基于NODE-GAM模型的太平洋中部大眼金枪鱼CPUE时空分布及其与环境因子的关系  

Spatio-temporal Distribution of CPUE of Central Pacific Bigeye Tuna and Its Relationship with Environmental Factors Based on NODE-GAM Model

作　　者：周淑婷 邹晓荣[1,2] 李东旭 ZHOU Shuting;ZOU Xiaorong;LI Dongxu(College of Marine Living Resource Sciences and Management,Shanghai Ocean University,Shanghai 201306,China;National Engineering Research Center for Oceanic Fisheries/Key Laboratory of Sustainable Exploitation of Oceanic Fisheries Resources,Ministry of Education/Scientific Observing and Experimental Station of Oceanic Fishery Resources,Ministry of Agriculture and Rural Affairs,Shanghai 201306,China)

机构地区：[1]上海海洋大学海洋生物资源与管理学院,上海201306 [2]国家远洋渔业工程技术研究中心/大洋渔业资源可持续开发省部共建教育部重点实验室/农业农村部大洋渔业资源环境科学观测实验站,上海201306

出　　处：《广东海洋大学学报》2025年第1期55-61,共7页Journal of Guangdong Ocean University

基　　金：远洋渔业高效捕捞与船载加工一体化智能装备技术服务项目(D-8006-23-0016)。

摘　　要：【目的】研究太平洋大眼金枪鱼(Thunnus obesus)的时空分布及不同环境因子对大眼金枪鱼单位捕捞努力量渔获量(CPUE)影响的规律,为大眼金枪鱼延绳钓的科学生产及渔业管理等提供参考。【方法】根据2018―2021年太平洋延绳钓作业数据,结合表层、100、200、300 m海水温度,海水表层盐度和海表层叶绿素a质量浓度等同期海洋环境数据,使用广义可加性神经遗忘决策集成模型(NODE-GAM)分析太平洋大眼金枪鱼CPUE时空分布及对环境因子的响应规律。【结果与结论】太平洋大眼金枪鱼渔获数量和CPUE值在年际分布和月度分布上的差异均很大。整体来看,太平洋中部大眼金枪鱼盛渔期为9―12月,低潮期为1―8月;2018―2021年间,渔场范围有逐年向东向南扩散的趋势,其中高产渔场区主要集中在赤道两侧、10°N―10°S之间的广阔海域。环境因子对大眼金枪鱼CPUE的相对重要性指数从大到小依次为200 m海水温度、100 m海水温度、300 m海水温度、海水表面温度叶绿素a质量浓度和海水表层盐度。NODE-GAM模型的均方误差为0.085,较广义可加性模型、随机森林和反向传播神经网络分别降低了63.0%、43.3%和29.2%。【Objective】To study the spatiotemporal distribution of Pacific bigeye tuna(Thunnus obesus)and the influence of different environmental factors on the catch per unit effort(CPUE)of bigeye tuna,providing references for scientific longline fishing and fishery management of bigeye tuna.【Method】According to the Pacific longline fishingdata from 2018―2021,combined with the marine environmental data such as the surface,100,200,and 300 m seawater temperatures,sea water surface salinity and sea surface chlorophyll a mass concentration of the same period,using Neural Oblivious Decision Ensembles for Generalized Additive Models(NODE-GAM),We analyzed the spatial and temporal distribution of CPUE of Pacific bigeye tuna and its response pattern to environmental factors.【Result and Conclusion】Pacific bigeye tuna catch and CPUE values varied greatly in both interannual and monthly distributions.Overall,the peak fishing season for central Pacific bigeye tuna is from September to December,and the low season is from January to August;during the period of 2018―2021,the fishery range has a tendency to spread from east to south year by year,in which the high yield fishery area is mainly concentrated on both sides of the equator,10°N―10°S.The environmental factors on CPUE of bigeye tuna,in descending importance order,were 200 m seawater temperature,100 m seawater temperature,300 m seawater temperature,seawater surface temperature chlorophyll a mass concentration and seawater surface salinity.The mean square error of the NODE-GAM model is 0.085,which is 63.0%,43.3%and 29.2%lower than that of the generalized additive model(GAM),the random forest model and the backpropagation neural network,respectively.

关 键 词：大眼金枪鱼 单位捕捞努力量渔获量 时空分布 环境因子 太平洋 NODE-GAM 

分 类 号：S977[农业科学—捕捞与储运]