Track Plates Detect the Endangered New Mexico Jumping Mouse

1. What they did

The New Mexico jumping mouse lives along streams in tall, dense plants in the southwestern United States, is active only 3 to 5 months a year, and hibernates the rest. It was listed as endangered in 2014. Surveys had relied on live traps, which can stress, injure or kill the mice. The authors wanted a gentler way to find out where jumping mice are present, using their unusually long feet and toes.

• Making a track reference (June to August 2015). Mice and voles caught in a live-trapping project were placed briefly (10 minutes or less) in a plastic box lined with sticky clear film and an inked felt pad. Their footprints were measured with calipers and the best prints were used to make a photo reference.
• Building a field track plate (June to September 2015). They tested inks and 4 enclosure designs: a folding Sherman trap, a double U-style gutter tube, a single K-style gutter and a modified plastic shoebox. Each held self-adhesive paper, an inked felt pad in the middle and bait (steel-cut oats and peanuts), with a shingle or roofing felt cover. They set 20 track plates at each of 3 sites for a week.
• Comparing track plates with live traps (2016 and 2017). At 16 sites in the two national forests they laid out 80 points 3 to 5 m apart along each stream and ran both live trapping and track plates at the same points, in random order, a few weeks apart. Each method ran for 3 days. Three trained reviewers checked every track plate; jumping mice counted as present at a point when all 3 reported a jumping mouse track.

2. What they found

Telling the tracks apart

• They collected 1 to 13 tracks each from 21 animals of 6 species: 5 jumping mice, 3 long-tailed voles, 3 Mogollon voles, 5 montane voles, 2 brush mice and 3 deer mice.
• The front toes of jumping mice were longer than those of voles and deer mice (a statistically significant difference). Front foot pad width and length did not differ.
• Jumping mouse hind prints and toes were over 20 mm long, compared with under 14 mm for the other species, so they could be recognised even from partial prints. Hind prints of voles and deer mice lacked the heel.
• Chipmunk and woodrat tracks were clearly larger, and shrew tracks clearly smaller.

Choosing a design

• A 1:1 mix of carpenter’s chalk and mineral oil made the best ink. Water dried too fast; graphite and charcoal printed poorly; carbon black printed well but was costly, hard to get and listed as possibly carcinogenic.
• An inkpad at the entrances left plates smeared and unreadable, so it went in the middle.
• The modified shoebox worked best: largest track surface, stable, light, stackable, easy to assemble, did not flatten vegetation and survived heavy rain and flooding. The Sherman trap and U-style tube rolled or collapsed, and the K-style gutter was bulky to carry.
• Plates needed checking every 24 hours so that tracks did not pile up too densely to read (overprinting).

Track plates versus live traps

• Jumping mice were caught at 10 of 16 sites and their tracks found at 9 of 16 sites.
• The two methods disagreed at only 1 site, where a mouse was caught but no tracks were found. Agreement was strong whichever method was used first.
• No injuries to animals were found from track plates, while live traps did injure or kill jumping mice.
• A track-plate enclosure cost about US$5, a live trap about US$28. Track plates could be checked at any time of day by smaller teams and spread over wider areas. In 2016, track plates were used at 66 sites across more than 1,000 km² of the Apache–Sitgreaves National Forests.

3. What they concluded

• Track plates are an effective, inexpensive, efficient and noninvasive way to detect New Mexico jumping mice.
• Low numbers of mice or too little ink may explain the one site where tracks were missed.
• Track plates show only presence. They cannot identify individuals or give information on age, sex or genetics, and where other jumping mouse species or subspecies overlap, tracks may only be identifiable to genus.
• Setting and checking track plates did not save much time compared with trapping, although checking was more flexible.
• Track plates make regional surveys and monitoring more feasible. The authors made a 14-minute video on how to assemble, deploy, interpret and archive track plates.

4. How tracking was or was not used

The whole study is about identifying footprints. The authors measured the fore and hind prints and toes of jumping mice and five other rodents, built a photo reference, and used it to identify jumping mouse footprints on ink track plates in the field. No trailing was involved. Identifications on track plates were compared with live captures at the same sites.

5. Tracker or observer certification

Reported but not validated. Three independent reviewers “trained in identification of tracks” checked each plate, using the photo reference, and a detection needed all three to agree. Who they were and how they were trained is not described, and their individual accuracy was not tested, although the close agreement with live trapping at site level supports the method as a whole.

6. Other technologies, skills or innovations

• A low-cost track plate made from store-bought materials: a modified plastic shoebox with two offset 5 × 5 cm entrances, drainage holes, clear self-adhesive film, a felt inkpad soaked in carpenter’s chalk and mineral oil, bait and a roofing-felt cover.
• Track plates are archived by sticking the film onto labelled white paper and storing it in plastic sleeves in a binder.
• Track plates float, so they can be recovered after floods.
• A 14-minute instructional video: https://www.youtube.com/watch?v=i2x0Ydc1XVM

7. Implications for conservation (from the paper)

• Survey sites for an endangered animal can be spread more widely, with less risk of injury, death or “take” under the Endangered Species Act.
• Better detection helps map the distribution and habitat use of the jumping mouse, which is needed for its recovery.
• Live trapping is still needed where individual, demographic or genetic information is wanted.