Effects of scale and land cover on loggerhead shrike occupancy

Abstract

ABSTRACTThe loggerhead shrike (Lanius ludovicianus) is a species of concern throughout its range because of severe population declines over the past 7 decades. Grassland habitat loss and fragmentation is widely viewed as contributing to the decline. Habitat associations have primarily been studied up to the territory scale, with few studies assessing shrike habitat selection at landscape scales. We conducted roadside passive‐active point counts for loggerhead shrikes in the coastal plain of South Carolina to evaluate detection probability and how land use and habitat fragmentation at multiple spatial scales influenced the occupancy of loggerhead shrikes. Detection probability increased with fair weather to 63% with clear skies, with temperature to 83% at 28°C, and with declining noise to 63% at no noise. High probability of detection at mean covariate conditions (P = 0.49 ± 0.08 [SE]) indicated support for the application of passive point counts in future loggerhead shrike monitoring. Occupancy was best predicted by percentage of pasture within 1 km, where predicted occupancy increased from 2 ± 0.02% when there was 0% pasture in a 1‐km radius, to 98 ± 0.03% occupancy when there was 43% pasture in a 1‐km radius, though our model‐averaged occupancy estimate was low (ψ = 0.17 ± 0.05). Extrapolation of our averaged model suggested that only 8% of the South Carolina coastal plain was occupied by loggerhead shrikes in the 2016 and 2017 breeding seasons. Our results highlight the importance of habitat beyond breeding territories and highlight the need for the establishment of higher concentrations of pasture and grassland at a 1‐km scale to increase the proportion of area in South Carolina occupied by shrikes. © 2018 The Wildlife Society.