# The Effect of Snowmobile Trails on Coyote Movements Within Lynx Home Ranges

> Do snowmobile trails open up deep-snow lynx country to coyotes? By following coyote trails for 322 km through the Montana snow, researchers found that coyotes stayed in lynx habitat all winter, but spent less than 8% of their travel on packed trails. They found firmer snow on their own, and mostly scavenged deer carcasses rather than hunting the hares lynx depend on.

Summary by Kersey Lawrence, Original Wisdom. Updated 9 November 2026. Page: https://www.originalwisdom.com/tracking-literature/2007-kolbe-squires-effect-snowmobile-trails-coyote/

## Citation and access

- **Authors:** Jay A. Kolbe, John R. Squires, Daniel H. Pletscher and Leonard F. Ruggiero
- **Year published:** 2007
- **Journal or source:** Journal of Wildlife Management
- **Full citation:** Kolbe, J. A., Squires, J. R., Pletscher, D. H. and Ruggiero, L. F. 2007. The effect of snowmobile trails on coyote movements within lynx home ranges. Journal of Wildlife Management 71(5): 1409–1418.
- **DOI:** 10.2193/2005-682
- **Access status:** Hosted (freely available elsewhere)
- **PDF:** https://www.originalwisdom.com/wp-content/uploads/2026/11/2007_Kolbe_and_Squires_etal_the-effect-of-snowmobile-trails-on-coyote-movements-within-lynx-home-ranges.pdf

## Tags

- **Species:** Mammals > American marten; Mammals > American red squirrel; Mammals > Bobcat; Mammals > Canada lynx; Mammals > Coyote; Mammals > Grey wolf; Mammals > Mountain lion; Mammals > Snowshoe hare
- **Region:** North America > United States
- **Tracking:** Feeding signs; Kill sites; Scat; Snow tracks; Track counts; Tracking; Tracks; Tracks and signs identification; Trailing
- **Behaviour and ecology:** Behaviour; Feeding; Habitat selection; Interspecific competition; Movement; Predation; Scavenging
- **Methods and technology:** Analysis of variance (ANOVA); Chi-square test; Geographic Information System (GIS); Global Positioning System (GPS); Linear regression; Mann–Whitney U test; Radio telemetry
- **Data type:** New field data
- **Observer qualification:** Not reported
- **Discipline:** Behaviour and ecology; Wildlife tracking
- **Year published:** 2007

## Detailed keywords

- **Species:** Studied: coyote (Canis latrans), and Canada lynx (Lynx canadensis) through their tracks on survey routes. Prey and food found at coyote feeding sites and in scats: snowshoe hare (Lepus americanus), deer and other cervids (a mule deer, Odocoileus hemionus, was the one cervid kill), grouse, red squirrel (Tamiasciurus hudsonicus), microtine rodents (voles), birds, and scavenged coyote, pine marten and skunk (Mephitis mephitis); bovids (family Bovidae) in scats only. Other tracks recorded on survey routes: mountain lion (Felis concolor), bobcat (Felis rufus), American marten (Martes americana) and wolf (Canis lupus). In the study area but not studied: white-tailed deer (Odocoileus virginianus), mule deer, moose (Alces alces), elk (Cervus elaphus), black bear (Ursus americanus), grizzly bear (Ursus arctos), blue grouse (Dendragapus obscurus) and ruffed grouse (Bonasa umbellus).
- **Location and habitat:** Clearwater River drainage near Seeley Lake, western Montana, United States (about 1,800 km²), between the Bob Marshall and Mission Mountain Wilderness areas. State, federal and private land with intensive commercial forestry, an extensive network of logging roads and heavy snowmobile use. Elevations 1,200 to 2,100 m: Douglas-fir, western larch, lodgepole pine and ponderosa pine at lower elevations; mixed conifer at mid-elevations; subalpine fir, whitebark pine and Engelmann spruce higher up. Winter snow depth averaged 69 cm.
- **Topics:** snowmobile trails and compacted snow; snow depth and snow supportiveness (how far a dropped weight sinks in); foot-load (body mass to foot area); coyote and lynx sympatry in winter; exploitation and interference competition; competition for snowshoe hares; forest roads and linear travel corridors; coyote winter diet; scavenging versus killing; paired used and random (nonuse) tracks; recreation management in lynx habitat.
- **Statistics or models:** Multi-response randomized block permutation procedures (MRBP) on paired coyote and random tracks; Mann–Whitney U tests; independent samples t-tests; chi-square goodness-of-fit tests; one-way analysis of variance (ANOVA); runs tests for serial correlation within animals; linear regressions (r²); Bezier smoothing of GPS tracks (ET GeoWizards in ArcGIS) and distances from segment centroids to the nearest compacted trail (Nearest Features in ArcView). Inconsistencies in the paper: the text gives 2,220 km of survey effort, but Table 1 and the survey-route snow tests use 2,200 km; lynx tracks are given as 32% of carnivore tracks, but 760 of 2,291 is about 33%; Table 4 gives cervids as 68% of 88 backtrack food items, but Table 3 lists 58 cervid sites (about 66%); and the test comparing feeding sites with coyote travel (t = 0.96, P = 0.92) and the snow supportiveness test among years (F = 0.79, P = 0.69) give P values that do not match their test values.
- **Data type:** New field data. Coyotes captured September 2001 to October 2003; track surveys and backtracking in the winters of 2002 to 2004 (backtracking January 2002 to March 2004); 30 years of snow-depth records from two Natural Resources Conservation Service snow survey stations used for comparison.

## Synopsis

### 1. What they did

Coyotes have a high foot-load: they carry more body weight on each square centimetre of foot than lynx do, so deep, soft snow is harder work for them. Researchers and managers had suggested that snowmobile trails, by packing the snow, might let coyotes into deep-snow lynx habitat in winter, where they could compete with lynx for snowshoe hares, the lynx’s main winter prey. The authors studied this near Seeley Lake, western Montana, an area with an established lynx population and heavy recreational snowmobile use.

