A multi‐scale occupancy model for the grasshopper sparrow in the Mid‐Atlantic. Issue 8 (5th September 2013)

• Record Type: Journal Article
• Authors: Irvin, Eric; Duren, Kenneth R.; Buler, Jeffrey J.; Jones, William; Gonzon, Anthony T.; Williams, Christopher K.
• Is Part Of: Journal of wildlife management. Volume 77:Issue 8(2013)
• Page Start: 1564
• Page End: 1571
• Publication Date: 2013-09-05
• Subjects: Ammodramus savannarum -- grasshopper sparrow -- landscape -- multi‐scale analysis -- occupancy modeling; Wildlife management -- Periodicals; Zoology -- Periodicals; 333.954
• DOI: 10.1002/jwmg.609
• Languages: English
• ISSNs: 0022-541X

Abstract

ABSTRACT: Identifying features of breeding habitat that influence occupancy and modeling the distribution of grassland birds is needed to direct conservation efforts to reduce population declines associated with habitat loss and fragmentation. Many recent studies on grassland bird habitat use incorporate both local and landscape attributes. However, few studies have determined the appropriate spatial scales at which to measure these relationships. We conducted roadside point counts within Delaware, USA, to determine the presence of grasshopper sparrows ( Ammodramus savannarum ). We quantified both land cover composition and configuration at local and landscape scales at our survey sites using data from the National Oceanic and Atmospheric Administration Coastal Change Analysis Program. We determined the spatial scales at which grasshopper sparrow presence was most strongly related to landscape metrics and modeled grasshopper sparrow habitat occupancy at multiple scales, while accounting for variation in detection. At the site scale, occupancy was negatively related to forest and shrub composition. At the landscape scale, grasshopper sparrow occupancy was positively related to the amount of grasslands and pastures, and negatively related to mean inter‐patch distance of grasslands and amount of low‐intensity development. Our model had good predictive accuracy (area under the receiver operating characteristics curve = 0.717). We present our predictive model applied to the Delmarva Peninsula, USA. © 2013 The Wildlife Society.