We examined timing of breeding, nest site selection and nest survival of Horned Larks (Eremophila alpestris), Savannah Sparrows (Passerculus sandwichensis) and American Pipits (Anthus rubescens) in an alpine habitat on Hudson Bay Mountain, BC, Canada in 2003–2007. These species partitioned their nesting niches temporally and spatially. We compared nest site characteristics among species using one-way ANOVA and logistic regression. Horned Larks (n = 103 nests) initiated breeding 2 weeks earlier (mean = 1 June) than Savannah Sparrows (n = 52, mean = 14 June) and American Pipits (n = 38, mean = 11 June). Horned Larks and American Pipits nested at similar elevations (means = 1714 and 1719 m, respectively); however, lark nests were more exposed (greater bare ground, rock and lichen/moss cover), with minimal nest concealment, while pipit nests, built into banks and soil mounds, had high concealment. Savannah Sparrows nested at lower elevation (mean = 1649 m) with greater dead vegetative cover. We assessed intraspecific habitat preferences for Horned Larks and Savannah Sparrows using logistic regression; both species chose nest sites with greater availability of their preferred habitat characteristics. We used model selection to evaluate effects of nest site characteristics, nest age, season and year on daily nest survival (DNS). Horned Larks displayed the lowest DNS of 0.954 ± 0.009 (n = 189 nests), which varied with year, season and nest age, but was not influenced by site characteristics. In contrast, DNS was highest for Savannah Sparrows (0.961 ± 0.014, n = 89) with strong responses to nest concealment, year and nest age. American Pipits exhibited an intermediate DNS (0.959 ± 0.009, n = 38), which varied with overhead concealment and elevation. Despite the simple structure of the alpine habitat, there was significant niche differentiation in nest site choices among these species. Preferences for nest concealment were positively related to nest survival in Savannah Sparrows and American Pipits but not Horned Larks, indicating how a common environment can differentially influence behavior and demography.
Read full abstract