Grassland Bird Nesting on Restored and Remnant Prairies in South Central Nebraska

Approximately 64% (402,000 km²) of mixed-grass prairie has been lost, especially to agricultural production, since European settlement in North America (Samson and Knopf 1996). This loss has reduced wildlife habitat; 77% of grassland birds tracked by breeding bird surveys have decreased populations from 1966 to 1996 (Sauer et al. 2000). The Platte River Whooping Crane Maintenance Trust has been restoring grasslands in south central Nebraska since 1986, measuring success primarily through native vegetation establishment and cover. While evaluating restored grassland as nesting habitat for migratory birds, we encountered some important differences in grassland bird use between restored and native grasslands. We present preliminary findings on both plant composition and structure, and grassland bird densities in natural undisturbed areas and restored sites.

The study area is located in the Platte River Valley in a landscape formerly characterized by grasslands and currently dominated by corn fields. Our study sites consisted of six former farmlands restored to grassland ranging from 5 to 15 years old, and six remnant grasslands moderately affected by common introduced grass species such as smooth brome (Bromus inermis) and Kentucky bluegrass (Poa pratensis); all restored and native plots were at least 40 ha and less than 1.5 km from the river. Restored grasslands were contoured to re-create former rolling terrain and planted with a high-diversity seed mix (100+ species). Management has typically consisted of light grazing (<1 AUM), and in some cases prescribed burning starting at least three years after restoration.

We identified the territory for individual males or breeding pairs, and located nesting sites of bobolink (Dolichonyx oryzivorus), dickcissel (Spiza americana), and grasshopper sparrow (Ammodramus savannarum) using Martin and Guepel's technique (1993). We selected these three species for evaluation because they are important in mixed-grass prairie ecosystems and have different habitat requirements. Bobolinks require areas with tall vegetation structure and density, while grasshopper sparrows prefer low and sparse vegetation; dickcissel was selected as an intermediate indicator, with nesting requirements of continuous herbaceous cover with woody perches. We used the spot mapping technique (Bibby et al. 2000) to determine breeding bird pairs between May and July. In each site, the locations of all birds (especially singing males) and nest sites are mapped in a square 16 ha plot during a series of weekly visits (>8 total) during the breeding season. We recorded a total of 242 nests in native remnants and 264 in restored sites.

We measured vegetation vertical structure and height; aerial cover by functional group (grasses, forbs, woody, litter, bare ground, and standing dead); litter depth, at paired random points 25 m away from nest sites to obtain a better description of site variation; and plant species composition, richness, and frequencies at site level. To obtain a more accurate description of vegetation at each nest site, we measured vertical cover using a density board (10 cm × 2 m, with 10 cm increments). Initial trials determined that the most accurate readings of visual obstruction occurred at 1 m height and 4 m away. To describe vegetation aerial cover we used five Daubenmire frames (20 × 50 cm) at each nest site, located at the center and 1 m away in each cardinal direction. Litter depth was measured at the center of each Daubenmire frame using a ruler to record thickness from ground level to the top of the litter layer. Species composition was sampled using the step point method (Owensby 1973), recording species encountered at basal level. Five 100 m transects were randomly established per site in July 2009, and individual species were identified at each step. We used ANOVA to compare treatments (site type).

Densities of the three grassland bird species were significantly different in restored and native sites (Figure 1). Bobolinks and grasshopper sparrows, the two species at the extremes of the structural gradient, had more than twice as many nests on native remnant sites (p < 0.001). Dickcissels, our indicator of intermediate vegetation structure, showed the opposite pattern, with three times...