Stock assessment of short-lived invertebrates using hierarchical Bayesian models

摘要: Fisheries for short-lived invertebrates typically have the following characteristics:(1) annual catch is comprised of a single age-class;(2) recruitments vary substantial inter-annually and the population dynamics are hence subject to large amounts of process error;(3) data contain large observation errors;(4) the resource is composed of several stocks;(5) several fleets may target or bycatch a species and stock; and (6) only catch and effort data are available. Thus, stock assessment for such fisheries is very difficult. Biomass dynamics (or production) models are a popular tool for stock assessment of finfish and shellfish fisheries that only have catch and effort data. An implicit assumption of most biomass dynamics models is that natural mortality is not very high so that a fairly large proportion of the biomass at the start of the next (annual) time-step consists of the biomass at the start of the current time-step. However items (1-2) for short-lived species may violate this assumption. Moreover the standard biomass dynamics model assumes that all of the error is in the observation process and that the dynamics are deterministic. This model has been shown inefficient for short-lived invertebrates, either due to its inappropriate assumptions or the traditional estimation methods.We construct two alternative versions of the standard biomass dynamics model (Standard) for shortlived species by ignoring the “old biomass” term (Annual), and assuming that the biomass at the start of the next year depends on density-dependent processes that are a function of that biomass (SR).