# Prey selection by caracal in the Kgalagadi Transfrontier Park

> What does a caracal really hunt? Following caracal tracks for 537 km with Kalahari San trackers and analysing 116 scats, researchers found that caracals in the Kgalagadi Transfrontier Park live mostly on rodents, especially springhare, but also take surprising numbers of other small carnivores, from Cape foxes to African wild cats. A few scats held sheep wool, mostly in the cold lambing season.

Summary by Kersey Lawrence, Original Wisdom. Updated 9 November 2026. Page: https://www.originalwisdom.com/tracking-literature/2004-melville-bothma-prey-selection-caracal-kgalagadi/

## Citation and access

- **Authors:** H.I.A.S. Melville, J. du P. Bothma and M.G.L. Mills
- **Year published:** 2004
- **Journal or source:** South African Journal of Wildlife Research
- **Full citation:** Melville, H.I.A.S., Bothma, J. du P. and Mills, M.G.L. 2004. Prey selection by caracal in the Kgalagadi Transfrontier Park. South African Journal of Wildlife Research 34(1): 67–75.
- **Access status:** Hosted (freely available elsewhere)
- **PDF:** https://www.originalwisdom.com/wp-content/uploads/2026/11/2004_Melville_and_Bothma_etal_prey-selection-by-caracal.pdf

## Tags

- **Species:** Mammals > Caracal
- **Region:** Africa > South Africa
- **Tracking:** Feeding signs; Scat; Tracking; Tracks; Tracks and signs identification; Trailing
- **Behaviour and ecology:** Behaviour; Feeding; Hunting; Intraguild interactions; Predation
- **Methods and technology:** Descriptive statistics
- **Data type:** New field data
- **Observer qualification:** Reported but not validated
- **Discipline:** Behaviour and ecology; Human–wildlife conflict; Wildlife tracking
- **Year published:** 2004

