Hundreds of millions of birds migrate every year through Israel between their high latitude breeding areas and their Afro-tropical over-wintering grounds. This migration flyway is one of the largest in the world, but its properties have only been rarely investigated. Many large birds (such as raptors, storks and pelicans) migrate using soaring and gliding by exploiting convective thermals that develop during the day and tend to concentrate in specific areas where thermals are intense, for example near mountain ranges.

Quantifying the migration of large soaring birds is important not only for better knowledge of their aero-ecology but also because they pose serious threat to aerial transportation. Indeed strikes of large birds have caused many severe and sometimes devastating accidents with large economic consequences and risk to human lives.

Application of remote sensing tools such as radars allowed quantifying bird migration on large spatial scales. However, no identification and quantification of soaring bird migration has been attempted so far using radar data.

In this study we tagged thousands of radar images from three different weather radars positioned in Mt. Meron in northern Israel, Bet-Dagan in central Israel, and Mt Aricha near Mitzpe-Ramon in southern Israel which together cover a large area in Israel. Using these images, we are developing a machine learning algorithm that uses three different parameters from the radars, as well as images from past time in the sequences that allow for identifying the movement of the birds. By now, the algorithm recognizes the bird flocks in the radar images with a 74% precision. In the coming weeks we will try improving it using deep learning methods.      

Inbal Schekler, Ph.D. student, Animal Flight Laboratory, Department of Evolutionary and Environmental Biology, University of Haifa

Website, Facebook , Tw