A collaborative project that developed spatially explicit boreal bird models in Alberta, Canada, to inform continental bird conservation.


For nearly two decades, BAM has worked to develop robust methods for estimating population sizes of North American boreal birds. In 2019-20, we continued to work on a project comparing population estimates of boreal birds in Alberta, derived from spatially explicit models and the Partners in Flight approach, which applies adjustments to North American Breeding Bird Survey (BBS) counts to get population estimates. We also quantified the effects of detectability, roadside bias, and other factors on these population estimates (see Box 1 for more details). In 2020, this work was published in The Condor.

Figure. Differences between Partners in Flight roadside sample-based population estimates and spacial model « pixel »-based population estimates with roadside count and habitat representation differences across 81 species in Northern Alberta, Canada. Red dots indicate species that are well-represented along roads; blue dots indicate species that are under-represented along roads. Filled dots indicate statistically significant differences between population estimates, P<0.05. (Species codes and values for species represented by unfilled dots can be found in the paper.)

Box 1. Comparing spatially explicit models (PIX) and the Partners in Flight (PIF) approach to estimate population sizes of boreal birds in Alberta, Canada.

Avian conservationists have relied on population estimates from Partners in Flight (PIF) that primarily use roadside data from the Northern American Breeding Bird Survey (BBS). However, BBS was not designated to estimate population sizes.

We compared the PIF approach with spatially explicit models (PIX) incorporating roadside and off-road point-count surveys for 81 landbird species in Bird Conservation Region 6 in Alberta, Canada. We found that time and detection distance adjustments explained average differences between the PIF and PIX estimates.

The variation in population estimates among species was explained mostly by differences in the roadside count and habitat representation assumptions. This variation was large enough to change the ranking of which species were needed to better understand and tease apart the complexity surrounding roadside point counts.

DOI: 10.1093/condor/duaa007, contact Péter Sólymos; solymos@ualberta.ca


This work was a collaborative effort with the Alberta Biodiversity Monitoring Institute (ABMI), Canadian Wildlife Service (Environment and Climate Change Canada), and United States Geological Survey.

Blog Post

Made in Alberta models help continental bird conservation, ABMI, June 17, 2020

Media Coverage

CBC News: Survey estimates much higher Alberta bird populations than thought, Jun 21, 2020