NASA - National Aeronautics and Space Administration

The response of brightness temperatures at different microwave frequencies to overland precipitation is investigated by using the Tropical Rainfall Measuring Mission (TRMM) Precipitation Radar (PR) and Microwave Imager (TMI) data. The Spearman correlation coefficients between observations at TMI channels or channel combinations and PR‐measured near‐surface rain are computed using 3 years of TRMM data. The results showed that the brightness temperature combinations from 19 and 37 GHz, that is, V19‐V37 (the letter V denotes vertical polarization, and the numbers denote frequency in GHz) or V21‐V37, can explain ∼10% more variance of near‐surface rainfall rate than can the V85 brightness temperature. Also, the global distribution of the above correlation revealed that over almost all of the tropical land area covered by TRMM satellite, the V19‐V37 channel has a closer response to the overland rainfall than does the V85 channel. This result is somewhat counterintuitive, because it has been long believed that the dominant signature of overland rainfall is the brightness temperature depression caused by ice scattering at high microwave frequencies (e.g., 85 GHz). To understand the underlying physics of this better low‐frequency response, data analysis and radiative transfer modeling have been conducted to assess the influence on brightness temperatures from clouds with different ice and liquid water partitions. The results showed that under the condition of low frozen water and medium liquid water in the atmospheric column, the signal from the V19‐V37 channel responded better to rainfall rate than did the one from the V85 channel. A plausible explanation to this result is that in addition to ice scattering signature, the V19‐V37 channel contains liquid water information as well, which is more directly related to surface rain than to ice water aloft. At heavy rainfall conditions, the V19‐V37, V37, and V85 channels all are correlated with near‐surface rain reasonably well, and the V37 or V21 channel becomes the top responder to surface rain as the amount of hydrometeors in the atmospheric column reaches very high values. Additionally, it is found that land surface type and 2 m air temperature have significant skills in characterizing rain cloud types, so that the V19‐V37 channel is more sensitive to surface rainfall for more vegetated warm surface, while the V85 channel is more sensitive to cold bare land. This finding implies that the above two parameters may be used to prioritize satellite observations at different channels, so that the channel that has the best rainfall sensitivity under a given condition receives the highest weight in retrieval algorithms.