# Habitat selection by the red fox Vulpes vulpes (L. 1758) in an Alpine area

> Where do red foxes spend the winter in the Alps? By mapping 922 fox scats along mountain footpaths every month for a year in Italy’s Gran Paradiso National Park, researchers found that foxes stick to forest and lower slopes in the cold months, likely drawn by the carcasses of chamois and ibex. The results suggest foxes may hold “vertical” home ranges that stretch up and down the mountain to cover their needs through the seasons.

Summary by Kersey Lawrence, Original Wisdom. Updated 9 November 2026. Page: https://www.originalwisdom.com/tracking-literature/2004-cagnacci-meriggi-habitat-selection-red-fox/

## Citation and access

- **Authors:** F. Cagnacci, A. Meriggi and S. Lovari
- **Year published:** 2004
- **Journal or source:** Ethology Ecology & Evolution
- **Full citation:** Cagnacci, F., Meriggi, A. and Lovari, S. 2004. Habitat selection by the red fox Vulpes vulpes (L. 1758) in an Alpine area. Ethology Ecology & Evolution 16(2): 103-116.
- **DOI:** 10.1080/08927014.2004.9522640
- **Access status:** Publisher link only
- **Where to find it:** https://doi.org/10.1080/08927014.2004.9522640

## Tags

- **Species:** Mammals > Red fox
- **Region:** Europe > Italy
- **Tracking:** Scat; Sign surveys; Tracking; Tracks and signs identification
- **Behaviour and ecology:** Behaviour; Habitat selection; Home range
- **Methods and technology:** Analysis of variance (ANOVA); Chi-square test; Geographic Information System (GIS); Kolmogorov–Smirnov test; Pearson correlation
- **Data type:** New field data
- **Observer qualification:** Not reported
- **Discipline:** Behaviour and ecology; Wildlife tracking
- **Year published:** 2004

## Detailed keywords

- **Species:** Studied: red fox (Vulpes vulpes). In the study area but not studied: Alpine ibex (Capra ibex, 3000 to 4000 animals) and chamois (Rupicapra rupicapra, 7000 to 8000 animals), whose carcasses are an important winter food for foxes; small mammals; marmots. Plants named in the habitat description: beech (Fagus sylvatica), cherry (Prunus avium), plum (Prunus domestica), larch (Larix decidua), Norway spruce (Picea abies), raspberry and blackberry (Rubus idaeus, Rubus fruticosus), bilberry (Vaccinium myrtillus), barberry (Berberis vulgaris) and wild roses (Rosa sp.). No large carnivores were present in the park during the study. Food items discussed from the authors’ earlier diet studies: ungulate carrion (mainly chamois), ungulate kids, marmots, small mammals, grasshoppers, ground beetles and fruit.
- **Location and habitat:** Gran Paradiso National Park (720 km², protected since 1921), Western Alps, Italy. All valleys of the park from 1000 m above sea level to the upper limit of vegetation (2500 to 2800 m). Alpine continental climate, with snow cover for 5 to 6 months a year. Forest (broad-leaved woods, conifer woods of larch and Norway spruce) at lower altitudes; shrubland, meadows, Alpine prairies, bushy and rocky prairies, debris and rocky areas higher up, with open habitats dominating above 1800 to 2000 m.
- **Topics:** habitat selection; habitat use versus availability; home range (range attendance); altitude belts; seasonal change (cold and warm seasons); snow cover; ungulate carrion; resting and denning sites; Resource Dispersion Hypothesis; “vertical” home ranges along the altitude gradient; scat transects on footpaths; indirect signs of presence versus radio tracking.
- **Statistics or models:** chi-square goodness-of-fit test; Bonferroni simultaneous confidence intervals (use versus availability); Attendance Index (scats per km of transect in each 25 ha sample square); Shannon Index of Diversity for habitat; Pearson product moment correlation; one-way analysis of variance (ANOVA); Discriminant Function Analysis (Wilks’ Lambda minimisation); Kolmogorov–Smirnov test for normality; habitat mapped with a Geographic Information System (GIS, IDRISI). Inconsistencies in the paper: (1) the Methods describe three altitude belts (1000-1500, 1500-2000 and over 2000 m), but Table 4 has columns for 1500-2000, 2000-2500 and over 2500 m and a “Pastures” column, while the text describing Table 4 refers to a lowest belt that has no column there; (2) the text says broad-leaved woods (whole year) and scrublands (whole year) differed between low and high attendance squares at the 0.1 level, but Table 5 gives P = 0.104 and P = 0.109; (3) in Table 5, several warm-season values are identical to the whole-year values; (4) the Methods mention eleven habitat types, while the habitat selection tests use eight; (5) the fox home range size used to choose the 25 ha squares (195.4 ± 152.8) is given without a unit.
- **Data type:** New field data: 13 fixed transects (30 km in total) walked monthly from August 1996 to August 1997; 922 fox scats mapped and removed. Cold season September 1996 to February 1997; warm season March to August 1997.

