Intelligent Velocity Azimuth Display (IVAD) Wind Estimation in Clear-Air With S-Band Polarimetric Weather Radars

Wind velocities approximated via velocity azimuth display (VAD) have been found to be contaminated by birds and enabled by insects. However, the widely used VAD wind profile (VWP) does not account for taxa, largely due to the challenge of distinguishing bird from insect echoes. This problem has been addressed by the recently developed bird-insect ridge classifier (BIRC). Hence, this work proposes an intelligent VAD (IVAD) that leverages BIRC to improve clear-air wind estimates by generating three new products, including the insects-birds ratio, bird-only VAD, and insect-only VAD. These products are analyzed for one-month periods containing nocturnal and diurnal bird migration. Wind bias is used as the evaluation metric, defined as the deviation of the predicted VAD from reference wind measurements obtained from the rapid refresh (RAP) model. Results show an inverse relationship between biases and the insects-birds ratio, such that increasing (decreasing) bird (insect) population was accompanied by larger biases. Furthermore, contaminated VADs showed improvements when insects-only signals were used instead of all biological echoes. We recommend that these products can be incorporated into the VWP. First, the insects-birds ratio can be used to identify whether a given height is bird dominated, mixed, or insect dominated. For the mixed case, improved wind estimates can be obtained from insect-only VAD. Otherwise, bird-only VADs can be obtained from bird-dominated heights, while insect-only VADs are obtained from insect-dominated heights. The former can be used to track birds while the latter tracks insects and the wind.