Bayesian assessment of the SNA1 snapper (Pagrus auratus) stock on the north‐east coast of New Zealand

Abstract

A statistical catch‐at‐age model was used in a Bayesian context to assess the SNA1 snapper (Pagrus auratus (Bloch & Schneider, 1801)) stock off the east coast of the North Island of New Zealand. The analysis differs from previous assessments by starting at an exploited population size in 1970 and using a Bayesian framework to describe uncertainty in the estimates of current stock status and projections of future stock status. The model was fit to two tagging biomass estimates and to catch‐at‐age data. Annual recruitment multipliers were estimated by assuming a temperature‐recruitment relationship and fitting to pre‐recruit trawl survey data for the Hauraki Gulf‐Bay of Plenty combined substock (HG‐BoP) and as individual parameters for the East Northland substock (EN). The combined HG‐BoP substock is estimated to be below the biomass level that will support maximum sustainable yield, but will rebuild under current estimated removals. The EN substock is estimated to be above the biomass level that will support maximum sustainable yield and will remain above this level during the next 20 years assuming current estimated removals continue. The initial fishing mortality rate in 1970 is a very important parameter in the model because it is inversely related to the rate of stock rebuilding. Unfortunately the initial fishing mortality rate is not well determined by the data in either assessment.