There is a present need for accurate, high-resolution global maps of the surface heat flux for the forcing and evaluation of numerical models of the ocean and atmosphere. One approach being applied is to use a combination of bulk flux models and input data collected from satellite measurements. For estimation of the latent heat flux, the most difficult input to retrieve accurately from satellites is the near-surface specific humidity. Initial work estimated the specific humidity on longer time scales through relations to the total column water vapor content. More recent work has explored direct correlations with brightness temperature measurements from the SSM/I satellite sensor.

This work explores the potential for improved satellite-based estimates of the specific humidity over the oceans using a combination of microwave imager and sounder data. Recently launched SSM/T-2 and AMSU-B microwave sounders have improved performance and increased potential for performing retrievals near the surface. An extensive set of matchups has been constructed between direct measurements of the near-surface humidity from research ships and measured brightness temperatures from the SSM/I, SSM/T-2, and AMSU-B sensors. The database covers a wide range of geographic regions and environmental conditions. The brightness temperature measurements from both individual sensors and combinations of the sensors were regressed against the corresponding specific humidity measurements to derive new retrieval algorithms. The results demonstrate that clear improvements over the existing accuracy of SSM/I only specific humidity retrievals are possible using a combination of SSM/I and SSM/T-2 data. This paper will summarize these comparisons and the accuracy of the resulting products. The potential spatial and temporal resolution of the product is also demonstrated.