Primary Cavity-Site Selection by Birds

Abstract

Current recommendations for snag management, which focus on characteristics of individual snags or stocking levels in forests managed for timber production, not only obscure the importance of forest characteristics surrounding a potential nesting site but also fail to meet the management objectives or abilities of the small landowner. During the summers of 1978 and 1979, a random sample of 816 snags (standing dead trees) was studied in central New York. Sixty-seven percent of the sample was in mature maple (Acer spp.)-ash (Fraxinus spp.)-elm (Ulmus spp.), the remainder was in 2nd growth woodland. A set of 21 characteristics of snags and 19 characteristics of forest sites was analyzed to determine which best predicted bird use. Use was defined as the presence of -1 bird cavity in a snag or a forest site. Forest characteristics were sampled in randomly located 0.049-ha circles (N = 61) in 1979. Stepwise logistic regression revealed that forest characteristics (total snag basal area, tree species diversity, and number of tree species) were more reliable predictors of bird use than were snag characteristics (diameter at breast height [dbh], amount of bark, height, and species). Snag management based on selecting suitable forest sites and on maintaining or creating suitable snags within those sites is recommended. Use of logistic regression models by field managers is discussed. J. WILDL. MANAGE. 50(4):576-583 Numerous bird species, especially cavity nesters, depend upon snags for roosting or nesting sites (Scott et al. 1977). Many of these species range throughout the northeastern United States where most of the forested land is privately owned and, therefore, where maintenance of cavity-nesting bird populations depends upon the management decisions of the landowner. If avian use of snags and forest sites can be associated with some easily measured variable, management for cavity nesters can be incorporated easily into forest management strategies (Hicks 1983). Agricultural trends in New York are typical of many northeastern states (Morris 1977). Once occupying 90% of New York lands, agricultural interests (excluding forestry) now occupy <33% of the state (Dwyer 1977). Natural reforestation has occurred on a large portion of the released northeastern farmland. Residual mature forests remain on sites least suitable for agriculture; many of these forests are riparian. In central New York, as elsewhere in the Northeast, American elms (U. americana) in mature forests were killed by Dutch elm disease (Ceratocystis ulmi) during the late 1950's and now stand as snags. This study evaluated the utility of snags to primary cavity-nesting birds in New York forests. Objectives were to differentiate between used and unused snags, using simple criteria, and to quantify important characteristics of the forest surrounding snags used by cavity nesters. Acknowledgment is due to J. Schmidt for collecting field data in 1978. W. M. Block, L. A. Brennan, and M. E. Richmond critically read the manuscript, and S. J. Schwager provided statistical advice. Financial support was provided by the Cornell Biol. Field Stn. at Shackelton Point, the N.Y. Coop. Wildl. Res. Unit, and Hatch Grant NYC-147419 (R. J. Gutierrez). R. W. Mannan, R. N. Conner, and R. F. Labisky provided numerous constructive comments.