The Bewick’s wren is a small resident passerine that breeds in a variety of
vegetation types, including brushy areas, scrub and thickets in open country,
and open and riparian woodlands (Kennedy and White 1997). This plasticity has
produced conflicting reports of habitat associations in the literature (e.g.,
dry vs. riparian, open woodlands vs. shrub thickets); however, the species
likely responds most strongly to the availability of nest sites. Bewick’s wrens
nest in cavities or opportunistically in crevices up to 10 m high. In the
eastern portion of its range, Bewick’s wrens often occur near human habitations,
particularly farmlands. Population declines of this species may be, in part, a
result of competition with house wrens, which destroy Bewick’s wren nests in
areas where the two species’ ranges overlap (Kennedy and White 1996). Bewick’s
wrens mainly occur in grassland scrub while house wrens occur in secondary
growth on abandoned agricultural land and residential areas. Both species are
able to exploit the full range of these habitat types, though, and populations
of both expanded as these latter habitat types increased. However, as scrub
habitats declined, Bewick’s wrens may have declined because its main source
habitat was no longer abundant.
Our model for Bewick’s wren contains five factors:
- successional age class
- interspersion of forest and open habitatsnag density.
The first suitability function combines landform, landcover, and successional
age class into a single matrix (SI1) that defines unique combinations of these
. We then directly assigned habitat suitability index score to
these combinations based on data from Hamel (1992) on the relative quality of
Bewick’s wren habitat based on vegetation type and successional age class.
We also considered interspersion of forest and grassland habitats (SI2) to be
important for this species, as Bewick’s wrens are most abundant in semi-open
areas containing ~60 percent woodland (Pogue and Schnell 1994;
. We used data from Pogue and Schnell (1994) to define suitability index values along
the diagonal axis of our interspersion table (i.e., where forest and grassland
totaled 100 percent) and completed the rest of the table based on these values.
We also included snag density (SI3) in our model of Bewick’s wren habitat
because as a secondary cavity nester, the species is believed to respond
strongly to nest site availability. We assumed higher snag densities would
decrease competition with other cavity nesters, improving habitat quality.
Specific data relating snag density to Bewick’s wren habitat suitability was not
available. Therefore, we adopted data on average snag density (16.4 snags/ha)
within home ranges of house wrens (Sedgwick and Knopf 1990, Rumble and Gobeille
), a similarly-sized secondary cavity nester, for use in building
a logistic function to quantify this relationship
To calculate the overall suitability index score, we first calculated the
geometric mean of the two suitability indices related to forest structure
attributes (SI1 and SI3), and then the geometric mean of this result and the SI
related to interspersion (SI2).
SI = ((SI1 * SI3)0.500 * SI2)0.500