Mesoscale wind stress-SST coupled perturbations in the Kuroshio Extension

Abstract

Mesoscale perturbations of wind stress (τMS) and sea surface temperature (SSTMS) in the Kuroshio Extension (KE) are analyzed using long-term high-resolution satellite observations. Mesoscale wind stress and SST perturbations are first extracted using a locally weighted regression (loess) method, and then analyzed using statistical methods, including linear regression, Singular Value Decomposition (SVD), and the inverse method. The coupling coefficient between mesoscale wind stress magnitude |τ|MS and SSTMS, and those between mesoscale wind stress divergence (curl) and downwind (crosswind) SST gradients, exhibit distinct seasonal variability, with large values in winter and small values in summer, consistent with previous studies. The three leading SVD modes for |τ|MS and SSTMS show highly consistent variability in both spatial patterns and temporal expansion coefficients, and the temporal expansion coefficients exhibit distinct seasonal and interannual variability; these modes are associated with inherent seasonal variability of |τ|MS and SSTMS and interannual variability of the KE jet dynamic state. Based on the observed coupling relationships between wind stress divergence (curl) and downwind (crosswind) SST gradients, an empirical model for mesoscale wind stress vector perturbations (τx, τy)MS = F(SST) is established; this model solves (τx, τy)MS from their divergence and curl estimated from SST gradient data using an inverse method. This newly established (τx, τy)MS = F(SST) model produces reasonably consistent solutions compared to the one established based on the empirical relationship between |τ|MS and SSTMS. Furthermore, this model could separate divergence-only and curl-only wind stress perturbations to study their respective effects, and thus is promising and will contribute to ocean dynamic analyses.