Comparing Satellite Salinity Retrievals with In Situ Measurements: A Recommendation for Aquarius and SMAP

The validation of sea surface salinity (SSS) retrievals from satellite measurements through
comparisons with in situ data is important for evaluating the retrieval, improving the underlying
geophysical model functions, and producing reliable estimates of measurement uncertainty.
The evaluation of in situ and satellite data is complicated by the different spatial and temporal
characteristics of the two measurements: satellite observations from Aquarius and SMAP are
continuous and global but with spatial resolution on the order of 50 – 100 km, while in situ data
(e.g. from Argo floats) are point measurements in both space and time. Although satellite data
are global, it may take several days to cover the entire globe. Hence, the probability of an exact
match in time and space between an in situ measurement and a satellite overpass is small.
Here, we define a meaningful definition of a matchup between satellite and in situ
measurements, primarily through the example of Aquarius/SAC‐D, SMAP, and Argo float data.
The analysis suggests that useful matchups between satellite and in situ salinity are obtained
by averaging all satellite observations within a space‐time window with a radius of 50 km and
± 3.5 days centered on the in situ report. This document is partitioned into two parts: In
Section I, the background of satellite salinity remote sensing is used to develop a rationale for
the chosen match‐up criteria, while Section II explores examples using Aquarius/SAC‐D and
SMAP in sufficient detail to generate a match‐up algorithm.