SCIENTIFIC COMMITTEE SEVENTH REGULAR SESSION 9-17 August 2011 Pohnpei, Federated States of Micronesia Developing a spatially adjusted CPUE for the albacore fishery in the South Pacific. WCPFC-SC7-2011/EB-WP-IP-05

Executive Summary 1. Standardised CPUE is regularly used as an index of abundance to tune the population dynamics models used to assess the status of tuna stocks in the WCPO. The primary purpose of CPUE standardizations is the removal of the variation in the data not attributable to changes in abundance. In the WCPO these indices have typically been generated to meet the model structures of MULTIFAN-CL and fine scale spatial variation is removed as part of the standardisation. As a result these indices have not been applicable for alternative population dynamics models such as SEAPODYM. 2. An adjusted CPUE index with spatial structure was developed for albacore (Thunus alalunga) in the south Pacific, based on operational logsheet data from distant-waters and domestic fleets. 3. To avoid bias due to targeting change, a cluster analysis was carried out for identifying fleets targeting albacore. These data were used in a GLM model to standardize CPUE using year-month, latitude, longitude, vessel, and longitude and latitude aggregated to 5 degrees resolution as factors. The predictor variable explaining most of the logged CPUE variability was the year-month factor (18.9%). The vessel factor explained 10.2% of the variation. The effort correction factor was defined as the exponential of the vessel factor. A total of 1421 vessel coefficients were extracted (p < 0.05 for 1142 of these) and used to generate the adjusted CPUE. This index was built using the correction factor for effort, estimated as the exponential of the estimate of the boat parameter from the GLM. This approach assures the equivalency of effort in all ships from the two fleets included in the analysis. 4. The average of the vessel correction factor was1.39 with a standard deviation of 0.69. Extreme values of the correction factor were defined as those whose value was outside the range of the 95% confidence interval. Rather than remove these data from the analysis we assign them the corresponding values from the limits of the 95% confidence interval. The removal of the influence of these extreme values on the correction factor was required as they increased or decreased the nominal effort to levels that could be considered abnormal. 5. The majority of the effort correction factors were larger than one and their overall effect was to increase the effort and consequently reduce the adjusted CPUE. The results of the temporal annual trend of the nominal and adjusted CPUE show higher …