An Investigation of the Relationship between Emission and Scattering Signals in SSM/I Data

Abstract

To provide guidance for the development of satellite microwave rainfall-retrieval algorithms, the basic relationships between emission and scattering signals in natural clouds must be understood. In this study, the relationship between two parameters observed from microwave satellite data—the polarization difference at 19 GHz D and the polarization-corrected temperature PCT—is investigated over the global ocean on a monthly and 5° (lat) × 5° (long) mean basis. Using data from January and July 1993, the occurrence frequencies and latitudinal variation and horizontal distribution of the D–PCT relationships are investigated. The D–PCT slope is studied by dividing the entire weather range into three regimes: nonprecipitation, light precipitation, and heavy precipitation. The analysis shows that small variation of PCT in the nonprecipitation regime could be achieved by employing a variable coefficient in the PCT definition equation. The slopes in the light precipitation regime are latitude dependent. Although the interpretation is inconclusive, it is felt that the differences in the fractional coverage and the rain layer depth in different latitudes is responsible for the latitudinal dependence. No clear latitudinal dependence of slopes in the heavy precipitation regime is found. The connection of the D–PCT relationship to the performances of an emission-based and a scattering-based rainfall algorithm are investigated using the Second WetNet Precipitation Intercomparison Project rainfall cases. The results of this study emphasize the necessity of incorporating the scattering signal in rainfall rate retrieval algorithms. Additionally, the D–PCT slope information can be used to help categorize precipitation types, which may be useful in determining the specific algorithm best used for a certain precipitation type and/or regime.