Interannual variation in bird–landscape relations: understanding sources of a pervasive conservation dilemma

Authors

  • Samuel K. Riffell,

  • Kevin J. Gutzwiller

Errata

This article is corrected by:

  1. Errata: Erratum Volume 121, Issue 7, 1174, Article first published online: 16 April 2012

S. K. Riffell (sriffell@cfr.msstate.edu), Dept of Wildlife and Fisheries, Box 9690, Mississippi State Univ., MI 39762, USA. – K. J. Gutzwiller, Dept of Biology, Baylor Univ., Waco, TX 76798, USA.

Abstract

Interannual variation in bird–landscape relations is pervasive, and it generates decision dilemmas for conservationists who use such relations to support landscape-level planning, acquisition and management. Interannual variation occurs in two ways. Model consistency refers to agreement among years in the variables and directions of relations in a model for a species. R2 consistency measures variation in the strengths of association between bird and landscape variables for a species among years. We examined how distributional and biological traits of species, model-selection certainty (likelihood the best model was selected in a given year), and model complexity (number of landscape variables) were associated with interannual variation in bird–landscape relations for 72 species in the Appalachian Mountains, USA. Abundant, narrowly distributed species, and cavity nesters exhibited higher model consistency, but model-selection certainty, not species’ traits or model complexity, had the most influence on model and R2 consistency. Increasing model-selection certainty by improving sampling methods and model-selection strategies may be the most effective way to cope with interannual variation in bird–landscape relations.

Ancillary