A comparison of point-count and area-search surveys for monitoring site occupancy of the Coastal California Gnatcatcher (Polioptila californica californica)

Abstract

Improving the efficiency of monitoring protocols prescribed by conservation plans can release typically limited funding for other management and conservation activities. We present an approach for optimizing protocols that considers the precision of parameter estimates, costs of implementation, and broader monitoring-program goals. In a case study of the Coastal California Gnatcatcher (Polioptila californica californica), we compared the efficiency of point-count surveys (with and without playbacks of vocalizations) and area-search surveys (with playbacks) for estimating site occupancy. Conducting an area-search survey of a 2.25 ha plot required an average of 19 min longer than conducting an 18-min point-count survey (15 min of silent observation followed by 3 min of playbacks) at the same location. However, the estimated detection probability (p) during a single visit was lower for point counts (0.41 ± 0.05) than for area searches (0.69 ± 0.05), while both methods generated similar occupancy (Ψ) estimates (0.34 ± 0.06). To obtain the specified level of precision for estimates of occupancy (i.e. with 10% coefficient of variation), the total survey effort (travel time + survey time) was projected to be 35% lower for area searches than for point counts because of differences in detection probability and, thus, in the required numbers of sites and visits per site. For point counts, detection probability increased from 0.35 ± 0.02 to 0.46 ± 0.03 visit-1 after playbacks were broadcast at the end of the count. Free use of playbacks is one of the factors that contributed to the higher detection probability of the area-search method, but playbacks may introduce a slight positive bias into occupancy estimates. Because there are tradeoffs in switching to area-search methods, the decision to switch protocols demands full consideration of monitoring-program goals and the costs and benefits of each survey approach.