We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
基于 ARIMA 的海洋尼诺指数对中西太平洋黄鳍金枪鱼年际 CPUE 的影响.
- Authors
王 啸; 刘文俊; 张 健
- Abstract
As a highly migratory pelagic fish, yellowfin tuna (Thunnus albacares) has high ecological and economic value. The Western and Central Pacific Ocean (WCPO) is the sea area with the highest tuna production of all oceans. In order to understand and predict the response of yellowfin tuna to climate change at different life stages in WCPO, we used the catch data of yellowfin tuna in purse seining and pelagic longlining and Oceanic Niño index (ONI) data from 1990 to 2020 in the WCPO to validate the applicability of general ARIMA (Autoregressive integrated moving average) model and dynamic ARIMA model, so as to explore the influence of the ONI on the interannual CPUE (Catch per unit effort) of yellowfin tuna. The results show that: 1) General ARIMA models could be used for long-term fitting of annual CPUE of yellowfin tuna in the WCPO, taking full account of the variability characteristics of annual CPUE of yellowfin tuna. 2) Compared with the general ARIMA model, the dynamic ARIMA model provided a better fit and a higher correlation between the fitted and true values, as well as smaller mean absolute and root mean square errors. 3) The influence of the ONI on the annual CPUE of yellowfin tuna differed between the northern and southern equatorial regions of the WCPO, with the ONI being a more critical factor and a better model fit relatively north of the equator. 4) The ONI had different impacts on the annual CPUE of yellowfin tuna in different fisheries in the WCPO, with a 1–2 years' lag in the ONI for the yellowfin tuna longline fishery in the WCPO, and a faster impact on the purse seine fishery during strong El Niño and strong La Niña events, without a lag
- Subjects
BOX-Jenkins forecasting; YELLOWFIN tuna; EL Nino; STANDARD deviations; TUNA fisheries
- Publication
South China Fisheries Science, 2023, Vol 19, Issue 4, p10
- ISSN
2095-0780
- Publication type
Article
- DOI
10.12131/20230007