- Capture. They trapped and radio-collared 25 adult coyotes inside known lynx home ranges. Twelve (4 males, 8 females) remained for sampling; the others died, dispersed, or lived where access was restricted.

- Track surveys. They surveyed three routes on snowmobile trails (111 km in all) twice a month from mid-December to late March, recording every carnivore track with a handheld Global Positioning System (GPS) unit. Every kilometre, and beside every coyote track, they measured snow depth and how far a 100 g brass weight dropped from 1 m sank into the snow.

- Backtracking. They located each coyote by radio, walked in to find their fresh trail, and followed it for 3 km, recording it with a GPS unit. A computer program set out a random “nonuse” track 2 to 3 km away, of a similar shape, which a second technician walked at the same time. Along both, they recorded snow depth, snow supportiveness, prey track crossings and whether the track followed a road or trail. Along coyote trails they also recorded every feeding site and the prey or carcass species.

- Diet. They sent 30 of 85 scats collected on the backtracks to a laboratory, where food items were identified by hair and bone fragments.

### 2. What they found

Coyotes and lynx in the same places

- Coyote tracks made up 65% of the 2,291 carnivore tracks found on survey routes, and lynx tracks 32%.

- Coyotes were found at the same elevations as the routes as a whole (mean 1,591 m), and lynx a little higher (1,626 m). The two largely overlapped.

- Coyotes stayed in deep-snow lynx habitat all winter, even though snow-free areas lay only 10 to 20 km away.

Use of snowmobile trails

- The team backtracked coyotes for 322 km (119 backtracks, an average of 2,705 m each) and walked 358 km of paired random tracks.

- Coyotes used packed snowmobile trails more than random tracks did, but only for 7.69% of their travel: 5.66% on forest roads packed by snowmobiles and 2.03% on other snowmobile trails. They traveled on them for a median of only 124 m at a time.

- They used unpacked forest roads about as much (4.62% of travel, median 102 m), even though snow there was no firmer than elsewhere.

- On average, coyote trails were no closer to packed snowmobile trails than random tracks were (368 m versus 339 m). This did not change with daily, monthly or yearly changes in snow depth or firmness.

- Coyotes clearly chose shallower, firmer snow than random, even when the stretches on packed trails were left out.

Feeding

- The team found 88 feeding sites, about one every 3.7 km. Most (77) were scavenging sites, mostly of deer and other hoofed animals.

- Only 11 were kills, and only 3 of those were snowshoe hares: one hare kill for every 107.3 km of travel. Two of the three were in late March.

- In the scats, 61% of food items were cervids (the deer family) and 12% snowshoe hares.

- Feeding sites were no closer to packed trails than random. Kill sites, including hare kills, were farther from packed trails than expected.

- Coyotes crossed lynx tracks at about the random rate, and hare tracks slightly more often than random (not significant).

### 3. What they concluded

- Coyotes in this area remained in lynx habitat throughout winter, but snowmobile trails did not appear to help them move about or find food.

- In winter, coyotes here were mainly scavengers and rarely killed hares. At an assumed 15 km of travel a day, a coyote would kill at most about one hare a week, far below what they would need to live on hares alone.

- Coyotes may be drawn to roads as cleared travel corridors, rather than to the packed snow on them.

- Coyotes cope with deep snow mainly by seeking out naturally shallower, firmer snow.

- Overall, the influence of snowmobile trails on coyote movements and foraging in winter appeared minimal on this study area.

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

Tracking was central to the study:

- Track surveys: all carnivore tracks along 111 km of routes were identified to species and mapped.

- Trailing: radio-collared coyotes were located and their fresh trails in the snow were followed backwards for 3 km at a time, 322 km in total. When a collared coyote traveled with others and the trails split, the team tried to tell which was the collared animal by track size or stride length; if they could not, they took the right-hand and left-hand trail in turn at each split.

- Signs along the trail: prey track crossings (snowshoe hare, red squirrel, lynx), feeding sites (kills and scavenging, with the prey or carcass species identified) and scats were recorded.

### 5. Tracker or observer certification

Not reported. The backtracking and track surveys were done by technicians whose tracking experience is not described. Telling the collared coyote’s trail apart from those of companions, and telling kills from scavenging, depended on their skill.

### 6. Other technologies, skills or innovations

- A simple measure of snow supportiveness: how far a 100 g brass weight dropped from 1 m sank into the snow.

- A computer program that generated random “nonuse” tracks shaped like real coyote trails, walked at the same time as each coyote backtrack.

- Data-logging Trimble GeoExplorer 3 GPS units that split each trail into segments, smoothed afterwards in a Geographic Information System (GIS).

- Finding collared coyotes by radio from a snowmobile, then walking in on existing trails to avoid packing new snow.

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

- Limiting packed snowmobile trails on this study area is unlikely to reduce competition between coyotes and lynx for food in winter.

- The results should help land managers assess the effects of snow-packing activities on coyotes and lynx.

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

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

- Backtracking on snow gave the researchers what radio collars alone could not: exactly where each coyote walked, how often they stepped onto a packed trail and for how far, and what they ate along the way.

- Pairing each real trail with a random walked track is a useful way to show what an animal chose compared with what was available.

## Cite this summary

Lawrence, K. 2026. Summary of: The Effect of Snowmobile Trails on Coyote Movements Within Lynx Home Ranges (Kolbe and Squires et al., 2007). Original Wisdom, Tracking Literature. https://www.originalwisdom.com/tracking-literature/2007-kolbe-squires-effect-snowmobile-trails-coyote/