Estimating red deer Cervus elaphus populations:: an analysis of variation and cost-effectiveness of counting methods

1. Different counting methods are currently used to estimate red deer populations in the open range in Scotland, but there are few data available to compare variation in estimates, or relative cost-effectiveness. 2. While it is impossible to determine the accuracy of counts (as real numbers are unknown), variation within and between different methods can be measured by repeat counts of the same area within as short a period as possible. 3. This study aimed to quantify the variation observed from repeat counts using each of four methods (ground, helicopter, infrared helicopter and dung-counting methods) at one of three study sites in late winters 2003, 2004 and 2005. Additional data from digital camera images of groups from counts in other areas of Scotland were also used to assess the accuracy of visual counts. 4. Coefficients of variation (CVs) within any method of between 5% and 16% were recorded, consistent with previous comparisons for red deer open range counts in Scotland. CVs were lowest for ground and helicopter counts. The infrequency of optimal conditions was likely to limit the applicability of infrared counts in Scotland. 5. In terms of cost-effectiveness, helicopter counting was the least labour-intensive, with costs of other techniques depending on the availability of existing manpower as an overhead cost. 6. It is concluded that helicopter counts are most likely to minimize errors while maximizing cost-efficiency. Accuracy can be improved by the use of digital photography for counting larger deer groups. Estimates are likely to be improved further by increasing the frequency of counts and using the same methods, counters and routes for repeat counts.