Remotely sensing cloud properties from microwave radiometric observations by using a modeled cloud database

Abstract

Abstract. As a first step for remote sensing cloud properties, a database of cloud genera has been established. This is derived from a microphysical model, and it considers the statistical profiles of four hydrometeor species for each cloud genus. From this database the corresponding radiative database is obtained making use of a radiative transfer model, so for each cloud genus the simulated microwave response at the special sensor microwave imager channels is found. The cloud and radiative databases allow the retrieval of the genera of the cloud and other relevant properties from satellite observations. An automatic cloud genus classifier has also been implemented. Several tests have been carried out, and the results are presented.

Neural-network approach to ground-based passive microwave estimation of precipitation intensity and extinction

“…Finally, it should be mentioned that in the last decade spaceborne radiometric techniques have been successfully applied to the rainfall retrieval on a global scale (e.g., Petty, 1995;Pierdicca et al, 1996;d'Auria et al, 1998). Microwave radiometry has proved to be a fairly accurate tool, even though its major drawback is represented by the poor spatial resolution (i.e., order of tens of kilometers) and low temporal sampling .…”

Potential of combined spaceborne infrared and microwave radiometry for near real‐time rainfall attenuation monitoring along earth‐satellite links

“…The decrease of high-frequency ¹ values for a cumulonimbus is mainly due to the graupel ice scattering. Analogous "gures can be shown for ¹ simulations over ocean, putting in evidence that while for a nimbostratus the surface e!ects are still signi"cant, for a cumulonimbus the results are almost equal to those over land due to the strong cloud opacity [24]. Moreover, it should be noted that the ocean surface is radiometrically cold so that for a nimbostratus low ¹ values at 19 GHz can be observed for moderate rainfall.…”

“…A critical aspect in determining these accuracies is the choice of the proper training data set. For what concerns the limitations of microwave radiometry in rain detection and estimation, sensors like SSM/I best capture convective-type rain over land where there is signi"cant ice scattering [24]. SSM/I ground resolution of the order of tens of kilometers is also a problem when observing cumulonimbus clouds which can have a horizontal size less than 10 km, even when organized in multicells or supercells [37].…”

“…cumulonimbus, nimbostratus, cirrus, etc. ), the cloud data set has been extended by means of a Monte Carlo statistical procedure, based on the use of the correlation matrix of the hydrometeor contents [19,24].…”

“…The double peak in the ice-content histogram is due to the modal values of crystal (non-precipitating) and graupel (precipitating) ice, the graupel one being the largest in terms of ice content. In order to associate ¹ 's to each cloud structure, a radiative transfer model, based on the fast and fairly accurate Eddington solution, has been used [24]. Within each layer, the temperature has been supposed to be linearly dependent on the height, and the gaseous absorption has been computed by means of the Liebe model.…”