基于物种分布集成模型的布兰斯菲尔德海峡南极磷虾栖息地研究  

Habitat of Antarctic krill(Euphausia superba)in the Bransfield Strait based on ensembled species distribution model

作　　者：陈洵子 朱国平[1,2,3,4] CHEN Xunzi;ZHU Guoping(College of Marine Sciences,Shanghai Ocean University,Shanghai 201306,China;Center for Polar Research,Shanghai Ocean University,Shanghai 201306,China;Polar Marine Ecosystem Laboratory,Ministry of Education Key Laboratory of Sustainable Exploitation of Oceanic Fisheries Resources,Shanghai Ocean University,Shanghai 201306,China;National Engineering Research Center for Oceanic Fisheries,Shanghai Ocean University,Shanghai 201306,China)

机构地区：[1]上海海洋大学海洋科学学院,上海201306 [2]上海海洋大学极地研究中心,上海201306 [3]上海海洋大学,大洋渔业资源可持续开发教育部重点实验室极地海洋生态系统研究室,上海201306 [4]上海海洋大学,国家远洋渔业工程技术研究中心,上海201306

出　　处：《水产学报》2022年第3期390-401,共12页Journal of Fisheries of China

基　　金：国家重点研发计划(2018YFC1406801);国家自然科学基金(41776185);中科院国际合作重点项目(183611KYSB20200059)。

摘　　要：南极磷虾为南大洋海洋生态系统中的关键物种,但目前关于该物种栖息地研究仍较为有限,这不利于对该物种分布的理解以及资源管理。实验利用南极磷虾资源密度数据与海表面温度、海表面叶绿素、海表面高度、净初级生产力和季节性海冰覆盖等5个环境因素,通过12种算法的集成模型对南极磷虾栖息地进行分析。结果显示,支持向量机、随机森林、多元自适应回归样条曲线、分类和回归树、多层感知机及递归分区和回归树等6种算法的性能优于其他算法;而尽管生物判别分析、主域分析以及最大似然估计模型的性能各月间有所不同,但整体上表现较差。其次,年际环境变化与南极磷虾潜在栖息地的时空模式之间存在一定的联系。研究表明,南极磷虾适宜栖息地主要位于布兰斯菲尔德海峡中部。本研究通过集成模型对南极磷虾栖息地的预测有助于南极磷虾资源的养护,探究南极磷虾的最佳渔场和未来可能成为重要渔区的空间。Species Distribution Models(SDM)are important tools for exploring future changes in global biodiversity.However,relatively few studies have applied multi-model approaches to assess biogeographic changes globally,especially in marine environments.Antarctic krill(Euphausia superba)is a key species in the Southern Ocean marine ecosystem,but researches on its habitat are still limited,which hampers the understanding of its distribution and resource management.To this end,based on the resource density data of Antarctic krill and five environmental factors including Sea Surface Temperature,Sea Surface Chlorophyll,Sea Surface Height,Net Primary Productivity and Sea Ice Coverage,through ensembled models of 12 algorithms,we analyzed the habitat of Antarctic krill.Our results showed that,of the 8 algorithms,the Support Vector Machines,Random Forests,Multivariate Adaptive Regression Splines,Classification and Regression Trees,Multi-Layer Perceptron,and Recursive Partitioning and Regression Trees consistently outperformed the others;while the Bioclim,Domain,and Maxlike had poor performance,although they also differed among months.Besides,there was a link between interannual environmental variability and the spatial and temporal patterns of potential habitat for Antarctic krill.The results showed that the suitable habitat for Antarctic krill is mainly located in the central Bransfield Strait.The prediction of Antarctic krill habitat by ensembled model is helpful for the conservation of Antarctic krill resources,and for the exploration of the best fishing ground for Antarctic krill and the space of important fishing area in the future.

关 键 词：南极磷虾 栖息地 集成模型 模型性能 布兰斯菲尔德海峡 

分 类 号：S931[农业科学—渔业资源]