Covariability of maritime surface wind divergence and vorticity extremes on synoptic timescales

Variability in the surface divergence and vorticity fields is investigated from 10 years of QuikSCAT and ASCAT-A scatterometer wind observations and ERA-Interim, ERA5, and NCEP CFSRv2 reanalyses wind fields over the global oceans. The probability distribution functions of these derivative wind fields are characterized by highly asymmetric distributions within mid-latitudes, skewed toward strong convergent and cyclonic wind events. Extremes of convergence and cyclonic vorticity more than 2 standard deviations from the mean are well correlated in the mid-latitude storm tracks, with cross-correlation coefficients greater than 0.5 in scatterometer winds and 0.8 in reanalysis winds. Earlier dynamical studies predicted a strong anti-correlation between divergence and vorticity associated with frontogenesis in extratropical cyclones, consistent with these observations. These studies also suggested that the spatial phasing of convergence and cyclonic vorticity extremes along atmospheric frontal boundaries in models are sensitive to boundary layer dissipation, which may explain the discrepancies in divergence-vorticity coupling noted between observations and reanalyses. The strong covariability of the wind divergence and vorticity also extends to extremes in the wind stress divergence and curl. This coincidence of extreme negative wind stress divergence and curl on synoptic time scales may have implications for time-dependent Ekman pumping in the ocean.