# Using spoor counts to analyse the effect of small stock farming in Namibia on caracal density in the neighbouring Kgalagadi Transfrontier Park

> Do caracals leave the Kgalagadi Transfrontier Park to raid Namibian sheep farms, as farmers believe? San trackers riding on the bumper of a slow vehicle read a year of caracal spoor along the border and deep inside the park. Caracals mostly stayed away from the border in the hot season but used it more, and crossed the border road, in the cold season, when wild prey is scarce and sheep are lambing.

Summary by Kersey Lawrence, Original Wisdom. Updated 9 November 2026. Page: https://www.originalwisdom.com/tracking-literature/2006-melville-bothma-spoor-counts-analyse-effect/

## Citation and access

- **Authors:** H.I.A.S. Melville and J. du P. Bothma
- **Year published:** 2006
- **Journal or source:** Journal of Arid Environments
- **Full citation:** Melville, H.I.A.S. and Bothma, J. du P. 2006. Using spoor counts to analyse the effect of small stock farming in Namibia on caracal density in the neighbouring Kgalagadi Transfrontier Park. Journal of Arid Environments 64(3): 436–447.
- **DOI:** 10.1016/j.jaridenv.2005.06.005
- **Access status:** Hosted (freely available elsewhere)
- **PDF:** https://www.originalwisdom.com/wp-content/uploads/2026/11/2006_Melville_and_Bothma_using-spoor-counts-to-analyse-the-effect-of-small-stock-farming-on-caracal-density-in-the-neighbouring-park.pdf

## Tags

- **Species:** Mammals > Caracal
- **Region:** Africa > South Africa
- **Tracking:** Spoor age estimation; Track counts; Tracking; Tracks; Tracks and signs identification; Trailing
- **Behaviour and ecology:** Behaviour; Movement; Road crossings
- **Methods and technology:** Analysis of variance (ANOVA); Chi-square test; Road transects
- **Data type:** New field data
- **Observer qualification:** Reported but not validated
- **Discipline:** Behaviour and ecology; Human–wildlife conflict; Wildlife tracking
- **Year published:** 2006

## Detailed keywords

- **Species:** Studied: caracal (Caracal caracal). Other felids in the area whose tracks had to be told apart from caracal tracks: lion (Panthera leo), leopard (Panthera pardus), cheetah (Acinonyx jubatus) and African wild cat (Felis sylvestris). Livestock on the neighbouring Namibian farms: sheep and cattle (small stock). Small mammals discussed as natural prey, not studied.
- **Location and habitat:** Management roads along the Namibian border near Mata-Mata, southwestern Kgalagadi Transfrontier Park (formerly Kalahari Gemsbok National Park), South Africa, next to small livestock farms in Namibia. Shrubby Kalahari Dune Bushveld in the Savanna Biome: arid savanna on undulating dunes of wind-blown sand over calcrete, 1000 to 1100 m above sea level, mean annual rainfall 153.5 mm, mainly in the hot season. Trees Acacia erioloba, Acacia haematoxylon and Boscia albitrunca; shrubs Grewia retinervis and Rhus tenuinervis; a well-developed grass layer.
- **Topics:** spoor counts as an index of relative abundance; caracal as a “problem animal”; livestock predation; park and farmland boundary; farmer beliefs about predators leaving the park (from an unpublished questionnaire); seasonal range use (hot season October to March, cold season April to September); roads used as movement corridors versus crossed; spoor orientation (longitudinal versus transverse); discrete track set length; back-tracking to avoid double counting; spoor age (only spoor under 24 hours old); seasonal prey availability and sheep lambing; statistical power and Type 1 and Type 2 errors in track surveys.
- **Statistics or models:** standard t-tests and least square means tests (spoor density); chi-square tests (spoor ratio; longitudinal versus transverse movements); analysis of variance (ANOVA) and t-tests (discrete track set length). Inconsistencies in the paper: (1) Table 1 and Table 3 give the hot and cold season mean track set lengths for transects 2, 3 and 4 the other way round, and Table 5 gives some different values again (for example transect 2 year-round 0.85 km in Table 5 but 0.70 km in Table 1; transect 4 cold 0.92 km in Table 5 but 0.85 km in Table 1). (2) Table 1 gives transect 2 a year-round spoor density of 6.60 per 100 km, lower than both its hot (8.20) and cold (7.95) season values, which an average cannot be. (3) The text says mean track set distance did not vary with distance from the border (Table 5), but Table 5 shows a difference between transects 1 and 3 (P = 0.01, year-round and cold season), and the ANOVA was significant year-round and in the cold season. (4) In Table 4 several significance labels do not match the size of the chi-square values (for example 0.00 marked P < 0.05 and 24.40 marked P ≥ 0.05). (5) The ANOVA is said to compare four road strips but reports 2 degrees of freedom between groups. (6) The conclusions describe a possible Type 1 error but describe it as a hypothesis being incorrectly rejected.
- **Data type:** New field data: monthly vehicle spoor count transects, June 2001 to July 2002 (26 surveys, 3290.4 km in total).

