Limits of precipitation nowcasting by extrapolation of radar reflectivity for warm season in Central Europe

Current precipitation nowcasting systems primarily use the extrapolation of observed radar reflectivity. The study focuses on both the extrapolation and the limits of predictability of precipitation using the concept of decorrelation time (DCT). The DCT analysis is based on reflectivity measurements from two radars covering the Czech Republic during the warm season (May-September) of four years (2009-2012). The analysis shows that that the mean DCT for the extrapolation forecast is 45.4 min, whereas the DCT for the persistent forecast is 13.4 min shorter. However, the DCT may increase or decrease by > 40% depending on current meteorological conditions. The investigation of the evolution of DCT in time during two storm events in the Czech Republic suggests that the DCT may significantly vary in time as a consequence of changing character of the precipitation systems.