ABSTRACT
Radars filter and select bird echoes within their beam in several ways. Longrange surveillance radars miss many birds behind the radar horizon. Moreover, the spatial behavior of the birds varies, thus further reducing detectability. Disentangling these effects requires empirical knowledge of the radar and general insight into the bird movements. A rough classification of bird echoes on the basis of direction, speed, and echo-type is necessary to quantify the total broad-front migration cohort by cohort. The extremely flat Dutch countryside and an ideal stacked beam S-band radar equipped with a sophisticated bird extraction system provided the database for the calculation of proportions of birds missed under typical migratory conditions. This improved the use of radar as an indicator of flight safety risk as well as a research tool for unravelling navigational strategies of migrants in the North Sea area. The key issue appears to be the selection of altitude and track direction in relation to landscape, wind, and flight phase.