Validation of AIRS/AMSU/HSB retrieved products

We describe preliminary comparisons of AIRS/AMU/HSB retrieved geophysical products with correlative data sets to constrain retrieval uncertainties. The results are relevant to the 70% of oceanic retrieval footprints within the latitude range from 40S to 40N where infrared retrievals are completed. Comparisons are further limited to those retrievals whose sea surface temperatures (SST) agree with forecast model SST to within +/-3 K. We present here comparisons with forecast model assimilations and dedicated radiosondes. Retrieved cloud cleared radiances and those calculated from weather forecast model output agree within 0.5 to 3 K, depending on cloud amount. Retrieved sea surface temperatures at night are compared against model output, with a resulting difference of 0.94 +/-0.95 K (a result skewed by the +/-3 K selection criterion). Retrieved temperature profiles are compared with model output, and with dedicated radiosondes. Temperature profile uncertainties vary from about 1.3 K just above the surface to less than 1 K in the troposphere. Total water vapor is compared against dedicated radiosondes. Under dry conditions retrieved total water vapor agrees with radiosonde total water to within 10%, with small biases. The current retrieval algorithm generates temperature profiles meeting the 1 K per km requirement of the AIRS system.