Home >  Methods >  Avian Data Analysis >  Guidance for Conducting Point Count Surveys



The case for standardization: Ornithologists created standards for conducting avian point-count survey data in 1991. Their goal was to ensure that data on avian abundance could be widely compared, and jointly stored and analysed by national data centres. However, for a variety of reasons, including changes in understanding about how to address incomplete detection rates, standard protocols have not been widely applied to surveys conducted across the boreal forest.

Different methods generate different results: Data collected using different survey protocols cannot be combined simply in boreal-wide analyses of avian density and population size. BAM analyses of 88 bird species indicated that surveys conducted with different count period or detection distances significantly changed the number of birds detected. For example, increasing the count period from 3 minutes to 10 minutes resulted in an average increase of 67% in avian counts. Similarly, increasing the survey radius from 50 m to an unlimited radius resulted in an average increase of over 175% in avian counts.

Use multiple time and distance intervals in point counts to facilitate data comparison: BAM has developed complex methods that allow adjustment of survey data from variable count periods and radii. However, applying certain standards in conducting avian point count surveys in the boreal region would make it easier to combine datasets in the future.

BAM recommendations for conducting point count surveys: BAM based its recommendations for conducting point count surveys on a thorough review of the literature and insights from BAM analyses. They address four aspects of point-count survey protocols that have exhibited the greatest variation over the past 20 years. Minimum recommendations are provided based on Ralph et al. (1995) . Additional recommendations are for surveys that will provide more accurate and precise estimates of abundance that can be calculated using more complex forms of removal or time-of-detection models, distance sampling, multiple-observer methods, and repeated counts. However, biologists should carefully evaluate their study objectives to determine whether the costs of the more complicated data collections will outweigh the sometimes small increases in accuracy and precision of abundance estimates.



Count Period

  • 5 min or 10 min interval (depending on travel time between points)
  • Record initial detections of each bird in 0-3, 3-5, and if applicable, 5-10 min subintervals
  • Subdivide the 5 or 10 min period into 4 or more equal time intervals
  • Record detections and redetections of each bird relative to each time interval

Count Radius

  • Record all observed birds to maximize the number of birds detected
  • Separate birds recorded within a 50-m radius from the count station from those initially detected beyond 50 m
  • Measure exact horizontal distances to birds, or
  • Record initial detections relative to 4 or more distance intervals (e.g., 10, 25, 50, 100, 200, >200 m)
  • Always include 50 m as one of the distance interval boundaries

Number of Observers at Count Station

  • Single observer
  • Use 2 or more observers (following multiple-observer protocols) but separate data among observers when compiled and analysed with surveys conducted by single observers

Number of Visits at each count station

  • Increase the number of statistically independent sampling stations instead of repeatedly counting a smaller number of stations
  • Revisiting sites may increase the number of detected species, and may help with detecting birds with different timing of singing.
  • Revisits should be as closely timed as possible when using repeated count methods.
This page last updated: Sep 09 2012