# Selection of bed-sites by roe deer Capreolus capreolus fawns in a boreal forest landscape

> Where do roe deer fawns hide while their mothers feed? Following 19 radio-collared fawns in Norway’s logged boreal forest, researchers found the fawns chose beds with taller vegetation and more canopy cover than nearby spots, yet scattered them across forest of every age, fields and bogs. Spreading out like this may force foxes and lynx to search everywhere.

Summary by Kersey Lawrence, Original Wisdom. Updated 9 November 2026. Page: https://www.originalwisdom.com/tracking-literature/1999-linnell-nijhuis-selection-bed-sites-roe/

## Citation and access

- **Authors:** John D. C. Linnell, Patricia Nijhuis, Ivonne Teurlings and Reidar Andersen
- **Year published:** 1999
- **Journal or source:** Wildlife Biology
- **Full citation:** Linnell, J. D. C., Nijhuis, P., Teurlings, I. and Andersen, R. 1999. Selection of bed-sites by roe deer Capreolus capreolus fawns in a boreal forest landscape. Wildlife Biology 5(4): 225-231.
- **DOI:** 10.2981/wlb.1999.027
- **Access status:** Hosted (freely available elsewhere)
- **PDF:** https://www.originalwisdom.com/wp-content/uploads/2026/11/1999_Linnell_and_Nijhuis_etal_selection-of-bed-sites-by-roe-deer-fawns-in-a-boreal-forest-landscape.pdf

## Tags

- **Species:** Mammals > Roe deer
- **Region:** Europe > Norway
- **Behaviour and ecology:** Behaviour; Habitat selection; Home range; Predation risk; Predator avoidance
- **Methods and technology:** Chi-square test; Mann–Whitney U test; Radio telemetry; Wilcoxon matched pairs test
- **Data type:** New field data
- **Observer qualification:** Not needed
- **Discipline:** Behaviour and ecology
- **Year published:** 1999

## Detailed keywords

- **Species:** Studied: roe deer (Capreolus capreolus) fawns, 19 radio-collared fawns from seven family groups, and their radio-collared mothers. Predators present in the study area: red fox (Vulpes vulpes), Eurasian lynx (Lynx lynx), occasional transient wolves (Canis lupus) and free-running domestic dogs. Fawn deaths during the study: two killed by red foxes, one by a domestic dog. Plants at bed-sites: Scots pine (Pinus sylvestris), Norway spruce (Picea abies), heather (Calluna sp.) and bilberry (Vaccinium sp.). Other hider species compared in the discussion: white-tailed deer (Odocoileus virginianus), mule deer (O. hemionus) and pronghorn (Antilocapra americana) in North America.
- **Location and habitat:** Åmot and Trysil municipalities, Hedmark county, southeastern Norway (61°10’N, 11°40’E). Three sites: Stener, Nordre Odden and Åsbygda, each a small enclave of settlement and small hay and silage fields within intensively logged boreal forest of Scots pine and Norway spruce below 900 m, a fine-grained mosaic of stands of different ages. Habitat types: forest (age classes I clearcuts to V mature stands), fields, bogs and gardens. Continental climate: January average -11 °C, July average 19 °C. Roe deer density below 1 per km².
- **Topics:** bed-site selection, hiding behaviour of a “hider” species, anti-predator behaviour, neonatal fawn mortality, concealment and visibility at fox height (30 cm), canopy cover, ground vegetation, habitat selection at two scales (micro-site and main habitat type), forest age classes, forestry and grazing management, mother-fawn behaviour, spreading bed-sites across habitats to increase a predator’s search area.
- **Statistics or models:** Wilcoxon paired test (bed-site versus random site 50 m north); Mann-Whitney U test; Kruskal-Wallis analysis of variance (ANOVA); compositional analysis of habitat use (Aebischer, Robertson and Kenward 1993), with missing values replaced as they recommend; minimum convex polygon home ranges (RANGES V program), calculated per family group because siblings moved together. Inconsistency: in the text, field versus forest bed-sites are given as vegetation height 76 ± 25 vs 57 ± 32 cm and visibility 35 ± 79 vs 164 ± 98 cm, while Table 1 gives forest vegetation height mean 58 (SD 32), field visibility mean 28 (SD 79) and forest visibility mean 156 (SD 87). Sample sizes also differ between measures (for example 133 forest bed-sites for vegetation height, 103 for visibility).
- **Data type:** New field data: fawns born June 1998, bed-site and habitat data collected 3 June to 10 July 1998. Adult females captured from 1995 to 1998.

## Synopsis

### 1. What they did

Newborn hoofed animals often die from predation, starvation or cold in their first weeks. Roe deer fawns are “hiders”: for their first four weeks they spend up to 80% of their time lying still in a bed-site while their mother feeds 50 to 1,000 m away. A well-hidden bed is their main protection against searching predators. Earlier studies of roe deer fawn bed-sites were done where predators of fawns were almost absent. This study was in southeastern Norway, where red foxes and Eurasian lynx were present.

