This monitoring is visulalised at http://www.flysafe-birdtam.eu/profile.php?radar=herwijnen. Collisions between aircraft and birds can cause extreme damages to the aircraft, in some cases leading to a crash. Therefore the Royal Netherlands Air Force (RNLAF) intensively uses this bird migration product for mission planning and last-minute changes to these plans. But bird migration data from weather radar is also highly relevant for e.g. disease monitoring, conservation, and ecology. An example of this is viewing the reactions of birds to the noise made by fireworks (http://horizon.science.uva.nl/fireworks/).
Research at KNMI on bird migration monitoring is carried out in close collaboration with the Institute for Biodiversity and Ecosystem Dynamics (IBED; http://ibed.uva.nl) of the University of Amsterdam, and RNLAF, as well as many other European research groups in the framework of the European Network on the Radar Surveillance of Animal Movement (ENRAM; see http://www.enram.eu). The work is focused on the following main topics:
The algorithm that is now operationally used was developed at KNMI (see Dokter et al., 2010; http://dx.doi.org/10.1098/rsif.2010.0116), and it uses volume data of radar reflectivity and radial velocity. The algorithm identifies bird echoes based on the texture (local spatial variation) of radial velocity, after which it produces an altitude profile of bird densities based on all bird echoes in a 25-km radius from the radar. This results in a time series of altitude profiles per radar.
The fact that most European radar data are now operationally collected centrally through the EUMETNET OPERA project (see http://www.eumetnet.eu/opera) makes this technique especially promising for e.g. continental-scale bird migration monitoring (see Shamoun-Baranes et al., 2014; http://dx.doi.org/10.1186/2051-3933-2-9). There are several challenges in adapting the algorithm to be applicable to all other radars in the OPERA network, related to data format issues, large differences in scan strategies, different settings of clutter suppression filters, and availability of radial velocity data and the quality thereof. Once these issues are solved, bird profile data from across Europe will become available. Bird information from such a large number of radars also requires new ways of visualizing these data (see Shamoun-Baranes et al., 2016; http://dx.doi.org/10.1371/journal.pone.0160106).
Dual-polarization radars have become more and more abundant in Europe over the last decade. This can be used to improve separation between birds, insects, precipitation, and other echoes (see here). This is not only helpful for obtaining more accurate bird density estimates, but also for obtaining more accurate wind velocity data for e.g. assimilation into numerical weather prediction models.
Dual-polarization can also help to try to separate different types of birds. Differences in bird shapes and behaviour will be linked to polarimetric variables and their texture. Furthermore, spectral analyses of signals from birds may yield information about their wing beat frequencies (see Dokter et al., 2013; http://dx.doi.org/10.1016/j.anbehav.2012.12.006). This is especially promising if vertically-pointing scans are preformed, so that birds will be in the measurement volume long enough to flap their wings several times.