Seasonality and spatial dependence of meso- and submesoscale ocean currents from along-track satellite altimetry
Along-track wavenumber spectral densities of sea surface height (SSH) are estimated from Jason-2 altimetry data as a function of spatial location and calendar month, to understand the seasonality of meso- and submesoscale balanced dynamics across the global ocean. Regions with significant mode-1 and mode-2 baroclinic tides are rejected, restricting the analysis to the extratropics. Where balanced motion dominates, the SSH spectral density is averaged over all pass segments in a region for each calendar month, and is fit to a 4-parameter model consisting of a flat plateau at low wavenumbers, a transition at wavenumber $k_0$ to a red power law spectrum $k^{-s}$, and a white spectrum at high wavenumbers that models the altimeter noise. The monthly time series of the model parameters are compared to the evolution of the mixed layer. The annual mode of the spectral slope $s$ reaches a minimum after the mixed layer deepens, and the annual mode of the bandpassed kinetic energy in the ranges $[2k_0,4 k_0]$ and $[k_0,2 k_0]$ peak $\sim$2 and $\sim$4 months, respectively, after the maximum of the annual mode of the mixed layer depth. This analysis is consistent with an energization of the submesoscale by a winter mixed layer instability followed by an inverse cascade to the mesoscale, in agreement with prior modeling studies and in situ measurements. These results are compared to prior modeling, in situ, and satellite investigations of specific regions, and are broadly consistent with them within measurement uncertainties.