Meetings: Documents

Impact of Atmospheric Boundary Layer Stability on Ocean Emissivity at L-Band: Understanding Regional and Seasonal Salinity Biases in Aquarius Data
[11-Apr-2016] Brown, S. and Misra, S.
Presented at the 14th Specialist Meeting on Microwave Radiometry and Remote Sensing of the Environment
The correction for wind-driven rough surface emission is one of the largest corrections that are required for satellite based sea surface salinity (SSS) retrieval. Missions such as SMOS, Aquarius and SMAP use models that describe the excess emission as a function of wind speed to remove this signal from the data in order to accurately estimate the sea water dielectric and hence SSS. Several studies have shown boundary layer stability is a key factor in air-sea interaction, including wave and foam formation and that wind speed alone may not be sufficient to describe the excess emission of the ocean. Analysis of Aquarius salinity biases with respect to in situ probes (ARGO) and an ocean model (HYCOM) have revealed systematic biases in the northern and southern high latitudes that are a half-year out of phase which each other, indicating a probable shortcoming in the physics of one of the correction models. These biases have been shown to be related to atmospheric boundary layer stability by correlating the biases with respect to air-sea temperature difference, a measure of boundary layer stability. The boundary layer is unstable when the atmosphere is colder than the ocean surface and stable when the air is warmer than the surface. There is a strong seasonal and regional dependence of airsea temperature difference, with winter months at high-latitudes having on-average unstable conditions and summer months having on-average stable conditions. Microwave sea surface excess emissivity has been shown to have a significant dependence on air-sea temperature difference at C-band and higher, with sensitivities upward of -0.4 K/C reported at C-band. The main reason is thought to be the influence of boundary layer stability on foam-formation. Monahan and O'Muircheartaigh (1986) show that foam fraction increases by about 10% per degree of air-sea temperature difference at a fixed wind speed and Salisbury et al. (2013), show a decrease in foam fraction for warm SSTs (> 15C). In this paper, we will describe a new ocean surface roughness geophysical model function derived from Aquarius data which explicitly accounts for atmospheric boundary layer stability. It is shown that the residual SSS retrieval error reaches 0.07 psu/delta-C if this correction is not accounted for. We will show an improvement of the Aquarius salinity retrievals after the correction is applied and the reduction of the regional and seasonal salinity differences that are observed in the Version 4 product. This correction will be implemented in the end-of-mission Version 5 data product.