# Refining population density estimation from track counts: improving daily travel distance estimates through trailing of large herbivores in the Kalahari, Botswana

> How far does an eland walk in a day? In Botswana's Kalahari, certified trackers followed the trails of eland, hartebeest, kudu, springbok and ostrich for 24 hours at a time, giving the first measured daily travel distances for these species there (eland: about 10.5 km). The results also showed that wide-ranging hunters and scavengers need a much larger correction than other species when day ranges are predicted from body size.

Summary by Kersey Lawrence, Original Wisdom. Updated 5 November 2026. Page: https://www.originalwisdom.com/tracking-literature/2025-gielen-araldi-refining-population-density-estimation/

## Citation and access

- **Authors:** Marie-Charlotte Gielen, Alessandro Araldi, Marie Jardeaux, Seitshiro Pule, Senxwai Mosololo, Xee Fire Seganaphohu, Duela Seganaphofu, Tebelelo Gabaikanye, Pogiso Ithuteng, Derek Keeping and Nicolas Schtickzelle (Gielen and Araldi share first authorship)
- **Year published:** 2025
- **Journal or source:** Movement Ecology
- **Full citation:** Gielen, M.-C., Araldi, A., Jardeaux, M., Pule, S., Mosololo, S., Seganaphohu, X. F., Seganaphofu, D., Gabaikanye, T., Ithuteng, P., Keeping, D. and Schtickzelle, N. 2025. Refining population density estimation from track counts: improving daily travel distance estimates through trailing of large herbivores in the Kalahari, Botswana. Movement Ecology 13: 53.
- **DOI:** 10.1186/s40462-025-00582-1
- **Access status:** Open access (CC BY-NC-ND 4.0)
- **PDF:** https://www.originalwisdom.com/wp-content/uploads/2026/11/2025_Gielen_and_Araldi_etal_refining-population-density-estimation-from-track-counts.pdf

## Tags

- **Species:** Mammals > Blue wildebeest; Mammals > Common duiker; Mammals > Eland; Mammals > Gemsbok; Mammals > Giraffe; Mammals > Greater kudu; Birds > Ostrich; Mammals > Red hartebeest; Mammals > Springbok; Mammals > Steenbok; Mammals > Warthog
- **Region:** Africa > Botswana
- **Tracking:** Footprints; Spoor age estimation; Track counts; Tracking; Tracks and signs identification; Trailing
- **Behaviour and ecology:** Behaviour; Daily travel distance; Movement
- **Methods and technology:** Akaike Information Criterion (AIC); Body mass–day range scaling; Bootstrapping; CyberTracker; CyberTracker software; CyberTracker Tracker Evaluation; Formozov–Malyshev–Pereleshin (FMP) formula; Geographic Information System (GIS); Global Positioning System (GPS); Indigenous knowledge; Linear regression; Population estimation
- **Data type:** Existing data; New field data
- **Observer qualification:** Certified trackers
- **Discipline:** Wildlife tracking
- **Year published:** 2025

## Detailed keywords

- **Species:** Trailed for day ranges: eland (Taurotragus oryx), red hartebeest (Alcelaphus buselaphus caama, spelled "Alcephalus" in the paper), greater kudu (Tragelaphus strepsiceros), springbok (Antidorcas marsupialis) and ostrich (Struthio camelus). Density estimated from track counts (11 species): the five above plus gemsbok (Oryx gazella), steenbok (Raphicerus campestris), common duiker (Sylvicapra grimmia), blue wildebeest (Connochaetes taurinus), warthog (Phacochoerus africanus) and giraffe (Giraffa giraffa). Existing day ranges used for the correction coefficients: African wild dog (Lycaon pictus), brown hyena (Hyaena brunnea), cheetah (Acinonyx jubatus), leopard (Panthera pardus), lion (Panthera leo), spotted hyena (Crocuta crocuta), aardwolf (Proteles cristatus), black-backed jackal (Canis mesomelas), African wild cat (Felis silvestris), yellow mongoose (Cynictis penicillata) and aardvark (Orycteropus afer). Also mentioned: porcupine (trailed from burrows in unpublished work).
- **Location and habitat:** Khutse Game Reserve and the southern part of the Central Kalahari Game Reserve, Botswana (between 23°–24°S and 24°–25°E); trailing covered about 3900 km² and the track survey about 2800 km² (2761.56 km²). Semi-arid Kalahari, almost entirely covered in sand, with grasses, shrubs and scattered trees; annual rainfall about 400 mm, falling from November to April. Pans hold water only briefly after rain; four artificial waterholes run all year. A double game fence along the southern and eastern borders of Khutse has many gaps. Track surveys ran on existing sand roads (59% of them). The trackers came from the Kaudwane community.
- **Topics:** improving day range (average daily travel distance) estimates for the Formozov–Malyshev–Pereleshin (FMP) formula; a two-day forward-trailing method developed with local trackers; whether day range stays stable within a season; correcting body mass–based day range predictions for Kalahari species by foraging strategy (social hunters and scavengers versus herbivores and solitary hunters); trailing versus GPS collars; seasonal density estimates; transect design and survey effort; integrating Indigenous knowledge with modern methods.
- **Statistics or models:** FMP formula, D = (π/2) × x / (S × M); body mass–day range scaling rules for Artiodactyla and Carnivora (Carbone et al.); linear regressions forced through zero, compared by Akaike Information Criterion corrected for small samples (AICc); linear regression of day range against date; nonparametric bootstrapping with bias-corrected and accelerated 95% confidence intervals; R version 4.3.2; ArcGIS 10.8 for cleaning trails.
- **Data type:** New field data: forward trailing between 15 May and 15 August 2023 (cold dry season; 61 trails kept); monthly track surveys over 12 months covering the 2021–2022 wet season and 2022 dry season (255 km per month). Existing data: day ranges for 15 other species from published studies (trailing, following habituated animals, GPS and radio tracking); body masses and scaling rules from the literature.

