SEARCH

SEARCH BY CITATION

Keywords:

  • Conservation;
  • geographic range;
  • grassland;
  • habitat;
  • songbird;
  • spatial autocorrelation

Abstract

Aim

The brown-headed cowbird is an obligate brood parasite known to exploit a large number of host species and use a variety of habitats. Much attention has been directed towards uncovering the fundamental factors that affect cowbird abundance; however, no study has evaluated these factors in the context of a biogeographic-scale analysis that takes into account spatial autocorrelation. Our primary objective was to compare the relative influence of geography, land cover and host species on the local abundance of cowbirds.

Location

Great Plains region of the USA.

Methods

We used data from the North American Breeding Bird Survey and the National Land Cover Database to examine the relationships between cowbird abundance and host species, land cover composition and geographic location of a survey route. Multiple regression models were developed for various combinations of these factors. To control for spatial autocorrelation, we used SAM 4.0 (Spatial Analysis in Macroecology) software to implement simultaneous autoregressive modelling of the error term. We then used a model comparison approach to identify the factors that most influence cowbird abundance.

Results

Among all models examined, host species richness was the single most strong predictor and the sole statistically significant predictor. Cowbird abundance increased with host species richness but did not change in any significant way with non-host passerine richness or abundance of host species. Models with land cover variables tended to have the poorest fit to the cowbird abundance data.

Main conclusions

Our results suggest that cowbirds may be attracted to areas with greater host richness and/or recruit better in such areas, although our data did not allow direct examination of either process. In a greater context, our study demonstrates the utility of a spatially based and geographically extensive analysis in finding range-wide factors that affect the local abundance of a species.