Distinguishing Tracks of Marten and Fisher at Track-Plate Stations

1. What they did

Fishers and American martens are rare, protected forest carnivores in much of the western United States, so their numbers cannot be followed through trapping records. Baited track-plate stations are a cheap, non-lethal alternative: an animal walks across a soot-covered aluminium plate and leaves a print on white contact paper. The two species are close relatives of similar size, though, and male marten tracks overlap in size with female fisher tracks. Field guides offered only anecdotal ways to tell them apart.

The authors set out to build a standardised way to measure these tracks and a formula to tell the species apart:

• A track library (1991 to 1993). They collected tracks from known adult individuals of two marten subspecies and two fisher groups: captive animals that walked across plates in their cages, and wild animals that walked across sooted plates in boxes as they were released from traps. An obstruction near the end of the plate encouraged animals to walk slowly, as they would at a station. Illegible tracks were excluded.
• A standard way to measure. They set rules for deciding whether a track came from a left or right foot, numbered the toe pads (1 to 5) and the interdigital (palm) pads (I1 to I4), and built a measuring grid anchored on the pads themselves, so that each track carries its own reference point. One observer measured photocopies of the tracks with calipers, a protractor and grid overlays.
• Choosing measurements. From 144 candidate measurements, they kept those that were easy to measure precisely and present on most tracks, then compared them between species and subspecies.
• Building and testing formulas. They used 80 tracks from 21 individuals to build discriminant functions (formulas that score a track as one species or the other), and tested them on 95 separate tracks never used to build them.

They also checked qualitative traits thought to help: hairier marten prints, separate palm pads in marten, and a missing thumb print in marten.

2. What they found

Track size

• About 25% of the track plates were discarded because the tracks were illegible.
• Overall track length and width overlapped between species, mainly between male marten (the large subspecies) and female fisher. Confidence intervals overlapped only for male M. a. actuosa and female Pacific fisher, but ranges overlapped for several combinations.
• How each single print was left caused more variation than differences between individuals or between plates (the track itself accounted for about 27% of variation, individuals and plates about 6% each). So the analysis was done on single tracks, not averages per animal.

Qualitative traits

• Of legible tracks, 22% were missing the thumb print (toe pad 1) and 15% were missing palm pad I2. About 80% of tracks missing the thumb print were marten.
• A connected print between palm pads I3 and I4, thought to mark fisher, was unreliable: only 27.5% of fisher tracks showed it fully connected, against about 2% of marten tracks.

Measurements and formulas

• Only four measurements differed between species in the single-variable tests: width of I2, height of I3, height of I4 and the combined area of I2, I3 and I4. Many more differed between subspecies.
• The final five candidate measurements all lay on the palm pad, near the centre of the grid.
• Several formulas classified every test track correctly. The simplest used only the total width of the central palm pad (I3). A three-measurement formula using easy straight-line measurements of the central and outer palm pads was just as successful.
• Formulas using only the four general track dimensions from an earlier key (overall length, toe spread, palm pad height and width) did less well, classifying 51% to 96% of test tracks correctly, depending on the combination.

3. What they concluded

• Tracks of adult fisher and American marten on sooted track plates can be told apart reliably with a few measurements of the palm pad, without more costly camera stations.
• They give a classification rule for unknown tracks believed to be adult marten or fisher: if (4.595 x width I3) + (3.146 x length I3) + (0.906 x width I4) minus 80.285 is above 0, the track is fisher; below 0, marten.
• They recommend the three-measurement formula over the single one, as insurance against errors by inexperienced measurers or poor photocopies, and because it should hold better for populations not in their sample.
• The palm pad measurements work best probably because there is less room for variation near the centre of the track; foot placement and weight shift spread the toes.
• The pad-anchored measuring grid is a key part of the method and could be adapted for other closely related species.
• Users must first be sure a track is from one of these two species, and that all needed parts are visible: fisher tracks with faint palm pads can be misclassified as marten. Each track on a plate should be classified separately rather than averaged, since both species may visit one station.
• The method has not been tested on juveniles; young fishers may have marten-sized tracks. With more work, it might separate subspecies, sexes and perhaps individuals.

4. How tracking was or was not used

The study is about tracks and signs identification: footprints left on sooted track plates with white contact paper. It turns identification into measurements of the toe and palm pads and a formula, rather than judgement by eye, and tests how well the traditional visual clues (connected palm pads, missing thumb print, hairiness) work. Tracks were collected from animals of known species and sex. No trailing was involved.

5. Tracker or observer certification

Not reported. One observer took all the measurements; who this was and their experience are not stated. Several named people collected tracks in the field. The authors note that observer and photocopy errors could increase misclassification by inexperienced users.

6. Other technologies, skills or innovations

• Sooted aluminium track plates (20 x 80 cm) partly covered with white contact paper, which records a “positive” print with more detail than the bare aluminium “negative”.
• A measuring grid anchored on the track’s own pads, set with T-square and triangle; measurements by electronic and digital calipers (to 0.01 mm), protractor (to 0.5 degrees) and gridded acetate overlays for areas.
• Tracks measured from photocopies, checked against originals.
• Rules, in order of reliability, for telling left from right feet.

7. Implications for conservation (from the paper)

• Both species are management indicator species for national forests in four Forest Service regions because of their link to old-growth forest, and fisher conservation is of growing concern. They are also possible competitors where their ranges overlap.
• A reliable, cheap, non-lethal way to tell their tracks apart supports regional surveys and monitoring of distribution and abundance, and studies of how the two species share space and time.
• Telling sexes apart from tracks could in future give population data that has so far needed trapping.