## Synopsis

### 1. What they did

Very few studies had looked at how red foxes use their ranges in mountains, where habitats change quickly with altitude and the winters are harsh. The authors expected that in the Alps a fox’s range would include a mix of habitats that provide food and shelter all year, that foxes would use these habitats differently through the year, and that the link between resources and habitat would be strongest in the severe cold season.

To test this across the whole of Gran Paradiso National Park in the Italian Alps, they used fox scats as signs of presence:

- Transects. They chose 13 transects, 30 km in total, all on footpaths, crossing the park’s main habitat types and three altitude belts (1000-1500 m, 1500-2000 m and over 2000 m). Footpaths were chosen so that scats would be equally easy to find in every habitat.

- Scat surveys. Every month from August 1996 to August 1997, they walked the transects, mapped every fox scat and then removed it. They collected 922 scats.

- Use versus availability. They compared how many scats were found in each habitat type and altitude belt with what would be expected if foxes used them in proportion to their availability, for the whole year, the cold season (September to February), the warm season (March to August) and each month.

- Attendance in squares. They divided the area into 25 ha squares, close to the average fox home range reported for the park. For each square crossed by a transect they calculated an Attendance Index (scats per km of transect) and measured the share of each habitat and altitude belt, then tested which habitat features set squares with high fox attendance apart from those with low attendance.

### 2. What they found

Habitats and altitudes used

- Over the whole year, foxes selected dense conifer woods and dense broad-leaved woods, and avoided almost all sparse woods and open habitats such as Alpine prairies, rocky prairies and shrubland.

- The same pattern held in both seasons, except that broad-leaved woods were selected only in the cold season.

- The lowest altitude belt (1000-1500 m) was selected and the highest (over 2000 m) avoided in both seasons. The middle belt was used in proportion to its availability, except in the warm season, when it was avoided.

- Month by month, scats were spread among habitats differently from what availability would predict in every month except November, when few scats were found.

- From September to April the highest belt was almost always avoided; from May to August it was used in proportion to its availability.

Fox attendance in the squares

- Over the year, only dense conifer woods were linked to higher fox attendance. In the cold season, attendance rose with broad-leaved woods and the lowest altitude belt, and fell with the highest belt. In the warm season no habitat feature was linked to attendance.

- Squares with high attendance had more conifer woods and more of the lowest altitude belt; squares with low attendance had more sparse conifer woods and rocky areas.

- The habitat features told high and low attendance squares apart correctly in 75.8% of cases for the whole year, 72.6% in the cold season and 80.6% in the warm season. In the cold season the lowest altitude belt, shrubland and broad-leaved woods mattered most; in the warm season it was rocky areas, sparse conifer woods and conifer woods, and altitude did not matter.

### 3. What they concluded

- In the cold season, foxes favour low altitudes and forest. The authors link this to ungulate carrion, mainly chamois, which their earlier diet studies showed to be the foxes’ main winter food. Many ibex and chamois die in winter and spring and are found at the bottom of the valleys, while the upper altitudes offer little food then.

- In summer, food is more plentiful and varied and spread across many habitats, so food probably shapes habitat use less, and no clear altitude pattern appeared.

- The steady preference for forest all year cannot be fully explained by food. Forest may also provide resting and denning sites, or better temperatures in both winter and summer.

- Because resources change over short distances along the altitude gradient, foxes may hold “vertical” home ranges that include several habitat types, some of which become key under harsh climate. Earlier radio tracking in one valley of the park recorded such up-and-down movements.

- The authors note limits of the method: scats cannot show activity patterns or individual ranging as radio tracking can, and foxes may use footpaths to travel and mark, which could bias results. They argue that scats suited this study because they allowed a large area and many individuals to be sampled, are not affected by a few individuals’ habits, and can be mapped precisely, whereas radio locations can carry large errors in mountains.

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

The study rests on signs of presence: fox scats were found, identified and mapped along fixed footpath transects every month for a year, then removed so that each survey counted only fresh scats. The paper does not report following trails or using footprints, and does not describe how scats were told apart from those of other species.

### 5. Tracker or observer certification

Not reported. The paper does not say who walked the transects or identified the scats, or what experience they had. Park wardens are thanked for their help.

### 6. Other technologies, skills or innovations

- An Attendance Index (scats per km of transect) for each 25 ha square, used to compare places that foxes visited often with places they visited rarely.

- Habitat mapping from a 1:25000 vegetation map and direct surveys, measured in a Geographic Information System (GIS).

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

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

- The study shows how a simple, repeated scat survey on footpaths can reveal seasonal habitat use across a whole national park, without capturing or collaring a single animal.

- The authors’ own caution is worth noting for anyone counting scats on trails: foxes use paths to travel and to mark, so scats on a path may say as much about the path as about the habitat around it.

## Cite this summary

Lawrence, K. 2026. Summary of: Habitat selection by the red fox Vulpes vulpes (L. 1758) in an Alpine area (Cagnacci and Meriggi et al., 2004). Original Wisdom, Tracking Literature. https://www.originalwisdom.com/tracking-literature/2004-cagnacci-meriggi-habitat-selection-red-fox/