It is known that strong deep convection lifts large amounts of ice to high altitudes, causing strong scattering of the upwelling radiance from below by the ice hydrometeors, leading to depression of the brightness temperature

To improve the current 2A12 rain retrieval algorithm (Kummerow et al. 2001) over land, it is important to evaluate the anomalies in the retrieved product. However, it is difficult to quantify these anomalies globally because there is no single rainfall product that can represent the ground truth. Products from rain gauges have reliable estimation of rainfall only over areas with dense rain gauge networks. Rainfall retrieved from the TRMM PR near surface reflectivity has global coverage between 36 N and 36 S, but suffers from uncertain attenuation correction, problems over complex terrain and the limit of minimum detectable signal (Iguchi et al. 2000). In this section, we start the discussion of the anomalies by comparing the rainfall estimates among products from TRMM TMI (2A12), PR (2A25) and rain gauges.