Remotely plucked hair genotyping: a reliable and non-invasive method for censusing pine marten (Martes martes, L. 1758) populations

1. What they did

Pine martens are secretive, live at low densities, and are hard to count. The usual way to monitor them is to collect scats along forest trails, but it is unclear how scat numbers relate to how many martens live in an area. Ireland’s martens also have very little genetic variation, which makes it hard to tell individuals apart by their DNA.

The researchers set out to:

• Build new DNA tests. They designed short, sensitive tests (real-time polymerase chain reaction, PCR) to confirm that a sample came from a pine marten and to tell whether it came from a male or a female.
• Find enough genetic variation. They tested 20 microsatellite markers (short, variable stretches of DNA) developed for other members of the weasel family, using tissue from 29 road-killed Irish martens and 7 from France, and chose the most variable ones.
• Collect hair and scats in the field. In April and May 2008 they fixed plastic hair tubes to trees in two plantation forests in County Waterford: 10 tubes at Curraghmore and 8 at Kill, on a grid of 500 × 500 m cells. Each tube was baited with chicken wings and marmalade, and had glue patches inside that plucked a few hairs as a marten pushed past to the bait. After four weeks of pre-baiting, they checked the tubes for seven hair-capture sessions. They also collected scats on transects between the tubes and recorded each sample’s position with a GPS unit.
• Check the results against live animals. In June 2008 they set a cage trap beside each Curraghmore hair tube and compared the DNA of the captured martens with the genotypes from the hair.

2. What they found

Genetic variation

• Irish martens were genetically very uniform: on average 2.29 versions (alleles) per marker, compared with 3.24 in the French samples. Their expected heterozygosity (a measure of genetic variety) was 0.35, against 0.59 in France.
• Eight markers were still enough to tell individuals apart, even close relatives.
• All 20 markers also worked on the stone marten sample, and six of them were variable.

Hair versus scats

• The hair tubes caught marten hair on 114 of 126 tube-nights (90.5%).
• From the last three sessions, 53 hair and 22 scat samples were tested. Almost all (97.3%) gave marten mitochondrial DNA, but hair gave much better nuclear DNA, which is needed to identify individuals and sex. One scat turned out to be from a fox.
• Scats did so poorly for individual identification in the first trial that the census used hair only.
• Hair samples amplified successfully 93.8% of the time, with a low error rate of 1.8% per reaction. Full genotypes came from 49 hair samples.

The census

• Seven individual martens were identified: five at Curraghmore (three males and two females) and two at Kill (one male and one female). Some were detected many times at several tubes, others only once.
• Live trapping caught five martens (three males and two females) at Curraghmore in a single night. Each matched one of the five hair genotypes at every marker, and the DNA sexing matched every animal’s known sex.
• Two Curraghmore males may be a parent and offspring, and so may the two Kill martens, but more markers would be needed to be sure.
• The density, about 2 martens per km², was within the range found in other Irish populations.
• The two Kill martens were also live-trapped in February 2009, and later radio-tracking data showed both lived within the 1 km² site.

3. What they concluded

• Even in a population with very low genetic variation, remotely plucked hair can reliably identify individual pine martens and their sex.
• Plucked hair may be a more reliable and cost-effective DNA source than scats for counting martens, and the methods may suit similar species such as the stone marten.
• Scats are still useful for showing where martens are and what they eat, especially when backed by genetic species identification. Other lab methods might make scats work for identifying individuals.
• Hair tubes are useful even without DNA work: how often they are visited points to the best places for cage traps or camera traps.
• Small woodlands can hold resident martens and could easily be overlooked in field surveys.
• More variable markers are needed, especially for studying relatedness.

4. How tracking was or was not used

The study did not use tracks. It relied on two kinds of sign: hair plucked by the hair tubes, and scats collected on transects between the tubes. The paper does not say how scats were recognised as possible marten scats in the field. One scat, collected as a marten sample, was shown by DNA to be from a fox.

5. Tracker or observer certification

Not reported. Nothing is said about who collected the scats or how they were identified in the field, although one of them proved to be a fox scat.

6. Other technologies, skills or innovations

• Hair tubes: lightweight plastic tubes (250 × 118 mm) wired to tree trunks, baited, with glue patches from a mouse trap inside to pluck hair. A commercial marten lure was put on a nearby tree.
• New real-time PCR tests for species and sex that work on short, degraded DNA, with the female (X) gene acting as a built-in check that the test ran.
• Redesigned, shorter primers for two markers, to work better on poor-quality DNA.
• A “multiple tubes” approach: each sample was tested repeatedly, and an allele was accepted only after it appeared at least twice (or three times for samples with only one allele).
• Handheld GPS (Garmin eTrex) to record sample locations.

7. Implications for conservation (from the paper)

• Basic information such as population size is needed to judge a population’s conservation status, but is hard to get for elusive carnivores.
• Genotyping plucked hair offers a way to “tag” individual martens permanently without catching or seeing them. It could be used to study genetic variation, population numbers and habitat use in landscapes changed by people.
• Non-invasive sampling can increase the sample size and area covered when studying where Ireland’s martens came from, where road kills and museum specimens are otherwise the main sources.
• Riparian woodland and hedgerows may serve as corridors between forest fragments, which matters for gene flow between populations.