Passive acoustic monitoring with AI‐based detection and identification reveal sooty grouse hooting patterns in western Oregon

Many bird species are monitored using auditory point count surveys during the breeding season. Autonomous recording units (ARUs) can be used to better understand the daily and seasonal timing of when a species is vocalizing, which can help align surveys with the time period when the maximum number of individuals are present. We used ARUs to improve our understanding of sooty grouse Dendragapus fuliginosus hooting frequency to align our current point count hooting surveys with the dates of peak hooting. From 2020–2023, we deployed ARUs at 121 locations in western Oregon, USA near known hooting males. We evaluated the performance of PNW‐Cnet, an AI‐based detection application, for identifying the hooting song, the main breeding vocalization of male sooty grouse. Precision (true positives/apparent detections made by PNW‐Cnet) was > 0.999, indicating the model was highly effective at avoiding false positive detections of hooting. Recall (true positives/true positives + false negatives) was 0.595, suggesting that while the model reliably identified hooting when detected, it missed a substantial portion of hooting events. We then used generalized additive models with random effects to examine breeding season patterns in hooting frequency relative to date, time, weather, elevation, and latitude. Expected hooting frequency peaked in late‐April, providing guidance on the optimal timing of point count surveys. Daily patterns revealed a rapid increase in hooting 30 min before sunrise, then leveling at a relatively constant hooting frequency up to at least six hours after sunrise (latest recording time). Our results suggest males continue to vocalize throughout the entire morning allowing surveys to be conducted beyond three hours after sunrise. Hooting frequency decreased with increased precipitation and at high temperatures, whereas elevation and latitude effects varied. ARUs provided detailed hooting data that will enhance future surveys and improve population monitoring efficiency of sooty grouse.