Monitoring Precipitation Contents and Extinction By Using A Ground-based Passive Microwave Technique
Abstract
In the last decades microwave radiometry has proved to be a valuable tool for retriev- ing atmospheric parameters both from ground-based and from satellite-borne plat- forms. Ground-based microwave radiometry has been mainly investigated for esti- mating temperature, water vapor and cloud liquid profiles in the absence of precip- itation. However, the increasing use of multifrequency radiometers in ground-based stations, especially for communication purposes, rises the question of their potential for retrieving also rainfall rate from ground. This capability might be also useful when weather radars are also present, since radiometric estimates might be used as a further constraint within the radar inversion procedure. One of the main problems of ground- based radiometry for rainfall retrieval is the possible impact of water layers on the receiving antenna whose measurements can be heavily contaminated. The estimate of atmospheric parameters by microwave radiometry may be approached by using both experimental and simulated data. The use of experimental measurements is limited by their scarcity or even their lack. The modeling approach is generally more versatile, even though it requests a thorough insight into the e.m. interaction between the mi- crowave radiation and the scattering medium. The radiative transfer theory has been so far the most used approach to take into account multiple scattering and vertical inho- mogeneity of the atmosphere in the presence of hydrometeor scattering. Following a previous work, the objective of this paper is to investigate about radiometric frequency sets and system configurations best suited for observing both stratiform and convective precipitation. We develop inversion algorithms for ground-based retrieval of surface rainrate, adopting a model-based approach. For both stratiform and convective precip- itation, we use a radiative transfer model (RTM), including spherical liquid, melt and ice hydrometeors. From the solution of the forward models, by varying the model pa- rameters in a physically-based manner, it is derived a synthetic database made by the downwelling simulated brightness temperatures, the associated mean radiative tem- perature and the total path-attenuation. Non-linear inversion algorithms are developed for different sets of frequency channels, observation geometries, cloud types, and pre- cipitation intensities. Examined and compared inversion techniques are both statistical regression estimators and feed-forward neural networks. A special care is devoted to
1 set up robust estimators in presence of a random noise which can be attributed to various sources, first of all effects of possible water-films on the antenna reflector. A maximum a posteriori probability discrimination technique is applied to separate stratiform from convective raining clouds. Simulation results are shown to illustrate the potential of the proposed models by selecting, for this study, a wide range of fre- quencies from 13.0 to 50.2 GHz. As an experimental validation, rain events occurred in Central Italy and observed at the ground-station of ITALSAT geostationary satellite near Rome (Italy), are considered. The ITALSAT station is equipped with 3 satellite beacons at 19.7, 39.6, and 49.5 GHz together with a multichannel radiometer at 13.0, 23.8, and 31.6 GHz, a raingage and a meteorological instrumentation. Results in terms of comparison between measurements and predictions are finally discussed. References Marzano F.S., E. Fionda, and P. Ciotti, SSimulation of radiometric and & cedil;attenuation measurements along earth-satellite links in the 10- to 50- GHz band through horizontally-finite convective raincellsT, Radio Sci., vol. 34, pp. 841-858, 1999. Marzano F.S., E. Fionda, P. Ciotti, and A. Martellucci, SRainfall retrieval from & cedil;ground-based multichannel microwave radiometersT, in Microwave Radiometry and Remote Sensing of the Environment, P. Pampaloni Ed., VSP Intern. Sci. Publisher, Utrecht (The Netherlands), pp. 397-405, 1999. 2- Publication:
-
EGS General Assembly Conference Abstracts
- Pub Date:
- 2002
- Bibcode:
- 2002EGSGA..27.2705M