HR: 0800h
AN: A51B-0754    [Abstracts]
TI: Validation of Aqua AIRS/AMSU/HSB Precipitation Estimates
AU: * Chen, F W
EM: fwchen@jansky.mit.edu
AF: Massachusetts Institute of Technology, Research Laboratory of Electronics, 77 Massachusetts Ave., Room 36-413, Cambridge, MA 02139-4307 United States
AU: Staelin, D H
EM: staelin@mit.edu
AF: Massachusetts Institute of Technology, Research Laboratory of Electronics, 77 Massachusetts Ave., Room 36-413, Cambridge, MA 02139-4307 United States
AB: This paper compares Aqua AMSR-E retrievals of precipitation rates with concurrent AIRS/AMSU/HSB estimates based on the ``opaque-channel'' algorithm described by Chen and Staelin (IEEE Transactions on Geoscience and Remote Sensing, 41, 2, pp. 410-417). This algorithm uses the opaque channels in the oxygen and water vapor absorption bands in addition to four window channels and was trained using NEXRAD data over the eastern U.S. The training set consisted primarily of data from convective precipitation. Its systematic tendency to overestimate summer precipitation rates at high latitudes was quantified, permitting correction of the initial opaque-channel estimates. The climate-dependent biases determined for $5^\circ\times5^\circ$ cells were applied to the initial 15-km opaque-channel retrievals using a neural network trained with AMSR-E data, and the results were validated using 15-km resolution NEXRAD data at midlatitudes and global AMSR-E statistics. For this purpose, the climate was characterized by latitude and by the three dominant principal components representing the atmospheric temperature profile channels. The corrected retrieval results agree well with the concurrent validation data available on a global scale.
DE: 3300 METEOROLOGY AND ATMOSPHERIC DYNAMICS
DE: 1640 Remote sensing
SC: Atmospheric Sciences [A]
MN: 2004 AGU Fall Meeting