Indirect effects of wind energy development on grassland bird habitat use and reproductive succes

Wind energy development is rapidly expanding in grassland habitat, but potential indirect effects to wildlife populations are largely unknown. As observed in other types of human development, wind energy infrastructure could alter plant and invertebrate communities, change soil and hydrology, generate visual and aural noise, and increase human activity in previously undeveloped areas. These changes to habitat can indirectly affect wildlife in many ways, for example, by altering trophic interactions or habitat use. Native grasslands in the United States have undergone extensive habitat loss and degradation, and subsequently, many grassland-associated bird species have experienced precipitous population declines. We assessed the indirect effects of wind energy development on the habitat use and reproductive success of two grassland songbirds, Horned Lark (Eremophila alpestris) and McCown's Longspur (Rhynchophanes mccownii) in Wyoming. We modeled bird density in 2012 as a function of habitat characteristics and wind energy development. We estimated bird density at two scales of development- between two wind farms and two undeveloped sites, and within wind farms as a function of proximity to nearest wind turbine. Despite a general characterization of Horned Larks as disturbance-tolerant, the numbers of Horned Lark within wind farms were substantially lower near turbines. McCown's Longspur numbers were higher near turbines in areas with high heterogeneity in bare ground, but showed no variation with heterogeneity of bare ground far from turbines. These responses indicate potential avoidance of turbines by Horned Lark, and a difference in habitat associations by McCown's Longspur near turbines. We evaluated nest placement of Horned Lark and McCown's Longspur in 2012 in relation to turbine proximity on three wind farm plots. The mean turbine proximity at each plot was no different than predicted by a random distribution, and the turbine-proximity values of observed nests were uniformly distributed, indicating no evidence of displacement of nests from turbines. Although Horned Lark occur in lower numbers near turbines, there was no evidence of avoidance of turbines in nest placement. We assessed the reproductive success of Horned Lark and McCown's Longspur in 2011 and 2012 on 3 wind farms and 2 undeveloped sites. We evaluated multiple indices of nesting productivity (clutch size, size-adjusted nestling mass, daily nest survival rate, and number of young fledged per successful nesting attempt), and modeled these reproductive metrics as a function of habitat and wind energy development. We quantified wind energy development using several measures of disturbance: turbine proximity, turbine density, distance to developed edge, and the amount of developed and reclaimed area within the neighborhood of the nest, and identified which measure was most predictive of reproductive success. Turbine density was repeatedly the measure of wind energy development most strongly associated with avian reproductive success. The nestling mass and, in 2011, nest survival of Horned Lark were negatively related to increasing turbine density. Annual variation in the effect of infrastructure on nest survival indicates that infrastructure may only have negative consequences in correlation with other biotic or abiotic events, e.g., during years of high predator abundance. The nest survival of McCown's Longspur was most strongly predicted by nest site and neighborhood habitat variables and turbine density within 1 km of the nest. Likelihood of McCown's Longspur nest survival decreased from 54% to 5% as vegetation density increased, and the inclusion of turbine density and grassland area in the neighborhood amplified this relationship resulting in a 28% to <1% chance of survival across the range of observed vegetation density values. Our research suggests there may be species- and activity-specific avoidance of wind turbines, and species-specific reductions in reproductive success as a function of turbine density. Given the unexpected negative response solely by the more disturbance-tolerant species in our study, we recommend an individual species-based approach be taken in assessing the responses of birds to wind energy development. Additionally, the standard measure of wind energy development, turbine proximity, may not adequately quantify the effects of infrastructure on avian reproduction, particularly as regions of high quality wind resources become more developed. The high ranking of turbine density in our models of reproductive success indicates that intensity of development rather than turbine proximity may have a greater influence on songbird productivity.