Abstract
Sea surface salinity (SSS) satellite measurements are validated using in situ observations usually made by surfacing Argo floats. Validation statistics are computed using matched values of SSS from satellites and floats. This study explores how the matchup process is done using a high-resolution numerical ocean model, the MITgcm. One year of model output is sampled as if the Aquarius and Soil Moisture Active Passive (SMAP) satellites flew over it and Argo floats popped up into it. Statistical measures of mismatch between satellite and float are computed, RMS difference (RMSD) and bias. The bias is small, less than 0.002 in absolute value, but negative with float values being greater than satellites. RMSD is computed using an “all salinity difference” method that averages level 2 satellite observations within a given time and space window for comparison with Argo floats. RMSD values range from 0.08 to 0.18 depending on the space–time window and the satellite. This range gives an estimate of the representation error inherent in comparing single point Argo floats to area-average satellite values. The study has implications for future SSS satellite missions and the need to specify how errors are computed to gauge the total accuracy of retrieved SSS values.
Highlights
In this work we looked at the way sea surface salinity (SSS) is sampled by remote sensing and how that sampling is validated by comparison with in situ data at level 2 (L2)
We focused on the Aquarius and Soil Moisture Active Passive (SMAP) missions in generating the L2 comparison values from the model
These missions provide a relatively straightforward sampling pattern and footprint size. Further work in this area will focus on Soil Moisture and Ocean Salinity (SMOS) using a similar set of methods
Summary
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Since 2009, three satellite missions have been launched to measure sea surface salinity (SSS), Soil Moisture and Ocean Salinity (SMOS) from the European Space Agency, Aquarius from NASA/SAC-D and Soil Moisture Active Passive (SMAP) from NASA These missions utilize sun-synchronous polar orbits with high inclinations, but differing spatial and temporal resolutions, and have provided continuous SSS measurement coverage of the global ocean [1]. In a couple of validation studies done using L2 satellite measurements and the ASD method [8,9,13], Argo float measurements are compared with averages of many L2 satellite values (Table 1), but there is no testing of how differences may depend on the spatial or temporal search window from which these averages are computed. There may be a space–time window to use for doing validation studies that minimizes the mismatch Such a result using real in situ and satellite data was found by [25].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
