# Density of Large Predators on Commercial Farmland in Ghanzi, Botswana

> How many cheetahs, leopards and brown hyaenas live on Botswana’s cattle and game farms? A San tracker and a biologist read fresh spoor on sand roads across the Ghanzi farmlands and found brown hyaenas far more common than assumed, and leopards much scarcer than farmers believe. The study also shows how killing predators can skew the very surveys meant to count them.

Summary by Kersey Lawrence, Original Wisdom. Updated 9 November 2026. Page: https://www.originalwisdom.com/tracking-literature/2012-boast-houser-density-large-predators-commercial/

## Citation and access

- **Authors:** Lorraine K. Boast and AnnMarie Houser
- **Year published:** 2012
- **Journal or source:** South African Journal of Wildlife Research
- **Full citation:** Boast, L. K. and Houser, A. 2012. Density of large predators on commercial farmland in Ghanzi, Botswana. South African Journal of Wildlife Research 42(2): 138–143.
- **DOI:** 10.3957/056.042.0202
- **Access status:** Hosted (freely available elsewhere)
- **PDF:** https://www.originalwisdom.com/wp-content/uploads/2026/11/2012_Boast_and_Houser_density-of-large-predators-on-commercial-farmland.pdf

## Tags

- **Species:** Mammals > African wild dog; Mammals > Brown hyena; Mammals > Cheetah; Mammals > Leopard
- **Region:** Africa > Botswana
- **Tracking:** Spoor age estimation; Track counts; Tracking; Tracks; Tracks and signs identification
- **Behaviour and ecology:** Behaviour; Habitat selection
- **Methods and technology:** Chi-square test; Detectability; Global Positioning System (GPS); Population estimation; Road transects
- **Data type:** New field data
- **Observer qualification:** Other experienced trackers
- **Discipline:** Human–wildlife conflict; Wildlife tracking
- **Year published:** 2012

## Detailed keywords

- **Species:** Studied: cheetah (Acinonyx jubatus), leopard (Panthera pardus) and brown hyaena (Hyaena brunnea). Also detected: African wild dog (Lycaon pictus), whose spoor was found off the transects, so no density could be estimated. Searched for but not detected: lion (Panthera leo) and spotted hyaena (Crocuta crocuta). Wild prey in the area: common duiker (Sylvicapra grimmia), steenbok (Raphicerus campestris) and greater kudu (Tragelaphus strepsiceros); game farms hold a mix of small and large antelopes. Also mentioned: cattle and feral domestic dogs. Inconsistency: the text says wild dog spoor was detected (off transect), while the Table 1 caption says wild dog spoor was not detected.
- **Location and habitat:** Ghanzi commercial farmlands, northwest Botswana (21.70°S, 21.62°E), the country’s largest commercial farming area. Hardveld with some sandveld; bush to open tree savanna. Three survey areas: mixed cattle farms and adjacent game ranches east of Ghanzi town (494 km²), and game ranches (408 km²) and cattle farms (435 km²) west of Ghanzi. Farms are divided by cattle or non-electrified game fences that predators can cross over or under.
- **Topics:** spoor surveys (spoor counts), spoor frequency and spoor density, calibration of spoor density to animal density, large predator density outside protected areas, farmland and land use (cattle farms versus game ranches), prey availability, human–predator conflict, persecution and removal of predators, landowner perceptions of leopard abundance, closed-population assumption, Botswana National Predator Management Strategy, trophy hunting quotas.
- **Statistics or models:** Road penetration, spoor frequency and spoor density calculated after Stander (1998). Cheetah and brown hyaena densities from the Funston et al. (2010) calibration equation; leopard density from the Stander (1998) equation, because the Funston equation is not suitable for low densities. Pearson’s chi-square test with post hoc standardized residuals (z scores), significance at P = 0.05. Inconsistency: the text gives brown hyaena spoor counts as game 81, cattle 78, mixed 98, while Table 1 gives mixed 81, game 78, cattle 98 (the total, 257, is the same). The ranges quoted in the text for each species (for example cheetah 0.00 to 0.58) are the range across the three surveys, not the 95% confidence intervals in Table 1 (cheetah 0.08 to 0.76).
- **Data type:** New field data: mixed farmland survey March to June 2007; game and cattle surveys run side by side February to July 2008.

## Synopsis

### 1. What they did

Large predators such as cheetahs, African wild dogs and brown hyaenas range widely and often live outside protected areas, on farmland. Botswana is a core country for cheetah and wild dog conservation, yet its national predator strategy relied on density figures from protected areas and on assumptions for farmland. This was the first study to estimate large predator densities on Botswana farmland.

The authors ran three spoor surveys on the Ghanzi commercial farmlands:

- Mixed farmland (March to June 2007): cattle farms and neighbouring game ranches east of Ghanzi town.

- Game ranches and cattle farms (February to July 2008): two surveys run at the same time west of Ghanzi, alternating between them each day.

