The black-and-white warbler is a Neotropical migrant that occurs throughout the eastern United States and southern Canada. The black-and-white warbler is a forest interior species and the 1.2 percent annual declines observed in the United States over the last 40 years are likely in response to increasing forest fragmentation (Sauer and others 2005). The species has a regional combined score of 16 in the WGCP, where it is a species needing management attention (Table 001) . However, it has a regional combined score of only 13 in the CH and is assigned a specific action code.

The black-and-white warbler is a forest-interior specialist that occurs in the mature deciduous hardwood forests of the eastern United States and Canada (Kricher 1995). Although not as sensitive to fragmentation as some species (e.g., cerulean warbler), the black-and-white warbler is typically absent from small woodlots (<7.5 ha; Galli and others 1976). Hamel (1992) suggested 550 ha was the minimum tract size this species would reliably occupy in the Southeast.

Few studies have focused exclusively on the habitat ecology of this species; however, Conner and others (1983) found black-and-white warblers to be associated with mature forest stands with high densities of large (>32 cm d.b.h.) trees. Although a ground-nesting species, black-and-white warblers are associated with high densities of hardwood saplings. Conversely, pine saplings negatively affect both the presence and abundance of black-and-white warblers.

Birds occupy upland and bottomland forests, but reach higher densities in the former, with oak-hickory and cove forests considered optimal (Hamel 1992). Nevertheless, successional age may be the most critical habitat factor influencing black-and-white warblers. Dettmers and others (2002) validated Hamel’s (1992) black-and-white warbler habitat suitability model and found it performed well due to the restriction of black-and-white warblers to older age-class forests. Thompson and others (1992) and Annand and Thompson (1997), though, commonly observed black-and-white warblers in sapling and clearcut stands in Missouri.

Our habitat suitability model for black-and-white warblers contains six variables:

• landform
• landcover type
• successional age class
• forest patch size
• percent forest in a 1-km radius
• canopy cover

The first suitability function combines landform, landcover, and successional age class into a single matrix (SI1) defining unique combinations of these classes (Table 028) . We directly assigned habitat suitability index scores to these combinations based on vegetation type and age class associations of black-and-white warblers reported by Hamel (1992). However, we assigned higher values to shrub-seedling and grass-forb stands based on data from Thompson and others (1992) and Annand and Thompson (1997).

Forest patch size (SI2) affects this species’ occurrence, with birds notably absent from small forest blocks. Therefore, we fit a logarithmic function (Figure 012) relating forest patch size to habitat suitability index scores derived from probability of occurrence data reported by Robbins and others (1989; Table 029 ). The relative value of a forest of a specific patch size is influenced by its landscape context. In predominantly forested landscapes, small forest patch sizes that may not be utilized in predominantly non-forested landscapes may provide habitat due to their proximity to large forest blocks (Rosenberg and others 1999). Thus we fit a logistic function to hypothetical data that captured this relationship (Figure 013) . We considered landscapes with <30 percent forest to be non-habitat (suitability index score = 0.000), while landscapes with >90 percent forest were considered excellent habitat (suitability index score ≥ 0.900; Table 030 ). We used the maximum value of either SI2 or SI3 to account for small patches in predominantly forested landscapes and large patches in predominantly non-forested landscapes.

Canopy cover (SI4) may also affect black-and-white warbler habitat quality. Thus, we included it as a factor in our overall suitability index model. King and DeGraaf (2000) and Prather and Smith (2003) reported higher densities of black-and-white warblers in forests with relatively open canopies, so we used their data (Table 031) to derive a quadratic function (Figure 014) and quantify the relationship between canopy cover and black-and-white warbler habitat suitability index scores.

We calculated the overall suitability index score as the geometric mean of the geometric mean of individual suitability index functions related to forest structure (SI1 and SI4) multiplied by the maximum suitability index score for either forest patch size or percent forest in the 1-km radius landscape.

Overall SI = ((SI1 * SI4)^0.500 * Max(SI2, SI3))^0.500

Black-and-white warblers occurred in all but 3 of the 88 subsections within the CH and WGCP. Not surprisingly, Spearman rank correlations based on all subsections and only subsections in which black-and-white warblers occurred produced similar results: significant (P ≤ 0.001 for both analyses) positive relationships (r = 0.54 and 0.53, respectively) between average HSI score and mean BBS route abundance. Negative binomial regression indicated HSI scores were positively related to black-and-white warbler abundance and that the HSI model was an improvement over a null model. The generalized R2 for this comparison was 0.40. Route-level analysis predicting black-and-white warbler abundance to average HSI scores within 3-km buffers around the 147 BBS routes in the CH and WGCP also confirmed the improvement of the HSI model over an intercept-only null model and elucidated the positive relationship between HSI scores and black-and-white warbler abundance. However, the strength of the relationship was weaker (generalized R2 = 0.16). We considered this model both verified and validated.