Owing to its cryptic coloration and crepuscular activity pattern, the whip-poor-will is one of the least-studied birds in North America (Cink 2002). Breeding habitat in the CH and WGCP consists of xeric deciduous and mixed forests with a sparse understory. In addition, birds are often associated with open areas, such as rural farmland, powerline and roadway rights-of-way, clearcuts and selectively logged forest, old fields, and reclaimed surface mines. Shaded forest stands with limited ground cover adjacent to open areas for foraging provide ideal whip-poor-will habitat. The species is usually absent from extensive areas of closed canopy forest, but no data exists on minimum or maximum forest patch size thresholds. Small isolated woodlots in a Maryland agricultural landscape are not used (Reese 1996, cited in Cink 2002). Grand and Cushman (2003) found whip-poor-wills in Massachusetts strongly associated with complex patch shapes and high contrast edges. Nests are placed on the forest floor, and hatching is synchronized with the full moon to optimize foraging time of adults. The species does not appear to be strongly territorial; home range sizes range from 2.8–11.1 ha.
The habitat suitability model for whip-poor-will contains four variables:
- successional age class
- the relative composition of forest and open habitats in the landscape
The first suitability function combines landform, landcover, and successional age class into a single matrix (SI1) that defines unique combinations of these classes
. We directly assigned suitability index scores to these combinations based on relative habitat rankings for vegetation and successional age class associations of whip-poor-wills reported in Hamel (1992).
Whip-poor-wills nest in forest and forage in openings. Therefore, they require landscapes with an interspersion (SI2) of these landcover types. We assumed a landscape with 70 percent forest and 30 percent open habitat was optimal (suitability index score = 1.000; Wilson, pers. comm.) and landscapes with a greater proportion of forest were more suitable than those with less forest cover provided some openings were present
; sensu Cooper 1981).
We calculated the overall suitability index score as the geometric mean of the two component factors.
Overall SI = (SI1 * SI2)0.500