Monitoring the small-scale distribution of sympatric European pine martens (Martes martes) and stone martens (Martes foina): a multievidence approach using faecal DNA analysis and camera-traps

1. What they did

European pine martens and stone martens look alike, eat similar food and share much of Europe, including northern Spain. Their scats cannot be reliably told apart by shape and size alone, which makes the two species hard to monitor. The researchers asked two questions: do both species really live side by side at a small scale, and how do camera traps and scat DNA compare as ways to detect them?

They worked in Montes do Invernadeiro Natural Park in north-west Spain, an area of 6700 ha divided into 67 grid cells of 1 km².

• Scat surveys. Two first surveys (two 400 m transects in each cell) found marten scats in 26 cells. From June 2004 to August 2006 they then walked 300 m transects along forest roads in those 26 cells every month, 218.7 km in all, with transects 700 m apart so that samples were more likely to come from different animals. They collected only fresh scats (strong smell, a mucus layer, no drying out) and told marten scats from those of red fox and wildcat by size and shape.
• DNA. They identified 90 fresh scats to species with a genetic test (PCR-RFLP) that gives a different banding pattern for each marten species.
• Camera traps. Every three months from June 2004 to October 2006, they set 10 to 20 cameras baited with chicken, in 22 cells: where marten scats had been found, and in good marten habitat even with no signs. Cameras stayed out 3 to 12 days (916 camera-trap-nights). The pale throat patch, or “bib”, is yellow in pine martens and white in stone martens, so they used its colour to tell the species apart, and its pattern, size and symmetry to tell individuals apart.

2. What they found

• No stone martens. Neither method detected a single stone marten.
• Scat DNA. 80 of the 90 scats (88.8%) were from pine martens; the other 10 (11.2%) gave no DNA result. Fresh scats were found in 17 of the 26 cells, and DNA confirmed pine martens in all 17. On average there were 5.3 samples per cell and 1.97 scats per month per kilometre surveyed.
• Camera traps. 126 photographs of four carnivores: pine marten 57.9%, red fox 33.3%, wildcat 5.6% and Iberian wolf 3.2%. Cameras detected pine martens in 9 of the 22 cells with cameras (40.9%), red fox in 36.4%, wildcat in 27.3% and Iberian wolf in 18.2%.
• Both methods together. Pine martens were confirmed in 20 cells (29.9% of the park): 17 by DNA, 9 by camera, 6 by both. Cameras found martens in some cells where the scat DNA had not.
• Individuals. From camera spacing, at least nine different pine martens were photographed; bib differences identified seven individuals.
• Habitat. Cells with pine martens had plenty of tree cover (conifer and broadleaf forest), permanent watercourses and much rocky ground and outcrops. Cells without them were mostly scrubland.
• Outside the park. Four scats from the edge of the nearest village and several dead mustelids showed that pine martens were also there.
• Fox scats. The DNA test did not work on fox scats, so any fox scat mistaken for marten in the field would have failed to amplify.

3. What they concluded

• Both methods agree: only the pine marten lives in the study area. This fits earlier findings that pine martens prefer uniform, well-preserved forest and that stone martens are pushed into disturbed, deforested and built-up areas where the two overlap.
• The authors suggest that the low level of human disturbance around the park may help explain why stone martens are absent, and call for studies across different levels of urbanisation and disturbance.
• Each method has strengths and weaknesses. Scat DNA covers large and small areas, but fresh scats take a lot of field effort to find, the DNA is often degraded, and scats from other species add laboratory cost. Camera traps record the whole carnivore community and can identify individuals, but need more effort to set up, check and revisit, so they suit smaller areas.
• Using both methods together in a small-scale survey can make the data more accurate and the study more ecologically meaningful.

4. How tracking was or was not used

Scats were the core of the study. Field teams walked transects along forest roads, where martens and other carnivores often leave scats as scent marks, and recognised marten scats by size and shape, separating them from red fox and wildcat scats. They judged freshness from smell, a mucus layer and lack of drying out, and collected only fresh scats for DNA. The cells surveyed were chosen because earlier surveys had found marten signs there (scats, footprints or sightings). The scats could be identified to the genus Martes by eye, but the two marten species were told apart only by DNA. No trailing was reported.

5. Tracker or observer certification

Not reported. The paper does not say who identified and collected the scats or how experienced they were. This matters because field identification decided which scats went for DNA testing. Three gamekeepers are thanked for help with fieldwork.

6. Other technologies, skills or innovations

• A genetic test (PCR-RFLP) on a short stretch of mitochondrial DNA, designed to work on degraded DNA from scats, that separates pine marten and stone marten.
• Baited camera traps (Moultrie, Canon Prima AF 9s and Stealth Cam), 16 triggered by infrared and four by a pressure plate under the animal’s weight. The bait was placed 70 cm above the ground to encourage martens to show their bibs.
• Telling individual martens apart in photographs from the colour, pattern, size and symmetry of the bib.

7. Implications for conservation (from the paper)

• Correct species identification is a key step in conservation, and scats alone can mislead when similar species share an area.
• Habitat management in these forests should aim mainly at conserving the pine marten, which seems to depend greatly, perhaps entirely, on undisturbed forest.