# Bed site selection of red muntjac (Muntiacus muntjak) and sambar (Rusa unicolor) in a tropical seasonal forest

> How do you find where a deer slept in a tropical forest? In Thailand’s Khao Yai National Park, researchers looked for little piles of shiny fruit seeds that muntjac and sambar bring up while chewing the cud, then cleared the leaf litter to read the hoof prints underneath. Both deer bedded under thick canopy, sambar preferred flatter ground, and the two species differed in which way their bed sites faced.

Summary by Kersey Lawrence, Original Wisdom. Updated 9 November 2026. Page: https://www.originalwisdom.com/tracking-literature/2009-brodie-brockelman-bed-site-selection-red/

## Citation and access

- **Authors:** Jedediah F. Brodie and Warren Y. Brockelman
- **Year published:** 2009
- **Journal or source:** Ecological Research
- **Full citation:** Brodie, J. F. and Brockelman, W. Y. 2009. Bed site selection of red muntjac (Muntiacus muntjak) and sambar (Rusa unicolor) in a tropical seasonal forest. Ecological Research 24(6): 1251–1256.
- **DOI:** 10.1007/s11284-009-0610-9
- **Access status:** Publisher link only
- **Where to find it:** https://doi.org/10.1007/s11284-009-0610-9

## Tags

- **Species:** Mammals > Red muntjac; Mammals > Sambar
- **Region:** Asia > Thailand
- **Tracking:** Beds; Feeding signs; Scat; Sign surveys; Track measurements; Tracking; Tracks; Tracks and signs identification
- **Behaviour and ecology:** Behaviour; Habitat selection; Interspecific competition; Predation risk
- **Methods and technology:** Analysis of variance (ANOVA); Bootstrapping; Chi-square test; Pearson correlation
- **Data type:** New field data
- **Observer qualification:** Not reported
- **Discipline:** Behaviour and ecology; Wildlife tracking
- **Year published:** 2009

## Detailed keywords

- **Species:** Studied: red muntjac (Muntiacus muntjak), also called common muntjac or barking deer, and sambar (Rusa unicolor; synonym Cervus unicolor). Food sign at bed sites: regurgitated seeds (pyrenes) of the fruit Choerospondias axillaris (Anacardiaceae), which the deer eat in large quantities in its fruiting season (June to October). In the area but not studied: Asian elephant (Elephas maximus), lesser mouse-deer (Tragulus kanchil), gaur (Bos gaurus, in the park but not seen within 1 km of the plot), gibbons (Hylobates spp.) and rodents (the only other known eaters of C. axillaris fruit), dhole (Cuon alpinus, relatively abundant), tiger (Panthera tigris, close to extirpation in Khao Yai). Leopards (Panthera pardus) do not naturally occur there. Comparison species from other studies include red deer (Cervus elaphus) and roe deer (Capreolus capreolus).
- **Location and habitat:** A 30 ha forest dynamics plot in the Mo Singto area, central western Khao Yai National Park (2,168 km²), northeastern Thailand. Seasonal moist evergreen forest, with a small area of secondary forest in the north of the plot from clearing before the park was created in 1962. Elevation about 700 to 900 m; rainfall about 2,200 mm a year, mostly May to October, with a pronounced dry season from November to April.
- **Topics:** bed site selection; within-habitat selection versus between-habitat selection; canopy cover; slope; aspect; concealment from predators; thermoregulation; habitat use versus availability; sympatric deer and possible avoidance or niche partitioning; ruminating and regurgitated seed piles as bed site sign; pseudoreplication (bed site clusters); hunting and predator hunting mode (ambush versus coursing) as possible influences. Inconsistency: the abstract and discussion say muntjac “disproportionately” chose west-facing sites and sambar east-facing sites, but the results report that the aspect of bed sites did not differ significantly from available aspects for either species (the significant difference was between the two species).
- **Statistics or models:** Two-tailed Student’s t tests with separate variance (canopy cover and slope, between years and between species); chi-squared goodness-of-fit tests (aspect); Neu-Byers method of use versus availability, with chi-squared tests and Bonferroni confidence intervals; Pearson correlation (canopy cover and slope); analysis of variance (ANOVA) (aspect against slope and canopy cover); bootstrap error bars (10,000 iterations) in the figures. Significance at 0.05; marginal significance between 0.05 and 0.10.
- **Data type:** New field data: 15 belt transects surveyed weekly for 10 weeks (mid-July through September) in 2003 and 2004. Available habitat measured at 375 canopy photo points and 762 slope and aspect grid points on the plot.

## Synopsis

### 1. What they did

Deer and other ruminants spend much of their time lying down to chew the cud, so where they choose to bed is an important part of how they use habitat. Most of what is known comes from temperate deer. Very little is known for tropical deer, and the authors knew of no study of sambar bed sites at all.

The researchers looked at bed sites of two deer that live side by side in Khao Yai National Park, Thailand: the small, usually solitary red muntjac (20 to 28 kg) and the larger sambar (about 109 to 260 kg), which live in small groups.

