The distribution and numbers of cheetah (Acinonyx jubatus) in southern Africa

1. What they did

Cheetahs are hard to count: they are elusive, live at low densities and range over huge areas, mostly outside protected areas. The International Union for Conservation of Nature (IUCN) had recently assessed the species and recommended moving the cheetah from “vulnerable” to “endangered”. These authors set out to make an independent, data-based assessment for Botswana, Namibia, South Africa and Zimbabwe, which hold most of the world’s remaining free-ranging cheetahs, and to show exactly where every record came from.

• Gathering records (2010 to April 2016). They collected cheetah records from 39 sources: Global Positioning System (GPS) and Very High Frequency (VHF) collar locations, camera traps, spoor surveys with experienced local trackers, direct sightings, questionnaires, government surveys and published studies. They also searched online photos, videos, social media and citizen science sites in English, German and Afrikaans, and kept only records they could verify.
• Mapping the range. Records went onto a 10 km by 10 km grid, with a buffer of neighbouring squares. In areas with no records, they mapped “possible” range: places where human and livestock densities were below levels that held more than 85% of known cheetah records, leaving out three deserts and salt pans and very small patches.
• Estimating numbers. They compiled 14 density estimates from local studies and applied them across ecoregions (broad habitat types), using existing counts for Zimbabwe and Kruger National Park.
• Using persecution records. On 185 commercial farms in Namibia they recorded how many cheetahs land managers killed or removed into captivity. A Leslie Matrix model (a population model based on age, breeding and survival) then estimated how many cheetahs an area would need to sustain that loss.

2. What they found

Records and range

• From nearly 20,000 verified observations plus collar data, free-ranging cheetahs were confirmed across about 789,800 km²: 55.6% in Namibia, 22.0% in Botswana, 11.9% in South Africa and 10.4% in Zimbabwe.
• Crowd-sourced records alone added 12.9% of the occupied grid squares and confirmed another 10.8%. Most came from well-visited protected areas, while research data came mainly from outside them.
• Only 18.4% of the known free range is in strictly protected areas (IUCN categories I to IV).
• Another 742,800 km² with low human and livestock densities may hold cheetahs.
• A managed population lived on small fenced reserves in South Africa (176 adults in July 2016).

Densities and numbers

• Local density estimates ranged from 0.18 to 0.90 cheetahs per 100 km², with an area-weighted mean of 0.48.
• They estimated 3,577 free-ranging adult cheetahs in confirmed range, with the most (1,451) in the Kalahari Xeric Savanna. If possible range were fully occupied, up to about 3,250 more could exist, for a maximum of 6,827.
• Their estimate is 11% lower than the IUCN’s 4,029 for the same four countries.

Persecution in Namibia

• 26.5% of land managers actively persecuted cheetahs, and 49.7% saw them as causing conflict.
• Managers trapped 245 cheetahs: 196 were killed, 32 put into permanent captivity and 17 translocated. This equals 0.30 adult cheetahs removed per 100 km² per year, including 0.10 breeding-age females.
• Ten farm owners removed 71.9% of these cheetahs, and one manager accounted for 36.0%. Wildlife farming and hunting operations removed the most; recreational land uses were the most tolerant.
• Under the most hopeful breeding and survival values (growth of 29.9% a year), about 0.67 adult cheetahs per 100 km² would be needed to sustain this loss. With less favourable values, the figure rose as high as 1.61.

3. What they concluded

• Their independent estimate supports the call to review the cheetah’s status and consider listing it as endangered.
• Cheetahs are highly concentrated: about 55% of the known population lives in about 400,000 km², much of it on unprotected farmland with high persecution. The authors warn that this stronghold could “hollow out”.
• The close match between the persecution-based density (0.67) and the measured density on Namibian farmland (about 0.7) suggests cheetahs there are, at best, holding their own in a productive area despite intense persecution.
• A few landowners cause most of the removals, so focusing efforts on persecution hotspots could greatly reduce losses.
• Crowd-sourced data are valuable, especially in busy protected areas, and could free researchers to work on unprotected land where most cheetahs live.
• The upper estimate is unlikely; it mainly shows where surveys are needed (for example central Botswana, south-eastern Namibia and northern South Africa).

4. How tracking was or was not used

The authors did not do tracking themselves. Spoor (track) surveys were one of their compiled data sources: they requested “intensive spoor surveys with experienced local trackers” and received data from 10 spoor survey programmes. Five of the 14 density estimates came from spoor surveys in Botswana and Namibia, analysed with a track-count formula (Funston et al. 2010, refined) or calibrated to the cheetahs’ daily travel distance (Keeping 2014). Spoor surveys covered by far the largest study area (109,612 km² in south-western Botswana).

5. Tracker or observer certification

Reported but not validated. The trackers in the spoor surveys are described only as “experienced local trackers”; no certification or test of their skill is reported in this paper.

6. Other technologies, skills or innovations

• A stepwise process for finding and verifying crowd-sourced records (blogs, news, social media, photographers), described in Appendix S1.
• GPS and VHF telemetry, camera traps and photographic identification of individual cheetahs.
• Presence data shown at a 10 km grid so that sensitive locations stay protected while all input data can be shown publicly.
• Global human population (LandScan 2015) and livestock (Gridded Livestock of the World) maps, ecoregion maps and the World Database on Protected Areas.
• A Leslie Matrix model in a spreadsheet, used to turn persecution numbers into a minimum density.

7. Implications for conservation (from the paper)

• The southern African cheetah population is probably more threatened than believed, because most animals live in two ecoregions overlapping commercial farmland with high persecution risk.
• Botswana’s plan to convert about 8,230 km² of Wildlife Management Areas to livestock production, in the core high-density range, would almost certainly increase conflict.
• Recording persecution hotspots matters, because a single land manager can remove large numbers of cheetahs.
• Range maps should show their evidence, so that areas based on expert opinion can be targeted for new surveys.