# American Marten, Fisher, Lynx, and Wolverine: Survey Methods for Their Detection

> How do you prove a wolverine, lynx, fisher or marten lives in a forest when you almost never see one? This classic 1995 Forest Service manual compares cameras, sooted track plates and snow tracking, and sets out how to read gaits, measure prints, and cast tracks in snow. It also warns that reading the manual is no substitute for practice in the field.

Summary by Kersey Lawrence, Original Wisdom. Updated 9 November 2026. Page: https://www.originalwisdom.com/tracking-literature/1995-zielinski-kucera-american-marten-fisher-lynx/

## Citation and access

- **Authors:** William J. Zielinski and Thomas E. Kucera
- **Year published:** 1995
- **Journal or source:** USDA Forest Service, Pacific Southwest Research Station, General Technical Report PSW-GTR-157
- **Full citation:** Zielinski, W. J. and Kucera, T. E. (eds.). 1995. American marten, fisher, lynx, and wolverine: survey methods for their detection. Gen. Tech. Rep. PSW-GTR-157. Albany, CA: Pacific Southwest Research Station, Forest Service, U.S. Department of Agriculture. 163 p.
- **DOI:** 10.2737/PSW-GTR-157
- **Access status:** Hosted: U.S. government work, not subject to copyright in the United States
- **PDF:** https://www.originalwisdom.com/wp-content/uploads/2026/11/1995_Zielinski_and_Kucera_american-marten-fisher-lynx-and-wolverine-survey-methods-for-their-detection.pdf

## Tags

- **Species:** Mammals > American marten; Mammals > Bobcat; Mammals > Canada lynx; Mammals > Fisher; Mammals > Mountain lion; Mammals > Wolverine
- **Region:** North America > United States
- **Tracking:** Beds; Caches; Digs; Gaits; Hair; Scat; Sign surveys; Snow tracks; Track casts; Track measurements; Track photographs; Tracking; Tracks; Tracks and signs identification; Trailing; Urine
- **Behaviour and ecology:** Behaviour; Food caching; Hunting; Movement
- **Methods and technology:** Camera traps; Detectability; Distribution mapping; DNA; Observer agreement; Power analysis; Survey costs
- **Data type:** Existing data; Protocol or guidelines; Review
- **Observer qualification:** Not needed
- **Discipline:** Wildlife tracking
- **Year published:** 1995

## Detailed keywords

- **Species:** Studied (survey targets): American marten (Martes americana), fisher (Martes pennanti), lynx (Lynx canadensis) and wolverine (Gulo gulo), together called MFLW. Track look-alikes discussed: bobcat (Lynx rufus), mountain lion, wolf (Canis lupus), coyote, domestic dog and cat, red fox (Vulpes vulpes), bears (Ursus), river otter (Lutra canadensis), badger, mink and weasels (Mustela), striped skunk (Mephitis mephitis), raccoon (Procyon lotor). Prey and bait: snowshoe hare (Lepus americanus), road-killed deer (Odocoileus), chicken, fish; porcupine quills in scat suggest fisher. Other examples mentioned: grizzly bear (Ursus arctos), tiger (Panthera tigris), European pine marten (Martes martes).
- **Location and habitat:** Western United States (Rocky Mountains, Cascade Range, Coast Range, Sierra Nevada), mainly coniferous forest; marten and fisher in late-successional forest, lynx in a range of forest stages, wolverine in mixed forest and open country. Track data come mainly from the Rocky Mountains and Alaska, with some from Michigan, Massachusetts, Maine and elsewhere; the lynx, bobcat and mountain lion discriminant functions use prints from Colorado. Notes on Alaska and Canada, where the species are more common and still trapped.
- **Topics:** detection (presence) surveys for rare forest carnivores; non-lethal survey methods; photographic bait stations (single-sensor, dual-sensor and line-triggered cameras); sooted track plates (enclosed box and unenclosed); snow tracking (searching for tracks on routes, and tracking at bait stations); 4 square mile sample units; regional surveys and project surveys; minimum survey effort and latency to first detection; baits and lures; carnivore foot and track morphology; track formulas; snow metamorphism and track distortion (melting, sublimation, settling); gaits (walk, trot, gallop, lope, bound; 2x, 3x and 4x patterns) and gait errors; Minimum Outline and Variable Outline measurements; trail measurements (stride, straddle, center straddle, trough); “Big Picture, Little Picture” approach; pedestal test; track drawing, plaster casting with Snow Print Wax and photography; Snow Tracking Quality (STQ) index; scat and hair collection; data forms and Natural Heritage reporting; failure to detect is not proof of absence; population monitoring and statistical power; winter safety and training; costs.
- **Statistics or models:** Discriminant functions: three-variable function to separate marten and fisher tracks on track-plate paper (Zielinski and Truex 1995); two functions from a small sample (3 bobcat, 6 mountain lion and 7 lynx prints) to separate bobcat, lynx and mountain lion trails in snow. Means and standard deviations of print and trail measurements. Discussion of Type I and Type II errors, statistical power and sample size for monitoring. Coefficient of variation of track measurements between observers (under 1 percent with Minimum Outline) and for one observer (7 percent).
- **Data type:** Protocol or guidelines (a field manual), with a review of earlier studies and compiled track measurements from the literature and the authors’ unpublished data; no new study. Inconsistencies in the report: (1) in the lynx table, hind print length (Variable Outline 7.7 cm, Minimum Outline 6.7 cm) is shorter than width (9.4 and 7.2 cm), while the text says hind print length generally exceeds width; (2) in the snow-tracking costs, “lost field days, 2 people at 2 days per survey” is costed as “2 pd at $75 = $450”, which does not add up; (3) the track-plate chapter calls the unenclosed plate the only method “capable of detecting all four species” but says in the same sentence that neither lynx nor wolverine has been detected on it; (4) the title page says 163 pages.

