Quality Assessment of Weather Radar Wind Profiles during Bird Migration

Abstract

Wind profiles from an operational C-band Doppler radar have been combined with data from a bird tracking radar to assess the wind profile quality during bird migration. The weather radar wind profiles (WRWPs) are retrieved using the well-known volume velocity processing (VVP) technique. The X-band bird radar performed range?height scans perpendicular to the main migration direction and bird densities were deduced by counting and normalizing the observed echoes. It is found that the radial velocity standard deviation (σr) obtained from the VVP retrieval is a skillful indicator of bird migration. Using a threshold of 2 m s?1 on σr, more than 93% of the bird-contaminated wind vectors are rejected while over 70% of the true wind vectors are accepted correctly. For high bird migration densities the raw weather radar wind vectors have a positive speed bias of 8.6 ± 3.8 m s?1, while the quality-controlled wind vectors have a negligible speed bias. From the performance statistics against a limited area numerical weather prediction model, it is concluded that all (significant) bird contamination is removed and that high-quality weather radar wind profiles can be obtained, even during the bird migration season.