OOS 61-3
Factors affecting avian fatality at onshore wind turbines in the contiguous United States

Thursday, August 13, 2015: 8:40 AM
316, Baltimore Convention Center
Julie Beston, USGS
James E. Diffendorfer, United States Geological Survey, Denver, CO

Wind offers a renewable source of energy that does not produce greenhouse gas emissions.  Turbines do, however, cause wildlife mortality through collisions with birds and bats.  While previous studies have shown that hub height explains some of the variation in mortality rates, much variation remains among facilities and among turbines within a facility.  We analyzed fatality records for 1226 turbines on monopole towers in the United States to determine whether environmental variables were correlated with the number of fatalities caused by turbines.  We fit a hierarchical Bayesian model to fatality counts from fatality monitoring studies.  We assumed the number of fatalities observed at each turbine was binomially distributed, with n corresponding to the actual number of individuals killed and p describing the observation probability.  We modeled nwith a Poisson distribution whose mean was a function of a variety of variables at the facility and turbine levels, as well as study timing and duration.  Environmental variables included land cover and turbine density at the facility, topography, latitude and longitude, and mean wind speed, and models were compared using DIC.


As found in previous work, taller turbines kill more birds, but in addition to hub height, preliminary analyses indicate that land cover explains significant variation in mortality rates. Turbines in agriculture or grassland kill fewer birds than other land cover types. Models that included turbine density or elevation are not well-supported.   Understanding the factors that affect levels of turbine collision mortality can inform the planning and placement of future wind energy development.  It can also be used to project the expected fatality for future development scenarios and explore potential impacts.