Landscape composition influences avian colonization of experimentally created snags

Landscape composition may have a substantial influence on species use of habitat at local scales and effectiveness of structural enrichment practices (e.g., augmenting existing amounts of snags or downed logs). If sufficient mature (i.e., cavity containing) habitat exists at the landscape scale, cavity nester demand for nest sites could be reduced due to the surplus of existing sites (habitat surplus hypothesis). Alternatively, if mature forest acts as source habitat for cavity nesters, increased amounts of mature forest in the landscape will inflate the demand for snags in clearcuts (habitat source hypothesis). In Oregon, USA. we evaluated cavity nesting bird colonization of experimentally created snags distributed across a gradient in amount of mature forest at the landscape scale. We sampled 28 plots, each of which represented one of six combinations of created snag density (similar to 0.5, 1, and 2 snags/ha) and spatial dispersion (clumps of 5-7 snags or dispersed individual snags). We tested whether early successional and mature forest species groups responded differently to plot-level treatments and if the percentage of mature forest (>40 years old) within a 1000 m radius surrounding each experimental plot influenced this response. Mature forest species more readily colonized created snags as the percentage of mature forest declined in the landscape, supporting the habitat surplus hypothesis. Estimated use by mature forest species increased from 7% to 17% as percent mature forest in the landscape declined from 58% to 13% (estimated for average levels of created snag density, 0.4 snags/ha, and a spatial distribution of 45 m). Colonization of snags by early seral species also increased with decreasing mature forest (from 3% to 9%), which may support the habitat source hypothesis. Our study suggests that creating snags in commercial harvest units is an effective practice for increasing structural complexity and maintaining nesting communities of cavity-dependent birds. At the local level, created snags should be spaced as widely as possible within individual harvest units. However, our results indicate that effectiveness of snag creation will be highest in the most intensively managed landscapes where the amount of structurally diverse habitat is limited. (C) 2012 Elsevier Ltd. All rights reserved.