Spatial and temporal variability of satellite-derived sea surface temperature in the southwestern Atlantic Ocean

Abstract

Abstract Mean fields, seasonal cycle, and interannual variability of the Southwestern Atlantic Ocean (SAO) sea surface temperature (SST) distribution are examined during an 18-year (January 1985–December 2002) period by means of monthly composite, around 9 km resolution satellite-derived SST data. Temporal evolution of the mean spatial value shows a very marked annual cycle, certain interannual variability and a weak increasing tendency (∼0.06 °C/decade). Spatial distribution of the mean temporal value shows the influence of the horizontal advection over the SST field. Over 90% of the temporal variability is explained by the annual harmonic, which reaches amplitudes of 7 °C over the shallowest zones. The inclusion of the semiannual harmonic improves the description of the seasonal cycle only in those regions of the domain where warming and cooling cycles show different durations. The strong correlation between the surface heat flux and the SST allows, by means of a single diffusion equation, the prediction of the shelf areas which do not stratify seasonally. The EOF analysis shows three modes (which together explain over 97% of the variance) associated with the stationary value, the amplitude and the phase of the annual harmonic, i.e., with the seasonal cycle. The fourth mode (0.22% of the variance) is the first one to show interannual variability, which is well correlated with the Southern Annular Mode (SAM).