BAM worked with Environment Canada, U.S. Geological Survey, and U.S. Forest Service to recommend minimum and enhanced standards for conducting point count surveys (including multiple time and distance intervals). The enhanced standards provide sufficient data to allow estimation of correction factors to account for detection bias in avian monitoring.
BAM conducted a gap analysis of its database to indicate which geographic areas and habitat types were under-represented in current sampling programs. They identified for Environment Canada almost 400 potential new sampling locations where Breeding Bird Atlas or other surveys could address these gaps, while recognising the need for road access to monitoring sites.
BAM is providing scientific support and access to its database to support development of bird-habitat models that will be used by the Joint Oil Sands Monitoring program (JOSM) in the oil sands areas of Canada. The models will be used to predict how changes in habitat from oil sands activity influence potential population size as well as to assess population trends within a larger regional context. BAM will also assist with coordinating data from other research and monitoring programs, and developing means to convert point count data (from human observers and recording devices) into density estimates.
Environment Canada is using portions of BAM's dataset to conduct a statistical analysis of options for design for a monitoring program for songbirds in proposed protected areas in the NWT. The results indicate how many sample points, visited over what frequency, will be necessary to provide sufficient information to understand local population trends through time.
Avian surveys conducted from locations on or near roads may significantly bias the resultant population estimates. Results of an analysis we conducted for Environment Canada showed that for the majority of boreal species, using data collected from roadside locations would result in overestimation of population size for many common and/or highly vocal species. Conversely, for a smaller number of species, particularly those associated with mature forest types or forest interiors, or less vocal species, roadside counts would result in underestimated populations.
Although not an official collaborative project, BAM was able to offer assistance in database quality control to the Breeding Bird Survey (BBS) staff of the Canadian Wildlife Service, Environment Canada. BAM staff generated or corrected geo-referenced locations for Breeding Bird Survey (BBS) stop locations while incorporating the BBS data into the BAM dataset. By assigning geo-referenced locations to the data, the information can be mapped and analysed spatially. We incorporated into the BAM dataset the BBS data from the years 1996 through 2009, based on availability at the 50 stop level for the bird survey data, resulting in inclusion of over 37,000 individual stop locations over a 13-year period and over 260,000 sampling events.