Boreal Avian Monitoring Strategy

Boreal Avian Monitoring Strategy

 

An effort to monitor the status, trends and distribution of boreal forest birds at a national scale.

Chestnut-sided Warbler. Paruline à flancs marron (Setophaga pensylvanica). Photo Credit: Charles M. Francis

Project Summary

BAM continues to support the design of monitoring and survey programs for boreal birds, including the development of a Boreal Avian Monitoring Strategy, an effort to monitor the status, trends and distribution of boreal forest birds at a national scale. This strategy is being led by Environment Canada and Climate Change (ECCC), and involves several BAM Contributing Scientists.

BAM supported ECCC in the development of a sampling scheme known as the Boreal Optimal Sampling Strategy (BOSS) and analyses were initiated to validate the use of proxy variables for sampling stratification (Wilgenburg et al., 2020). The BOSS design is a randomized hierarchical sampling design stratified by political jurisdictions, ecoregions, and habitat. The BOSS design also incorporates access costs into sample unit selection such that all else being equal, study areas are more likely to be selected if they are less expensive to access. These traits make the BOSS design more efficient than similar alternative spatially balanced sample designs (Van Wilgenburg et al., 2020). The immediate goals for a Boreal Avian Monitoring Strategy are to improve representation of sampling in relation to existing gaps in space and covariates to improve estimates of species’ population sizes, distributions and habitat relationships. In the long-term, the goal is to establish a spatially balanced sample across the boreal for estimating trends in species population sizes. A preliminary target is to sample ~5000 primary sampling units across Canada, representing ~90,000 new point-count sampling locations (See Figure 1). Sample sizes were allocated to spatial strata (based on the intersection of political jurisdiction and ecoregions) based on an analysis of proxy variables thought to contribute to spatial and temporal variation in avian community composition, with greater sampling allocated to strata that are predicted to be more variable.

 

Project Partners

This project is led by partners at Environment Canada and Climate Change (ECCC).  For more information please contact us. 

Publications

Van Wilgenburg, S. L., Mahon, C. L.,Campbell, G., McLeod, L., Campbell, M., Evans, D., Easton, W., Francis, C. M., Haché, S., Machtans, C. S., Mader, C., Pankratz, R. F., Russell, R., Smith, A. C., Thomas, P., Toms, J. D., and Tremblay, J. A., 2020. A cost efficient spatially balanced hierarchical sampling design for monitoring boreal birds incorporating access costs and habitat stratification. PLoS ONE. https://doi.org/10.1371/journal.pone.0234494

bSims: methods and products to support survey design

bSims: methods and products to support survey design

BAM continues to develop tools and methods to help researchers design better monitoring programs.

BAM team member Péter Sólymos created the bSims R package to help researchers design better monitoring programs. The bSims package is a bird point count simulator, which was first presented as a teaching tool at a workshop at the American Ornithological Society 2019 Meeting in Anchorage, AK. The package has since evolved into a standalone tool that (1) allows for easy testing of statistical assumptions and exploring the effects of violating these assumptions and (2) aids survey design by comparing different options. The package presents a spatially explicit mechanistic simulation framework that is based on statistical models widely used in the analyses of bird point count data (i.e., removal models, distance sampling). The workflow involves (1) interactive exploration of multiple setups, e.g., comparing roadside vs. off-road sampling; (2) the settings can be copied from the web application and used in the command line tool to conduct more extensive simulations. 

The bSims package implements the following main functions for simulation: (1) initialize and (2) populate the landscape, (3) animate (individual behaviours described by movement and vocalization events), (4) detect (the physical aspect of the signal transmission), and (5) transcribe (the counts by distance and time intervals).