Intercomparison of Error Characteristics Across Remote Sensing Salinity Products in the Gulf of Mexico, a Rver-Influenced System
[27-Aug-2018] Vazquez-Cuervo, J., Fournier, S., Dzwonkowski,B., and Reager, J.T. .
Presented at the 2018 Ocean Salinity Science Team and Salinity Continuity Processing Meeting
Four satellite derived sea surface salinity (SSS) datasets are evaluated in the Gulf of Mexico (GoM), three from NASA's Soil Moisture Active Passive (SMAP) mission and one from the European Space Agency's Soil Moisture Ocean Salinity (SMOS) mission. The first two SMAP products are from Remote Sensing Systems (REMSS) with different resolutions of 40 km and 70 km, while the third, a 40 km product, is produced by the Jet Propulsion Laboratory. All products are compared against in situ
data from buoys and the World Ocean Database (WOD) in the GoM, a coastal/semi-enclosed basin strongly influence by river discharge. All datasets reproduce the intraseasonal and seasonal variabilities associated with freshwater river plumes. Comparisons with 7 buoys indicate that Root Mean Square Error (RMSE) can be as low as 0.2 pss at the seasonal time scale. Collocated SSS and in situ
observation from the WOD were aggregated by distance from the coast to identity spatial variability in product performance. Within 100 km from land RMSE reached 1.0 pss for all the products. Greater than 100 km from land biases decreased to approximately 0.1 pss. Similar to the time series evaluation, the products generally performed well with RMSE between 0.2 and 0.5 pss and biases between 0.1-1.0 pss. Clear spatial structure in product performance was observed with the quality of the products decreasing as the coast was approached.