Hooded Warbler
Ecoregional Scale Conservation Planning

Made possible through a partnership with the National Wetlands Research Center

Acadian Flycatcher (Empidonax virescens)
The Acadian flycatcher is a long-distance migrant that occurs throughout most of the eastern United States. While Acadian flycatcher populations have declined in the northern portion of their range (particularly the Appalachians) over the last 40 years, populations in the South, particularly along the Atlantic and Eastern Gulf Coastal Plains, have increased (Sauer et al. 2005). However, in the WGCP Acadian flycatcher numbers have declined Table 001 (Table 005) . The USFWS classifies this species as a Bird of Conservation Concern in the WGCP Table 001 (Table 001) . Similarly, PIF considers the species a planning and responsibility species in the CH (regional combined score = 16). In the WGCP, the Acadian flycatcher has a regional combined score of 17, warranting management attention Table 001 (Table 001) .
Relative abundance of Acadian Flycatcher, derived from Breeding Bird Survey data, 1994 - 2003.
image courtesy of www.whatbird.com

Natural History:
The Acadian flycatcher is a forest-interior species associated with water throughout most of its range - bottomland hardwood and cypress forests in the Southeast and riparian forests and ravines in the deciduous forests of the Midwest and Northeast (Whitehead and Taylor 2002). Acadian flycatchers occur in numerous forest types and a wide number of tree species are used for nesting; however, birds are typically associated with mesic forest stands and avoid upland oak-hickory sites (Klaus et al. 2005). Breeding territories are small and average ~1 ha (Woolfenden et al. 2005). Acadian flycatchers typically nest in midstory trees and large shrubs in mature forests. Canopy cover is typically dense (>95 percent; Wilson and Cooper 1998), and the understory is usually sparse (Bell and Whitmore 2000, Wood et al. 2004).

Acadian flycatchers are particularly susceptible to forest fragmentation. Aquilani and Brewer (2004) only found Acadian flycatchers in forest tracts >55 ha in north central Mississippi. Blake and Karr (1987) observed no Acadian flycatchers in woodlots smaller than 24 ha. In east Texas, Acadian flycatchers were not present in riparian buffer strips <70 m wide (Conner et al. 2004). Similar effects have been observed in Missouri (Peak et al. 2004) and Indiana (Ford et al. 2001).

Even in large forested tracts (>600 ha), nest predation and parasitism rates may be 10–20 percent higher if the surrounding landscape is highly fragmented. Nevertheless, Fauth and Cabe (2005) observed no significant effects of parasitism in their Blue Ridge study area where 75 percent of a 10-km buffer around their study site was forested, including 45 percent >250 m from an edge. Disturbance, whether natural (e.g., tornado or pest outbreak) or anthropogenic (e.g., silvicultural treatments- thinning, selective harvesting, clearcutting, and prescribed burning) reduced Acadian flycatcher numbers and productivity in most landscapes (Artman et al. 2001, Duguay et al. 2001, Robinson and Robinson 2001, Twedt et al. 2001, Prather and Smith 2003, Blake 2005).

Model Description:

Our Acadian flycatcher model contains seven factors related to density:

  • Landform
  • Landcover type
  • Successional age class
  • Distance to water
  • Canopy cover
  • Forest patch size
  • Percent forest in a 1-km radius window

The first suitability function combines landform, landcover, and successional age class into a single matrix (SI1) defining unique combinations of these classes Table 006 (Table 006) . We directly assigned suitability index scores to these combinations based on habitat suitability data from Hamel (1992) on the relative quality of different vegetation types and successional stages for Acadian flycatchers. However, we reduced suitability index scores for sapling and evergreen habitats based on data from Hazler (1999).

Acadian flycatchers are typically found near water (Whitehead and Taylor 2002). Therefore, we fit an inverse logistic function to describe the relationship between suitability index scores for Acadian flycatchers and increasing distance to water (SI2) Figure 002 (Figure 002) . Acadians often align at least one edge of their 1-ha territory along a stream or wetland (Woolfenden et al. 2005). Assuming a circular home range, the diameter of the home range (112.8 m) represents the furthest distance from water a bird could be within the home range. Based on this assumption, we assigned all locations ≤100 m from water suitability index scores of 1.000 Table 007 (Table 007) . Acadian flycatchers may also utilize sites >100 m from water but likely occur at lower densities there. Thus, we considered areas 400 m from water (a distance of four home range diameters) to have a suitability index score one-quarter the optimal (i.e, 0.250) and sites ≥ 500 m from water to be non-habitat (i.e., suitability index score = 0.000).

The Acadian flycatcher habitat suitability model also included canopy closure (SI3) as a factor because of the strong affinity of Acadian flycatchers for closed-canopy forests (Prather and Smith 2002). For this factor, we utilized a logistic function Figure 003 (Figure 003) to extrapolate between known break points in the canopy cover-relative density relationship Table 008 (Table 008) .

We also included forest patch size (SI4) as a model parameter because of the sensitivity of Acadian flycatchers to fragmentation (Robbins et al. 1989) and increasing edge density (Parker et al. 2005). We used a logarithmic function Figure 004 (Figure 004) to describe the relatively quick increase in suitability of a forest patch with increasing area (Robbins et al. 1989; Table 009 (Table 009)) . Nevertheless, forest patch size is influenced by the percent of forest in the landscape. 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 et al. 1999). Thus we fit a logistic function Figure 005 (Figure 005) to hypothetical data Table 010 (Table 010) that captured this relationship. Landscapes with <30 percent forest were considered to provide poor habitat (suitability index score ≤ 0.100), while landscapes with >70 percent forest were considered excellent habitat (suitability index score ≥ 0.900). The maximum value of either SI4 or SI5 was used to assess area-sensitivity and to account for small patches in predominantly forested landscapes and large patches in predominantly non-forested landscapes.

To calculate the overall suitability index score, we determined the geometric mean of SI scores for forest structure attributes (SI1 and SI3) and landscape attributes (maximum value of SI4 or SI5 and SI2) separately and then the geometric mean of these means together.

Overall SI = ((SI1 * SI3)0.500 * (Max(SI4, SI5) * SI2)0.500)0.500