Camera trapping for pine martens: challenges, and how to tell who’s who

Student Spotlight - Sian Green

PhD Research at the University of Durham

Twitter: @SianGreen92   Website: https://www.mammalweb.org/en/ 

Those of you following UK mammal news this summer will have seen the exciting news that a population of pine martens (Martes martes) has been confirmed in the New Forest. This is big news for a species that was once thought to be almost extinct in England, and with only small remnant populations hanging on in Scotland. This New Forest population is not the only British pine marten population outside of Scotland, with conservation efforts having translocated pine martens to Wales and the Forest of Dean as part of efforts to restore pine martens to more of their native range. Individuals have also been recording at other locations including Shropshire.

Pine martens have made the headlines a few times with hopes they might help to control grey squirrel populations, but there is still a lot we don’t know about this elusive member of the mustelid family. When re-introducing any species to an area, monitoring is very important in order to determine whether it was a success, and a new, stable population has been created. It is also important to understand what impacts the species might be having on its environment. However, pine martens are not always the easiest species to monitor, as I found out during my survey of the Forest of Dean with the Gloucestershire Wildlife Trust following the release of the first animals of that reintroduction project.

When it comes to monitoring elusive mammals, camera traps are usually a researcher's friend as they can be left in the field for weeks at a time recording animals that move in front of them. After rotating 15 pairs of camera traps around 74 locations over four months, we hoped to have enough recordings of pine martens to be able to estimate their density. As we already knew how many martens had been introduced to the forest, this presented a unique opportunity to test this camera trapping methodology and establish whether it was a suitable method for monitoring martens and accurately establishing their density. Unfortunately, the pine martens had other ideas, and were only detected on 3 occasions at 2 different sites, which didn’t give us much data to work with. (Thankfully, this survey did have another purpose which was more successful, so it wasn’t a completely failed experiment!)

Not to be deterred, working with the Gloucestershire Wildlife Trust again, we have gone in for round two with the pine martens, but this time trialling a slightly different camera trap setup. Having learned that a systematic grid of 1km spaced camera traps around the Forest was not adequate for monitoring pine martens at a low density we needed a way to increase the chances of detecting a pine marten at a camera trap station. This way was eggs and peanut butter. Camera traps only have a relatively small field of view and so it could be very easy for a pine marten to pass within meters of a camera but not be detected. Much like students, providing free food is usually a good way to get pine martens to visit a site, and not only get them to visit but get them to stay in front of the camera for a little longer.

Baiting sites like this can violate assumptions for certain types of camera trap data analysis, making it unsuitable for answering some questions, but it can open up other possibilities. Pine martens have a bib of creamy coloured fur extending from under the chin down the chest. At first glance, this bib may look very similar on all pine martens, but a closer look shows they each have individually unique markings. This can include patches of brown fur in the middle of the patch of lighter fur, and different patterns of fur colouration around the edge of the bib. Using camera traps for individual ID has often been used in the study of large felids, such as tigers and leopards, which have unique coat patterns but has rarely been attempted for other types of animals.

If we can identify individual pine martens, we can build up detailed behavioural information on individuals, such as which areas of the forest they are visiting, if they are interacting with other pine martens (especially important around breeding season), if they have successfully bred and had kits, and whether individuals are healthy. Set up in the correct configuration, individually identified animals in camera trap footage can also be used to estimate population density using a mark-recapture technique. However, the first step is to determine whether we can collect good enough quality camera trap footage in order to be able to identify pine martens, and to establish the best settings and positioning of a camera trap in order to maximize the chances of getting suitable footage.

How you can help

Classifying camera trap footage can be time consuming, and with many conservation projects short on time and funds there is often a delay in processing camera trap footage after it is collected, and some may even go unanalysed. One solution to this problem is recruiting members of public as volunteers, or ‘citizen scientists’ to help classify the footage. Camera trapping citizen science projects have become increasingly popular for classification of species in large scale camera trap projects, with the dual benefits of engaging the public in the scientific process and assistance processing large data sets. Asking people to identify animals to an individual level is a somewhat more difficult challenge, and few projects have involved this kind of task so far. But with an increase in pine marten populations in the UK, what better time to try? We have created the first ever camera trapping pine marten identification project on the citizen science platform MammalWeb and are inviting all wildlife enthusiasts to help us in trying to identify individuals from the Forest of Dean. The challenge is a tricky one, but with patience and attention to detail it is possible (from some footage – if all you get is a blurry marten tail it is definitely not!)

But how does one go about trying to individually identify a pine marten? Well, think of it as a little like ‘Guess who?’ but for pine martens! We have created a number of profiles for known pine martens in the Forest of Dean and it's best to start by eliminating the martens it is definitely not. For example:

Question 1) Is it a male or female?

Males tend to be larger than females with slightly more robust features, whereas females tend to be a little smaller with more slender features. In some footage, sexual characteristics can be seen – making the answer to this question much more obvious.

Question 2) Are there any distinctive brown patches on the pale fur of the bib?

Some animals have distinctive patterns of brown fur on their bib, which can help identify them. We have drawings of the bib patterns on the underside of the chin and chest of each pine marten to see if they can be matched up.

Want to try it?

Have a look at the sequence of photos below and see if you can identify which pine marten you think it might be out of these four options:

FD03 - Female	FD05 - Female	FD10 - Male	FD11 - Male

Answer: This pine marten is FD10! Well done if you got it right. We can tell from the third photo in the sequence this animal is male. Then if we look closely at the first photo in the sequence we can just about see most of a triangular patch of dark fur on the bib, matching the one in the diagram.  If you want to try another head to the Forest of Dean pine marten project on MammalWeb for more martens to try and identify.