Influence of the small-scale variability of the raindrop size distribution on radar power laws

The strong spatial and temporal variability of rainfall, and hence of the raindrop size distribution, makes it difficult to compare rain-related variables (e.g., rain rate, radar reflectivity) at different scales. This issue is of importance when disdrometer data collected at the point scale are used to derive the parameters of power laws used to convert weather radar measurements (corresponding to a large sampling volume) into rain rate. To investigate this issue, data from a network of 16 optical disdrometers (OTT-Parsivel) deployed in March 2009 over a typical weather radar pixel (about 1 x 1 km²) in Lausanne, Switzerland, are used. DSD measurements corresponding to about 40 rain events (convective, mixed, stratiform) are analyzed at the point and radar pixel scale to quantify the influence of the small scale variability of the DSD on commonly used power laws (e.g., Z-R, R-K_DP). It is shown that the parameters of these power laws can significantly vary between the point and the pixel scales. For the Z-R power law, this variability can lead to discrepancies of the order of 20 % in the total (event) rain amount. These results demonstrate that the variability of the DSD can have significant influence on the uncertainty affecting radar rain-rate estimates.