利用多种空间插值法分析西北太平洋柔鱼渔场

Analysis of fishing ground for Ommastrephes bartramii in the Northwest Pacific based on different statistical interpolation methods

柔鱼（Ommastrephes bartramii）是中国远洋鱿钓渔业的重要经济种类之一。以2012年7月-10月柔鱼单位渔船日捕捞产量（CPUE）作为渔场分布的指标,利用地统计学中全局多项式、局部多项式、径向基函数、反距离权重和普通克里格等5种插值方法对柔鱼渔场空间分布进行分析与比较。研究表明,普通克里格插值法结果为最优,其他依次为反距离权重插值法、径向基函数插值法、局部多项式插值法、全局多项式插值法;以普通克里格插值法,对7月-10月各月柔鱼CPUE进行预测,获得7月-10月柔鱼中心渔场随月份的迁移分布情况;与实际作业位置以及CPUE比较发现,各月实际和预测CPUE均值相差不大,各月CPUE预测值在0.51-1.95t·d^-1,误差控制在0.15 t·d^-1以下。研究认为,普通克里格插值法在柔鱼中心渔场预报中具有较好的可靠性,同时结合海洋环境因子对柔鱼渔场分布变化进行了讨论。

Ommastrephes bartramii is one of the most important economic species caught by Chinese squid jigging fleets. Based on the abundance index（ CPUE,catch per day in one fishing vessel） of O. bartramii from July to October in 2012,we applied five statistical interpolation methods（ global polynomial,local polynomial,radial basis function,inverse distance weighted and ordinary Kriging） to analyze the fishing ground of O. bartramii in the Northwest Pacific. The results show that ordinary Kriging was the best method for analyzing CPUE distribution,followed by inverse distance weighted,radial basis function,global polynomial and local polynomial. The optimal method of ordinary kriging was considered as an example to predict the CPUE distribution from July to October,so as to obtain the change of monthly fishing ground distribution. Little difference was found between actual CPUE and predicted CPUE; the mean monthly CPUEs varied from 0. 51 t·d^-1to 1. 95 t·d^-1and their absolute errors were less than 0. 15 t·d^-1. Finally,we discussed the distribution of fishing ground and its variation by combining with marine environmental factors. It is indicated that ordinary Kriging in-terpolation method was suitable for forecasting the fishing ground of O. bartramii.