Predicting habitat use by bats to protect bats and inform wind energy development

Although wind turbines are a clean, renewable source of energy, sometimes they incidentally kill bats and birds in large numbers. In 2012 for example, an estimated 600,000 bats died due to encounters with turbines at wind... [ view full abstract ]

Although wind turbines are a clean, renewable source of energy, sometimes they incidentally kill bats and birds in large numbers. In 2012 for example, an estimated 600,000 bats died due to encounters with turbines at wind energy facilities in the U.S. alone. Migratory species such as the Mexican free-tailed bat (Tadarida brasiliensis) and hoary bat (Lasiurus cinereus) have the highest mortality at wind energy facilities. Arizona has both high species richness of bats and a high proportion of migratory species (bats that migrate long distances or regionally) that creates a high risk of mortality from interactions at wind energy facilities. Our objectives are to determine the species composition, examine bat use, study topographic features on the landscape that might influence bat movement (e.g., long distance migration), and identify elevational movements (e.g., regional migration) by bats. Our study area encompasses open grassland and shrubland in northern Arizona in areas where wind energy development might be suitable. We are sampling bat activity to determine habitat use and migratory patterns by bats in northern Arizona. Across our study area, we are deploying 34 acoustic detectors (Song Meter SM3BAT) at randomly-selected points that represent a range of measures for each habitat covariate (e.g., slope, aspect, elevation). We surveyed points during the summer of 2015, and we are surveying points during spring, summer, and fall of 2016 and 2017. We are using SonoBat 3 software to identify bat calls to species. We analyzed data from the summer of 2015 using linear regression in R, and we separated the collected calls in to 5 groups: all call files, low frequency calls, high frequency calls, Q25 group (calls with characteristic frequencies around 25 kHz), and less than 15 kHz calls. In these preliminary results, we found that there is higher bat activity for all groups in valleys and lower slopes than other topographic features. Using our models, we will create a map that predicts bat use in areas of northern Arizona that may also be suitable for wind energy development.