基于GLBM模型的中国大陆阿根廷滑柔鱼鱿钓渔业CPUE标准化

西南大西洋阿根廷滑柔鱼既是西南大西洋生态系统中的重要种类,也是鱿钓渔业的重要捕捞对像.单位捕捞努力量渔获量(CPUE)是表示渔业资源状况及其丰度的常用指标.根据2000-2010年中国大陆鱿钓船在西南大西洋的生产统计数据和海洋卫星遥感获得的海洋环境数据(表温,表温水平梯度,海面高度,叶绿素浓度),利用基于贝叶斯的广义线性模型(GLBM),分未加入固定交互选项、加入固定交互选项和加入随机交互选项3种情况对中国大陆西南大西洋阿根廷滑柔鱼鱿钓渔业的CPUE进行标准化.根据偏差信息准则(DIC)值最小来确定最佳贝叶斯模型.结果表明,包含纬度、海表温度、表温水平梯度、海面高度、月×纬度、月×经度及年×纬度变量且加入随机交互项的GLBM模型为最适.标准化后的CPUE较名义CPUE小,年间变化平缓.与广义线性模型(GLM)和广义加性模型(GAM)标准化的CPUE比较,GLBM模型更能反映其资源丰度的真实水平.研究认为,2001-2010年间经GLBM模型标准化后的CPUE呈现逐年下降的趋势.

CPUE Standardization of Illex argentinus for Chinese Mainland squid-jigging fishery based on generalized linear Bayesian models

Illex argentinus is not only one of the important species in the marine ecosystem, but also a fishing target of the most important squid-jigging fishery in Southwest Atlantic Ocean. Catch per unit effort (CPUE) is an important index for fishery abundance, and the standardization of CPUE is an important content in the fishery stock assessment. Based on the catch data from Chinese mainland squid jigging fishery in the southwest Atlantic ocean and marine environmental factors (sea surface temperature, SST; horizontal gradient of sea surface temperature, GSST; sea surface height, SSH) derived from ocean satellite during 2000 to 2010, the Generalized Linear Bayesian models (GLBM), including the interaction terms excluded, fixed effects interaction terms and random effects interaction term, are used to standardize the CPUE for the Chinese Mainland squid-jigging fishery of I. argentinus in the southwest Atlantic Ocean, and the best model is selected based on the lowest DIC (Deviance Information Criterion). The results indicated that the models with random effects interaction term including the variance of latitude, SST, GSST, SSH, month×longitude, year×latitude and month×latitude had the best fit for the Chinese mainland squid-jigging fishery of I. argentinus in the southwest Atlantic Ocean. The CPUE standardized by GLBM is smaller and fluctuated lower than the nominal CPUE. Compared with the result by generalized linear models (GLM) and generalized additive models (GAM), the GLBM seemed to be best for standardization of CPUE for the Chinese Mainland squid-jigging fishery of I. argentinus in the southwest Atlantic Ocean, and our result showed that the standardized CPUE of I. argentinus has been decreased during 2000 to 2010.