Relationships among Weather, Habitat Structure, and Ponderosa Pine Forest Birds

Abstract

Avian community structure during the breeding season in a ponderosa pine (Pinus ponderosa) forest of northern Arizona was influenced by weather and a series of timber harvest treatments. Fewer birds and bird species were present after a winter with the heaviest snowfall on record and low temperatures than after milder winters. Bird density was greater (P < 0.05) on the light and medium cut plots than on the untreated plot. A cluster analysis of bird densities over plots and the 3-year study period indicated treatment effects were more important in determining bird community composition than weather effects. J. WILDL. MANAGE. 50(2):253-260 Breeding bird populations are affected by many factors that may or may not be interactive (Haila et al. 1980, Sabo 1980), such as habitat alteration (Anderson and Shugart 1974, Szaro and Balda 1979a,b) and critical climatic events that are not directly associated with habitat (Fretwell 1972, Smith 1982). Climatic factors have received increasing attention in bird community studies (Wiens and Rotenberry 1981). Winter and spring weather conditions can influence the onset of breeding, permanent resident survivorship, and the influx of summer residents into an area (Monson 1973). Moreover, severe or variable winter weather conditions frequently may overshadow food availability as the major determinant of wintering bird distribution (Laurance and Yensen 1985). Yet bird populations in the ponderosa pine forests reflect changes in both habitat configuration and weather conditions. To understand the determinants of bird community structure in ponderosa pine forests of northern Arizona, we conducted a study to examine responses of bird species to habitat alteration, examine effects of winter and spring weather conditions on breeding bird populations, and determine the relative impact of these factors on bird populations. We thank K. E. Evans, C. E. Grue, R. M. King, R. R. Roth, and J. Verner for critical reviews. Special thanks to C. M. Snyder for patience with many rewrites of this manuscript and to R. D. Babb for drafting the figures. STUDY AREAS AND METHODS Study Areas Five areas within a 21-km radius on the Coconino National Forest, Coconino County, 4363 km south of Flagstaff, Arizona, were studied from 1973 to 1975. The areas had a wide variety of disturbance including clearcut, heavy cut, medium cut, light cut, and untreated plots. All study areas had been cut before this study beg n except the light cut plot, which was tre ted in early spring 1974. No trees had been commercially removed from the untreated area for 60 years. All study areas were set up as 15ha gridded plots except for the clearcut area, which encompassed 45 ha. Habitat manipulation differed on the 4 treated study areas (Fig. 1) (Brown et al. 1974). Large trees and small, dense tree thickets had been selectively removed from the light cut plot. The medium cut plot had been treated to create strips of trees (36 m wide) alternating with cleared areas (18 m wide). Rows of uncut trees were left at intervals of 120 m to break up visual continuity. The heavy cut plot was severely thinned and had slash piled in windrows spaced at regular intervals. The clearcut plot had slash and debris windrowed in piles 51.5 m high. Since treatment in spring 1967, shrubby Gambel oak (Quercus gambelii) had grown next to the slash piles. Alligator juniper (Juniperus deppeana) was removed from all treated sites. For a more complete description of the study plots see Szaro and Balda (1979a).