## Synopsis

### 1. What they did

In southern Africa, caracals are seen as “problem animals” because they habitually kill small livestock. Namibian farmers next to the Kgalagadi Transfrontier Park told the researchers in a questionnaire that predators from the park regularly cross the border onto their land to kill small livestock, especially in the cold season. If that were true, there should be more caracals along the park’s western border than further inside the park. The authors tested this with spoor counts.

- Two long routes. One followed the Namibian border for 114.6 km. The other, 138.5 km long, ran into the interior up to 30 km from the border. The interior route was split into three parts by distance from the border: 0 to 10 km, 10 to 20 km and 20 to 30 km. With the border road, this gave four transects.

- Monthly counts. From June 2001 to July 2002, experienced Kalahari San trackers sat on a tracking chair on the front bumper of the research vehicle and scanned the sandy management roads at a steady 12 km/h. Both routes were driven on two days in a row each month: 26 surveys over 3290.4 km.

- What they recorded. For each set of caracal tracks: where it entered and left the road, how far it ran along the road, and whether it followed the road (at least 10 m) or simply crossed it. Only spoor less than 24 hours old, as judged by the tracker, was counted.

- Avoiding double counts. When a caracal left the road and came back, the trackers back-tracked the spoor to check whether it was the same animal. Track sets more than 0.5 km apart were treated as different caracals.

- Comparisons. They compared spoor density (sets of tracks per 100 km), how often caracals walked along versus across the roads, and the length of track sets, between transects and between the hot season (October to March) and the cold season (April to September).

### 2. What they found

Spoor density

- Year-round and in the hot season, spoor density was significantly higher on the two transects furthest inside the park (10 to 30 km from the border) than on the border road. In the hot season, the interior transects had at least twice the spoor density of the border road (3.39 per 100 km on the border road; 8.15 to 10.89 inside).

- In the cold season, there was no difference between the border and any interior transect. Spoor density on the border road rose to 6.40 per 100 km.

Walking along roads or crossing them

- Caracals walked along the roads far more often than they crossed them: 20.3 to 1 in the hot season and 8.1 to 1 in the cold season.

- Year-round, caracals tended to walk along the border road and the nearest interior transect more than the transects further in. In the cold season the interior transects seemed to be used less than those near the border.

- Crossings of the roads differed significantly between seasons; walking along the roads did not.

- No caracal spoor was seen crossing the border road into or out of Namibia in the hot season. In the cold season, crossings of the border road did occur.

Track set length

- Mean track set length ranged from about 0.6 to 1.2 km. The authors report that the distance caracals travelled along the roads did not change with season or with distance from the border, though some of their tables show differences (see detailed keywords).

### 3. What they concluded

- Caracals likely change how they use the park’s border area through the year: they mostly avoided it in the hot season and used it more, and crossed the border road more, in the cold season.

- This may reflect less natural prey in the park in the cold season (small mammals breed and are most plentiful in the hot, rainy season), together with sheep lambing on the Namibian farms in the cold season, which offers extra prey.

- Spoor counts are useful for detecting relative area use by caracals, provided the spoor is equally easy to detect in all areas, and they confirm that spoor counts mainly pick up large changes.

- The data set was small, so the results may contain statistical errors. Any management based on them should be applied with caution. The authors recommend long-term road track counts of caracals and other larger predators along the border with Namibia and in the nearby interior of the park.

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

Tracking was the whole method. Trackers riding on the front of a slow-moving vehicle found and identified caracal spoor on sandy roads, telling it apart from the spoor of lion, leopard, cheetah and African wild cat. They judged spoor age and recorded only spoor under 24 hours old, following Stander’s (1998) guidelines. They measured how far each track set ran along the road and whether the animal followed or crossed the road. Where a caracal left the road and returned, the spoor was back-tracked to the point where it first left, so the same animal was not counted twice.

### 5. Tracker or observer certification

Reported but not validated. The trackers are described as “experienced Kalahari San trackers” who are “highly accurate and reliable”, citing Stander et al. (1997), a study that tested Kalahari San trackers. The trackers in this study are not named, no certification is given, and their own skill was not tested here.

### 6. Other technologies, skills or innovations

- A tracking chair mounted on the front bumper of the research vehicle, so the tracker could scan the road at 12 km/h.

- Spoor orientation (along versus across the road) and track set length used as simple measures of how caracals use roads as movement corridors.

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

- Caracals may move to the park border and onto farmland in the cold season, when natural prey is scarce and sheep are lambing, so conflict with farmers may be seasonal.

- Because the data set was small, management decisions based on these results should be made with caution.

- Long-term road track counts of caracals and other larger predators along the Namibian border and in the nearby park interior are recommended.

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

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

- The study shows how much a road spoor count can record beyond a simple tally: spoor age, direction relative to the road, distance followed, and back-tracking to separate individuals all depend on the tracker’s skill.

- The trackers’ accuracy is supported only by a citation to an earlier test of San trackers, so this paper is an example of where testing or certifying the trackers in the study itself would strengthen the findings.

## Cite this summary

Lawrence, K. 2026. Summary of: Using spoor counts to analyse the effect of small stock farming in Namibia on caracal density in the neighbouring Kgalagadi Transfrontier Park (Melville and Bothma, 2006). Original Wisdom, Tracking Literature. https://www.originalwisdom.com/tracking-literature/2006-melville-bothma-spoor-counts-analyse-effect/