## Detailed keywords

- **Species:** Studied: caracal (Caracal caracal). Prey found in scats or targeted in hunts: scrub hare (Lepus saxatilis), springhare (Pedetes capensis), ground squirrel (Xerus inauris, hunts only), Brant’s whistling rat (Parotomys brantsii), striped mouse (Rhabdomys pumilio), Namaqua rock mouse (Aethomys namaquensis), short-tailed gerbil (Desmodillus auricularis), hairy-footed gerbil (Gerbillurus paeba), Highveld gerbil (Tatera brantsii), African wild cat (Felis silvestris), black-backed jackal (Canis mesomelas, hunts only, none successful), bat-eared fox (Otocyon megalotis), Cape fox (Vulpes chama), striped polecat (Ictonyx striatus), yellow mongoose (Cynictis penicillata), steenbok (Raphicerus campestris), dorper sheep (Ovis aries, wool in scats), kori bustard (Ardeotis kori), redcrested korhaan (Eupodotis ruficrista), unidentified birds, unidentified beetles (Coleoptera). Plants in scats: camel thorn (Acacia erioloba), grey camel thorn (Acacia haematoxylon), black thorn (Acacia mellifera), Aptosimum elongatum, tsamma melon (Citrullus lanatus) and Kalahari sour grass (Schmidtia kalihariensis). Mentioned but not studied: leopard (Panthera pardus) and cheetah (Acinonyx jubatus), at low density in the area.
- **Location and habitat:** Southwestern Kgalagadi Transfrontier Park, South Africa, along the South African and Namibian border near Mata-Mata: an area stretching 60 km north from Mata-Mata rest camp to O’Kuip windmill and about 20 km into the park. Arid savanna (Shrubby Kalahari Dune Bushveld) on undulating dunes at 1000 to 1100 m, mean annual rainfall 182 mm, temperatures from -10°C to 45°C. Camel thorn, grey camel thorn and shepherd’s tree (Boscia albitrunca) with Grewia and Rhus shrubs and a well-developed grass layer. Sheep farms lie across the border in Namibia.
- **Topics:** caracal diet and prey selection; scat analysis (hair cuticle scale imprints and hair cross-sections, teeth and jaw fragments); reconstructing hunts from spoor; hunting success; carnivores as prey; diurnal activity; predation on small livestock (sheep) and the cold lambing season; birds, beetles and tsamma melons in the diet; comparison of scat analysis with spoor-tracking; comparison with other caracal diet studies in southern Africa and Saudi Arabia. Inconsistencies in the paper: (1) the text says springhare, Cape fox and Lepus (hare) were hunted 23, 11 and 15 times with 2, 3 and 1 successes, but Table 1 gives Cape fox 10 hunts and 2 successes, and scrub hare 14 hunts; (2) the text gives 10.1% hunting success, but Table 1 shows 27 successes in 327 hunts (8.3%); (3) the text says 23 of 74 identified hunts (31.1%) were on carnivores, but Table 1 adds up to 24 (32.4%, the figure given later in the Discussion); (4) rodents made up 60.9% of prey items identified (Results and Table 2), but the Discussion calls this 60.9% “of scats”; (5) the abstract and Results say wool was in eight scats, but the Discussion says “four of the six scats that contained wool” were from the cold season; (6) the black-backed jackal row in Table 1 is shifted into the scat columns, although the text says jackals were not found in scats and were only recorded from tracking; (7) springhare are said to make up 6.1% of kills recorded, a figure that does not appear in Table 1 (which gives 7.0% of all hunts, and 2 of 27 kills).
- **Statistics or models:** Descriptive statistics only: percentage occurrence in scats, percentage of prey items, percentage of all ingested items, hunting success. Sample size of 116 scats judged adequate by comparison with earlier sample-size studies, but too small for seasonal comparisons.
- **Data type:** New field data, June 2000 to July 2002: 116 fresh scats collected while following caracal spoor, and 537 km of spoor followed, with 327 hunting attempts recorded. The introduction mentions questionnaires done during the study (caracals seen as killers of small livestock), and the discussion cites observations by Namibian farmers, but neither is reported in detail.

## Synopsis

### 1. What they did

Caracals are widely seen as a problem animal where sheep farms border protected areas, and the border between South Africa’s Kgalagadi Transfrontier Park and Namibian farmland is one such place. The authors set out to find which prey caracals use in the park, and whether they move out onto the farms to take small livestock.

They used two methods between June 2000 and July 2002:

- Following spoor. Experienced Kalahari San trackers helped follow and interpret fresh caracal tracks over 537 km. Every hunting attempt the tracker interpreted as such was recorded, successful or not, and the prey targeted was identified where possible. A hunt counted as successful only when prey remains were found.

- Scat analysis. They collected 116 fresh scats while following caracals. In the laboratory they washed out the undigested remains, compared teeth and jaw fragments with reference keys, and identified prey hair under the microscope from its cuticle scale patterns and cross-sections, checking at least 20 hairs from each scat against reference hair from known prey.

They then compared what the two methods showed, and compared the caracals’ diet with other studies in southern Africa.

### 2. What they found

From scats

- Of 37 suitable mammal prey species in the area, caracals used 16.

- Rodents made up 60.9% of all prey items identified. Springhare were the most important single prey, found in 37.1% of scats, followed by Highveld gerbils (31.9%), Brant’s whistling rats (22.4%) and striped mice (19.8%).

- Five species of smaller carnivore made up 10.7% of prey items: African wild cat, bat-eared fox, Cape fox, striped polecat and yellow mongoose.

- Steenbok, the only wild antelope found, appeared in just 3 scats.

- Sheep wool was found in 8 scats (6.9%). Only three of these were near the Namibian border; the others were more than 13 km into the park, and one about 23 km from the border.

- Birds were in 13.0% of scats, including kori bustard and redcrested korhaan. Beetles were in 23.3% of scats. No reptile remains were found.

