Rain-Induced Near Surface Salinity Stratification and Rain Roughness Correction for Aquarius SSS Retrieval

The effect of rain on surface salinity stratification is analyzed to develop a rain roughness correction scheme to reduce the uncertainty of Aquarius sea surface salinity (SSS) retrieved under rainy conditions. Rain freshwater inputs may cause large discrepancies in salinity measured by Aquarius at 1-2 cm within the surface and the calibration reference SSS from HYCOM (SSSHYCOM) a few meters below the surface. We used the rain impact model (RIM) to adjust SSSHYCOM to reflect near surface salinity stratification caused by freshwater inputs accumulated from rain events that occurred over the past 24 h before Aquarius measurements (SSSRIM). When calibrated with SSSRIM, the residuals, i.e., the difference between measured and model predicted brightness temperature TB, are considered as rain-induced roughness. It was found that rain-induced roughness is larger at lower wind speeds, and decreases as wind increases. The Combined Active Passive algorithm is used to retrieve SSS with (SSS
_{CAP_RC}
) or without (SSS
_CAP
) rain roughness correction. We find that the simultaneously retrieved wind speed with rain roughness correction has significantly improved agreement with the NCEP wind speed with the rain-dependent bias reduced, self-justifying our rain correction approach. SSS retrieved is validated with salinity measured by drifters at a depth of 45 cm. The difference between satellite retrieved and in situ salinity increases with rain rate. With rain-induced roughness accounted for, the difference between satellite retrieval and drifter increases with rain rate with slope of -0.184 psu (mm h
^-1
)
^-1
, representing the salinity stratification between the two depths (1-2 cm versus 45 cm).