The hooded warbler breeds in a variety of habitats, from mixed-hardwood forests in the northern portion of its range to cypress-gum swamps in the South. Regardless of forest type, hooded warblers prefer mesic sites in large forest tracts (>15 ha; Evans-Ogden and Stutchbury 1994). Although nest success in small forest patches is not significantly lower than in large patches (Buehler and others 2002), females may avoid small fragments and males use edge less than its availability (Norris and Stutchbury 2002, Norris and others 2000). Occupancy of a site by a nesting pair increases with shrub height and the percentage of vegetation between 1 and 2 m.
Hooded warblers nest in shrubs within small forest clearings or in the dense understories of closed-canopied forests. Thus, territories often include a mix of open and closed canopy conditions. Gaps created by tree fall or selective logging are particularly attractive (<0.5 ha; Annand and Thompson 1997, Moorman and others 2002, Whittam and others 2002), and hooded warblers colonize these sites within 1–5 years. Nest sites in Canada had denser ground vegetation, fewer tree stems, less small tree basal area, and greater large tree basal area than control sites (Whittam and others 2002). Bisson and Stutchbury (2000) concluded canopy gaps and understory vegetation density were the most important factors affecting site selection. Repeated burning, which removed understory vegetation, reduced hooded warbler abundance in Ohio (Artman and others 2001). Hooded warblers are a common cowbird host, which may explain their sensitivity to fragmentation (Donovan and Flather 2002).
The habitat suitability model for hooded warblers contains seven variables:
- land cover
- successional age class
- small (<2.5 cm d.b.h.) stem density
- canopy cover
- forest patch size
- percent forest in a 1-km landscape
The first suitability function combines landform, landcover, and successional age class into a single matrix (SI1) defining unique combinations of these classes
. We directly assigned suitability index scores to these combinations based on relative habitat quality rankings from Hamel (1992) for hooded warblers in the Southeast.
Hooded warblers occur in dense understories in mature forested habitats. Therefore, we included both small stem density (SI2) and canopy cover (SI3) in our model. We fit a logistic function
linking small stem density to suitability index scores based on data from Annand and Thompson (1992) and Moorman and others (2002;
. We assumed the average stem density measured at nest sites by Moorman and others (2002; 4700 stems/ha) was representative of ideal habitat conditions for hooded warblers (i.e., suitability index score = 1.000) and no upper threshold existed above which habitat suitability declined (i.e., stems could not be too dense). We also fit a logistic function
to data from Annand and Thompson (1997;
to link canopy cover values to suitability index scores.
We included forest patch size (SI4) as a model predictor because of the negative effect of fragmentation on this species. We used an exponential curve
to predict hooded warbler habitat suitability from forest patch size based on data from Evans-Ogden and Stutchbury (1994) and Kilgo and others (1998;
. The suitability of a specific forest patch is influenced by the percent of forest in the landscape (SI5). Small forest patches that would be otherwise unsuitable may be occupied when in close proximity to a large forest block or in a predominantly forested landscape (Rosenberg and others 1999). We assumed landscapes with <10 percent forest were non-habitat (suitability index score = 0.000) and landscapes >90 percent forest were optimal (suitability index score = 1.000) and fit a logistic function to this data to capture the relationship between forest in the landscape and habitat suitability
. We used the maximum suitability index score from either S4 or SI5 to account for the higher suitability of small forest patches when they occur in a heavily forested landscape.
The overall suitability index score was calculated as the geometric mean of the geometric mean of the suitability index values from the landform, landcover, and successional age class matrix, small stem density, and canopy cover functions (SI1, SI2, and SI3) multiplied by the maximum value of either the forest patch size or percent forest in the 1-km radius landscape functions (SI4 and SI5).
Overall SI = ((SI1 * SI2 * SI3)0.333 * Max(SI4, SI5))0.500