- Finding bed sites. They walked 15 belt transects (each 500 m by 4 m, covering 10% of a 30 ha forest plot) once a week for 10 weeks, from mid-July through September, in 2003 and 2004. Both deer regurgitate piles of clean, shiny seeds of Choerospondias axillaris fruit after ruminating on their beds, so the team looked for these seed piles together with the body imprints of the deer.

- Identifying the species. They measured the tracks and, where present, the scat at each site. They carefully removed the leaf litter around each bed and looked for hoof prints that had registered through the litter. Because the two deer differ so much in size, the species was easy to tell from track or scat size. They did not use the size of the body imprint, because it varies so much with how the animal lay.

- Measuring each site. At each bed site they measured canopy cover (from hemispherical “fish-eye” photographs), slope and aspect (the direction the slope faces). Groups of 2 to 7 beds within 10 m of each other were treated as one site.

- Comparing with what was available. They compared bed sites between years, between the two species, and with the habitat available across the plot.

### 2. What they found

Bed sites found

- 2003: 50 muntjac bed sites in 35 clusters and 21 sambar bed sites in 16 clusters.

- 2004: 46 muntjac bed sites in 41 clusters and 20 sambar bed sites in 18 clusters.

Canopy cover

- Both species bedded under very closed canopy (average cover about 0.88 to 0.92, where 1 is completely closed).

- Muntjac beds were in slightly more open places in 2004 than in 2003. Sambar beds did not differ between years.

- The two species did not differ from each other in canopy cover.

- Both avoided the more open parts of the forest. Muntjac used fully closed canopy more than expected from its availability. For sambar the overall difference from available habitat was only marginally significant.

Slope

- Slope did not differ between years for either species, and the difference between the two species was only marginal.

- Muntjac used slopes in proportion to what was available.

- Sambar favoured flatter ground (0 to 10% slope): 41% of their bed sites, compared with 17% of the plot.

Aspect

- Aspect did not differ between years for either species.

- The two species differed: muntjac used a higher share of west-facing sites, and sambar of east-facing sites.

- Neither species’ choice of aspect differed significantly from the aspects available on the plot.

### 3. What they concluded

- Like most deer, both muntjac and sambar choose bed sites with high cover, which may be for concealment or for keeping cool.

- Sambar’s choice of flatter ground may save energy getting to and from beds, or give better long-range visibility.

- The difference in aspect could mean the smaller, solitary muntjac avoids the larger, social sambar, or that the two species divide some unknown resource between them.

- The study looked only within forest, so it cannot say whether muntjac here prefer grassland for bedding, as they do on Hainan Island, China. The authors note that Khao Yai’s grasslands are kept open by burning or mowing and generally lack shrubs for cover.

- Bed site choice is likely to differ where hunting is heavier (near the edges of the park and in other parks), and where the mix of predators differs. Ambush hunters such as large cats and coursing hunters such as dhole may push deer to choose differently.

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

Tracks and signs identification was central to the study. Bed sites were found by their sign: piles of regurgitated seeds and the body imprints of resting deer. The species at each bed was identified from measured hoof prints, including prints found by clearing the leaf litter to reveal tracks that had registered through it, and from scat size. The authors cite a field guide to the mammal tracks of Thailand (Kanjanavanit 2004) and a guide to the mammals of Khao Yai for telling the two deer apart. They explain why they did not use the size of the body imprint. No trailing was reported.

### 5. Tracker or observer certification

Not reported. The paper does not say who searched the transects or identified the tracks, or how their identifications were checked. The authors state that species could “easily be ascertained” by track or scat size because of the large size difference.

### 6. Other technologies, skills or innovations

- Hemispherical canopy photographs taken with a Nikon D70 digital camera and fish-eye lens, analysed with HemiView 2.1 software.

- Using a food sign (regurgitated fruit seeds left after ruminating) as a reliable marker of bed sites.

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

- Hunting is illegal in Khao Yai and low in the central area where the study took place, but much higher near the park edges. In other parks hunting has wiped out many large animals. Deer there are likely to choose bed sites differently in response to the greater risk.

- The predators present matter: tigers are close to extirpation in Khao Yai, while dhole remain relatively abundant. In parks with more large cats, deer bed site choice may differ.

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

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

- A neat example of combining signs: the seed piles tell you an animal rested and ruminated there, and the tracks, uncovered beneath the leaf litter, tell you who it was.

- The authors chose not to use the size of the bed imprint to identify species, because it changes so much with how the animal lay. Trackers reading beds may find the same caution useful.

## Cite this summary

Lawrence, K. 2026. Summary of: Bed site selection of red muntjac (Muntiacus muntjak) and sambar (Rusa unicolor) in a tropical seasonal forest (Brodie and Brockelman, 2009). Original Wisdom, Tracking Literature. https://www.originalwisdom.com/tracking-literature/2009-brodie-brockelman-bed-site-selection-red/