We used catch rates to infer relative survival rates. We tested the hypotheses that fingerlings and yearlings, or yearlings and yearlings-plus survive at the same rate.
2. Ocean recovery and hatchery return rates by year class for the three age-at-release types. We estimated the percentage of total recovery that occurred at ages 2 through 5 years.
3. Length at age. Analysis of mean fork length (FL) at age for fish returning to TRH had been made by Hankin (1990) for tag-code groups from the 1977, 1978, and 1979 BYs. His results are presented in this report to provide a comparison with our results for the 1983, 1984 and 1986 BYs. He compared mean FL at ages 2, 3, and 4 years based on separation of confidence intervals (+95%).
We calculated confidence intervals (+95%) for the mean FL at age for tag-code groups from the 1983, 1984 and 1986 BYs, but observed overlap. We used analysis of variance to determine if there were differences among the three release types. Then we used a Tukey Studentized Range Test to determine which of the release types were different (Zar 1984).
The results of comparisons of fish based on age at release are as follows:
1. Survival to adulthood, as shown by the paired-sample sign test, was greater for yearlings than for fingerlings and greater for yearlings-plus than for yearlings (p = 0.05, n = 5, and x = 0 for both tests). Survival to adulthood, as suggested by total mean recovery rate comparisons, was several times greater for older fish. The mean recovery rate of the five BYs was 5.6 times greater for yearlings than fingerlings, and 2.1 times greater for yearlings-plus than for yearlings.
2. The percentages of recoveries by year class for fingerlings taken in the ocean fisheries were: age 2 - 2%, age 3 - 87%, and age 4 - 11%. The percentages for yearlings were: age 3 - 70% and age 4 - 30%. Percentages for yearlings-plus were: age 3 - 30% and age 4 - 70%.
The percentages of returns to TRH by year class for fingerlings were as follows: age 2 - 57%, age 3 - 38%, and age 4 - 5%. Percentages of returns for yearlings were: age 2 - 26%, age 3 - 62% and age 4 - 12%. The percentages of return for yearlings-plus were: age 2 - 15%, age 3 - 73%, and age 4 - 12%.
3. Based on separation of confidence intervals (+95%), Hankin (1990) made the following conclusions: i) the mean FLs at age for fingerlings returning to TRH were greater at ages 2 and 3 than yearlings and yearlings-plus; ii) yearlings were always larger than yearlings-plus at age 2 and 3, and iii) there were no differences among types at age 4. The 1979 BY releases were affected by the El NiZo that occurred in 1983 and resulted in reduced growth. Our results for the 1983, 1984, and 1986 BYs were similar to Hankin's results. We observed that fish released at a younger age were larger at age 2 and generally larger at age 3. We had results from only one BY available for age 4 fish, and these results did not show any differences among release types.
Based on study results, we made the following conclusions:
1. Fish released at an older age survive better to adulthood and in turn contribute more to fisheries and spawner escapements.
2. Fish released at an older age contribute more to fisheries and escapement as older fish (four- and five-year-olds).
3. Greater fraction of fish released at a younger age mature earlier and return to TRH as two-year-old fish.
4. Fish released at a younger age grow faster until age 4, when all age-at-release types achieve equal size.
Hankin, D.G. 1990. Effects of month of release of hatchery-reared chinook salmon on size at age, maturation schedule, and fishery contribution. Oregon Dept. of Fish and Wildlife Inform. Rep. No. 90-4 37 p.
Siegel, S. 1956. Nonparametric Statistics for the Behavior Sciences. McGraw-Hill Book Company, Inc., 312 p.
Zar, J.H. 1984. Biostatistical Analysis. 2nd ed., Prentice Hall, Englewood Cliffs, New Jersey. 718 p.
Hds/Ods
= return rate to TRH divided by ocean catch rate for the group released at a downstream site, and
