Using dogs to find cats: detection dogs as a survey method for wide-ranging cheetah


M. S. Becker, S. M. Durant, F. G. R. Watson, M. Parker, D. Gottelli, J. M'soka, E. Droge, M. Nyirendam, P. Schuette, S. Dunkley, R. Brummer


Journal of Zoology


Rapid global large carnivore declines make evaluations of remaining populations critical. Yet landscape-scale evaluations of presence, abundance and distribution are difficult, as many species are wide-ranging, occur only at low densities and are elusive. Insufficient information-gathering tools for many large carnivore species compounds these challenges. Specially trained detection dogs have demonstrated effectiveness for carnivore surveys, but are untested on extremely sparse, wide-ranging species, such as cheetah Acinonyx jubatus. In this study, we conducted the first rigorous cheetah survey using detection dogs in a key transboundary area in the remote Liuwa-Mussuma Transfrontier Conservation Area (TFCA) in Western Zambia. We proposed to (1) evaluate the effectiveness of detection dog versus spoor surveys in detecting cheetah presence; (2) extract and analyze DNA from scat samples to estimate minimum population size and genetic effective population size; (3) determine the extent of cheetah occurrence in the unprotected transboundary corridor. Two detection dog teams surveyed 2432 km(2) containing 74 randomly located transects in the transfrontier area. Twenty-seven cheetah scats were detected and confirmed by genetic analysis, while no cheetah spoor was detected, clearly demonstrating the superiority of detection dogs in detecting cheetah presence. Combining scat samples with opportunistically collected samples, we estimated 17-19 cheetahs, an effective population size of 8-14 and a density of 5.9-6.6 per 1000 km(2). Cheetah utilized key transfrontier areas outside of the national park; however, because utilization appears low, improved connectivity and protection for these areas are critical. Approximately one third of Africa's estimated cheetah resides in protected areas, with 87% in transboundary areas. Our study demonstrates the efficacy of detection dog survey methods in providing information on cheetah across large landscapes. It will have particular value in areas where other survey means may be impossible, such as TFCAs, where size, remoteness and lack of accessibility often make traditional survey methods difficult or cost prohibitive.



How is this information collected?

This collection of Montana State authored publications is collected by the Library to highlight the achievements of Montana State researchers and more fully understand the research output of the University. They use a number of resources to pull together as complete a list as possible and understand that there may be publications that are missed. If you note the omission of a current publication or want to know more about the collection and display of this information email Leila Sterman.