Avian Point Count, Habitat, and Covariate Data for Subarctic Bird Abundance, Seward Peninsula, Alaska, 2012-2014

Bird surveys consisted of a 10 min point count wherein observers recorded all birds detected by sight or sound within a 250-m radius. We conducted surveys from 21 May to 21 June, 2012–2014. We restricted surveys to times when weather was amenable (no more than light rain and winds <20 km h-1). We measured various aspects of vegetation at each survey point once during the 3-year study. From the survey point, we measured distance (m) to the nearest dwarf shrub (<0.5 m tall), low shrub (0.5-1.0 m), and tall shrub (>1.0 m) in four quadrants, based on cardinal directions (i.e. NE, SE, SW, NW). These four measurements were combined into an average distance for each shrub-height category. We measured additional vegetative characteristics at 10–15 subsampling points within each bird survey area. Subsample points were placed at 5 m intervals along two or three 20 m transects with randomly selected orientations (0-359 degrees). The first transect originated at the bird survey point and the second (and third, in a few areas) was randomly placed within 250 m of the survey point. At each subsample point, we measured visual obstruction (VO), an index of vegetation height and density, from a distance of 2 m and at a height of 0.5 m. Additionally, within a 0.5 x 0.5 m2 quadrat frame centered on each subsample point, we visually estimated percent cover and measured height of the tallest specimen (to the nearest cm) in each of the following categories: alder, dwarf birch, ericaceous shrub (e.g. Empetrum spp., Vaccinium spp.), herbaceous plants, lichen (e.g. Cladonia spp.), and willow, as well as percent bare ground. Percent cover was categorized as 0%, 1-5%, 6-25%, 26-50%, 51-75%, 76-95%, or 96-100% of the quadrat frame; we converted this to the midpoint of each category (0, 2.5, 15, 37.5, 62.5, 85, and 97.5). At one study block with low diversity in structure and composition of plants, subsample vegetation surveys were not completed at 10 of 39 survey points. For these points, we imputed the mean value from the other 29 points in the block.