Song recordings and environmental factors affect the response rate of Tropical Screech-Owls to conspecific playback: the importance of carefully designed protocols

Abstract

Reproducing conspecific vocalization has been widely used to detect owls in the field due to the nocturnal habits and cryptic nature of most of them. Surveying and monitoring owl populations using playback and other methods are important given the key roles these birds play in the environment and the ecosystem services they provide, especially Neotropical owls that are still poorly understood. We described Tropical Screech-Owls (Megascops choliba) response to playback and examined playback- and environmental-related variables affecting it, at one of the last lowland Atlantic forest remnants, in southeastern Brazil. Our playback protocol consisted of three repetitions of 2-min broadcasting of one recording of the species’ A-song, interspersed by 5-min silent listening periods. We used nine recordings randomly selected by sampling point to avoid pseudoreplication. The most common response (77%, N = 85) was an undescribed variation of the A-song, followed by other two voices that are possibly undescribed. The response rate to playback was 83.4% (N = 102 samples). Response rate was negatively affected by wind speed (especially wind bursts), and positively affected by greater cloud cover and lower ambient light. Recordings used for playback were more influential than any environmental variable. Recordings with fewer phrases, uneven and longer intervals between phrases, and higher-pitched songs elicited more responses. These characteristics are usually associated with lower quality individuals. We argue that playback protocols, whatever their goals, should be carefully designed and tested to improve their efficiency.