Rain Rate Retrieval Using the 183-wsl Algorithm

One of the main open scientific challenges lasting since a few decades is the compelling need for a substantial improvement of techniques adopted to retrieve precipitation intensities and their duration in time. This is particularly pressing for those precipitation episodes that interest very small areas, where also light rains are often trigger disastrous flooding events. Note that an accurate knowledge of the terrain features such as morphological and orographic aspects is crucial. In this context, satellite sensors with their large number of spectral bands and their wide spatial coverage can be easily employed for a global research and monitoring strategy. In particular, the increasing spatial resolutions of the new passive microwave (PMW) sensors and the constantly increasing number of orbiting platforms are providing data at an unprecedented temporal frequency. Usually, PMW estimation algorithms for the retrieval of rain rates or of other atmospheric parameters (e.g., Grody et al. 2000, Ferraro et al. 2005) are based on two different approaches: scattering and absorption. Scattering methods infer rain intensities by exploiting the brightness temperature depression due to the frozen hydrometeors located in the upper part of the cloud. The scattering signal (or index) is strictly linked to the probability of melting ice crystals into the clouds and their conversion into rain droplets. As demonstrated by Bennartz et al. (2002) the measure of the scattering signature can be efficiently used to select and classify rain rate intensities distinguishing conditions ranging from norain to heavy rain. The quantitative comparison with co-located radar data shows the robustness of the technique in classifying heavier precipitation characteristic of a convective system, where the amount of scatterers is relatively large. The second approach is founded on the water vapor properties to absorb and successively emit the radiation centered on specific absorption bands (Grody 1991, Staelin et al. 1999, 2000). As to this method, the retrieval at low and high frequencies must be considered differently. If the retrieval is carried out at low frequencies, i.e. around the water vapor band at 22.235 GHz, the scattering is weak and the absorption dominates the extinction. Nevertheless, Wang et al. (1989) note that this absorption band provides only enough sensitivity to measure the total column water vapor and the high emissivity background masks the atmospheric contribution over land. On the other hand, at higher frequencies the scattering effect increases also into the strong absorption bands such as at 183.31 GHz. The presence of cold clouds can depress the brightness temperature particularly for the frequencies located farther from the center of the absorption line (Greenwald et al. 2002, Burns et al. 1997). In this work, we report the results of two new fast algorithms over land and ocean to estimate rain rates using PMW opaque frequencies of the Advanced Microwave Sounding Unit module B (AMSU-B) of the National Oceanic and Atmospheric Administration (NOAA). Our choice of using these frequencies is mainly founded on considerations reported in section 2 and other studies not reported here. In section 3 ___________________________________