Search

Increasingly, data from weather surveillance radars are being used by biologists investigating the ecology and behavior of birds, insects, and bats in the aerosphere. Unfortunately, these radars quantify echoes caused by layered biological targets such as migrating birds in a manner that introduces bias in radar measures. We investigated the performance of a bias-adjustment algorithm that adjusts radar measures for vertical variation of reflectivity, nonstandard beam refraction, and spatial displacement of radar targets. We evaluated the efficacies of four variations of this algorithm by their ability to increase correspondence between radar reflectivity measured at two weather radar sites and the ground density of migrating birds measured during two autumn seasons and two spring seasons among 24 hardwood forest sites along the northern coast of the Gulf of Mexico. The algorithm integrated close-range reflectivity data from the five lowest elevation angle sweeps to derive high-resolution vertical profiles of reflectivity (VPRs) that closely corresponded to the observed vertical target density profiles based on a vertically oriented portable radar. The radar reflectivity of birds aloft near the onset of migratory flight was positively correlated with the bird density on the ground. All four radar data adjustment schemes that we tested produced significant improvement in the accuracy of bird density estimates relative to unadjusted radar data. In general, adjusting reflectivity based solely on the VPRs derived using observed refractive conditions yielded the most accurate radar-based estimates of bird density. [ABSTRACT FROM AUTHOR]