Study On Integrated Habitat Index For Albacore Tuna (Thunnus Alalunga) In The High Seas Of The Southeast Pacific Ocean

Albacore tuna（Thunnus alalunga）is one of the most productive tuna species and an important part of tuna fishery resources.It is widely distributed in temperate and subtropical waters of world oceans.The research on the abundance of albacore tuna resources in the south Pacific Ocean is the main task of the West and Central Pacific Fisheries Commission（WCPFC）and Inter-American Tropical Tuna Commission（IATTC）.China also pays more attention to the sustainable utilization of albacore tuna resources in the southeast Pacific Ocean,and it is particularly important to understand the impact of marine environmental variables on the distribution of albacore tuna resources.Based on the fisheries data and hook depth data of 31 stations surveyed in the high seas of the southeast Pacific Ocean from October 1,2022 to December 23,2022,this paper establishes a hook depth calculation model and obtains the actual Catch Per Unit Fishing Effort CPUE（unit:individuals/thousand hooks）of albacore tuna.Using cluster analysis,combined with environmental factors such as 40～320m water layer,temperature,salinity,chlorophyll a concentration,and dissolved oxygen concentration,the range of preferred environmental variables for the albacore tuna to inhabit in the high seas of the southeast Pacific Ocean are obtained.The K-S test（Kolmogorov-Smirnov test）and P-P plot are used to test whether the CPUE obeys the normal distribution,and the correlation between each environmental variable and CPUE is analyzed.The VIF method is used to remove the multicollinearity environmental variables.Quantile Regression（QR）and Gaussian Mixture Model（GMM）are used to establish prediction models for CPUE of albacore tuna in each water layer and the whole water body,respectively.VAR model is used to evaluate the prediction results of the two models.The integrated habitat index（IHI）is established,and the environmental data of the verification site are input into the two models to calculate the CPUE.The significant difference between the predicted CPUE and the actual CPUE of each water layer and the whole water body obtained by the two methods is tested by the Wilcoxon test.Then,by calculating the Spearman correlation coefficient between the predicted IHI and the actual CPUE of each water layer and the whole water body of the site,the prediction ability of the two methods to the CPUE of each water layer and the whole water body is analyzed.After comparison,it is concluded that QR prediction ability is higher than GMM.Finally,the spatial distribution of albacore tuna is predicted by IHI model（QR）.The results show that:（1）In the calculation model of longline hook depth in the southeast Pacific high seas,the factors affecting the hook depth are wind speed,drift angle and hook number.All three factors are significantly correlated with the hook depth ratio（P values are0.00,0.00,0.00,respectively）.（2）The most preferred water layer for the habitat of albacore tuna in the high seas of the southeast Pacific Ocean is 160～240 m;the optimum salinity range is 35.19～35.45;the optimum chlorophyll a concentration is 0.034～0.088μg/L;the optimum dissolved oxygen concentration is 3.45～6.25mg/L;（3）The IHI_QRestablished by quantile regression method has better prediction ability in 200～240 m water layer,and the Spearman correlation coefficient is 0.703.The prediction ability of 80～120m,160～200m and 240～280m water layers are good,and the Spearman correlation coefficients are 0.512,0.538 and 0.502 respectively.The IHI_GMMestablished by the GMM model has good predictive ability in the 160～200 m and 200～240 m water layers,and the Spearman correlation coefficients are 0.521 and0.688,respectively.The QR（Spearman correlation coefficient is 0.557）is better than GMM（Spearman correlation coefficient is 0.504）for the prediction of the entire water body.（4）For the QR and GMM,the environmental variables affecting the distribution of albacore tuna are mainly temperature and salinity.Based on the VAR model,temperature and salinity have the stronger effect on the actual CPUE,dissolved oxygen concentration has a strong effect on the actual CPUE,and chlorophyll a concentration has the least effect on the actual CPUE.It is indicated that the most important factors affecting the actual CPUE of albacore tuna in the southeast Pacific Ocean are temperature and salinity.（5）The high IHI index distribution area of albacore tuna predicted by QR model is14°～19°S,96°～106°W。...