Short-eared Owls (Asio flammeus) are a globally distributed species, but whether their breeding ecology varies biogeographically is unknown and results in a lack of regionally relevant knowledge to inform conservation needs. We investigated the breeding ecology of Hawaiian Short-eared Owls (A.f. sandwichensis), or Pueo, using data from both focal study sites and incidental observations across the state of Hawai‘i. At focal sites, we found that greater visual obstruction readings were top predictors of the use of a site for Pueo nesting. However, these same metrics did not necessarily translate to increased nest survival; nests initiated earlier in the season and with higher percent vegetation cover had higher daily nest survival rates. Breeding Pueo diet at focal study sites included a variety of rodent, bird, and insect species, but diet composition did not vary significantly among nests. Across both focal and incidental state-wide observations, we found that Pueo nesting season spans from November through July, and breeding habitats ranged from coastal vegetation on nearby atolls to high elevation native wet forest. This study is the first to describe Pueo breeding ecology, providing a baseline for management actions in Hawai‘i and filling a geographical knowledge gap for this widely distributed species.

Fire disturbance is a driving force for biodiversity in forested ecosystems, but a new era of megafires that result from land use legacies and climate change has led to prolonged negative consequences for forest specialists. The forest owl community in the Sierra Nevada is presumably adapted to shorter-interval fires of low-moderate severity, but we have yet to quantify how individual species in this guild of avian predators respond to disturbance over space or time. In this study, we leveraged automated detections from passive acoustic surveys in the Sierra Nevada and occupancy models to 1) examine species-specific associations with burned habitat and fire legacies and 2) quantify the effect of novel fire disturbance on forest owl populations. We found that fire disturbance had variable effects of forest owl species, depending on both severity and time. Generally, low severity fire had delayed but posive effect on site occupancy for forest owl species, but high-severity fire generally had an immediate and delayed negative effect on occupancy for most species. Great orned owls were the expection, and they showed a delayed, postive relationship with areas burned at higher severity. Thus, we found evidence that species may benefit from fires that reflect the natural disturbance regime of the region, but fires characteristic of nodel megafires pose concern for forest obligates. More frequent and extreme forest fires may threaten the biodiverstiy of this guild of forest predators, and it is imperative to understand spatial and temporal responses of different species to disturbance to document potentially novel communities in a post-megafire landscape.

The California spotted owl (Strix occidentalis occidentalis) – an older forest species of conservation concern – has declined in many forests due to loss of habitat via severe wildfire and vegetation change. While population trends and factors responsible for trends have been well studied in the Sierra Nevada, less is known about their population status at the range boundary in southern California - a region undergoing rapid environmental change. Therefore, we conducted extensive spotted owl nighttime surveys during the 2022 breeding season in the Los Padres, Angeles, and San Bernardino National Forests, and compared remotely-sensed data on vegetation conditions and disturbance between currently occupied and vacant territories, and at vacant territories during historical occupancy versus after the loss of territorial owls. We made 1,913 visits to call points, surveying an area of 1,100 km2, yet located only eight occupied territories. Only three of the 13 historically occupied territories we surveyed remained occupied (23%). Vacant territories had a lower basal area of deciduous trees compared to occupied territories, and vacant territories had greater drought-related tree mortality after the loss of territorial owls compared to during historical occupancy. Our results coincide with other recent work indicating declines in spotted owl populations in southern California.

The range expansion of the Barred Owl (Strix varia) into western North America over the last century has emerged as a major threat to the Northern Spotted Owl (S. occidentalis caurina) and likely to other western forest species as well. A better understanding of Barred Owl natal dispersal, one of the primary driving mechanisms of range expansion, is needed for the implementation of effective management. Our research team attached satellite-GPS tags to juvenile Barred Owls in the Coastal Redwood Region of northern California to characterize their habitat use as they disperse from natal territories. These tags allow for fully remote tracking of owls and have the potential to provide locations up to 1.5 years past the date of deployment, making this study the first of its kind. We collected location data from 31 Barred Owl juveniles during dispersal and conducted a multi-scale habitat selection analysis to identify landscape features which juvenile Barred Owls are using. We found evidence for selection of shorter forest stands, drainages, and areas of relatively lower elevation. Our findings have implications for understanding patterns of Barred Owl spread and identifying potential dispersal corridors, and thus for managing further spread and recolonization of removal areas.

Barred Owls (Strix varia) have recently expanded westward from eastern North America, leading to substantial declines in Northern Spotted Owls (Strix occidentalis caurina). Passive acoustic monitoring (PAM) represents a potentially powerful tool for tracking range expansions like the Barred Owl’s, but further methods development is needed to ensure that PAM-informed occupancy models meaningfully reflect population processes. We used a combination of PAM data, GPS-tagging, and active surveys to identify patterns of vocal activity that best reflect resident occupancy and subsequently estimated resident occupancy rates by correcting site occupancy estimates using vocal activity thresholds. The proportion of survey nights with confirmed vocalizations (VN) and the number of ARUs within a sampling cell with confirmed vocalizations (VU) were strongly indicative of Barred Owl residency. Applying thresholds of VN > 0.17 and VU = 3 to our occupancy analysis, we were able to minimize false positive and negative errors, and we estimated the occupancy rate for resident owls to be 0.45. Our findings provide a scalable framework for monitoring Barred Owl populations throughout their expanded range and, more broadly, a basis for converting site occupancy to resident occupancy in PAM programs—thus ensuring changes in occupancy rates reflect changes in populations.

Concern for potential effects on California spotted owls can constrain forest restoration projects intended to reduce large, severe wildfires and drought-related tree mortality in the Sierra Nevada. Thus, call-based spotted owl surveys are typically conducted as part of the planning stage of forest restoration projects with a goal of achieving a 0.95 probability of detecting owls at occupied territories. Call-based surveys, however, require extensive and potentially hazardous nighttime work, typically across two years. As part of the development of a one-year, acoustically-aided survey protocol for spotted owls, we created a tool to help prioritize areas for differing survey strategies (full/twilight/acoustically aided/no surveys). The tool provides predictions of the probability of occupancy based on historical occupancy information and/or remotely-sensed estimates of tree height. Areas with shorter trees (primarily within a large, severe fire footprint) had a low probability of being occupied. In unburned landscapes, areas with more tall trees and more frequent historical occupancy were more likely to be occupied. Using these predictors, managers can more effectively identify areas not needing owl surveys, areas where one year of acoustically-aided surveys are appropriate, and areas where call-based surveys can be conducted without nighttime work, while maintaining a detection probability of 0.95.