Is the Tropical Zonal Bias in Aquarius-Argo Salinity Comparisons Due to the Difference in Measurement Depth?

Abstract:

We compare Aquarius 3.0 sea surface salinity with Argo upper mixed layer salinities, using both the standard Aquarius level-2 product and the Combined Active Passive (CAP) algorithm. Both algorithms show a zonal bias that increases with latitude in either hemisphere; however, the negative bias observed in the tropics is significantly reduced in the CAP version. In previous work using collocated TRMM precipitation data, we showed that the mean freshening bias due to rain-induced salinity stratification between the typical Argo measurement depth of 2-7 m and the Aquarius penetration depth of 0-2 cm is limited to less than 0.03 PSU, which is not enough to explain the tropical zonal bias seen in the standard algorithm [Drucker and Riser, 2014]. The reduction of the tropical bias with CAP is consistent with this finding and suggests that the zonal biases are likely due to retrieval error rather than the difference between Aquarius and Argo measurement depths.