Research Document - 2009/070
Petrale sole (Eopsetta jordani) in British Columbia, Canada:Stock Assessment for 2006/07 and Advice to Managers for 2007/08
By Paul J. Starr
Abstract
Information pertaining to Petrale sole (Eopsetta jordani) in British Columbia was reviewed and updated for inclusion in a delay-difference stock assessment model. This model was used to determine the status of this coastwide stock and to provide quantitative advice on levels of catch and the associated risk relative to selected management performance indicators.
A range of model uncertainties was explored through sensitivity runs which varied model assumptions which could not be easily reconciled through inspection of the model fits to the data. Four pairs of alternative model assumptions were investigated: a) estimating M, the rate of instantaneous natural mortality through the use of mean weight data sampled from the fishery or fixing M at the preferred value of 0.20 and dropping the mean weight data; b) varying the age of knife-edged recruitment between age 6 and age 7; c) applying a single CPUE series for the entire model period, effectively assuming that the fishery catchability has been constant for 40 years or splitting the CPUE series between 1995 and 1996 in recognition of the severe management restrictions that were applied at that time; d) estimating alternative standardised CPUE series based on different data selection criteria: one set of criteria was suggested by fishing industry representatives to optimise the data for Petrale sole and the other data set used criteria that allowed more peripheral data into the data set.
Model results showed that within the range of the criteria investigated, the effects of fixing or estimating M and the age of knife-edge recruitment were relatively minor, with the management advice almost identical across these options. However, the effect of splitting the CPUE series was major, with the model estimating a drop in catchability in recent years and consequently being much more optimistic about stock status. Finally, the CPUE series optimised for Petrale sole indicated a much stronger recent rebuild for this species and thus also provided more optimistic advice.
The split CPUE series model runs using the CPUE series optimised for Petrale sole predicted that the stock would increase over the next 5 years as well as stay above the Bref and Bmin reference points with a high probability (greater than 90%) at removals equal to the 2006 TAC (600 t). The split CPUE series using the wider data selection criteria did not predict that the stock would increase over the next five years, but there was an 80 to 100% probability that the stock would stay above the Bref and Bmin reference points (again at levels of removal equal to the 2006 TAC). On the other hand, the runs which assumed a single CPUE series with constant catchability over the past 40 years had a low probability of increasing at the beginning of 2012 (31–41%) while there were even lower probabilities of exceeding Bref (9 to 43%) at levels of removal equal to the 2006 TAC. The probability for the single CPUE runs exceeding the Bmin reference point were acceptable, ranging from 64–81%. The lack of capacity to predict a stock size increase over the next 5 years at levels equivalent to the 2006 TAC in several of the model runs was a result which may originate from the use of mean recruitment to drive the predictions. There is some evidence from the model fits that recruitment over the most recent 10 years is about 10% above the mean which may mean that the stock projections are conservative.
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.
- Date modified: