Radar observations of birds have been used since 40 years for ornithological studies, in the field of air traffic security to help in reducing bird strike risk, in radar meteorology, as bird echoes can sometimes mimic weather, and for ecological and environmental applications. Nowadays, there is a growing interest in the detection and quantification of bird migration through radar sensors as well as in tracking, identifying and individual body size and even species of a given bird (radar aerocology). Each radar system has its advantages and limitations in detecting birds but meteorological radars appear very suitable for the remote sensing of these type of targets, also of those with a small size, as in the case of the passerine species, because of the use of high directivity antennas and conical beam. Despite the standard products of operational weather radars have a coarse resolution (e.g. 1 km × 1° maps are provided by the WSR-88D of NEXRAD), radars usually collect high-resolution data which are then spatially and temporally averaged by the signal processor before forming the final products. Therefore, low-cost algorithms operating at raw data level could be in principle implemented in an operational context to allow bird identification and counting. This paper addresses the use of weather radar as a remote sensing instrument of the characteristics of birds with the ambitious aim of classifying them by species. In fact, the echo power returned by a bird in flight, as measured by an high resolution radar, shows periodic fluctuations in time and a corresponding modulation of the Doppler spectrum width, which are dependant on the flight pattern, hence on bird type. In view of that, an electromagnetic model of a flying bird has been developed and it is described here. The model suggests that the bird echo modulation is owed to wing flapping rather than to the periodic expansion and contraction of bird's breast. Moreover, it provides a more realistic dependence of the radar cross section on both frequency and wave polarization. Finally, it works as an "ornithological-electromagnetic" reference. The model considers a geometry constituted by a hollow spheroid with two stylized wings. The external dimension of the bird is maintained as well as the weight, by modifying the size and thickness of the spheroid. The wings are modeled with two tapered slabs of muscle tissue that mimic the real wing structure. The simulations are compared with measurements purposely collected by the S-band weather radar sited at Spino d'Adda (Italy). Results obtained with a preliminary, non optimized, model confirm the role of wings on radar cross section fluctuations and are in quantitative good agreement with measurements.
Weather radar: a tool for the remote sensing of birds
2011
Abstract
Radar observations of birds have been used since 40 years for ornithological studies, in the field of air traffic security to help in reducing bird strike risk, in radar meteorology, as bird echoes can sometimes mimic weather, and for ecological and environmental applications. Nowadays, there is a growing interest in the detection and quantification of bird migration through radar sensors as well as in tracking, identifying and individual body size and even species of a given bird (radar aerocology). Each radar system has its advantages and limitations in detecting birds but meteorological radars appear very suitable for the remote sensing of these type of targets, also of those with a small size, as in the case of the passerine species, because of the use of high directivity antennas and conical beam. Despite the standard products of operational weather radars have a coarse resolution (e.g. 1 km × 1° maps are provided by the WSR-88D of NEXRAD), radars usually collect high-resolution data which are then spatially and temporally averaged by the signal processor before forming the final products. Therefore, low-cost algorithms operating at raw data level could be in principle implemented in an operational context to allow bird identification and counting. This paper addresses the use of weather radar as a remote sensing instrument of the characteristics of birds with the ambitious aim of classifying them by species. In fact, the echo power returned by a bird in flight, as measured by an high resolution radar, shows periodic fluctuations in time and a corresponding modulation of the Doppler spectrum width, which are dependant on the flight pattern, hence on bird type. In view of that, an electromagnetic model of a flying bird has been developed and it is described here. The model suggests that the bird echo modulation is owed to wing flapping rather than to the periodic expansion and contraction of bird's breast. Moreover, it provides a more realistic dependence of the radar cross section on both frequency and wave polarization. Finally, it works as an "ornithological-electromagnetic" reference. The model considers a geometry constituted by a hollow spheroid with two stylized wings. The external dimension of the bird is maintained as well as the weight, by modifying the size and thickness of the spheroid. The wings are modeled with two tapered slabs of muscle tissue that mimic the real wing structure. The simulations are compared with measurements purposely collected by the S-band weather radar sited at Spino d'Adda (Italy). Results obtained with a preliminary, non optimized, model confirm the role of wings on radar cross section fluctuations and are in quantitative good agreement with measurements.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
