Abstracts of Fisheries Management Reports, Technical Circulars and Project Reports of the Fisheries Branch Daiva O. Zaldokas & Debra L. Aird B.C. Fisheries Branch, 2204 Main Mall, U.B.C., Vancouver, B.C. V6T 1Z4 Fisheries Technical Circular No. 91 1992 1) Ludwig, B.W., and R. Clough. 1989. Performance of trout fed a variety of feeds at British Columbia trout hatcheries, 1985-87. Prov. B.C. Fish. Manage. Rep. No. 93:22p. 2) Parkinson, E.A., J.M.B. Hume, and R. Dolighan. 1989. Size selective predation by rainbow trout on two lacustrine Oncorhynchus nerka populations. Prov. B.C. Fish. Manage. Rep. No. 94:14p. 3) Ward, B.R., and J.C. Wightman. 1989. Monitoring steelhead trout at the Keogh River as an index o stock status and smolt-to-adult survival: correlations with other data sources. Prov. B.C. Fish. Manage. Rep. No. 95:21p. 4) Parkinson, E.A. 1990. An evaluation of adaptive management and minimal sampling as techniques for optimizing rainbow trout stocking rates. Prov. B.C. Fish. Mange. Rep. No. 96:14p. 5) Oliver, G.G. 1990. An evaluation of special angling regulations for cutthroat trout in the lower St. Mary river. Prov. B.C. Fish. Mange. Rep. No. 97:19p. 6) Walters, C., J. DiGisi, J. Post and J. Sawada. 1991. Kootenay Lake fertilization response model. Prov. B.C. Fish. Mange. Rep. No. 98:36p. FISHERIES MANAGEMENT REPORTS Ludwig, B.W., and R. Clough. 1989. Performance of trout fed a variety of feeds at British Columbia trout hatcheries, 1985-87. Prov. B.C. Fish. Manage. Rep. No. 93:22p. The dietary performance and cost efficiency of production diets currently used for culturing trout at Provincial trout hatcheries were compared to several other diets available in British Columbia. Seven studies were conducted over three brood years at three different hatcheries. Six starter diets (BioDiet, Silver Cup Softmoist, Oregon Moist Pellet (O.M.P.), Ewos and Rangen's Softmoist) were tested. Of the starter and grower diets tested, BioDiet and BioMoist resulted in the best performance in terms of survival and production, while for the grower diets, Silver Cup provided the lowest cost per kg of fish produced. Several other diets including O.M.P., Ewos and Rangen showed promise but require further testing. BioDiet was recommended as the starter feed. Based strictly on production costs, the recommended grower feed would be Silver Cup, but a final recommendation cannot be made until a comparison of the relative survival in lakes for fish fed BioDiet and Silver Cup is completed. Parkinson, E.A., J.M.B. Hume, and R. Dolighan. 1989. Size selective predation by rainbow trout on two lacustrine Oncorhynchus nerka populations. Prov. B.C. Fish. Manage. Rep. No. 94:14p. Size distributions of prey from the lake and in the diet of large (>350 mm) rainbow trout feeding on O. nerka were compared in two lakes. In both lakes, O. nerka made up over 98% of the fish in both rainbow trout stomachs and midwater trawl catches. In Kootenay Lake, where there were no anadromous sockeye, kokanee (the non-anadromous morph) that were 100 - 200 mm fork length made up 42% of trawl catches, but in Quesnel Lake, which supports both kokanee and a large run of anadromous sockeye, trawl catches consisted on 95% age-0+ O. nerka (<80 mm fork length). Although the size of prey in Quesnel Lake (104 mm) was smaller than in Kootenay Lake (139 mm), most of this difference appeared to be due to differences in the size of the trout predators which averaged 540 mm in Quesnel Lake and 713 mm in Kootenay Lake. Few fry (<80 mm) were observed in the stomach samples of rainbow trout caught in either lake. In both lakes, the largest prey taken by rainbow trout predators were less than about 1/3 of the predator length. Ward, B.R., and J.C. Wightman. 1989. Monitoring steelhead trout at the Keogh River as an index o stock status and smolt-to-adult survival: correlations with other data sources. Prov. B.C. Fish. Manage. Rep. No. 95:21p. Population estimates of adult steelhead trout, obtained at the Keogh River on northern Vancouver Island, were compared to catch statistics and underwater fish counts collected elsewhere in British Columbia. Also, survival rates from smolt-to-adult for Keogh steelhead were compared to rates obtained at another research station, Snow Creek, Washington. Similar trends were found in all comparisons. Angler catch of wild fish in the Keogh River and neighbouring streams was closely correlated with the run size estimates of the Keogh River adult steelhead (n=13). Annual catch of wild fish on Vancouver Island followed a similar trend as the winter run of the Keogh but the statistical correlation was weak unless years were sequentially paired (n=6). Representative winter-run streams to the south displayed catch statistics that weakly correlated with Keogh run sizes unless also sequentially paired. Of the larger rivers, there was close correlation between Nimpkish River catch, on northern Vancouver Island, but not Cowichan River catch on southern Vancouver Island, and Keogh run size. Underwater counts of steelhead in two summer-run streams had similar trends but failed to