Techniques for trapping and tracking stoats (Mustela erminea); a review, and a new system

1. What they did

Stoats are small, fast, wide-ranging and often active at night, so watching them directly is rarely possible. The authors reviewed the three simplest indirect ways of studying them: kill-trapping, live-trapping and recording footprints. For each, they set out what information it gives and its advantages and disadvantages, drawing on studies of other species where work on stoats was lacking.

They then described three tools suited to rugged country far from base, developed or introduced for work in New Zealand:

• The Fenn trap, a humane kill-trap from England, introduced to New Zealand by Carolyn King in 1972, with instructions on tunnels, setting, baiting and spacing.
• The Edgar live-trap, a wooden tunnel trap with a large, dark nest box, designed by R. L. Edgar. A mercury switch lets the trap be checked from a distance.
• A footprint tracking tunnel using a two-part chemical dye: one part in an ink pad, the other sprayed on paper. Animals walking through leave permanent prints.

They tested parts of these methods in the field and finished by showing how the three could be combined in a single study.

2. What they found

Kill-trapping (Fenn traps)

• In Fiordland National Park, 44 baited traps caught 9 stoats, 11 rats and 12 other animals, while 44 unbaited traps caught only 1 stoat over the same 76 nights.
• In a fresh versus stale bait trial, traps with fresh bait caught 14 stoats and those with stale bait caught 7 (67% with fresh bait). This difference was not statistically significant, but fresh bait caught significantly more other species, mostly opossums.
• At Craigieburn Forest Park, unbaited Fenns set for 3 years caught 166 stoats, a steady average of 4.6 a month. Catches varied greatly between trap sites (from 0 to 31 stoats per trap), because stoats rarely visit some parts of their home range.
• Female stoats are smaller and have smaller home ranges, so traps spaced too widely catch mostly males.
• An area attractive to stoats and larger than about 10 to 20 hectares probably could not be kept permanently clear of them without prohibitive effort.

Live-trapping (Edgar trap)

• About 30 to 40 hectares must be live-trapped for each stoat observed, so a worthwhile study area needs at least 200 hectares. Walking round traps twice a day limits how much one person can cover.
• In the Orongorongo Valley, 7 marked stoats gave 36 captures and recaptures in 2748 trap-rounds. Most (29) came in early 1972, when rodents were plentiful after a heavy beech seedfall. Stoat numbers then fell sharply, and in 1974-75 the newest version of the trap caught only 3 stoats (none recaptured), plus 27 rats, 4 hedgehogs and 2 opossums.
• When stoats were plentiful they were caught easily: one released stoat ran ahead and entered the next trap before the trapper.
• Stoats and rats almost always carried the dead-mouse bait into the dark nest box and stayed there. Captives were anaesthetised with ether inside the nest box, so they never saw the operator and were handled only while unconscious. They were seldom unconscious for more than 2 to 3 minutes.
• An ear-tag on one male lasted at least two and a half years (February 1972 to September 1974).

Footprint tracking

• Tracking tunnels can show distribution, activity, regular movements and habitat use, and with marked animals, home range use and social behaviour. They disturb animals little and record every animal that passes, from large insects upward, but give no information on diet or body weight. Without marked animals, more activity is easily confused with more animals.
• The chemical ink gives sharp, permanent blue-black prints that develop in seconds and are not spoiled by damp, heat or sun. The ink stays wet for 10 to 30 days.
• A tunnel check takes only a minute or two, and prints can be identified and analysed later (for example, mouse, rat and mustelid prints, fore and hind feet).
• Where mice are numerous they cover the papers quickly and hide other prints.
• In the Orongorongo Valley, baited mouse traps caught 0.5 mice per 100 trap nights, while unbaited tunnels were tracked by mice at 1.6 per 100 tunnel nights. Mice seemed at least as willing to enter unbaited tunnels as baited traps.

3. What they concluded

• All three tools imitate an open-ended tunnel, like the burrows and runways where stoats hunt. Stoats have very acute senses, though, so trap bias can be reduced but never eliminated.
• The three methods complement each other and tend to cancel out each other’s limitations. A complete study needs information from both live and dead animals.
• They outlined a hypothetical integrated study: tracking tunnels first show where resident stoats are; live-traps are then placed there to catch and mark them; and an outer ring of Fenn traps picks up dispersing animals and gives data on population structure and breeding.
• Radio-telemetry gives more information about a few individuals, but it still depends on reliable live-trapping, so these simpler methods remain useful alongside it.
• For quick surveys of offshore islands, footprint tunnels would be more efficient than traps for detecting small numbers of rats, mice and stoats.

4. How tracking was or was not used

Footprint tracking was one of the three methods reviewed and the subject of the authors’ new system. Prints on the tunnel papers were identified to animal group (mouse, rat, mustelid) and to fore and hind foot. The authors also reviewed earlier ways of recording tracks: carbon dust, talc, real snow, sand transects and ink pads. They mention that formulas from snow-tracking studies should be consulted for interpreting tunnel data, and that toe-clipping could give individual stoats distinctive footprints, though this had not yet been tried on small mustelids and raises humane concerns. No trailing was reported.

5. Tracker or observer certification

Not reported. The paper does not say who identified the footprints on the tracking papers or how reliable the identifications were.

6. Other technologies, skills or innovations

• The two-component ink: ferric nitrate, polyethylene glycol, detergent and water in the ink pad, with tannic acid in ethanol sprayed on brown Kraft paper. The full recipe is given, with adjustments for wet and dry climates.
• Tracking tunnels for stoats 50 x 8 cm, with a removable aluminium tray holding an ink pad between two papers, so animals are recorded in either direction.
• The Edgar live-trap: a wooden trap with a novel trigger, a nest box that doubles as an anaesthetising chamber, and a mercury switch wired to a checkpoint so a battery tester can show from the track whether the trap has closed.
• Fenn traps set in tunnels, with signal devices that drop when a trap is sprung so it can be checked from a distance.
• Horses suggested as the best transport for long trap rounds where they are allowed.

7. Implications for conservation (from the paper)

• These methods have potential as management tools, especially for identifying nest predators and for wildlife surveys of islands.
• Many of New Zealand’s offshore islands are vital sanctuaries for rare native species that cannot survive alongside introduced predators. Tunnels could be used for first surveys, Fenns added if control is needed, and tunnels used again to check whether control worked.
• Tracking tunnels have been used to identify some predators at South Island robin nests, but the tunnels themselves improved the birds’ fledging success, so they may not suit studies where the birds’ breeding rate matters.
• To protect a nesting area, 40 to 50 Fenn traps over a hectare or two, set permanently and re-baited, placed to meet newcomers travelling in.