Spatial extent of human-intrusion effects on subalpine bird distributions

Low levels of human intrusion have become ubiquitous, yet the distance at w hich they affect bird distributions remains unclear. By testing for changes in bird abundance, we assessed uht ther low lo cls of intrusion altered bi rd distributions within and beyond intruded sites. In Wyoming subalpine for ests, we experimentally implemented intrusions within circular 1.0-ha (113- m diameter) sites for 1-2 hr (Snowy Mountains, 1989-1993) or for 5 hr (Pole Mountain, 1991-1993) each week during 10 consecutive weeks of the breeding season. The intrusions did not displace birds during most years, with the following exceptions. Mean abundances for Mountain Chickadee (Parus gambeli ) in the Snowy Mountains (1992) and at pole Mountain (1993), and mean abund ances for American Robin (Turdus migratorius) and Hermit Thrush (Catharus g uttatus) in the Snowy Mountains (1989), were 46-57% low er within intruded sites than they were within control sites. Intrusion did not influence abun dances outside of the 1.0-ha intruded sites. Minimum detectable effect size s (R(2)s for the intrusion effect) were 18-32%: effects of this magnitude a nd larger were detectable with a probability of 0.80. We would therefore ha ve readily detected moderate and large abundance changes had they occurred. The spatial extent of intrusion effects on distributions was thus limited to the actual sires of intrusion, and the effects occurred infrequently. Kn owledge about the distance at which low levels of intrusion do and do not a lter bird distributions is essential for protecting intrusion-sensitive spe cies and avoiding unnecessary restrictions on landscape use by the public.