## Synopsis

### 1. What they did

The Formozov–Malyshev–Pereleshin (FMP) formula converts track counts into population density, but it needs each species' average daily travel distance (day range). Good measured day ranges exist for few Kalahari species. Day ranges predicted from body mass are too short for Kalahari species, and until now were corrected with a single coefficient (Keeping 2014). The authors set out to improve both measured and predicted day ranges.

- Forward trailing, developed with local expert trackers: on day 1, a target animal or group was spotted from a vehicle, usually around midday or in the afternoon, and its location, group size, sex and age, behaviour and a photograph were recorded in a CyberTracker app. On day 2, the team began trailing from that spot at least 4 hours before the time of the sighting, walking exactly along the animal's footsteps with GPS fixes every second. The trail ended where the target was seen again, or had just fled, at the same time as the day before, giving exactly 24 hours. If the target was not reached in time, trailing continued until the trackers judged, from the trail's freshness, that 24 hours of travel had been covered. A vehicle followed at a distance for safety, with a recorder on the roof watching for the target.

- Species and team: eland, red hartebeest, greater kudu, springbok and ostrich, none of which had measured day ranges in the area, were trailed between May and August 2023 by three Indigenous trackers from Kaudwane (two certified Trailing Professional, one Trailing III) and two data recorders. Trails were checked and cleaned; running caused by the observers was removed, and trails with uncorrectable biases were discarded.

- Stability: for each species, they checked whether day range changed over the three months.

- New correction coefficients: they combined their five species with existing measured day ranges for 15 other Kalahari species (20 in total) and compared each with its day range predicted from body mass, testing whether the correction differed between social hunters and scavengers and the other species.

- Case study: monthly driven track surveys over a year, by three certified trackers recording only fresh tracks, were converted to seasonal density estimates for 11 herbivore species.

### 2. What they found

New day ranges

- They collected 71 trails and kept 61 (one or two per working day).

- Average day ranges: eland 10.47 km (12 trails), red hartebeest 5.84 km (13), greater kudu 5.84 km (18), ostrich 5.16 km (5) and springbok 1.79 km (13).

- For the 50 trails where the target was found again, timing was within half an hour of 24 hours.

Stable within the season

- No general trend in day range over the cold dry season. Ostrich day ranges increased by about 97 m per day, but with only five trails this could be spurious; springbok showed a small, nearly significant decrease.

Two corrections are better than one

- A model with separate corrections for the two foraging groups fitted far better than one shared correction (adjusted R² 0.97).

- Measured day ranges were longer than predicted for both groups, but much more so for social hunters and scavengers: the correction factors were 4.698 for social hunters and scavengers and 1.917 for herbivores and solitary hunters, a difference of nearly 2.5 times.

- Springbok was the only species that travelled less than predicted, probably because springbok rarely leave pans in the cold dry season. Ostrich fitted the herbivore line almost perfectly.

