LAUSR

For population studies that rely on aural or visual observations of individuals, methods that induce a detectable response can greatly improve sample sizes and reduce costs. Numerous authors have reported improved detections using broadcast surveys in forested areas for various species of raptors; however, to our knowledge, none have attempted to quantify the effects of broadcast surveys on the detection probability (p) in the context of an occupancy study for multiple species. Our objective was to determine whether broadcasting conspecific and heterospecific calls (broadcast surveys) would increase p of raptor species compared to passive surveys. This comparison was accomplished by estimating p during both types of surveys using a multiscale occupancy framework. We conducted 8400 surveys for Cooper's Hawk (Accipiter cooperii), Swainson's Hawk (Buteo swainsoni), White-tailed Hawk (Geranoaetus albicaudatus), Harris's Hawk (Parabuteo unicinctus), Red-tailed Hawk (Buteo jamaicensis), and Great Horned Owl (Bubo virginianus) in 2015 and 2016 in south Texas. We conducted 10-min passive surveys in both years, 10-min broadcast surveys in 2015 including calls from each of the target hawk species, and 10-min broadcast surveys in 2016 using calls from Great Horned Owls. Our results suggest that p is improved with multiple-hawk broadcast surveys (0.16, SE 0.04) over passive surveys (0.04, SE 0.02) for Harris's Hawks; p was not affected in broadcast surveys for the other hawk species considered. Great Horned Owls had a higher p during broadcast surveys using Great Horned Owl calls (0.35, SE 0.04) than during passive surveys (0.19, SE 0.02). Our results do not support the use of broadcast surveys to improve the detection probability of multiple species of hawks. However, our results do indicate that using broadcast surveys for occupancy studies focused on Harris's Hawks or Great Horned Owls should significantly reduce the effort required over passive surveys, resulting in improved power and reduced costs.