## Synopsis

### 1. What they did

By the 1990s American martens, fishers, lynx and wolverines had disappeared from much of their former range in the western United States. With trapping mostly closed, managers no longer had trapping records to tell them where these animals still lived. This manual, written by several authors for the USDA Forest Service and partner agencies, sets out standard, non-lethal ways to find out whether these four species are present in an area.

It describes three methods, in “cookbook” fashion, without favouring one:

- Photographic bait stations: automatic 35-mm cameras (single-sensor and dual-sensor) and cheaper line-triggered cameras at baits.

- Sooted track plates: aluminium plates coated in soot, either in a plywood box with sticky white paper that keeps a clear print, or open on the ground.

- Snow tracking: searching roads and trails on skis or snowshoes for tracks, or checking the snow around bait stations.

It also proposes a common survey design: 4 square mile (about 3.2 by 3.2 km) sample units, how many devices or kilometres of route to use in each, how long to keep searching, and forms for recording and reporting results, so that many local surveys can be combined into regional distribution maps. Costs, equipment lists and safety advice are given for each method.

### 2. What they found

Comparing the methods (from earlier studies)

- No study had compared all the methods. Cameras and snow tracking had detected all four species; track plates had detected only fishers and martens (and bobcats).

- Studies disagreed on which method works best. One comparison found snow tracking the most effective for martens when snow allowed; another, in Montana, found snow tracking the least effective, because it missed martens, fishers and a wolverine that cameras or plates detected, and depends on ideal snow and well-trained technicians.

- Identifying an animal is harder from tracks than from a photo of the whole animal. Snow tracking and cameras need more training than track plates, and snow tracking needs a lot of experience.

Recommended minimum effort in each sample unit (stop as soon as the target species is found)

- 35-mm cameras: at least two, 1 mile apart, for at least 28 days.

- Track plates or line-triggered cameras: at least six, 0.5 mile apart, for 12 nights, checked every 2 days; repeat in the other season if nothing is found.

- Snow tracking: at least 10 km on foot in one day, three times a winter (for lynx, three winters, because lynx numbers cycle); or at least two bait stations, checked every few days for at least 30 days. The 30 days come from one study where wolverine tracks first appeared at bait stations after an average of 26.7 days.

Reading tracks in snow (chapter 5)

- Snow changes tracks: melting and wind can enlarge prints up to four times in area, and settling can shrink them. A melted bobcat or coyote track can be mistaken for a lynx.

- Gait matters: mistaking a trot for a walk makes an animal seem larger (a bobcat becomes a lynx), and mistaking a slow gallop for a walk can turn a lynx into a mountain lion. Trackers should follow the trail both ways and measure where the animal walked.

