# Botswana’s wildlife losing ground as Kalahari Wildlife Management Areas (WMAs) are dezoned for livestock expansion

> From the air, Botswana’s dezoned Kalahari wildlife areas looked almost empty, with no carnivores seen at all. On the ground, local trackers counting tracks along sand roads told a different story: nearly 3,900 large herbivores and around 50 large carnivores, including leopards, cheetahs, brown hyaenas and wild dogs, living in land set aside for cattle ranches. The authors call for trackers’ ground surveys to inform decisions before wild land is given up.

Summary by Kersey Lawrence, Original Wisdom. Updated 9 November 2026. Page: https://www.originalwisdom.com/tracking-literature/2019-keeping-kashe-botswanas-wildlife-losing-ground/

## Citation and access

- **Authors:** Derek Keeping, Njoxlau Kashe, Horekwe (Karoha) Langwane, Panana Sebati, Nicholas Molese, Marie-Charlotte Gielen, Amo Keitsile-Barungwi, Quashe (/Uase) Xhukwe and !Nate (Shortie) Brahman
- **Year published:** 2019
- **Journal or source:** bioRxiv (preprint, Short Communication, not peer-reviewed)
- **Full citation:** Keeping, D., Kashe, N., Langwane, H., Sebati, P., Molese, N., Gielen, M.-C., Keitsile-Barungwi, A., Xhukwe, Q. and Brahman, !N. 2019. Botswana’s wildlife losing ground as Kalahari Wildlife Management Areas (WMAs) are dezoned for livestock expansion. bioRxiv preprint 576496.
- **DOI:** 10.1101/576496
- **Access status:** Open access (CC BY 4.0)
- **PDF:** https://www.originalwisdom.com/wp-content/uploads/2026/11/2019_Keeping_and_Kashe_etal_wildlife-losing-ground-as-wildlife-management-areas-are-dezoned-for-livestock-expansion.pdf

## Tags

- **Species:** Mammals > African wild dog; Mammals > Blue wildebeest; Mammals > Brown hyena; Mammals > Cheetah; Mammals > Gemsbok; Mammals > Greater kudu; Mammals > Leopard; Mammals > Lion; Birds > Ostrich; Mammals > Red hartebeest; Mammals > Spotted hyena
- **Region:** Africa > Botswana
- **Tracking:** Spoor age estimation; Track counts; Tracking; Tracks; Tracks and signs identification
- **Behaviour and ecology:** Behaviour; Connectivity; Daily travel distance; Migration; Movement
- **Methods and technology:** Aerial surveys; Body mass–day range scaling; Bootstrapping; Distribution mapping; Formozov–Malyshev–Pereleshin (FMP) formula; Global Positioning System (GPS); Population estimation
- **Data type:** Existing data; New field data
- **Observer qualification:** Reported but not validated
- **Discipline:** Behaviour and ecology; Wildlife tracking
- **Year published:** 2019

