9.1 Evaluating Parameters for Species-Based Classification of Bird Radar Echoes for Wind Energy Site Assesment

Thursday, 14 January 2016: 11:00 AM
Room 346/347 ( New Orleans Ernest N. Morial Convention Center)
Sheila Werth, University of Massachusetts, Amherst, MA; and S. Frasier
Manuscript (497.9 kB)

Wind energy is one of the fastest-growing segments of the world energy market, offering a clean and abundant source of electricity. However, wind energy facilities can have detrimental effects on wildlife, especially birds and bats. Monitoring systems based on marine navigation radar are often used to quantify bird and bat migration near potential wind sites, but the ability to reliably distinguish between bats and different varieties of birds has not been practically achieved. This classification capability would enable wind site selection that protects more vulnerable species, such as bats and raptors. In this work, we analyze radar parameters for their species-based classification potential in echoes from avifauna.

During the 2014 fall migration season, the UMass X-Pol weather radar [1] was used to collect low elevation observations of migrating birds as they traversed through a fixed antenna beam. The radar was run during the night time, in clear-air conditions. Data was coherently integrated, and detections of biological targets exceeding an SNR threshold were extracted. Detections without some dominant frequency content (i.e. clear periodicity, potentially the wing beat frequency) were removed from the sample in order to isolate observations suspected to contain a single species or bird. For the remaining detections, parameters including the polarimetric products and the Doppler spectrum were extracted at each time step over the duration of the observation. Clustering techniques were used to determine the extent to which observations fell within distinct groups, based upon the extracted parameters.

The presence of strong clusters of avian radar echoes, based upon selected parameters, would suggest the potential for a broad, species based avian classification algorithm. Such a classification scheme could ultimately help select and monitor wind sites in order to minimize harm to at-risk bird and bat species.


[1] V. Venkatesh, S. Palreddy, A. Hopf and K. Hardwick, "The UMass X-Pol Mobile Doppler Radar: Description, Recent Observations and New System Developments," in Geoscience and Remote Sensing Symposium, 2008. IGARS 2008., Boston, 2008.

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner