ACCURACY AND PRECISION OF TECHNIQUES FOR COUNTING GREAT BLUE HERON NESTS

Managing colonially nesting wading bird populations is often hampered by the lack of reliable information on colony sizes and population status. Therefore, we evaluated 9 techniques for accuracy and precision of counting great blue heron (Ardea herodias) nests in South Carolina. We examined accuracy by comparing percent deviation of technique estimates from corresponding ground counts. Precision of the different techniques was evaluated by comparing 95% confidence intervals for predicting individual colony size and the statewide nesting population. Techniques included point count, perimeter count, 40% sampling with 4-m-wide transects, 20% sampling with 2- and 4-m-wide transects, aerial estimate, aerial count, aerial photographic count, and post-nesting ground count. Error rates varied among techniques and for colonies containing < 320 nests. Ninety-five percent confidence intervals were wide and not within 20% of estimate totals for any technique. In large (> 320 nests) colonies, post-nesting ground counts, aerial estimates, and all transect techniques had acceptable confidence interval widths (< 20%). Except for 40% transect sampling, the approximate 95% confidence intervals for the statewide nesting population were wide (> 20%). Although counting techniques were not precise enough to be used for an estimate of the statewide nesting effort, managers using a combination of aerial counts of small (< 50 nests) colonies and ground counts of other colonies should be able to detect a 15% change in nesting between years.