# Estimating Ungulate Numbers in a Forest by Track Counts

> Can you count deer and wild boar by their tracks instead of sending a line of people through the forest? Comparing drive counts with track counts across five forest districts in the early 1970s, this study found that the answer depended on the species: tracking around blocks worked best for red deer, and counting tracks along transects showed promise for roe deer and wild boar. The catch, then as now, is knowing how far animals travel each day.

Summary by Kersey Lawrence, Original Wisdom. Updated 9 November 2026. Page: https://www.originalwisdom.com/tracking-literature/1976-dzieciolowski-estimating-ungulate-numbers-forest/

## Citation and access

- **Authors:** Ryszard Dzięciołowski
- **Year published:** 1976
- **Journal or source:** Acta Theriologica
- **Full citation:** Dzięciołowski, R. 1976. Estimating ungulate numbers in a forest by track counts. Acta Theriologica 21(15): 217-222.
- **DOI:** 10.4098/at.arch.76-20
- **Access status:** Hosted (freely available elsewhere)
- **PDF:** https://www.originalwisdom.com/wp-content/uploads/2026/11/1976_Dzieciolowski_estimating-ungulate-numbers-in-a-forest-by-track-counts.pdf

## Tags

- **Species:** Mammals > Red deer; Mammals > Roe deer; Mammals > Wild boar
- **Region:** Europe
- **Tracking:** Track counts; Tracking; Tracks
- **Methods and technology:** Formozov–Malyshev–Pereleshin (FMP) formula; Linear regression; Population estimation
- **Data type:** New field data
- **Observer qualification:** Not reported
- **Discipline:** Wildlife tracking
- **Year published:** 1976

## Detailed keywords

- **Species:** Studied: red deer (Cervus elaphus), roe deer (Capreolus capreolus) and wild boar (Sus scrofa). White-tailed deer and other deer are discussed from earlier North American studies.
- **Location and habitat:** Forest habitat in five state forest districts: Józefów, Kobiór, Iława, Augustów and Płaska. The paper does not name the country, but the author’s institute is in Warsaw and the analysis was run at the Polish Academy of Sciences, so the districts are in Poland.
- **Topics:** censusing wild ungulates; drive census (people drive the animals out of a forest block and count them) as the reference count; counting tracks around the areas that were then driven; counting track crossings along transect lines; conversion factors between track counts and animal numbers; daily range or daily track length of animals; spatial distribution of animal populations; game management. Inconsistency in the paper: the abstract says that for red deer “the regression of drive census upon track counts was statistically significant”, and the results say the number of tracks per length of transect “may be used as a conversion factor for estimating the number of red deer”. Tables 2 and 3 show instead that for red deer only tracking around the driven areas (x1) was significant; track counts on the transect (x3) and tracks per length of transect (x4) were not significant for red deer.
- **Statistics or models:** Multiple regression of drive results on tracking around driven areas (x1), length of transect (x2), number of track crossings (x3) and tracks per length of transect (x4), plus each variable alone (simple linear regression), with correlation coefficients, standard errors and significance. Run on an ODRA 1204 computer at the Polish Academy of Sciences. The discussion describes Formozov’s (1932) formula, later supplemented by Pereleshin (1950): Z = 1.57 s / (d m), linking animals per area unit (Z) to tracks recorded (s), route length (m) and mean length of the daily track (d).
- **Data type:** New field data, collected 1971 to 1973: 35 samples for red deer, 39 for roe deer and 26 for wild boar.

## Synopsis

### 1. What they did

Game managers need to know how many animals live in an area. Driving animals out of a forest block and counting them (a drive census) was seen as the best way to estimate numbers, but it takes a great many people. Counting tracks is far cheaper, so the author asked whether track counts could stand in for drives.

From 1971 to 1973, in five state forest districts, the author compared three ways of counting red deer, roe deer and wild boar:

- Drives of forest areas, used as the reference count.

- Tracking around the areas before they were driven. The methods for this follow an earlier study (Pucek et al. 1975).

- Track counts along transects. Tracks were plotted on sketches to count how many crossed the transect lines.

The results were compared with multiple regression: 35 samples for red deer, 39 for roe deer and 26 for wild boar.

### 2. What they found

Totals (Table 1)

- Red deer: 70 by drives, 116 by tracking around the driven areas, and 584 track crossings on 128,270 m of transect.

- Roe deer: 139 by drives, 75 by tracking, and 687 crossings on 146,790 m of transect.

- Wild boar: 35 by drives, 12 by tracking, and 214 crossings on 103,900 m of transect.

How well the methods matched

- The best method differed between species.

- Red deer: the count from tracking around the driven areas was the only significant predictor of the drive count (correlation 0.522). The number of tracks on the transect was not significant on its own.

- Roe deer and wild boar: the number of tracks crossing the transect was the significant predictor of the drive count (correlation 0.350 for roe deer and 0.389 for wild boar). Tracking around the driven areas was not.

- All the correlations were modest. The multiple correlation was 0.529 for red deer, 0.478 for wild boar and 0.356 for roe deer.

- Tracks per length of transect, which the author expected to matter, was not significantly related to drive results for any species.

Earlier studies

The author reviews earlier attempts to turn track counts into animal numbers. In North American studies, counts of deer track crossings matched a known population poorly in one enclosure study (correlation 0.20), and in another could show that populations differed in size but not by how much. Formozov’s formula, used in the Union of Soviet Socialist Republics (USSR), links animal numbers to the number of tracks, the route length and the average length of an animal’s daily track. Such methods have stalled on the difficulty of knowing how far animals travel each day.

### 3. What they concluded

- Basic research on how each animal population is spread across the land is needed before track counts can be reliably used to census wild ungulates.

- Track counts along transects offer promise as a census method for roe deer and wild boar.

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

Tracking was the subject of the study. Tracks were counted around forest areas before they were driven, and track crossings along transect lines were plotted on sketches and counted. Both counts were then compared with drive counts of the same species. The paper gives no details of how the tracks were found or identified, and refers to Pucek et al. (1975) for the tracking procedures. It does not say whether the tracks were in snow or other ground.

### 5. Tracker or observer certification

Not reported. The paper does not say who counted the tracks or how skilled they were.

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

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

- The study shows that a count of tracks is not yet a count of animals. Turning one into the other depends on knowing how far each species moves in a day and how the animals are spread through the forest, which is the knowledge of animal behaviour that trackers build in the field.

## Cite this summary

Lawrence, K. 2026. Summary of: Estimating Ungulate Numbers in a Forest by Track Counts (Dzięciołowski, 1976). Original Wisdom, Tracking Literature. https://www.originalwisdom.com/tracking-literature/1976-dzieciolowski-estimating-ungulate-numbers-forest/