Lagrangian reconstruction of sea surface salinity to extract small-scale variability from SMAP

As the resolution of observations and models improves, emerging evidence indicates that ocean variability associated with submesoscale features (1-10 km) is of fundamental importance to ocean circulation, air-sea interaction, and biogeochemical cycling. Recent work has shown that, in many regions, salinity variability dominates over temperature variability in forming submesoscale and mesoscale density fronts. The resolution of current L-band satellite radiometers (~40 km) does not allow for the observation of submesoscale features in sea surface salinity (SSS). The present work is motivated by the need to (i) develop techniques to exploit satellite salinity measurements in order to extract signals with the highest possible spatial resolution; and (ii) better characterize surface density variability, and in particular the role of salinity, across the submesoscale to mesoscale range (1-100 km).

In this study we investigate the reconstruction of SSS fields at ~5 km scales from 25 km resolution gridded observations from the Soil Moisture Active Passive (SMAP) satellite. We apply a Lagrangian reconstruction method that consists of advecting the SMAP SSS field using currents from satellite altimetry. We compare the reconstructed, high-resolution SSS field with in situ thermosalinograph data to demonstrate the improvement of the advected field compared to the original low-resolution field, focusing on the representation of fronts. Two regions with different dynamics are considered: (i) the California Current system dominated by the coastal upwelling and eddy formation, and (ii) the Gulf Stream, with a dominance of advection of tracers.