## Detailed keywords

- **Species:** Studied (large herbivores): gemsbok (Oryx gazella), hartebeest (given as “Alcephalus buselaphus” in the table, a misspelling of Alcelaphus buselaphus), kudu (Tragelaphus strepsiceros), ostrich (Struthio camelus; a bird, counted by the authors among the “large herbivores”) and wildebeest (Connochaetes taurinus). Studied (large carnivores): African wild dog (Lycaon pictus), brown hyaena (Parahyaena brunnea), cheetah (Acinonyx jubatus), leopard (Panthera pardus), lion (Panthera leo) and spotted hyaena (Crocuta crocuta). Livestock (cattle) discussed as the cause of habitat loss. Elephants mentioned only in passing.
- **Location and habitat:** Kalahari ecosystem, Botswana: four dezoned Wildlife Management Areas (WMAs) totalling 8,268 km²: GH10 and GH11 (west of the Central Kalahari Game Reserve, CKGR) and SO2 and KW6 (the southern link towards the Kgalagadi Transfrontier Park, KTP). GH11, SO2 and KW6 were surveyed; GH10 was not. The “Schwelle”, the rangelands between KTP and CKGR, is described as the core area for semi-migratory wildlife; the free-ranging Kalahari ecosystem stretches about 700 km from Two Rivers to Kuke Corner without fences.
- **Topics:** land use change; dezoning of Wildlife Management Areas for fenced private livestock ranches; habitat loss; edge effects; landscape connectivity between protected areas; wildebeest migration routes along fossil drainages; gene flow and resilience in drought; limits of aerial surveys (3 to 4% coverage); limits of small GPS-collar samples; track surveys by expert local trackers as an alternative; IUCN Red List status; cattle economy versus tourism; Botswana’s 2014 hunting moratorium; community-based natural resource management (CBNRM).
- **Statistics or models:** Formozov–Malyshev–Pereleshin (FMP) formula (Stephens et al. 2006) linking track interceptions of 24 hours or less over a known transect length to average day range; empirical day ranges for large carnivores, gemsbok and wildebeest, and allometric (body mass) day ranges for hartebeest, ostrich and kudu (Keeping 2014); bootstrapping for 95% confidence intervals; Jolly’s method II (ratio method) for aerial survey population estimates; extrapolation from surveyed to unsurveyed WMA (GH10). Inconsistencies: the abstract gives “50 large carnivores” while the text says 53 and the Table 1 estimates add up to 53 (the abstract figure may be rounded). Abstract and text say about 3,900 / 3,878 large herbivores, which matches the table sum but includes ostrich, a bird.
- **Data type:** New field data: track surveys along sand trails and cutlines, at opportunistic periods between 2014 and 2017 (GH11 transects sampled twice, SO2-KW6 once). Existing data: national aerial survey records from 2012 and 2015.

## Synopsis

### 1. What they did

The Kalahari of Botswana, with the Kgalagadi Transfrontier Park (KTP), the Central Kalahari Game Reserve (CKGR) and the unfenced rangelands between them, may be the largest wild ecosystem left south of the Sahara. In 2014, the same year as Botswana’s hunting moratorium, four long-established Wildlife Management Areas (WMAs) covering 8,268 km² were officially dezoned so they could become fenced, private livestock holdings. When the paper was written, the change had been approved on paper but not yet carried out on the ground.

The authors argue that a common belief, that wildlife has already lost access to these areas, rests on thin evidence: aerial surveys never cover 96 to 97% of the Kalahari, and GPS-collar studies follow very few animals. So they went and counted the wildlife using tracks:

- Track surveys (2014 to 2017). In three of the four dezoned areas (GH11, SO2 and KW6), two expert local observers sat on the front of a four-wheel-drive vehicle driving along little-used sand trails and cutlines. They recorded every crossing of large herbivore and carnivore tracks that was still identifiable, estimated each track’s age to the nearest 24 hours, and recorded its position with the Global Positioning System (GPS).

- Density estimates. Using only tracks 24 hours old or less, they applied the Formozov–Malyshev–Pereleshin (FMP) formula, which turns track crossings per distance into density using how far an animal travels in a day. Day ranges came from earlier studies for carnivores, gemsbok and wildebeest, and from body size for hartebeest, ostrich and kudu. Bootstrapping gave confidence intervals.

- Scaling up. They assumed the unsurveyed area (GH10) had similar densities and estimated numbers for all 8,268 km².

- Comparison with aerial surveys. They estimated populations in the same areas from the 2012 and 2015 national aerial surveys, and mapped track detections (all ages) against aerial sightings.

### 2. What they found

Wildlife in the dezoned areas (track survey estimates, Table 1)

- About 3,878 large herbivores: gemsbok 1,579, hartebeest 865, ostrich 716, kudu 647 and wildebeest 71. Gemsbok, which are sensitive to disturbance, were the most numerous.

- About 53 large carnivores: brown hyaena 18, leopard 18, cheetah 11, African wild dog 4, lion 1 and spotted hyaena 1. Most of these rank Vulnerable or worse on the International Union for Conservation of Nature (IUCN) Red List.

Tracks versus aerial surveys

- The aerial surveys recorded no large carnivores at all, and very few antelopes. In GH11, for example, tracks gave 367 gemsbok detections against 13 from the air; in SO2-KW6, 374 hartebeest detections against 2.

- Aerial estimates jumped around between years because they rested on so few sightings: kudu 0 in 2012 and 96 in 2015; wildebeest 209 in 2012 and 0 in 2015. Ostrich was the one species where an aerial estimate (2,100 in 2015) was higher than the track estimate (716).

