An integrated stock assessment for the Patagonian toothfish (Dissostichus eleginoides) in Division 58.5.2 using CASAL

This paper follows preliminary work in 2005 and early 2006 in developing an integrated assessment for Patagonian toothfish in Division 58.5.2. It focuses on developing an integrated assessment using CASAL and demonstrates that all of the data available for assessments, including surveys, fishery catch-at-length data, standardised catch-per-unit-effort (CPUE) series for trawl grounds, and length-binned mark-recapture data, can be incorporated into the model. Earlier difficulties in using CASAL have been resolved. The results presented here indicate a general downward bias in the expected recaptures showing that the mark-recapture observations indicates a smaller stock size than the other datasets. This is consistent with the problems discussed of the potential bias of a single-area model in trying to cover a highly spatially structured stock. It is recommended that the mark-recapture data not be included in the assessment of D. eleginoides in Division 58.5.2 until the spatial structure of the mark-recapture program can be incorporated appropriately in the assessment. The assessment of long-term annual yield using CASAL, 2,306 tonnes, is less than that using the GYM, 2483 t. The reasons for this difference are discussed. We conclude that an integrated assessment using CASAL is possible for Dissostichus eleginoides in Division 58.5.2. However, as expected, the assessment will be sensitive to the inclusion of different datasets and to the choices of parameters used in both the stock assessment and projections. An important outstanding issue, relevant to stock assessments generally, is the need to provide an estimate of natural mortality. It is suggest that the assessment not include tagging data at this stage and that yields be estimated by integrating across uncertainties in natural mortality. It is also recommended that work be undertaken to understand the spatial dynamics of D. eleginoides in the region and, therefore, how best to incorporate mark-recapture data in the assessment. Until that time, recruitment surveys provide the best means of establishing current stock status.