A Comparison Of Integrated Habitat Index Models For Ablacore Tuna(thunnus Alalunga) In Waters Near Cook Islands

Albacore tuna(Thunnus alalunga) is one of the important catch species in world tuna fisheries, and it’s also one of the important fisheries resources in the deep-sea tuna longline fisheries of China. So it is necessary to study in depth to achieve efficient and sustainable use of the resources. In order to quantify its’ abundance and resource distribution status better, we use the environment data and catch data in high resolution to predict effectively the integrated habitat index(IHI) model of albacore tuna. Because of the close relationship between the distribution of the resources and environmental factors, it is necessary to apply the related environmental factors to the study of predicting CPUE and integrated habitat index for albacore tuna.In this study, we used the environment data(including sea surface temperature, sea surface height, primary productivity, sub water temperature in 150 m and their interactions) and the corresponding albacore tuna catch data(including the operation time, position, the number of being capture and the number of hooks of each day) obtained in April, 2014 to build the predicting model of catch per unit fishing effort(CPUE) by generalized additive model(GAM), quantile regression model(QRM) and support vector machine(SVM), then we estimated the predicting CPUE, calculate the integrated habitat index of three models and evaluate whether it is significant correlation with the measured CPUE. We compared the predicting ability of three models. In addition, we input the environment data in the same style obtained in May, 2014, to the three models, then we got the corresponding predicting CPUE, and we evaluated whether it was significant correlation with the measured CPUE by Wilcoxon rank test, and determined the best model.The results showed that:1) The IHI value of three models in each area was different. The predication ability of GAM model was the best in area 1, the predication ability of SVM model was the best in area 3, while the predication ability of QRM model was the best both in area 2 and the entire area.2) In the entire area, the higher value of IHI was defined in 10o00′S~12o30′S, 157o00′W~159o30′W by GAM model, the higher value of IHI was defined in 10o00′S~12o30′S, 156o30′W~159o30′W by QRM model, while the higher value of IHI was defined in 10o30′S~12o30′S, 157o00′W~159o00′W by SVM model.3) The predicting CPUEs, which output from three models by inputting 20 test grids of environment data obtained in May, 2014, had significant relationship, with the measured CPUEs by Wilcoxon rank test, except GAM model. The predicting model based on QRM was the best. The model based on SVM was the second.The model based on GAM was relatively weak because of the non significant relationship in area 3.4) The key environmental variables, which influence the albacore tuna’s distribution in each area, were different from three models. The key environmental variable of GAM was sea surface height. The key environmental variables of QRM were sea surface temperature, sea surface height, primary productivity, sub water temperature in 150 m and their interactions. The key environmental variables of SVM were sea surface temperature and sea surface height.5) The model’s characteristic was different from each other. The predicting model based on GAM was good at choosing the key environmental variables. The predicting model based on QRM had the highest predicting ability among three models. The predicting model based on SVM was good at predicting the central fishing grounds.