Analytical and computer model of a Doppler weather radar system

Abstract

With advances in Doppler weather radar, severe storm and tornado detection has improved greatly. However, the resolution limitations of deployed radar systems can still limit severe storm detection. In the case of larger tornadoes, characteristic abrupt changes in wind direction can usually be detected between adjacent range-angle bins. However for smaller tornadoes, the rotating cell may be contained within one bin. In this case, a wind directional change cannot be detected, and the tornado may go undetected. The purpose of this research is to develop analytical and computer simulation models of typical Doppler weather radar measurements. These models can be used to determine how various factors affect the reflectivity, velocity, and spectrum width measurements that are commonly used in storm detection algorithms. The models account for convolution in azimuth due to the radar antenna pattern, convolution in range due to the radar pulse shape, randomness of the weather events and measurements, variations in the measurements between radar pulses, and the addition of noise to the measurements. Using both analytical and simulation models allows for simulated data to be generated, as well as equations that predict the behavior of the data. Therefore, the analytical and simulation models can be used to test the other's accuracy. Additionally, the analytical model can be used to create future algorithms (e.g. resolution enhancement), and the simulated data can be used as a test for these algorithms.