Each survey area had two fixed transects along sand roads, kept as straight as possible to avoid counting the same animal twice. A San tracker with a predator biologist searched for spoor from a vehicle driven slowly (about 10 to 13 km/h) from sunrise until 11:30, when spoor is easiest to see. Every large predator spoor was identified and recorded with the date, a Global Positioning System (GPS) location and the number of animals. Only fresh spoor (less than 24 hours old) was used, and road sections disturbed by vehicles or rain in the previous 24 hours were left out. Where several tracks of one species were found on the same transect, the team judged whether they came from the same or a new animal, so that each animal was counted only once a day. Animals travelling together were counted individually.

The spoor counts were converted into densities with calibration equations from earlier studies of predator populations of known size, and the surveys were compared with chi-square tests.

### 2. What they found

Effort

- In total the transects were driven for 3535 km (1026 km mixed, 1269 km game, 1240 km cattle), covering 1337 km² of farmland.

Spoor found

- Cheetah, leopard, brown hyaena and wild dog spoor were found. The wild dog spoor was off the transects, so no wild dog density could be estimated. No lion or spotted hyaena spoor was found.

- Spoor counted: 38 cheetah, 7 leopard and 257 brown hyaena.

Densities (per 100 km², mean of the three surveys)

- Brown hyaena: 2.18 (range 1.82 to 2.38 across surveys). This is far higher than the earlier estimate of 0.01 to 0.1 per 100 km², and similar to the Kgalagadi Transfrontier Park and the Makgadikgadi Pans.

- Cheetah: 0.21 (range 0.00 to 0.58), similar to the presumed 0.15 to 0.56.

- Leopard: 0.10 (range 0.00 to 0.25), lower than the assumed 0.95 to 1.5, and in contrast to local landowners who believe leopards are abundant.

Differences between farm types

- Cheetah spoor: 23 on mixed farmland, 13 on game ranches and only 2 on cattle farms, a significant difference.

- Leopard spoor: 6 on game ranches, 1 on mixed farmland and none on cattle farms. This difference was also significant, but there were too few leopard spoor to meet the test’s assumptions.

- Brown hyaena spoor did not differ between the three areas. (The text and Table 1 give the hyaena counts per area in a different order.)

Removals of predators

- Two leopards were killed in the cattle study area in the three months before that survey, and one leopard was shot in the mixed farmland area during the survey.

- Ten cheetahs had been removed from the game ranch area between November 2006 and June 2007 because of conflict with people.

### 3. What they concluded

- Measured densities differed from the figures that had been assumed for Ghanzi farmland, showing the risk of managing predators on assumptions rather than surveys.

- Brown hyaenas occur evenly across cattle and game farms, probably because, as scavengers, they find food on all farm types. Their wide movements may make their spoor easier to find than the calibration allows for, so their density may be overestimated, but the area still appears important for them.

- Cheetahs and leopards were found mostly on game ranches and mixed farmland, most likely because of prey. Cattle farms had no stocked game, plus feral dogs and poaching, and generally fewer wild game were seen there. The lower cheetah density on game ranches than on mixed farmland may reflect the cheetah removals of the year before.

- The leopard result may be biased low because leopards were killed during and just before the surveys, which breaks the assumption of a closed population. Landowners’ belief that leopards are abundant may be inflated by the high level of conflict.

- The variable results for cheetahs and leopards show the need for repeated surveys, and for surveys in other farming areas to confirm the land use preference and its cause.

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

Tracking was the survey method. A San tracker found and identified large predator spoor along sand road transects from a slow-moving vehicle, aged it (only spoor less than 24 hours old was counted), counted the number of animals, and judged whether repeated tracks of one species on a transect belonged to the same or a different animal. No trailing is reported.

### 5. Tracker or observer certification

The trackers were tested on spoor identification before they were hired for this study; details of the test and results are not given. No certification is reported. The authors cite Stander et al. (1997) as having shown that San trackers can reliably and accurately identify and age spoor. The tracker or trackers are not named; Vivien Kent, Adam Camm and Service are thanked for their help with the survey.

### 6. Other technologies, skills or innovations

- GPS locations recorded for every predator spoor.

- Calibration equations (Stander 1998; Funston et al. 2010) that turn spoor density into animal density.

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

- Farmland remains important habitat for large predators, especially cheetahs, and Ghanzi appears important for brown hyaenas.

- Human–predator conflict on Ghanzi farmland is still high; better livestock management and community education are needed for people to live alongside predators.

- Leopards, the predator farmers rank as most problematic, may be scarcer on this farmland than thought. Monitoring of predator off-take and trophy size, together with future surveys, is needed to find population trends.

- Accurate density estimates are needed to target conservation and conflict work, set trophy hunting quotas and guide national predator policy.

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

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

- Spoor counts let a small team estimate the density of rare, persecuted predators across farmland, where other methods would be hard to use.

- The study depends on the tracker’s judgments: aging spoor and deciding whether repeated tracks belong to one animal or several. The trackers were tested before hiring, but the test and its results are not reported, which makes it harder for readers to judge the reliability of the counts.

## Cite this summary

Lawrence, K. 2026. Summary of: Density of Large Predators on Commercial Farmland in Ghanzi, Botswana (Boast and Houser, 2012). Original Wisdom, Tracking Literature. https://www.originalwisdom.com/tracking-literature/2012-boast-houser-density-large-predators-commercial/