Presented at the 2015 Aquarius/SAC-D Science Team MeetingSalinity microwave remote sensing is extremely challenging because there are many corrections required to obtain the smooth ocean surface brightness temperature from which SSS is derived. The largest error source is the surface warming due to oceanic winds, and the AQ/SAC-D baseline approach uses the AQ-scatterometer (Scat) measurement of ocean radar backscatter to infer L-band excess ocean emissivity associated with wind roughness. This paper describes a robust MWR ocean roughness correction technique and presents results of associated SSS retrievals (with MWR corrections applied) using the AQ baseline SSS retrieval algorithm. Further, comparisons are made between: the MWR based salinity retrievals, the AQ L-2 V3.0 SSS product, and the Hybrid Coordinate Ocean Model (HYCOM) surface salinity. Results are presented, which demonstrate that the two SSS retrievals (Scat and MWR roughness corrections) are nearly identical over low to moderate wind speeds, but they diverge slowly at higher wind speeds. Since MWR and Scat corrections are statistically independent, the possibility exists to reduce random errors by combining these data sets before SSS retrieval. SSS retrievals are performed using the simple average of Scat and MWR roughness corrections, and results of HYCOM comparisons for this approach show promising improvements at high wind speeds.