correlate with Keogh stock size, even with years paired. Survivals at the smaller Snow Creek in northwest Washington were similar to survivals obtained at the Keogh in 4 of 6 years, and differences in the other two years were possibly due to smolt size, handling effects and anomalous oceanographic conditions. Implications for future stock monitoring are discussed including the effects of introductions of hatchery fish and regulation changes on the validity of use of results from the Steelhead Harvest Analysis for interpretation of steelhead abundance. Parkinson, E.A. 1990. An evaluation of adaptive management and minimal sampling as techniques for optimizing rainbow trout stocking rates. Prov. B.C. Fish. Mange. Rep. No. 96:14p. Following a change in stocking rate of +40% Kentucky Lake and -40% on Alleyne Lake, length of age-3 rainbow trout declined by over 40 mm in Kentucky Lake and increased by over 50 mm in Alleyne Lake. Length at age 2 remained the same in Kentucky Lake but increased by 25 mm in Alleyne Lake. Higher stocking rates were associated with higher ages at harvest but variation in stocking rate had little effect on either the length of creeled fish or CPUE. An intense fishery, which rapidly removes recruitable sized fish was suggested to be the factor that decouples CPUE and fish size from stocking rate. High variation in CPUE and length at capture among years suggests that more years of data collection and larger management perturbations will be needed if the effects of changes in management policy are to be detected. Oliver, G.G. 1990. An evaluation of special angling regulations for cutthroat trout in the lower St. Mary river. Prov. B.C. Fish. Mange. Rep. No. 97:19p. Annual studies of wild westslope cutthroat trout (Oncorhynchus clarki lewisi) were conducted at the lower St. Mary River in the East Kootenay for a ten year period from 1980 to 1990. The effects of progressive fishery regulation changes on the population were examined by underwater surveys and standardized angling techniques. Age and growth characteristics of the cutthroat population were also documented. After pollution abatement in the late 1970's, the lower river supported a largely unexploited fluvial trout population (adult spawning and juvenile rearing in tributaries; adult rearing in the mainstem river) with a mean length of 342 mm and a range from 260 to 440 mm. Estimated trout abundance was 118 per km of stream channel. Angling regulations in 1980/81 included catch and possession limits of 4 fish and a 'flyfishing' only restriction for approximately 18.5 km. In 1982, the total estimated population was 6400 cutthroat trout >20 cm. From 1984 to 1990, population trends were monitored using a 3.3 km "index section" of the river. In the presence of an expanding fishery, numbers of catchable wild fish declined from a peak of 96 per km in 1986 to 47 per km in 1990. Despite further restrictive regulations including a 2 fish catch limit, a bait ban, a winter trout closure and a 30 cm minimum size limit, the estimated population in 1989 was 2921 fish and mean size had dropped to 271 mm in contrast to 342 mm 7-8 years earlier. An experimental shift in 1988 to a catch and release regulations on a 13 km section of river resulted in an increase in the proportion of larger fish after a single year although further sampling and counts are needed to confirm this improvement. Only 34% of the population exceeded 300 mm in the section with a kill fishery, but in contrast 51% of the population were >300 mm within the trout release section, although with the small size (n=4,2) a difference could not be statistically confirmed. A comparison of counts in a 4 km catch and release "index site" after two years however, indicated a significant difference as the percentage of trout >300 mm increased from 18% (prior) to 33% after regulation. Future intensive management options to improve the age structure, and thereby fishery quality, in the kill zone include catch-and-release or an increased minimum size limit of 35 cm. The latter would permit a small harvest of trophy fish and diversify the fishery. Walters, C., J. DiGisi, J. Post and J. Sawada. 1991. Kootenay Lake fertilization response model. Prov. B.C. Fish. Mange. Rep. No. 98:36p. This report describes a computer simulation model developed to make predications about the response of Kootenay Lake plankton and fishes to restoration of higher nutrient loading. The model was developed using the Adaptive Environmental Assessment (AEA) workshop process, where a team of scientist with programming experience works with an interdisciplinary participant group of scientists and managers to translate the ideas, data, and policy concerns of that group into a working model for quantitative prediction. While group ultimate aim of the AEA process is to produce useful models for management, the initial aim in the workshop development is to clarify research priorities by directing attention to processes that are key to the model predictions but are difficult to quantify, to discrepancies between historical data and model predictions, and