Evaluation of sign surveys as a way to estimate the relative abundance of American mink (Mustela vison)

1. What they did

American mink are an invasive species in Europe and a threat to native wildlife, so managers need a cheap, reliable way to monitor them. Mink are nocturnal and elusive, and are mostly monitored through their signs. The authors asked whether surveys for scats and footprints can show the relative abundance of mink (how many there are compared with another place or time).

• Sign surveys. Trained surveyors walked both banks of 500 m river sections, searching up to 5 m from the water for mink scats and footprints, and removed every sign so it would not be counted twice. They worked on the River Teign (surveyed 11 times), the River Exe (once) and 44 spaced sites in central England (5 times), between November 1998 and March 2000. Surveys waited at least 3 days after major rain.
• Three possible sources of bias. They tested whether the surveyor (seven surveyors in four pairs, each covering one bank), the number of marking places (tree roots, boulders, bridges and soft verges), and the month affected how many sections had signs.
• Comparison with trapping. On the River Teign they ran eight live-trapping sessions with 40 baited traps, marked each mink with a microchip, and compared the number trapped with the proportion of sections with signs.

2. What they found

Surveyors and marking places

• The two banks did not differ in habitat or marking places.
• No pair of surveyors differed in how often they found mink signs.
• Scats were not more common in sections with more tree roots, boulders or bridges. Footprints were more common in sections with more soft verges.
• Footprints were found more often than scats.

Month

• The proportion of sections with signs changed greatly through the year. It peaked in the breeding season (January to March) and in June, when mothers are still with their kits, and was lowest in April and May (gestation and weaning), and low in October and December on the Teign.
• The Teign and central England gave similar results in the same months, except in October 1999, when central England had many more positive sections.
• The authors point out that a river with 20% of sections positive tells you little unless you know the month: a dense population in the gestation season and a sparse one in the breeding season could give the same figure.

Trapping versus signs

• Twelve different mink (five males, seven females) were trapped on the Teign, with few recaptures. Trapping suggested two to four adult mink on the river, a very low density.
• The number of mink trapped was only weakly related to the proportion of sections with signs.
• In the breeding season many sections had signs even in months when no mink were trapped, so signs showed mink were present when trapping did not. The chance of detecting signs seemed high (about 0.5) even at low density.
• Otter signs were found in more sections than mink signs (yearly average 55% versus 36% on the Teign), even though there were more mink than otters.

Cost

• Monitoring 10 km of river for a month would take about 80 hours and £650 by trapping (plus about £900 for traps), but only about 15 hours and £110 by sign survey.

3. What they concluded

• Of the three factors, only month had a substantial effect. If surveyors are trained and each section has enough marking places, the proportion of sections with signs probably depends mainly on how many mink there are and how they are behaving.
• Sign surveys can compare the relative abundance of mink between areas if the surveys are done in comparable seasons and the sites are laid out in the same way. Continuous surveys and spaced (discontinuous) surveys should not be compared directly.
• At low mink densities, sign surveys may be better than trapping for showing presence and relative abundance. At higher densities trapping may do better, and only trapping can at present estimate absolute numbers.
• Sign surveys cannot be used to compare otters with mink, because otters leave more signs.
• Surveys should not be done in heavy flood conditions, and surveyors should be trained and of comparable motivation and experience.
• Further work is needed on a range of mink densities, on the effects of diet and weather, on how fast signs decay and are replaced, and on track boards like those used for martens.

4. How tracking was or was not used

The whole study rests on tracks and signs identification. Surveyors searched river banks for mink scats and footprints and had to tell them apart from otter signs. The paper describes the difference: mink scats are more compact, often corkscrew-shaped, about 6 to 8 cm long and 0.9 cm across, often contain fur or feathers, and smell unpleasant when fresh, while otter scats smell sweet and musky. Mink footprints look more star-shaped and less rounded than otter footprints, with five toes around a central pad (sometimes only four register), and claw marks are occasionally visible. The authors estimate that mink scats last about a month in the open (based on otter studies), while footprints probably disappear sooner through rain, flooding and trampling, so each survey reflects roughly the previous month. No trailing was reported.

5. Tracker or observer certification

Reported but not validated. The surveyors are described as trained and “of comparable experience”: training meant surveying alongside an experienced surveyor for at least five sections with signs, then being helped with sign identification for at least 3 days. No certification or test of identification accuracy is reported. The study did compare how often different surveyors found signs and found no difference, but this measures detection rates, not whether the identifications were correct.

6. Other technologies, skills or innovations

• Live trapping with baited traps, anaesthesia and microchip marking.
• Habitat recorded with a modified version of the River Habitat Survey.
• A cost and time comparison of sign surveys and trapping.

7. Implications for conservation (from the paper)

• Because mink threaten native wildlife, a reliable and efficient way to monitor them is of great importance.
• Sign surveys are much cheaper and quicker than trapping (about 6 times cheaper and 5 times faster), so regional monitoring over large areas has to rely on them.
• Sign survey results must be read with the season in mind, and surveys compared only when done at comparable times of year and with similar site layouts.