- Measuring only the part of the foot that would touch hard ground (the “Minimum Outline”) gave much more consistent results: three trained measurers differed by less than 1 percent on the same footprint.

- Species accounts give print sizes, gaits, trail habits and signs. Lynx: hairy feet, indistinct prints, a trough of hair drag, mostly walking. Wolverine: five toes, 1-3-1 toe spacing, sideways 3x lope with a wide straddle, caches and scent marks. Fisher and marten tracks overlap in size and are hard to tell apart in snow; marten trails wander and often go under the snow.

- Front prints of adults range from about 5 by 4 cm (marten) to 16 by 11 cm (wolverine). Lynx (about 10 kg) and mountain lions (up to 75 kg) leave prints of about the same size.

Costs (for five sample units, no target species found)

Snow tracking about $4,250; track plates about $5,936 (two seasons); line-triggered cameras about $5,799; 35-mm cameras about $7,448 to $7,500.

### 3. What they concluded

- The methods are for detecting presence only, not for counting animals or measuring population change.

- Failing to detect a species does not mean it is absent, until more is known about how often each method detects animals that are there.

- Each method has strengths and weaknesses; managers should choose by their goals, budget, staff, experience and how reliable the snow is, and should use more than one method where possible and publish the comparisons.

- Population monitoring needs careful planning, enough statistical power and advice from a statistician before it starts; for such rare species it may not be affordable at all.

- Reading the manual is no substitute for field practice: newcomers should first work with experienced technicians.

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

Tracking is one of the three main methods. Chapter 5 is a full snow-tracking guide covering: route choice, snow conditions and timing; carnivore foot shapes and track formulas; how snow changes tracks; gaits and common gait errors; measuring prints and trails; following trails in both directions to find clear prints and a true walk; the “Big Picture, Little Picture” approach (setting, time and surface first, then light and touch); the pedestal test for snow-covered tracks; drawing, plaster casting and photographing tracks; collecting scat and hair; and a Snow Tracking Quality rating for how well the snow holds tracks. Track plates are also a form of tracks identification, and a discriminant function is given for telling marten from fisher prints. The authors stress that decisions about rare species often rest only on track evidence and may be challenged, even in court.

### 5. Tracker or observer certification

No certification is described. The manual recommends training from a biologist experienced in tracking these species, local experts such as trappers, tracking seminars (including Halfpenny’s “Field Verification of Rare Species”) and tracking books, and says that becoming a good tracker takes time. Doubtful track-plate and snow-track identifications should be checked by experienced biologists, and photographs, forms or casts can be sent to the chapter’s senior author.

### 6. Other technologies, skills or innovations

- Trailmaster single-sensor (infrared beam) and dual-sensor cameras, Manley dual-sensor cameras, line-triggered 110 cameras, and early remote video.

- Enclosed track-plate boxes with sticky white Con-Tact paper that record a positive print in fine detail.

- Snow Print Wax for casting in snow, polarized or yellow glasses and low-angle flashlight for seeing tracks, and folding rulers for photographing trails.

- A method for measuring tracks from photographs, correcting for scale and parallax.

- The authors expected DNA from hair and scat to help identify species and individuals in future, and recommend collecting them.

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

- Martens, fishers, lynx and wolverines are indicators of healthy forest ecosystems, and knowing where they live is a key need for their conservation and for decisions under the Endangered Species Act.

- Standard surveys whose results are pooled can map current distribution and, repeated over time, show changes in distribution.

- Project areas are usually small compared with these animals’ home ranges, so surveys should cover a whole township (36 square miles) around each project, and nearby populations should be considered even when nothing is found.

- Surveys where nothing is found should be recorded as carefully as detections.

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

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

- The manual’s own comparison shows that snow tracking can be the most or the least effective method depending on snow and on the skill of the people doing it, which makes tracker skill a key factor in survey results.

- Its advice to verify track identifications through casts, photographs and experienced reviewers, because results may be challenged even in court, shows why tested and recognised tracking skill matters for rare-species surveys.

## Cite this summary

Lawrence, K. 2026. Summary of: American Marten, Fisher, Lynx, and Wolverine: Survey Methods for Their Detection (Zielinski and Kucera, 1995). Original Wisdom, Tracking Literature. https://www.originalwisdom.com/tracking-literature/1995-zielinski-kucera-american-marten-fisher-lynx/