to alternative hypotheses that would give the same predictions. The model appears to give credible quantitative predictions of basic nutrient dynamics and budgets (phosphorus inputs, outputs, horizontal and vertical transport, epilimnetic concentration patterns, responses of concentration to fertilization). However, there are major uncertainties in the model relationships that translate nutrient concentrations into predictions of phytoplankton and grazing zooplankton production, the utilization of zooplankton production by Mysis relicta and kokanee salmon (Oncorhynchus nerka), and the effect of changing kokanee and Gerrard rainbow trout (Oncorhynchus mykiss). With the best available parameter estimated, the model "reconstructs" recent population declines in kokanee abundance as a combined effect of reduced food availability (on growth, fecundity, and time spent in sizes vulnerable to trout prediction) and depensatory predation by trout. While the calculations attribute the reduced food availability partly to reductions in nutrient loading to the lake, the main impact is due to competition with Mysis for grazing zooplankters, particularly Daphnia. The model does not predict favourable responses of kokanee to lake fertilization; instead it predicts that fertilization will mainly enhance Mysis abundance and hence possibly even hasten the kokanee decline. Further, calculated prediction rates by Gerrard trout are presently high enough to cause further kokanee declines, and search efficiencies by the trout are high enough to cause even higher mortality rates, so the kokanee may soon be unable to recover even if their growth rates are substantially enhanced by fertilization. Thus it appears that the key uncertainties now are about how the zooplankton community (grazer species and size composition, Mysis) will respond to fertilization. At least four processes were identified that may counter the pessimistic baseline prediction that fertilization will enhance Mysis to the detriment of kokanee: (1) cannibalism preventing further Mysis increase; (2) increased export of Mysis over the West Arm sill under fertilization conditions, (3) increases in predation on Mysis by kokanee and other fishes; and (4) reduced relative importance of Mysis predation on early spring development of grazing zooplankton populations, under more fertile conditions. One or more of these processes apparently does operate in Kennedy Lake, where K. Hyatt and K. Cooper have measured positive responses of both Neomysis and fish to fertilization. All indications from the model are that the kokanee are now caught in a squeeze between competition and predation that will result in collapse of the kokanee population (and hence the prized Gerrard stock) if no action is taken soon to control either one or both sides of the squeeze. Control of predation could not easily be accomplished by reduction in the Gerrard stock, due to its high recruitment rate per spawner, even if such reductions were politically feasible. Control of other predators is likely impractical for the same reasons. Hence it appears that the best management options are those related to kokanee productivity and competition with Mysis. We conclude that it is still worthwhile to conduct a major fertilization experiment in the lake. It will be particularly critical during this experiment to closely monitor responses of Mysis (abundance, size, distribution, export to West Arm), grazing zooplankton (seasonal abundance, size distribution), growth of kokanee, and characteristics (size, number) of kokanee appearing in the stomachs of major predators - Gerrard trout, dolly varden (Salvelinus malma). The situation with kokanee is now desperate enough, and the risk of its recovery being jeopardized by Mysis is high enough, that extreme options should now be considered to reduce Mysis abundance. These options include introduction of Mysis predators (such as deep water sculpin), use of bubblers or hydrodynamic oscillations to push Mysis toward the surface near the West Arm sill, and use of high fertilizer loads near the sill to reduce water clarity and induce Mysis to spend more time near the water surface (where they are vulnerable to export over the sill). Parkinson, E.A. 1990. Impaired school formation at low density: A mechanism for depensatory mortality in sockeye salmon. B.C. Fish. Br., Fish. Mgmt. Rep. No. 99:17p. A simple encounter model was used to simulate the process of school formation at dawn for different densities of juvenile sockeye salmon. The results suggest that the process of school formation may be severely inhibited by a low encounter rate among juvenile sockeye at the lowest densities commonly observed in natural populations. Encounter rates of predators with schools are insensitive to density because, under a given set of conditions, asymptotic school density is similar over a wide range of initial densities of prey individuals. Enhanced school formation may reduce predation risk at high sockeye density and provide a mechanism for the multiple equilibrium hypothesis of sockeye dominance cycles.