Habitat Selection at Different Scales for a Declining Aerial Insectivorous Bird as Determined by Autonomous Recording Technology

Elly Knight
Erin Bayne
Resource Date:
Page Length

Conservation management is impeded by the lack of baseline data for many non-passerine, cryptic, or nocturnal species that are inadequately sampled by traditional monitoring programs. The recent rise of bioacoustic technology, including autonomous recording units (ARUs) and automated signal recognition software provide an opportunity to use archived bioacoustic datasets to fill baseline data gaps for rare and/or nocturnal species.  We extracted detections from a large bioacoustic database using automated signal recognition software and boosted regression tree models to build regional home range selection and territory selection models for the Common Nighthawk (Chordeiles minor ), a declining aerial insectivore for which there is minimal existing data. We found Common Nighthawk home range selection and territory selection to be explained by different environmental variables. Home range selection was primarily explained by landscape scale geographic and climate variables and some avoidance of wetland areas. Territory selection was also strongly influenced by landscape scale climate variables, proportion of seismic lines, and areas with minimal poor fen. Mean January temperatures and the proportion of pine forest were the only environmental variables that had relative influence (> 3.5) for both home range and territory selection, with the marginal effect of pine forest increasing sharply after a threshold of approximately 30% pine in the surrounding 3 km for both selection models. The
importance of landscape scale variables relative to local scale variables was higher for both home range and territory selection, although the magnitude of importance of landscape scale variables was higher for home range selection. Our results provide wildlife managers with guidance on where Common Nighthawks may be found in the boreal forest during the breeding season, with selection for cold, dry, northern landscapes, pine forests, and avoidance of wetland areas which is contrary to results for other biomes. The strong influence
of landscape scale variables emphasizes the importance of landscape scale conservation for highly-mobile species with large home ranges. Our case study also highlights the value of archived bioacoustic datasets for conservation of understudied species.