CPUE Standardization of Longline Fishing Thunnus Alalunga in South Pacific Ocean

Abstract：Albacore Thunnus Alalunga is a highly migratory species and one of the main target species in the world fisheries with great economic value and abundant resources.Based on the longline albacore fishery data collected by Western and Central Pacific Fisheries Commission (WCPFC) from 2000 to 2012,in combination with the environment data retrieved from remote sensing website,generalized linear models (GLM) and generalized additive models (GAM) were used to standardize the catch per unit effort (CPUE) of T.alalunga.The GLM modeling indicated that there are eleven variables influencing the CPUE greatly,which included months of year,latitude,sea surface temperature (SST) with the horizontal gradient (GSST),sea surface salinity (SSS),sea surface height (SSH),eddy kinetic energy (EKE),the concentration of dissolved oxygen in 100-metre depth,and interaction between year and latitude,month and latitude,and month and longitude.According to the Akaike Information Criterion (AIC),we selected the optimal GAM in which the maximum deviance explained for CPUE was 58.9％,and the highest cCoefficient of determination between independent variable and dependent variable was 58.4％.The GAM standardized CPUE had the same trend as the GLM standardized,and the two standardized CPUEs were obviously lower than or closely to the norminal CPUE except in year 2011.The inter-annum and inter-month variation of T.alalunga density varied in a large range in South Pacific Ocean from 2000 to 2012.T year;the highest value of density appeared from June to August,in 2001.T;the standardized average CPUE of GLM and GAM were 11.48 inds./khooks,12.41 inds./khooks and 13.54 inds./khooks and 18.63 inds./khooks,respectively.GAM tended to be more appropriate than GLM for such standization as was found in this study.