The radiative consistency of Atmospheric Infrared Sounder and Moderate Resolution Imaging Spectroradiometer cloud retrievals

The consistency of cloud top temperature (T-C) and effective cloud fraction (f) retrieved by the Atmospheric Infrared Sounder (AIRS)/Advanced Microwave Sounding Unit (AMSU) observation suite and the Moderate Resolution Imaging Spectroradiometer (MODIS) on the EOS-Aqua platform are investigated. Collocated AIRS and MODIS T-C and f are compared via an "effective scene brightness temperature" (T-b,T-e). T-b,T-e is calculated with partial field of view (FOV) contributions from T-C and surface temperature (T-S), weighted by f and 1-f, respectively. AIRS reports up to two cloud layers while MODIS reports up to one. However, MODIS reports T-C, T-S, and f at a higher spatial resolution than AIRS. As a result, pixel-scale comparisons of T-C and f are difficult to interpret, demonstrating the need for alternatives such as T-b,T-e. AIRS-MODIS T-b,T-e differences (Delta T-b,T-e) for identical observing scenes are useful as a diagnostic for cloud quantity comparisons. The smallest values of Delta T-b,T-e are for high and opaque clouds, with increasing scatter in Delta(Tb,e) for clouds of smaller opacity and lower altitude. A persistent positive bias in Delta T-b,T-e is observed in warmer and low-latitude scenes, characterized by a mixture of MODIS CO2 slicing and 11-mu m window retrievals. These scenes contain heterogeneous cloud cover, including mixtures of multilayered cloudiness and misplaced MODIS cloud top pressure. The spatial patterns of Delta T-b,T-e are systematic and do not correlate well with collocated AIRS-MODIS radiance differences, which are more random in nature and smaller in magnitude than Delta T-b,T-e. This suggests that the observed inconsistencies in AIRS and MODIS cloud fields are dominated by retrieval algorithm differences, instead of differences in the observed radiances. The results presented here have implications for the validation of cloudy satellite retrieval algorithms, and use of cloud products in quantitative analyses.