Maps representing near-national waterfowl density and abundance of 13 individual species and 4 species groups (goldeneyes, mergansers, scoters, and scaup). Predictive models were built using Boosted Regression Tree analysis, data from the Waterfowl Breeding Population and Habitat Survey, and several environmental datasets. Methods used to create 17 species-level models are described in Barker et al. 2014 (ACE 9(2): 7). Guild-Level Maps (cavity nesters, ground nesters, and overwater Nesters) were produced by summing species-level observations and then built models at the guild level. The prediction of total waterfowl abundance was produced by summing the predictions from guild-level models (cavity nesters, ground nesters, and overwater nesters). Species specific model uncertainty are also provided.

Species specific mean projections and uncertainty estimates (coefficient of variation) of80 boreal-breeding songbird species generated using boosted regression tree models for the current period (based on climate data from 1961-1990) and three future time periods (2011–2040, 2041­–2070, 2071–2100).