- Finding the fawns. Adult females were caught in box traps at winter feeding sites and fitted with radio-collars. In June, the researchers checked them for signs of having given birth, then searched for the fawns and fitted each with a small radio-collar. They followed 19 fawns from seven mothers.

- Recording bed-sites. Each fawn was located about once a day and approached until it was seen, until it was 2 to 3 weeks old and began to run off. The bed-site was marked and examined the next day, once the fawn had moved on. After fawns began to run, the researchers only came close enough to see which habitat they were in.

- Measuring cover. At each bed-site and at a random site 50 m to the north, they recorded habitat type and forest age class, the height of the tallest vegetation, the distance to the nearest tree, its trunk diameter and the height of its lowest branch, and canopy cover. They placed a model fawn in the bed and measured how far away it disappeared from view at fox height (30 cm), from four directions. They recorded the main ground cover in a 50 x 50 cm square.

- Habitat use and availability. For each family group’s home range they compared the habitats used for beds with the habitats available, read from forest maps. They estimated the availability of forest age classes by walking transects at two sites.

### 2. What they found

Hiding at the bed-site

- In the forest, fawns chose bed-sites with taller vegetation (median 60 cm versus 38 cm at random sites), lower visibility (median 145 cm versus 250 cm) and more canopy cover (median 78% versus 58%) than nearby random sites.

- They did not choose sites closer to trees, or under trees with lower branches or thicker trunks.

- In the fields, where cover is even, bed-sites did not differ from random sites.

- Bed-sites in fields gave better cover than bed-sites in the forest: taller vegetation and lower visibility.

- Canopy cover increased with forest age, but vegetation height and visibility at bed-sites did not differ between forest age classes.

Ground cover

- Fallen branches and tree trunks were chosen early in the season, when other plants had not grown much.

- Later, as herbs covered more of the ground (from 11% to 37% at random sites), the preference for wood disappeared, grasses and low shrubs were avoided, and bare low ground cover (lichens, moss, leaf litter, soil) was chosen.

- No selection was found for bushes or herbs.

Main habitat types

- Fawns bedded in almost every habitat in their home ranges: forest, fields (pasture and grain) and bogs. Only gardens and roads were not used.

- Forest was preferred over fields and other habitats, which were used about equally.

- Even so, fields and bogs were used often. One family group (#58) had most of its bed-sites (78%) in a field.

- All forest age classes were used, from recent clearcuts to over-mature stands, in roughly the same proportions as they were available (31%, 24%, 25% and 20% use for classes I, II, III and IV-V, against 27%, 19%, 29% and 26% availability).

- Three fawns died during the study: two were killed by red foxes and one by a domestic dog.

### 3. What they concluded

- Where predators are present, roe deer fawns pick bed-sites that hide them better than expected, mainly through dense ground vegetation rather than trees and bushes.

- The soft low ground cover at the bed itself probably gives comfort and lets fawns lie low, getting the most from the surrounding plants.

- Fields hide a lying fawn well from fox height but give no cover to the mother. Predators may watch the mother approach and use her to find the fawn, which may be why fields were not favoured.

- Unlike a study in France, where fawns strongly chose forest openings, fawns here used all forest ages. The authors link this to the richer understory of Scandinavian conifer forests compared with European deciduous forests.

- Because fawns could hide well almost everywhere, all habitats may simply have been equally good. Another possibility, not exclusive of the first, is that spreading beds across all habitats forces a predator to search everywhere, raising the cost of hunting fawns.

### 6. Other technologies, skills or innovations

- Radio-collars from Televilt International: 30 g transmitters on elastic collars for newborn fawns, swapped for 100 g transmitters on expanding collars at 10 days old.

- A “model” fawn placed in each bed to measure from what distance it was hidden at fox height.

- A flat mirror with a grid to measure canopy cover from directly above the bed-site.

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

- If fawns depended on very specific habitat for bed-sites, keeping that habitat available would matter when planning forestry or grazing. In this study fawns used all forest ages and hid well in each, so the authors judge the effect of forest age on hiding to be small.

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

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

- Fawn beds in this landscape were found in every habitat, from clearcuts to mature forest, fields and bogs, mostly in spots with tall ground vegetation over moss, lichen or litter. Beds are a sign trackers look for, and this study suggests they should not be expected in one habitat type only.

## Cite this summary

Lawrence, K. 2026. Summary of: Selection of bed-sites by roe deer Capreolus capreolus fawns in a boreal forest landscape (Linnell and Nijhuis et al., 1999). Original Wisdom, Tracking Literature. https://www.originalwisdom.com/tracking-literature/1999-linnell-nijhuis-selection-bed-sites-roe/