Densities

- Estimated numbers in the 2761.56 km² survey area (wet season / dry season): steenbok 290 / 226; gemsbok 223 / 204; common duiker 223 / 68; red hartebeest 154 / 83; ostrich 105 / 64; greater kudu 90 / 75; springbok 76 / 3; giraffe 71 / 53; eland 50 / 21; warthog 44 / 17; blue wildebeest 24 / 1.

- Steenbok, duiker and gemsbok were the most abundant, and wildebeest and springbok notably few. Dry season densities were lower for all species.

### 3. What they concluded

- Trailing works: their forward-trailing protocol captured detailed 24-hour movements of large herbivores. It is a real, non-invasive alternative to GPS collars, at about 110 US dollars per trail.

- Advantages of trailing: many independent individuals or groups can be sampled, effort can be spread over space and time, movement can be recorded in fine detail, and Indigenous knowledge is involved and valued. Trackers can also interpret behaviour along the trail, such as resting, foraging, running, fighting, digging and mating.

- Two correction coefficients: these let practitioners correct body mass–based day ranges for Kalahari species according to foraging strategy. Brown hyenas, jackals and wild dogs range widely in search of food, while felids, which stalk or ambush, travel less.

- Choosing day ranges: stand-in values should come from similar places and times. For example, spotted hyenas in the Masai Mara travel about half as far each day (12.4 km) as those in the Kalahari (26.5 km). Seasonal effects on day range need further research; duiker may move more in the wet season.

- Survey design matters too: springbok, present as at least 3–4 herds, were estimated at only 3 in the dry season because transects skirted rather than crossed the pans where they live. Eland may be underestimated because large herds use areas with few transects.

- Adapting the method: smaller animals with small home ranges may need back-trailing after rain, as done for steenbok; burrows can serve as starting points for nocturnal species; camera traps can help spot targets; GPS collars may still suit rare or dangerous species.

- Communities: track-based methods can join Indigenous knowledge with modern science, reinforcing cultural identity, recognising knowledge and providing employment.

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

- Trailing: three certified Indigenous trackers led two-day forward trailing of individual animals or groups, walking exactly along their footsteps for 24 hours and judging from the trail's freshness when 24 hours had been covered if the animal was not found in time.

- Protocol design: the trailing protocol was designed in consultation with local expert trackers, based on their knowledge of animal behaviour.

- Track counts: in the case study, three certified trackers on a moving vehicle (two in front, one on the roof) recorded fresh tracks (24 hours or less) of herbivores along sand roads each month.

### 5. Tracker or observer certification

- Certified trackers. Trailing: two trackers certified as Trailing Professional and one as Trailing Level III (CyberTracker Conservation's Tracker Certification System). Track survey: three trackers from the Kaudwane community, certified with a mean Track & Sign score of 98.0 ± 1.75%, full-time trackers with several years of research experience.

- Track detection in the survey was assumed to be close to perfect, given clear sand roads, morning surveys and these trackers' skill.

- The acknowledgements thank the Kaudwane trackers Supula Monnaanoka, Sokwa Poditaroma, Mosepele Mamou, Mpho Tshiama, Meno Tshiama, Moisapodi Gana and Phalatsa Nkadima.

### 6. Other technologies, skills or innovations

- CyberTracker app: a custom app recorded sightings and logged GPS fixes every second along each trail.

- Data cleaning: trails were cleaned in ArcGIS, with observer-induced running removed.

- Combining approaches: trailing combined with body-mass scaling rules, and suggestions for camera traps or GPS collars where trailing is not practical.

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

- Scale of use: about 170,000 km² of Botswana's Kalahari conservation landscape has been surveyed with the FMP formula since 2008, a cost-effective alternative to aerial counts.

- Land-use planning: these data have supported analyses of habitat suitability and connectivity and guided land-use planning.

- Low-density areas: better day ranges strengthen track data for monitoring wildlife in low-density environments.

- Participation: participatory approaches can improve attitudes towards wildlife, promote coexistence and improve conservation outcomes, especially where people bear high costs of living alongside wildlife.

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

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

- The method depends on trailing skill: staying exactly on one animal's or group's trail for 24 hours and judging from the trail's freshness when a full day of travel has been covered.

- Trailing records not only how far animals move but also what they did along the way, information the authors note is hard to obtain by other means.

## Cite this summary

Lawrence, K. 2026. Summary of: Refining population density estimation from track counts: improving daily travel distance estimates through trailing of large herbivores in the Kalahari, Botswana (Gielen and Araldi et al., 2025). Original Wisdom, Tracking Literature. https://www.originalwisdom.com/tracking-literature/2025-gielen-araldi-refining-population-density-estimation/