We study the effect of land use parameters, habitat scale, and weather on presence of bird species in Kentucky’s upper Green River watershed, an area where large-scale riparian habitat restoration is currently in progress (funded by the USDA Conservation Reserve Enhancement Program). Ten sampling sites were established along the riparian corridor of the Green River. We surveyed birds by monitoring their calls with 20 automated recording stations (Song Catchers) during spring and summer in 2006-2007. The recording protocol was 3 minutes per hour, 24 hours per day, and the stations were deployed during nine to 14 days per season, dependent upon battery life. A subsample of the recorded sounds was analyzed by ear to generate a control set of identified bird calls.  This data set was then used to create a recognizer to optimize the performance and test the efficiency of SongScope, an acoustic detection software package.

Results/Conclusions  

Detection parameters and acoustic analysis settings were determined to generate positive identifications of neotropical birds. Our test of the recognizer for Common Yellowthroat yielded zero Type 1 errors and zero Type II errors.  This procedure is essential prior to embarking on batch signal detection of data volumes in this acoustic autosampling autoanalysis system.