HELICOPTER-BASED LINE-TRANSECT ESTIMATES OF SOME SOUTHERN AFRICAN GAME POPULATIONS

The results of helicopter-based counts of some South African animal sp ecies in the Northern Cape Province are presented. A distance-based es timator is used to provide a correction for those individuals present but not seen. Two different types of helicopter with a different numbe r of observers are compared. The results indicate significant undercou nting even for the largest species. Increasing the number of observers provided a less biased and more precise estimate. Visibility profiles for some species are presented. For most species the proportion of th e population seen beyond 200 m from the helicopter is small. An estima tor for the variance of total counts is derived, based on the visibili ty profiles and is evaluated for a selection of species. The results i ndicate that attempts to count animals up to 400 m from the aircraft p rovide an estimate which is both more biased and less precise than est imates obtained by counting animals in the first 200 m only. We also s uggest that the proportion of animals seen with small helicopters with fewer observers, as generally used in game counting on wildlife ranch es, is lower than in this study.