印度洋北部鸢乌贼CPUE标准化初步研究

Preliminary standardization of Sthenoteuthis oualaniensis in northern Indian Ocean

本研究根据中国远洋渔业协会鱿钓技术组和公海围拖网技术组提供的2017—2019年印度洋北部鸢乌贼3种捕捞方式(灯光敷网、灯光罩网和鱿钓)的生产统计数据,结合时空因子(年、月、经纬度)和环境因子(海表温度、海表盐度、光合有效辐射、风速和流速),利用广义线性模型(GLM)找出主要影响因子,并以广义可加性模型(GAM)找出因子变化规律并进行CPUE标准化研究。结果表明:GLM模型分析得出,年、月、纬度、海表温度、风速、流速以及作业方式因子均为显著性变量,对CPUE的影响较大。将上述的显著性影响因子加入GAM模型,根据AIC准则,包含这7个因子的GAM模型的AIC值最小为最优模型,对CPUE的总偏差解释为21.6%,其中作业方式对CPUE的影响最大,解释了5.6%的总偏差,随后依次是年、纬度、月、海表温度、流速、风速。名义CPUE与标准化CPUE年间变化趋势一致,且月间变化趋势整体较为相似。研究表明,基于GLM模型和GAM模型标准化的CPUE能够较全面、真实地反映印度洋鸢乌贼资源密度的变化情况,其中作业方式、海表温度、流速和风速对印度洋鸢乌贼资源密度影响较大。因此在后续的渔情预报研究中,可以将这些因子综合考虑进渔情预报模型中,以提高预测精度。

According to the catch data of three fishing types(light lift net, light casting net and jigging) of Sthenoteuthis oualaniensis in northern Indian Ocean during 2017—2019 provided by Squid Jigging Technology Group of China Overseas Fisheries Association and High Seas Seine and Trawl Technology Group, spatial-temporal factors(year, month, latitude, and longitude) and environmental factors(sea surface temperature, SST;sea surface height, SSH;sea surface salinity, SSS;photosynthetically active radiation, PAR;wind speed, WS;and current speed, U) were analyzed. The main influencing factors are identified by using generalized linear model(GLM), and a generalized additivity model(GAM) is used to find out the variation pattern of factors and perform a CPUE standardization study. The results showed that the year, month, latitude, SST, WS, U and fishing types are all significant variables that significantly influence CPUE. Adding the factors of this significant effect to the GAM model, according to the AIC guidelines, the AIC value of the GAM model containing these seven factors is the minimum optimal model, and the total deviation of CPUE is interpreted as 21.6%, of which the fishing types has the most significant impact on CPUE, explaining the total deviation of 5.6%, followed by year, latitude, month, SST, U, WS. Nominal CPUE are consistent with standardized CPUE trends over the years, and the monthly trends are generally similar. The results showed that the CPUE based on GLM model and GAM model standardization can reflect the change of the resource density of Indian Ocean squid more comprehensively and realistically, among which the fishing types, SST, U and WS factors have a significant influence on the resource density of Indian Ocean squid. Therefore, in the subsequent fishery forecasting research, these factors can be considered into the fishery forecast model to improve the accuracy of prediction.