南极半岛南极磷虾栖息地适应性研究

南极磷虾(Euphausia superba)作为南极生态系统中的关键物种，其时空分布研究对发展磷虾渔业和理解南大洋生态系统均有重要的作用，相关研究有利于磷虾种群的资源评估以及中心渔场的探测。本文利用海表温度（Sea Surface Temperature，SST）、海面高度（Sea Surface Height，SSH）、海表面叶绿素（Sea surface chlorophyll，SSC）、海冰面积覆盖(Sea Ice Coverage，SIC）等环境因子，分别采用神经网络拟合和一元非线性拟合方法，并结合最小值法、最大值法、连乘法、算术平均法、几何平均法、加权算术平均法等构建了磷虾栖息适宜指数 (Habitat Suitability Index，HSI) 模型。结果表明，神经网络模型预报结果更符合磷虾实际栖息分布情况，而一元非线性拟合预测结果较为连续。最大值法和最小值法计算结果差异较大，容易引进较大的误差。连乘法的预测效果较好，算术平均法、几何平均法和加权算术平均法的预测结果相似，且较为稳定。

Habitat suitability of Antarctic krill (Euphausia superba) in the Antarctic Peninsula

As the keystone species in the Antarctic ecosystem, the study on spatiotemporal distribution of Antarctic krill (Euphausia superba) provides an important role for development of krill fishery and understanding of the Southern Ocean ecosystem. Such studies are helpful for stock assessment of krill population and detection of key fishing ground. To this end, using environmental variables, such as Sea Surface Temperature (SST), Sea Surface Height (SSH), Sea surface chlorophyll (SSC), and Sea Ice Coverage (SIC), based on Neural Network (NN) model and Univariate Nonlinear (UN) model with combining with six algorithms, i.e., minimum, maximum, continued multiplication, arithmetical average method, geometric method, and weighted arithmetic averaging method, this study constructs the habitat suitability index (HSI) model of krill in the Antarctic Peninsula. The results show that, compared to UN model, the prediction performance of NN is better for describing the habitat distribution of krill, but the habitat of krill is continuous for the prediction of UN model. The large difference occurs in the results between minimum and maximum methods, which could introduce a large bias. The prediction performance of continued multiplication is better than other methods. The results of arithmetical average, geometric, and weighted arithmetic averaging methods are similar and stable.