A yield-per-recruit model is developed for the Alaska troll chinook salmon fishery, incorporating hooking mortality, fishing mortality growth, natural mortality, and maturation rate. The yield-per-recruit model provides a conceptual framework for evaluating the effects of size limits, gear restrictions, and time-area closures. Landed value per recruit is also calculated using a price function which tracks ex-vessel prices through the fishing season. Growth is allowed to vary depending on the age of maturity. Natural mortality is specified as an inverse function of fish age. A size related hooking mortality function is constructed based upon mortality observed under trolling conditions. A relationship between hook size and mouth size is hypothesized to explain the observed increase in hooking mortality with fish length. Fishing mortality is estimated from an analysis of micro-wire tag recovery data. Results of the model are presented as response surfaces of yield with respect to size and fishing mortality. Adding hooking mortal i ty to a yield-per-recruit model causes the yield response surface to decline with increasing fishing mortality so that a global maximum appears on the surface. Response surfaces of yield with respect to size limit and hooking mortality are also presented. The model demonstrates that the optimum size limit is a continuous increasing function of fishing mortality rate and is a continuous decreasing function of hooking mortality. Based on the most likely estimates of fishing and hooking mortalities, a size 1imit of 24.0 inches would maximize yield in weight of chinook salmon in the Alaska troll fishery and a size limit of 26.5 inches would maximize the landed value of the catch. Gear restrictions to reduce hooking mortality had a negative impact on yield in all cases investigated since the decrease in CPUE associated with the gear restriction outweighed the advantages of reduced hooking mortality. Time-area closures in high-density shaker areas would probably increase yield from the fishery by only 2% to 7%. These relatively minor increases in yield must be balanced against the negative socio-economic impacts of the time-area closures. The most serious shortcoming of the current configuration of the model is the assumption that fishing mortality is independent of maturity stage. More precise estimates of fishing mortality rates by maturity group and age would allow more precise estimation of optimum size limits from the model.
Key words: yield-per-recruit model, computer simulation, hooking mortality, chinook salmon, Alaska troll fishery, shaker mortality, eumetric yield, landed value-per-recruit, gear restrictions, time-area closures, optimization, size 1imit.
Figures - uploaded by
Fritz FunkAuthor contentAll figure content in this area was uploaded by Fritz Funk
Content may be subject to copyright.