科研之友
微信
新浪微博
Facebook
分享链接
摘要
A biomass-based length-cohort analysis (LCA) was examined for its performance in estimating total stock biomass and fishing mortality (F) for a population in equilibrium. We compared two LCA methods-(1) a numbers-based LCA that relies on catch numbers at length as input data and (2) a new biomass-based LCA that relies on catch biomass at length-by applying both to simulated data that generally followed characteristics of Pacific chub mackerel Scomber japonicus in Korean waters. We investigated the effects of using two length aggregation methods, two data simulation methods, and three expressions of weight growth (i.e., exponential growth over the length increment of the analysis, arithmetic mean growth, and geometric mean growth). Biomass and F were estimated from the new model and compared with those estimated from the numbers-based LCA model. A sensitivity analysis was performed to understand which aspect of the model parameters controlled the results. Sensitivity to estimates of terminal F was also examined. Results of the analysis showed that the allometric power coefficient beta was the most important model parameter and that sensitivity of biomass-based LCA to terminal F was similar to that of numbers-based LCA. Monte Carlo simulation was used to evaluate the equilibrium assumption and to calculate confidence intervals for biomass estimates. We applied the biomass-based LCA model to data for northern rock sole Lepidopsetta polyxystra in the eastern Bering Sea. A comparison of biomass-based LCA population biomass estimates with northern rock sole research survey biomass estimates showed good agreement between the two types of estimates. The biomass-based LCA described in this study is unique in that it incorporates growth explicitly into the estimation procedure, which can be useful under data-deficient situations when simple length composition data and estimates of total catch biomass from just one sampling year are the only available data.
- 出版日期2010-5
