HIRS-AMTS Satellite Sounding System Test—Theoretical and Empirical Vertical Resolving Power

Abstract

Theoretical and empirical analyses of the vertical resolving power of the High resolution Infrared Radiation Sounder (HIRS) and the Advanced Moisture and Temperature Sounder (AMTS) is carried out. First, the infrared transmittance weighting functions, and associated radiative transfer kernels, are analyzed through singular value decomposition. The AMTS is found to contain several more pieces of independent information than HIRS when the transmittances are considered, but the two instruments appear to be much more similar when the temperature-sensitive radiative transfer kernels are analyzed. The instruments are also subjected to a theoretical analysis using the methods of Backus and Gilbert. From this analysis, it is found that the two instruments should have very similar vertical resolving power below 500 mb but that AMTS should have superior resolving power above 200 mb. In the layer 200?500 mb, the AMTS shows a badly degraded spread function which may or may not be a realistic assessment of vertical resolving power there. An empirical method for assessing vertical resolving power is also developed and compared with theory. This method involves a measure of vertical resolving length determined from retrievals carried out with minimum information and Backus-Gilbert inverse solutions. This test shows that the Backus-Gilbert spread function does not represent the minimum separation of resolvable signals in an atmospheric temperature profile. By the empirical evaluation, the HIRS and AMTS should have similar vertical resolution in the troposphere with vertical resolving length around 2?3 km, quite consistent with theory. The AMTS exhibits a slight advantage over the HIRS above 200 mb, a result which is also qualitatively consistent with theory. In the layer 200?500 mb, the two instruments appear to have very similar resolution with AMTS showing a slight advantage over HIRS. This is totally contrary to theoretical results which indicate that AMTS is very poor in this layer.