Recent Advances in the Salinity Retrieval Algorithms for Aquarius and SMAP (Poster)
[16-Dec-2016] Meissner, T. and Wentz, F.J.
Presented at the 2016 AGU Fall Meeting
"Our presentation discusses the latest improvements in the salinity retrievals for both Aquarius and SMAP since the last releases. The Aquarius V4.0 was released in June 2015 and the SMAP V 1.0 was released in November 2015. Upcoming releases are planned for SMAP (V 2.0) in August 2016 and for Aquarius (V 5.0) late 2017.
The full 360o look capability of SMAP makes it possible to take observations from the forward and backward-looking direction at the same instance of time. This two-look capability strongly aids the salinity retrievals. One of the largest spurious contaminations in the salinity retrievals is caused by the galaxy that is reflected from the ocean surface. Because in most instances the reflected galaxy appears only in either the forward or the backward look, it is possible to determine its contribution by taking the difference of the measured SMAP brightness temperatures between the two looks. Our result suggests that the surface roughness that is used in the galactic correction needs to be increased and also the strength of some of the galactic sources need to be slightly adjusted. The improved galaxy correction is getting implemented in upcoming Aquarius and SMAP salinity releases and strongly aids the mitigation of residual zonal and temporal biases that are observed in both products.
Another major cause of the observed zonal biases in SMAP is the emissive SMAP mesh antenna. In order to correct for it the physical temperature of the antenna is needed. No direct measurements but only a thermal model are available. We discuss recent improvements in the correction for the emissive SMAP antenna and show how most of the zonal biases in V1.0 can be mitigated.
Finally, we show that observed salty biases at higher Northern latitudes can be explained by inaccuracies in the model that is used in correcting for the absorption by atmospheric oxygen. These biases can be decreased by fine-tuning the parameters in the absorption model."