Presented at the 2014 AGU Fall MeetingSea-surface temperature (SST) plays an important, and yet to be fully understood, role in sea-surface sailinty (SSS) retrievals. The Version-3 release of Aquarius/SAC-D salinity retrievals applied an empirically derived adjustment to SSS that is a function of SST. This adjustment was derived after noticing regional salinity biases relative to modeled and in situ salinity observations. These SSS biases correlate well with climatological SST maps. While the ΔSSS(SST) adjustment has already been implemented in the ADPS standard processing, there is great value in determining the physical basis of this bias adjustment. Understanding the root causes of this adjustment will enable improved Aquarius's salinity retrievals, as well as ensure that no true SSS-SST correlations or variability are being removed by the adjustment. There are several factors contributing to the ΔSSS(SST) adjustment, which may be classified into direct, correlated, and indirect effects. Direct effects include the dependence of the sea-water dielectric constant on SST and the dependence of the roughness model on SST. Correlated effects include a small mis-modeling of the atmospheric oxygen absorption, which correlates with SST. Indirect effects include other errors sources such as reflected galactic radiation and RFI that produce ocean-basin-wide anomalies that are being misinterpreted as an SST effect. This presentation will demonstrate how the ΔSSS(SST) adjustment was derived, as well as present our findings regarding the physical basis of this adjustment.