Possible link between interannual variation of neon flying squid (Ommastrephes bartramii) abundance in the North Pacific and the climate phase shift in 1998/1999

Abstract

The relationship between interannual variation in abundance of the autumn cohort of the neon flying squid (Ommastrephes bartramii) and ocean environmental changes in the central North Pacific was examined. We focused on the change in subsurface ocean state during the 1998/1999 climate shift. Changes in catch per unit effort (CPUE) of the neon flying squid derived from long-term driftnet survey was compared to that in ocean environments related to the feeding conditions of the squid. A four-dimensional variational (4D-VAR) ocean data assimilation product was used as an accurate estimate of the dynamic state in the North Pacific. Correlation analysis indicated that the squid CPUE was highly related with the Pacific Decadal Oscillation (PDO) in winter. In January, the correlation field with the entrainment rate (ENT), the proxy for the nutrient-rich water supply entering the mixed layer, showed a good agreement with the main spawning and nursery ground of the autumn cohort (MSNGAC). The nutrient-rich water supply in the MSNGAC in early winter was mainly induced by the deepening of the mixed layer forced by surface latent heat cooling and turbulent mixing, while the basin-scale wind stress curl and the horizontal advection were less affected. These results suggest that the amount of newly supplied nutrient-rich water mass in early winter could affect the primary productivity throughout the winter and the resultant feeding conditions of the juvenile squid. We assumed that this process would determine the stock levels of the neon flying squid in the following summer. We further attempted to reconstruct the changes in neon flying squid CPUE during 1994–2006 by applying regression analysis to several parameters. The result showed that ENT, surface and subsurface temperatures, and the PDO index in February were good predictors for estimating the squid CPUE time series. In addition, the subsurface temperature in the MSNGAC in the preceding autumn was also a good predictor. This indicates that the squid CPUE could be predicted four months earlier than its nursery period, as the same accuracy as that using February data. The accurate prediction in early period could be useful for the squid stock management in the future.