This presentation examines the relative merits of precipitation estimation from different types of satellite sensors and different satellite orbit configurations, in terms of both current and future spacecraft technology. At the outset, some of the basic physics and properties of modern space sensors are reviewed, to help establish the basis for why we should proceed simultaneously with precipitation retrieval in both the microwave and optical-infrared spectrums. Examples of precipitation retrieval based on Special Sensor Microwave/Imager (SSM/I) measurements and METEOSAT infrared measurements for the Valtellina-1987, Genova-1992 (see Figure 1), and Piemonte- 1994 extreme flood events are presented to help place these concepts into an application oriented framework. A brief overview is given of how an SSM/I rain-profile retrieval algorithm developed at the Institute of Atmospheric Physics in Frascati, Florida State University in Tallahassee, and the University of Rome, based on a combined cloud-radiative transfer model, is being used to develop the physical underpinnings for a transfer calibration procedure which can be applied to lower fidelity rain retrievals from METEOSAT infrared image sequences. The merits of this combined approach within the framework of a mesoscale model-satellite rain prediction-monitoring system applied to flood-producing mountain storms is addressed in the companion paper by E.A. Smith, A. Mugnai, and G.J. Tripoli. A brief review is also presented concerning the respective advantages and disadvantages of high inclination and low inclination low-altitude spacecraft orbits versus high-altitude geosynchronous spacecraft orbits, vis a vis the problem of rainfall monitoring. This discussion considers: a), the sun-synchronous U.S. Defense Meteorological Satellite Program (DMSP) satellites which carry the SSM/I passive microwave instruments; b), the non-sun-synchronous NASA/NASDA Tropical Rainfall Measuring Mission (TRMM) satellite (to be launched in late 1997) which will carry a microwave radiometer, a visible-infrared imager, and a precipitation radar, but will be in a limited coverage 35 degree inclination orbit; c), the future polar platform MetOp of the European Space Agency (ESA), which will probably carry the Multifrequency Imaging Microwave Radiometer (MIMR); and d), ESA's Earth Explorer Mission on Precipitation, which is presently under study. High resolution aircraft radiometer measurements from the NASA Advanced Microwave Precipitation Radiometer (AMPR) and the Millimeter Wave Imaging Radiometer (MIR) obtained in February, 1993 during the Tropical Ocean Global Atmosphere - Coupled Ocean- Atmosphere Response Experiment (TOGA-COARE), are used to demonstrate the inherent correlation (actually anti-correlation) between precipitation features in the low and high microwave frequency range. These results are used to advance arguments on why a high frequency microwave radiometer mounted on a geostationary satellite, would be a useful tool for monitoring precipitation, based on the rain signatures emanating from upper levels within the cloud column. The presentation concludes with remarks on why, given the current spacecraft technology, it is important to combine optical-infrared spectrum measurements from geosynchronous satellites such as METEOSAT, with cm-mm spectrum measurements such as from SSM/I. Such a strategy takes advantage of the fact that geosynchronous measurements, although not directly responsive to precipitation signals, address the fundamental space and time scales of precipitation, while the passive microwave measurements respond directly to precipitation microphysics, and although not available on a continuous basis, can be used periodically with transfer-calibration procedures to improve geosynchronous rain estimates.

Figure 1: Comparison of rainfall estimates using the 1555 UTC SSM/I overpass and the closest METEOSAT thermal infrared image (1551 UTC) for the 27 September 1992 Genova flood event. Top panels are the SSM/I 85.5 GHz vertically polarized image (left) and the corresponding METEOSAT 11.5 micron image (right). Bottom-left panel shows the SSM/I-based rainfall estimate using the rain-profile retrieval algorithm developed in Frascati-Rome-Tallahassee, whereas the bottom-right panel shows the METEOSAT-based estimate using a modified Negri-Adler-Wetzel technique. Bottom panels represent enlarged views of the areas delimited by the black squares in the top panels. Raingage measurements are superimposed as black symbols. The images are displayed in Mercator projection. [Taken from V. Levizzani, F. Porc—, F.S. Marzano, A. Mugnai, E.A. Smith, and F. Prodi: Investigating SSM/I Microwave Algorithm to Calibrate METEOSAT Infrared Instantaneous Rainrate Estimates. In press on Meteorological Applications.]