- Tracks record animals that moved through over a period of time, not just at the moment a plane flies over, so detection rates were much higher.

Wider effects

- Wildlife numbers inside the dezoned areas are already reduced by edge effects from neighbouring livestock areas, and these WMAs currently buffer the core WMAs. Fenced ranches in the Kalahari are often overstocked, and cattle are released outside the fences when grazing runs out, so the real impact will likely be larger than the numbers above.

- Dezoning SO2 and KW6 removes the southern and most direct link between KTP and CKGR, leaving one narrowed linkage. Dezoning GH10 and GH11 cuts a wildebeest migration route along fossil drainages used in dry years.

- The authors note that cattle make up less than 2% of Botswana’s gross domestic product (GDP), while tourism, mostly based on wildlife and wilderness, contributes about 11.5%. The WMAs were dezoned without an Environmental Impact Assessment.

### 3. What they concluded

- The dezoned areas still hold roughly 3,900 free-ranging large herbivores and 50 large carnivores, which will be consumed, displaced or become potential conflict animals once livestock move in. Wildlife that was effectively public will become de facto private property for a small elite.

- The losses go beyond habitat: reduced connectivity between protected areas limits seasonal movements and gene flow, weakening the resilience of wildlife in a drought-prone environment.

- At the scale of single WMAs, Botswana’s aerial survey returns too few detections to be very useful.

- The authors urge collaboration with expert local trackers, and ground surveys by Kalahari trackers, to assess wildlife in vulnerable habitats before decisions to give them up for livestock are made.

### 4. How tracking was or was not used

Tracking was the main method. Two expert local observers, seated on the front of a slow-moving vehicle, identified the tracks of large herbivores and carnivores crossing sand trails and cutlines, aged each track to the nearest 24-hour period, and recorded its GPS position. Only tracks still identifiable after weathering were counted. All tracks were used to map detections; only tracks 24 hours old or less were used for density estimates. No trailing was reported.

### 5. Tracker or observer certification

Reported but not validated. The observers are described as “expert local observers” and “expert local trackers”, but no certification or test of their skill is reported in this paper. Several co-authors are members of community natural resource trusts in Bere, Kacgae, Ngwatle and Zutshwa, and the author contributions credit six of them with study design and fieldwork; the paper does not say which two did the track counts. The authors cite their earlier comparison of trackers with aerial survey and distance sampling (Keeping et al. 2018), but do not describe it as a test of these observers.

### 6. Other technologies, skills or innovations

- Vehicle-based track transects with observers seated over the front of a four-wheel-drive vehicle.

- GPS recording of each track crossing.

- The FMP formula with day ranges from body mass–day range scaling (Keeping 2014) and bootstrapped confidence intervals, a way of estimating density from track counts without seeing the animals.

- Maps comparing track detections with aerial sightings (Figure 1).

### 7. Implications for conservation (from the paper)

- Botswana keeps 42% of its land for conservation, but the dezoning of 8,268 km² of free-ranging wildlife estate has received little attention compared with elephants and the Kavango–Zambezi area (KAZA).

- The boundaries of WMAs can be changed at any time under Botswana’s Wildlife Conservation and National Parks Act, and these were dezoned without an Environmental Impact Assessment, which the authors see as a poor sign for free-ranging wildlife.

- Decisions about giving up wildlife habitat should be informed by ground surveys by Kalahari trackers, not only by coarse aerial survey data.

### 8. Implications for tracking (Original Wisdom's observations)

_These are Original Wisdom's observations, not the authors'._

- The paper is a clear example of tracks showing what aerial surveys miss: carnivores that were never seen from the air, and antelope detected in their hundreds where planes saw a handful.

- Track counts by local trackers are presented here as evidence for land use decisions. This makes the reliability of the trackers central, and the paper does not report a test of the observers’ skill, so readers cannot judge it from this paper alone.

## Cite this summary

Lawrence, K. 2026. Summary of: Botswana’s wildlife losing ground as Kalahari Wildlife Management Areas (WMAs) are dezoned for livestock expansion (Keeping and Kashe et al., 2019). Original Wisdom, Tracking Literature. https://www.originalwisdom.com/tracking-literature/2019-keeping-kashe-botswanas-wildlife-losing-ground/