西北太平洋柔鱼渔场非参数栖息地适宜性指数模型

Nonparametric Habitat Suitability Index Model for Ommastrephes bartramii Fishing Ground in the Northwest Pacific Ocean

【目的】建立西北太平洋柔鱼(Ommastrephes bartramii)渔场非参数的栖息地适宜性指数(HSI,habitat suitability index)模型,为提高柔鱼渔场预报精度提供参考。【方法】利用西北太平洋海域的柔鱼生产数据和海表温度、叶绿素a浓度、表温梯度强度以及水深100 m的Argo浮标水温等海洋环境因子的适宜度指数值(SI,suitability index),使用随机分类粒子算法,构建非参数的栖息地适宜性指数模型,估算西北太平洋柔鱼渔场各环境因子在不同月份中适宜度指数系数的空间分布。【结果】随着空间位置不同,各月份不同环境因子适宜度指数系数分布具有较大差异,其中SST的SI系数的高值区大面积出现在11月份,叶绿素浓度SI系数的高值区出现8月份南部渔场和10月份北部渔场,次表层(100 m)温度的高值SI系数在7-9月份均有较大面积分布,而Grad的SI系数在整个渔汛期均处于较低水平。基于该模型,本研究对2014年7-11月渔获产量进行预报检验,预报结果与实际渔场在空间分布一致,曲线下面积(AUC,Area Under Curve)月平均为0.791,高于算术平均HSI(arithmetic mean HSI,AM HSI)模型。【结论】非参数的HSI模型的渔场预报能力在一定程度上优于AM HSI模型。

【Objective】To establish the nonparametric Habitat Suitability Index(HSI)model of neon flying squid(Ommastrephes bartramii)and provide a reference for improving the forecast accuracy of fishing ground in the Northwest Pacific.【Method】Based on the production data of neon flying squid and marine environmental factors including sea surface temperature,chlorophyll-a concentration,surface temperature gradient intensity and Argo buoy water temperature at 100 m depth underthe sea surface,nonparametric HSI model was constructed and the spatially distributed suitability index(SI)coefficients of environmental factors were estimated from July to November with the help of random particle classification algorithm.【Result】SI coefficients of environmental factors varied considerably in different seasons and regions.Among them,high SI coefficients of SST occupied a large area in November and high SI coefficients of chlorophyll-a concentration in the southern fishing area in August and northern fishing area in October.High SI coefficients of T100 were distributed in a large area from July to September.In the meanwhile,SI coefficients of Grad were at a low level during the whole fishing season.It showed that the predicted fishing grounds in 2014 from July to November were consistent with the spatial distribution of the production data.The monthly average Area Under Curve(AUC)value of the nonparametric HSI is 0.791 and higher than that of arithmetic mean HSI(AM HSI).【Conclusion】The fishing ground forecast capacity of the non-parametric HSI model of neon flying squid fishing ground is superior to that of the AM HSI.