PRECIPITATION ESTIMATION THROUGH SATELLITE SYSTEM OVER THE MAJOR MEDITERRANEAN ISLANDS

Reliable and accurate precipitation measurement or estimation is crucial for disaster monitoring and water resource management. Nowadays the scientific community expects significant improvements in precipitation monitoring by the continuous technological evolution of satellite-rainfall estimate systems which are able to produce data with global coverage and thus can provide low-cost information even in scarcely populated areas or places where for economic reasons ground measures are missing. The most recent satellite mission is the Global Precipitation Measurement (GPM), which is an international constellation of ten partner satellites, and a source of rainfall estimates at high spatial and temporal resolution (Huffman et al., 2017). In this context, the aim of this study is to test satellite-precipitation GPM-IMERG products against data provided by dense raingauges over the two major islands of Mediterranean Sea, i.e. Sardinia and Sicily. The two islands are characterized by a complex morphology and by small spatial scale and long see-land transition borders. Moreover, they experience different precipitation types, originated by convective and stratiform systems as well as by the interaction of steep orography in the coasts with winds carrying on humid air masses from the Mediterranean Sea. For these reasons, the two islands can be considered as interesting test sites for satellite-precipitation GPM-IMERG product in the European mid-latitude area and in general for complex domains. The GPM-IMERG post real-time “Final” run product (last version V4 released in spring 2017) at 0.1° spatial resolution and half-hour temporal resolution has been selected for the two years period 2015-2016. Evaluation and comparison of the selected product are performed with reference to data provided by the raingauges network of the two islands. Both GPM and raingauges data have been aggregated at hourly time scale, then the raingauges data have been spatially interpolated and resampled at the GPM resolution grid. In order to obtain general information about the performances of estimates related to the entire two islands, features of rainfall spatial distribution and the influence of the aggregation time scale have been investigated using statistical and graphical tools.