Weather Radar Performance at Long Range Simulated and Observed

Abstract

A large set of high-vertical-resolution reflectivity profiles was used to simulate the performance of radars in the United Kingdom weather radar network. In particular, limitations in the estimation of surface precipitation due to incomplete beam filling and variability in the reflectivity profile were investigated. Marked seasonal variations in range performance were found and detection failures were shown to make a large contribution to accumulation underestimation at long range. Although some current methods of correcting radar data for the effect of range may be capable of reducing mean biases, the ratio between radar measured precipitation rate and the rate at ground level is subject to large uncertainty at long range due to variability in reflectivity profile shape on the scale of individual 5-km pixels. The results of the simulation experiments were validated by comparison with monthly averages of data from two radars in the operational network. It was concluded that long-term averages of radar data are a useful tool for monitoring radar performance.