Application of Gradient-Dependent Optimal Interpolation in Fishery Analysis of Neon Flying Squid (Ommastrephes bartramii) in the Kuroshio–Oyashio Confluence Region

Abstract

A key issue in fishery forecasting is the collection of high-precision subsurface environmental data. A data assimilation method, named gradient-dependent optimal interpolation, was used to construct the near-real-time vertical temperature and salinity structure of a squid fishery ground based on Argo observations. The results were verified by truth-finding comparisons and applied to analyze the relationship between neon flying squid and the subsurface environment in the Kuroshio–Oyashio Confluence Region. The temperature and salinity differences between the constructed results and survey data were less than ±0.5 °C and ±0.02, respectively. Most of the relative analysis errors were less than the observational errors. Statistical analysis revealed that the most suitable temperature for squid was 18–24 °C at the near-surface (<5 m), although the squid can endure a temperature range from 11 to 12 °C at a depth of 300 m. There was an obvious thermocline in the fishery ground, with a thermocline depth of 65 m and a mean strength of approximately 0.10 °C/m. The regressive relationship between vertical temperature (thermocline parameters) and squid catch per unit effort (CPUE) followed the exponential (Gaussian) function. The most suitable salinity was 33.0–34.2 at depths shallower than 300 m.