- Plant material made up 20.5% of the items found, mostly Kalahari sour grass (in 38.8% of scats). Tsamma melon seeds were in three scats.

From following spoor

- 327 hunting attempts were recorded, but the target prey could be identified in only 74 (22.6%).

- The paper reports a hunting success of 10.1% (see the inconsistencies in the detailed keywords).

- The prey most often hunted were springhare, hares and Cape fox. Caracals also tried to catch black-backed jackals and ground squirrels, which never showed up in the scats; none of these attempts succeeded.

- About a third of the identified hunts were on other carnivores, and seven of these succeeded, involving six species.

- On two occasions a caracal stalked and killed a kori bustard roosting on the ground, and one tracked caracal apparently ate a tsamma melon.

- No hunts on reptiles were seen.

### 3. What they concluded

- Rodents are the caracal’s main prey in the park, and springhare are especially important. The authors link the high number of rodents to two good rainy seasons, and suggest that springhare fill the role that rock hyrax play for caracals elsewhere.

- Smaller carnivores, some weighing 2.5 to 8 kg, seem to take the place of antelope as larger prey here. With few small antelope other than steenbok, and few leopards and cheetahs, these small predators may be more available to caracals.

- Prey that are active by day (Brant’s whistling rats, ground squirrels, yellow mongooses), together with sightings of caracals active in daylight, show some daytime activity, especially in the cooler winter months.

- Caracals near the border occasionally eat small livestock. Most wool-containing scats were found in the cold season (April to September), when the sheep lamb, which matches what Namibian farmers report. The authors see this as opportunistic, when newborn lambs are vulnerable, and conclude that caracals living mostly in the park do cross onto the farms.

- Beetles probably supplement the diet, and tsamma melons may supply moisture.

- The two methods complement each other: scat analysis reveals small items such as insects and small rodents that tracking misses, while following spoor shows hunting success and which prey are hunted.

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

Tracking was one of the two main methods. Kalahari San trackers followed fresh caracal spoor for 537 km and interpreted each hunting attempt from the tracks, recording the prey targeted where they could tell, and a kill only when prey remains were found. All 116 scats were collected while following caracal spoor, so the scat sample also came from tracking. The authors note that small prey eaten whole leave no remains, so tracking probably underestimates hunting success, and they think most of the hunts with unidentified prey were on rodents.

### 5. Tracker or observer certification

Reported but not validated. The trackers are described as “experienced Kalahari-San trackers”, with no names, certification or test of their skill given; the acknowledgements thank J. Januarie for contributing greatly to collecting the field data. The authors support the method by citing earlier Kalahari studies that used trackers, including Stander et al. (1997).

### 6. Other technologies, skills or innovations

- Hair identification from scats: gelatine imprints of hair cuticle scale patterns, and hair cross-sections made by embedding hair bundles in a wax mixture, compared with reference keys and with reference hair from known prey prepared the same way.

- Caracal hair in scats was treated as a sign of grooming, confirming the scat came from a caracal, not as food.

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

- Although caracals are widely condemned as a problem animal, they help control small mammals that eat the natural grazing that livestock depend on.

- Caracals that live mostly in the park do cross into neighbouring farmland and occasionally take small livestock, mainly during the cold lambing season.

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

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

- The study shows what each method sees. Scats revealed small rodents and beetles that the trackers never recorded, while the tracks revealed hunts that left nothing in the scats, such as failed attempts on jackals and ground squirrels, and how often hunts succeeded.

- Reading a hunt from spoor has limits: the target could be identified in fewer than a quarter of the 327 hunts, and a kill was only counted when remains were found.

## Cite this summary

Lawrence, K. 2026. Summary of: Prey selection by caracal in the Kgalagadi Transfrontier Park (Melville and Bothma et al., 2004). Original Wisdom, Tracking Literature. https://www.originalwisdom.com/tracking-literature/2004-melville-bothma-prey-selection-caracal-kgalagadi/