[18-Feb-2020] Schanze, J.J., Springer, S.R., Thompson, E.J., Lagerloef, G.S.E, and Schmitt, R.W.
Presented at the 2020 Ocean Sciences MeetingIn this study, we leverage concurrent salinity and meteorological observations from the Salinity Processes in the Upper Ocean Regional Study - 2 (SPURS-2) field program in a high-resolution model of upper ocean stratification following precipitation. Rain-induced freshwater lenses are prevalent features in the inter-tropical convergence zone (ITCZ), monsoon areas, and high latitudes, with 100 freshwater lenses identified in 40 days of in situ observations. A polarized radar provides information of precipitation and the wind field, the fields of which are used to force Regional Ocean Modelling System (ROMS) simulations at a horizontal resolutions of 100m at 150 levels. Here, we use shipboard measurements of freshwater lenses gathered during two SPURS-2 to study the physical processes leading to the creation and dissipation of surface salinity stratification by precipitation. In addition to regular shipboard thermosalinograph measurements at 5 m depth, the Underway Salinity Profiling System provides measurements at 3 m and 2 m, and the 'salinity snake' instrument continuously samples the top 1-2 cm. These observations are leveraged to force a suite of ROMS models to elucidate the processes present for different cases of freshwater lenses. The models are forced using ancillary information from ship-borne polarimetric Doppler precipitation radar. We are thus able to generate statistical information about both the prevalence and the persistence of haline surface stratification in the ITCZ. Our results indicate that rain-induced stratification can persist for tens of hours in low wind, where mixing is reduced. Since near-surface stratification can influence air-sea fluxes, we investigate these impacts in the model. Based on these findings and the general statistics of the prevalence and persistence of freshwater lenses, we suggest how the effects of rain-induced stratification may be parameterized in global models.