Response of Breeding Passerine Birds to Rangeland Alteration in a North American Shrubsteppe Locality

Publisher Website
Google Scholar
Abstract
(1) During 1980, a large area of sagebrush (Artemisia tridentata Nutt.)-dominated rangeland in southern Oregon, U.S.A., was sprayed with the herbicide 2,4-D, the dead shrubs removed, and crested wheatgrass (Agropyron cristatum (L.)) planted as part of a range improvement programme carried out by government agencies. We monitored breeding bird populations in the treated area for 3 years prior to and 3 years following the alteration. (2) The treatment produced major changes in habitat composition and structure. Before treatment, sagebrush cover was 19-24% and grass cover 1-4%. Afterwards, sagebrush cover decreased to 4-12%, while grass cover increased during the post-treatment period to 10-57% as the wheatgrass became established and cheatgrass (Bromus tectorum L.) invaded. The area of bare, unvegetated ground also decreased, and the height profile of the vegetation changed substantially. (3) There was no evidence of an immediate response by the bird populations to these habitat changes. In the year following the treatment, abundances of sage sparrows (Amphispiza belli (Cassin)) did not change, while the density of Brewer''s sparrows (Spizella breweri Cassin) decreased. During the following 2 years, however, Brewer''s sparrows fluctuated in abundance, sage sparrow abundance gradually decreased, numbers of horned larks (Eremophila alpestris (Linnaeus)) increased, and vesper sparrows (Pooecetes gramineus (Gmelin)) first appeared on the area. These density changes were not related to population changes occuring elsewhere in this shrubsteppe region. (4) Results from earlier work on unaltered rangelands in the region were used to predict the population densities that would be expected on the basis of the structure and floristic composition of the study area following the treatment. Horned lark densities approximated the predicted values, but for both sparrow species the fit was poor. Incorporating a 1 year time lag improved the fit of observed to predicted values in a few cases, but overall made no significant difference. (5) The failure of such bird populations to respond rapidly to major habitat changes may relate to time lags produced by site tenacity of breeding individuals. Such time lags complicate attempts to formulate management policies on the basis of short-term before and after surveys. They also suggest that conducting field experiments in ecology may require longer periods of post-treatment study than has normally been the case.