Management and Conservation Note Relationship of Obligate Grassland Birds to Landscape Structure in Wisconsin LES D. MURRAY, 1,2 Department of Wildlife Ecology, University of Wisconsin-Madison, 1630 Linden Drive, Madison, WI 53706, USA CHRISTINE A. RIBIC, United States Geological Survey, Wisconsin Cooperative Wildlife Research Unit, Department of Wildlife Ecology, University of Wisconsin-Madison, 1630 Linden Drive, Madison, WI 53706, USA WAYNE E. THOGMARTIN, United States Geological Survey, Upper Midwest Environmental Sciences Center, 2630 Fanta Reed Road, La Crosse, WI 54603, USA ABSTRACT Conservation plans for grassland birds have included recommendations at the landscape level, but species’ responses to landscape structure are variable. We studied the relationships between grassland bird abundances and landscape structure in 800-ha landscapes in Wisconsin, USA, using roadside surveys. Of 9 species considered, abundances of only 4 species differed among landscapes with varying amounts of grassland and forest. Landscape variables explained ,20% of variation in abundances for 4 of the 5 rarest species in our study. Our results suggest landscape-based management plans for grassland birds might not benefit the rarest species and, thus, plans should incorporate species-specific habitat preferences for these species. ( JOURNAL OF WILDLIFE MANAGEMENT 72(2):463–467; 2008) DOI: 10.2193/2006-556 KEY WORDS abundance, bird, composition, conservation, grassland, habitat, landscape, Wisconsin. Grassland birds have declined across North America and most species are considered to be of conservation concern in the Midwest (Knopf 1994, Herkert 1995, Peterjohn and Sauer 1999, Knutson et al. 2001). Loss and degradation of grassland habitat has been cited as a major factor contributing to the decline of grassland birds (Martin and Finch 1995). Thus, much research has been conducted on the relationship of grassland birds to habitat characteristics. Research has provided a good understanding of the influence of vegetation structure and composition on grassland birds within grassland patches (e.g., Wiens 1969, Sample 1989, Madden et al. 2000). In addition, research on grassland birds has suggested abundances are related to grassland patch size (e.g., Herkert 1994, Vickery et al. 1994, Winter and Faaborg 1999) and the surrounding landscape (Ribic and Sample 2001, Bakker et al. 2002, Fletcher and Koford 2002). Relationships of grassland bird abundances to patch size and landscape context, however, are variable among years, sites, and regions (Johnson and Igl 2001, Bakker et al. 2002, Winter et al. 2006). Despite the variability and lack of biological under- standing of relationships of birds and landscape attributes, landscape context has been included as part of the selection criteria for potential management areas (Sample and Moss- man 1997, Fitzgerald et al. 1998). Areas with high amounts of grassland and little forest are thought to be good areas for grassland birds relative to areas with less grassland or more forest. Landscape attributes also have been used to predict regional abundances of grassland birds (Thogmartin et al. 2006). We examined the relationship between landscape structure and grassland bird abundances and the ability to use landscape variables to predict grassland bird abundances in southwest Wisconsin, USA. First, we compared abundances in 800-ha landscapes classified into 6 landscape types based on the amounts of grassland and forest in the areas. Second, we examined the amount of variation in grassland bird abundances explained by landscape variables measured at 2 scales. STUDY AREA Our study took place in Dane, Grant, Greene, Iowa, and Lafayette counties in southwest Wisconsin. The approx- imately 620,000-ha area was contained in the rolling hills of the southwestern uplands in the Driftless Area of Wisconsin (Martin 1965). Historically the area was composed of a mixture of oak savanna, tallgrass prairie, southern oak forests, and lowland hardwood forests. The area currently is composed mostly of grassland–herbaceous land covers (e.g., pasture, hay, Conservation Reserve Program fields; 57%), row crops (22%), and deciduous forest (18%; 1992 National Land Cover Dataset [NLCD]; Vogelmann et al. 2001). METHODS We overlaid a grid of 800-ha cells on the 1992 NLCD. We used a cell size of 800 ha to match the smallest spatial extent used to predict grassland bird abundances by Thogmartin et al. (2006). We combined the grassland–herbaceous and pasture–hay categories of the NLCD into a general grassland category because of high error rates in classifica- tion of these cover types (Thogmartin et al. 2004, Environmental Protection Agency 2006). We then calcu- lated percent composition of grassland and forest for each 800-ha cell. We then categorized landscapes based on proportions of grassland and forest. We classified proportion grassland into low (0.30–0.45), high (0.60–0.80), and very 1 E-mail: murray.445@osu.edu 2 Present address: School of Environment and Natural Resources, The Ohio State University, 2021 Coffey Road, Columbus, OH 43210, USA Murray et al.  Grassland Birds and Landscape Structure 463