Probability and latency of vocal responses of the Little Owl (Athene noctua) in a low-density farmland population: methodological implications for monitoring

Reliable monitoring of nocturnal birds requires survey protocols that account for imperfect detection and temporal variation in vocal activity. We evaluated how the probability of vocal response to playback and response latency varied throughout the night and breeding season in a low-density farmland population of the Little Owl ( Athene noctua ) in central Poland. Seventy-two previously known breeding sites were surveyed three times during spring 2022 using a standardised playback protocol. Each survey was assigned to one of six relative night periods between evening and morning civil twilight. GEE models showed that night period significantly affected the probability of vocal response to playback, whereas breeding season had no detectable effect. Cox proportional hazards models likewise demonstrated significantly shorter response latencies immediately before sunrise than during the early part of the night. Territorial responses were recorded during 105 surveys (48.6%), whereas any vocal response occurred during 111 surveys (51.4%). The probability of vocal response increased during the middle of the night and reached its highest value immediately before sunrise. Median response latency was 109 s, and 90% of territorial responses occurred within 331 s after playback initiation. Our results demonstrate that the timing of playback surveys strongly influences the probability of detecting Little Owls in low-density agricultural landscapes. Surveys conducted immediately before sunrise maximise detectability, whereas breeding season had comparatively little influence within the study period. These findings provide practical guidance for optimising playback-based monitoring of Little Owls and may also improve survey protocols for other nocturnal territorial birds.