Detection ratios of riparian songbirds
This article is part of a larger document. View the larger document here.Abstract
This paper presents preliminary results from the first year of a two-year study designed to evaluate bias in a typical songbird survey by examining differences in detection ratios among species, cover types, and time of the season. Detection ratios, calculated as number of individuals detected during a 15-25 minute fixed-width transect survey divided by the number of individuals shown to be present through intensive nest searching and territory mapping, were obtained for riparian songbirds on 22, 1.5-ha plots in southeastern Oregon. During intensive territory mapping and nest finding, overall, 80 percent of territories were identified after 3 hours of effort per plot, the estimated number of territories stabilized after about 9 hours, and nests or probable nests were located on 204 (78 percent) of the 261 identified territories. The pooled detection ratio for all species was 0.46 (±0.03 se). Species-specific detection ratios ranged from 0.59 in Song Sparrows to 0.20 in MacGillivray’s Warbler. The detection ratio was higher in willow/meadow cover type than in aspen, possibly due to the more frequent nest loss in willow/meadow and presumably higher detectability of re-nesting pairs, the higher density of individuals and species in aspen, and the greater structural complexity of aspen. Total detections were higher earlier in the season (17-22 May) than later (31 May-19 June), suggesting a decline in detectability with nesting phenology. Further study to understand the visual and auditory detectability of individual territory-holders as a function of distance from the observer, stage of nesting, paired status, date, time of day, and various other factors, will help evaluate potential sources of bias and aid in identifying rapid survey methods that are likely to be most efficient.