Occupancy Modeling of Woodpeckers: Maximizing Detections for Multiple Species With Multiple Spatial Scales

Abstract

Abstract Numerous forest birds benefit from woodpecker presence or have similar habitat requirements. Monitoring populations of forest woodpeckers can be useful for management decisions regarding these and other forest species. Usefulness of monitoring efforts depends on methods used and the quality of resulting parameter estimates. Estimating the proportion of area occupied by a species can be an attractive and affordable alternative to abundance or survival estimates. The purpose of this study was to assess the distribution and area of occupancy for pileated woodpeckers (Drycopus pileatus) and American three-toed woodpeckers (Picoides dorsalis) in north-central Idaho, and to compare occupancy estimates using silent point counts, playback surveys, and playback surveys that incorporated estimates of detection probability (p). We used a hierarchical multiscale framework that allowed estimation of occupancy at two spatial scales and applied a removal design such that repeat visits to sampling stations was not necessary to estimate p. The initial naïve estimate of occupancy (using presence–absence data) for pileated woodpecker was 0.39, which increased to 0.59 using playback surveys. The corrected estimate of occupancy at the 1-km2 unit scale was 0.70. The naïve estimates of occupancy for American three-toed woodpeckers using silent point counts and playback surveys were 0.14 and 0.34, respectively. The unbiased estimate of occupancy at the 1-km2 unit scale was 0.71. Detection probabilities are known to vary spatially and temporally for numerous reasons. Thus, comparisons of naïve estimates of occupancy to monitor forest woodpeckers would be imprudent and could lead to poor management decisions. We recommend incorporating detection probability for monitoring wildlife species and show how this can be done within a single sampling framework for species that utilize the landscape at disparate scales.