Projecting boreal bird responses to climate change: the signal exceeds the noise

• DOI: 10.1890/13-2289.1
• Publication Date: 2014-07-17T21:20:54Z
• Authors: D. Stralberg; S. M. Matsuoka; A. Hamann; E. M. Bayne; P. Sólymos; F. K. A. Schmiegelow; X. Wang; S. G. Cumming; S. J. Song
• Keywords: Species distribution

For climate change projections to be useful, the magnitude of must understood relative uncertainty in model predictions. We quantified signal-to-noise ratio projected distributional responses boreal birds change, and compared sources uncertainty. Boosted regression tree models abundance were generated for 80 boreal-breeding bird species using a comprehensive data set standardized avian point counts (349,629 surveys at 122,202 unique locations) 4-km climate, land use, topographic data. changes abundance, we calculated ratios examined variance components related choice global (GCM) two distribution (SDM) uncertainty: sampling error variable selection. also evaluated spatial, temporal, interspecific variation these The mean across increased over time 2.87 by end 21st century, with signal greater than noise 88% species. Across species, represented largest component (0.44) change. Among evaluated, GCM (mean = 0.17) was most important 66% (mean= 0.12) 29% selection =0.05) 5% Increasing number GCMs from four 19 had minor effects on results. range characteristics differed markedly, reinforcing individuality species' challenges one-size-fits-all approaches adaptation. discuss usefulness different conservation depending strength noise, as well dominant source prediction