Detection of Drought, Flood and Snow Anomalies with 37GHz Passive Microwave Space-borne Data

Abstract

Europe is a continent with diverse climatic conditions. The dominant climates are the Oceanic, the Mediterranean and the Continental ones. The western part of Europe has an oceanic climate, southern Europe has a Mediterranean climate and eastern Europe has a continental climate. Because of such heterogeneities, a vast range of extreme climatic events might occur in different areas. We define extreme climatic events the droughts, floods and heavy snowfall. Those events will be generically referred to in this research as anomalies. The purpose of this study is the identification of these extreme climatic events in the area of Europe, with the use of Special Sensor Microwave Imager (SSM/I) data at 37GHz frequency. The data that are used are Brightness Temperature (TB) values. The detection of the events will be achieved with the Polarization Difference Brightness Temperature (PDBT). The PDBT values can be related to changes to surface wetness and the surface geometry. It could be used as an indicator of an anomaly, because the higher the values of PDBT the higher the surface wetness. The methodological steps of the work consist in a statistical analysis of the SSM/I time-series, in the design of a detection algorithm of the anomalies under investigation and on the debate of its performance. The analysis of the temporally long SSM/I data will provide a first understanding of the data sensitivity to events under investigation and of their distribution for the statistical modelling of the Normalized Polarization Difference Brightness Temperature (NPDBT) indicator. The calculation of the NPDBT exploits the same principles as the well-known z-score index. The detection of the anomalies will be then achieved through thresholding the NPDBT index. Further information for the detection of anomalies is provided by the soil moisture time series from the Soil Moisture Active Passive (SMAP) sensor and the precipitation data from the Global Satellite Mapping of Precipitation (GSMAP). The soil moisture data appear to be more useful for the dry events, whereas the precipitation data for the flooding and the heavy snowfall events.