Landscape Modification and Patch Selection: The Demography of Two Secondary Cavity Nesters Colonizing Clearcuts

--Current forestry practices radically modify habitat, particularly by increasing the frequency of early seral stages. We examined the demography of two secondary cavity nesters, the Mountain Bluebird (Sialia currucoides) and the Tree Swallow (Tachycineta bicolor), with respect to landscape changes. The study area contained amosaic of clearcuts ranging from 3 to 27 years in age. The initial distribution of both species was uneven with respect to patch age. Experimental addition of nest boxes: (1) increased the number of nesting pairs of both species, and (2) expanded the age range of occupied patches. Nest-site availability was the primary limiting factor in young patches, but vegetation structure was increasingly important in determining occupation in older patches. Mountain Bluebird patch occupation and Tree Swallow density were best predicted by the proportion of trees 1 to 3 m in height. Although some patches contained multiple pairs of Mountain Bluebirds in natural cavities, experimental patches typically contained no more than one pair, even though excess nest boxes were available. The presence of nonbreeding floaters was an indication of nest-site limitation and suggested that Mountain Bluebirds ometimes defer breeding attempts rather than nest close to conspecifics. At low densities, Tree Swallows also were over-dispersed, but pairs saturated available nest boxes as density increased. Clutch size and number of chicks fledged from successful nests did not vary with patch age for either species, although swallows initiated clutches later in younger compared with older patches. Nest predation varied consistently with patch age; mean nest predation for both species was 86% (n = 30 pairs) in the youngest patches, 37% (n = 34) in mid-age patches, and 25% (n = 22) in the oldest patches. The low densities of both species in older patches was not associated with low productivity, suggesting that habitat selection occurred independently of potential productivity in this novel habitat. Received 21 June 1996, accepted 10 February 1997. LANDSCAPE MODIFICATION can have various effects on bird populations. For example, habitat loss may reduce overall population size (Lynch and Whigham 1984) and decrease population resilience after stochastic events (Freemark and Merriam 1986). Habitat fragmentation has the potential to create source/sink dynamics that cause independent populations to become dependent on source populations (Pulliam 1988). Edge effects can increase nest predation for passerines (Wilcove 1985, Paton 1994, Yahner and Mahan 1996), thus creating ecological traps (Gates and Gysel 1978, Ratti and Reese 1988, Terborgh 1992). Most studies of the effects of landscape modification on birds have focused on species in fragmented 4E-mail: rholt@unixg.ubc.ca original habitat, rather than on species colonizing the newly created habitat. Clearcuts in mature forest increase the frequency of early seral stages in the landscape. Forest management may mimic forest landscape patterns traditionally produced in firedominated ecosystems (Bunnell and Kremsater 1990). Species that naturally inhabit burns may colonize clearcuts as an alternate habitat because the role of fire has been reduced or eradicated in many forest regions (Hutto 1995). However, habitats such as clearcuts may superficially resemble traditional habitats but differ in subtle attributes (Bunnell and Kremsater 1990) that can influence population demography (e.g. Zwickel and Bendell 1972, Dunning and Watts 1990). Habitat quality cannot always be assessed from population density because reproductive success can differ independent of density (Van