Research Document 2017/015

Stock assessment of the coastwide population of Shortspine Thornyhead (Sebastolobus alascanus) in 2015 off the British Columbia coast

By Starr, P.J. and Haigh, R.

Abstract

A new stock assessment is presented for coastwide (Pacific Marine Fisheries Commission areas 3 and 5 combined) Shortspine Thornyhead (SST) based on a delay-difference production model fit to five fishery-independent surveys, a CPUE series derived from commercial catch rates, and an annual mean weight series derived from unsorted commercial catch samples.  Growth rates, natural mortality and selectivity were assumed externally from the model.  The stock assessment was characterised by considerable uncertainty associated with conflicting growth models and a range of plausible natural mortality estimates and sizes for full recruitment.  In addition, there was conflict in the fitted index series, with the biomass indices generally showing little contrast while the mean weight at age series showed increases after a sharp initial drop.  The stock assessment was conducted in a Bayesian framework, where the best fit to the data was used as the starting point for a search across the joint posterior parameter distributions using the Monte Carlo Markov Chain (MCMC) procedure.  A composite reference scenario (model average) was selected to represent this stock, consisting of 12 model runs which spanned plausible hypotheses with respect to growth, either based on age-length pairs from the Department of Fisheries and Oceans (DFO) biological database or based on a published growth model from the US National Marine Fisheries Service (NMFS).  The composite reference scenario also included three values for instantaneous natural mortality M (0.03, 0.06, 0.08) as well as a range of lengths at which knife-edge recruitment (k) to the fishery occurred (k = 29, 24, 21 cm). Three of the 12 runs assumed the DFO growth model in conjunction with three M values and k=29 cm. The remaining nine runs were based on the NMFS growth model and included three values of M and three values for knife-edge recruitment (k=29, 24, 21 cm). An MCMC posterior for the composite scenario was constructed by pooling 1000 MCMC samples from each of the selected runs to give a posterior of 12,000 samples, thus giving equal weight to each run. The composite reference scenario was evaluated against two BMSY-based reference points consistent with the DFO Precautionary Approach policy and the exploitation rate at B_MSY (u_MSY). The model-average median estimates of current stock status (B₂₀₁₆/ B₀) ranged from 40% to 141% B₀ with a median estimate of 79%, indicating that current biomass is well above all reference points. Three-year projections at the level of current removals resulted in a biomass decline which remained above reference levels.  The stock assessment provides a decision table which evaluates the probability of the model average case staying above three reference points across a range of 11 constant catches.  However, because the six abundance index series show little contrast, the stock assessment is unable to resolve productivity uncertainties, with similar fits to the data across the investigated range of natural mortality and age-at-knife-edge recruitment parameters.  Therefore, this assessment was unable to provide advice on the absolute size of the stock and the level of equilibrium yield.

Complete document (PDF, 8,412 KB)

Accessibility Notice

This document is available in PDF format. If the document is not accessible to you, please contact the Secretariat to obtain another appropriate format, such as regular print, large print, Braille or audio version.