Abstract: The US operates a system of ~140 S-band doppler radar towers that continuously monitors the airspace around the majority of the United States. These radar towers are capable of tracking large aggregations of flying species like free-tailed bats. Free-tailed bats (Molossidae), through their voracious insect consumption, are thought to provide immense levels of ecosystem services; but their movements, and their ecosystem service provision, have historically been difficult to track/study in space and time. We introduce ‘BATS’, or Bat-Aggregated Time Series, an open-source Python toolkit that streamlines and automates the process of downloading, classifying, and aggregating time series of free-tailed bats across large landscapes. BATS first downloads data from NOAA’s weather radar data repositories hosted on Amazon Web Services; it then classifies the pre-processed radar data using a semi-supervised artificial neural network algorithm pre-trained to detect free-tailed bats. With an accuracy rate of 93% and a false discovery rate of 0.035, BATS is highly effective in identifying free-tailed bats in NEXRAD radar data. Furthermore, BATS is capable of distilling six months of radar data (3.5Tb) into a single 15Mb-sized map of bat occurrence quickly, contingent on compute resources used. BATS will help scientists and stakeholders identify areas of high occupancy of bats and other flying species at the landscape level over long periods of time. This ability has the potential to increase our understanding of the economic and agricultural value of these species.
