ABSTRACT

This chapter describes an analysis of the rainfall spatial variability in a small area with a high-density rain gauge network. Radar rainfall estimations were compared and evaluated with the rain gauge data. Statistical measurements of discrete and continuous validation scores were calculated for the radar estimates at hourly and daily time step. Probabilistic distribution functions were calculated for the Bias with the purpose of knowing the rainfall uncertainty over a small area. The hit rate indicates the occasions when the categorical estimation correctly determined the occurrence of rainfall event or non-event. The probability of detection (POD) is the likelihood that the event would be estimated by the radar, increasing with the time step, with the daily data. For the hourly time step, the Pixel 1 POD was higher than the POD for Pixel 2 and the average of 4 Multisensor Precipitation Estimates pixels.