Evaluation of Bias in Roadside Point Count Surveys of Passerines in Shrubsteppe and Grassland Habitats in Southwestern Idaho1

Breeding passerine abundances in Great Basin shrubsteppe and grassland habitats were surveyed in southwestern Idaho by using 73 pairs of 200-m radius circular point counts. Points were placed along roads and paired with points 400 m away from roads but in similar habitat. Grassland species such as Horned Larks (Eremophila alpestris) and Western Meadowlarks (Sturnella neglecta) accounted for 65 percent and 21 percent of the total number of individuals counted in both habitat types. Typical shrubsteppe bird species such as Sage (Amphispiza belli) and Brewer's sparrows (Spizella breweri), and Sage Thrashers (Oreoscoptes montanus) were less common (11 percent of the total counted). Except for Western Meadowlarks (P < 0.05), all comparisons between the number of individuals of a species counted at points onversus off-roads were statistically insignificant (P > 0.20). Meadowlarks were likely over-sampled along roads because of the presence of adjacent fences, which provide conspicuous song perches. Passerine birds are a conspicuous element of Great Basin shrubsteppe habitats (Rotenberry and Wiens 1978, 1980; Smith and others 1984; Wiens and Rotenberry 1981). Because of several important ecological attributes that these species exhibit, monitoring their abundances over relatively large spatial scales can be of interest. For example, because their position is relatively high in the food chain, and because several species may exhibit rather narrow habitat associations, songbirds may serve as sensitive indicators of ecosystem disturbance or other habitat changes. Before any monitoring scheme can be implemented, however, its methodology must be verified. One sampling method widely used throughout North America is the USDI Fish and Wildlife Service's Breeding Bird Survey (BBS) (Bystrak 1981, Droege 1990). Each survey route is a series of 50 3-min point counts conducted along a 25-mi (40-km) stretch of road. One potential bias of surveying from roads, however, is the inevitable edge, or discontinuity in habitat, created by the road itself. The potential difference between the species and their abundance detected from roads and those detected in similar habitat away from roads ("roadside bias") has only rarely been examined (Hutto and Hejl, Keller and Fuller, Ralph and others, in these Proceedings). We systematically surveyed breeding passerine abundances at paired onand off-road sites throughout the Snake River Birds of Prey Area (SRBOPA) in southwestern Idaho. We asked if abundances of species surveyed along roads were the same as those surveyed in similar habitat but away from roadsides. Our results provide estimates of any This paper was not presented at the Workshop on Monitoring Bird Populations by Point Counts but is included in this volume because of its relevance. 2 Director, Natural Reserve System, Department of Biology, University of California, Riverside, California 92521; Research Ecologist, Raptor Research and Technical Assistance Center, USDI Bureau of Land Management, 3948 Development Ave., Boise, Idaho 83705 USDA Forest Service Gen. Tech. Rep. PSW-GTR-149. 1995 bias due to sampling along roads associated with BBS or similar large-scale surveys conducted in shrubsteppe habitats.