基于灰色系统的西北太平洋秋刀鱼资源丰度预测

Predicting the Abundance of Pacific Saury Based on Grey System in the Northwest Pacific

【目的】建立西北太平洋秋刀鱼资源丰度预报模型,为其资源的可持续开发与管理提供参考依据。【方法】根据2000—2013年西北太平洋秋刀鱼(Cololabis saira)渔业数据及海洋环境数据,包括产卵场与索饵场表面温度(sea surface temperature, SST)、太平洋年代气候震荡(Pacific Decadal Oscillation,PDO)和厄尔尼诺指数(Trans-NinoIndex,TNI),建立西北太平洋秋刀鱼资源丰度灰色预报模型。【结果】比较建立的5种灰色系统模型的相对误差和相关系数,选择最优预报模型。最优模型结果表明,影响西北太平洋秋刀鱼资源丰度的环境因子重要性依次为TNI、PDO、产卵场SST、索饵场SST。【结论】基于产卵场SST、PDO、TNI的GM(1,4)模型,相关系数高达0.7598且平均相对误差为36.64%,可作为西北太平洋秋刀鱼资源丰度预报模型,能够预报其资源状况变动,建议使用该模型指导渔业企业生产。

【Objective】To establish a model for predicting abundance of saucy in the Northwest Pacific Ocean, and provide suggestions for its sustainable development and management. 【 Method 】According to the Cololabis saira fishery data and environmental data, including sea surface temperature(SST), the Pacific Decadal Oscillation(PDO), the El-Nino Index(TNI) during 2003-2013 in the Northwest Pacific Ocean, we constructed a grey model to predict the abundance of Pacific saury. The optimal model was confirmed among five models by relative error and correlation coefficient. 【Result】The results derived from the selected model showed that the order of importance of affecting the abundance of Pacific saury for environmental factors were: TNI, PDO, SST on spawning ground, and SST on feeding ground. 【Conclusion】The optimal model was GM(1, 4) with correlation coefficient = 0.7598 and relative error = 36.64%, and its factor were SST on spawning ground, PDO and TNI. It can be used to predict the abundance of Cololabis sairs and guide the operation in the Northwest Pacific Ocean.