Can trackers count free-ranging wildlife as effectively and efficiently as conventional aerial survey and distance sampling? Implications for citizen science in the Kalahari, Botswana

Estimating wildlife abundance is central to conservation. We compared two widely practiced standards for counting animals - aerial strip surveys and ground line transects - with interpreted counts of animal tracks. At equal sampling intensity in semiarid savanna with good visibility all three methods produced similar population estimates and precision for six large herbivores. This comparison adds empirical support for the use of track count data to estimate population density rather than being restricted to ambiguous indices of relative abundance. Although expected to capture more species than aerial surveys, we found line transects limiting because encounter rates by direct sightings were relatively low; a minimum threshold 40 observations was achieved for only 1/3 of antelope species in 648.4 km of transect. By contrast, animal track counts returned exceedingly high encounter rates that allowed estimation of abundance for the entire large predator-prey community and mapping density-distributions more completely. Unlike aerial surveys conducted by Botswana's wildlife authority, the track survey provided opportunity to involve local people in the research process. The track survey cost 40% less than the aerial survey, and could be reduced a further 3-fold if trackers collected data autonomously without motor vehicles. Counting animals by their tracks is ultimately constrained to regions with appropriate substrates. However, in suitable environments like the Kalahari, we suggest that a citizen science driven by expert local trackers could ultimately replace conventional wildlife counts, generating knock-on benefits to conservation beyond improved data.