Identifying individual tigers from their paw prints has been controversial. Exciting new methods combine field work and cutting edge statistics from the software giant JMP to show how it can be done.
The majestic and enigmatic tiger – arguably our planet’s most iconic animal and the favourite subject of our childhood books and dreams.
There are only about 3,200 tigers left in the wild. The situation is critical. With all the energy, funding, expertise and public goodwill that is poured into their conservation, surely populations must be recovering?
Recent reports from the third International Tiger Day were conflicting, and most free-ranging tigers are still highly endangered.
So what’s the problem? Surely all we need to do is provide good monitoring to find out exactly where tigers are, and their numbers, and then protect them from illegal killing. How difficult can it be?
Let’s look at the recent history of tiger monitoring in India, as an example. Indian tigers used to be monitored by groups of foresters and naturalists who would walk through the forests and count footprints. This ‘pugmark technique’ as it was known, was not very scientific – it was really based on best ‘guesstimates’ of the number of tigers out there from the number of footprints, or ‘pugmarks’, seen. Some of the people undertaking the survey were experienced, some not, and methods varied too.
On the plus side, footprints were much easier to find than tigers themselves, a huge area of tiger territory was covered, and many local people were engaged in this conservation effort, so hearts and minds (and pockets) were won for the tiger.
About ten years ago there was a radical shift. Influential scientists quite rightly decided that the system should be more scientific, more objective, better controlled. They suggested that camera-traps, where tigers were photographed walking along trails, would provide a much more reliable way to count them.
Thousands of foresters and naturalists (along with the pugmarks they counted) were effectively disenfranchised from the business of monitoring.
Imagine the shock waves then, when 5 years later, census figures using camera-traps appeared to show that tiger populations had shrunk by around 50%. Many experts agreed that this was at least in part due to the change in monitoring technique, but nobody could be sure. That’s the problem in relying heavily on one technique.
Here we are, yet another 5 years on. Has camera-trapping lived up to it’s initial promise for counting India’s tigers effectively? Yes and No. Camera-traps, when used correctly, by experts, in relatively high-density tiger areas, can provide very reliable data. But they’re expensive, and need to be set up and maintained very precisely – the expertise to do this isn’t always available when it’s needed.
Some tigers appear to have an opinion on this issue too – they’ve learned where cameras are hidden and changed their paths to avoid them.
What other options are open to us? Biologists have monitored individuals in small populations with radio or satellite collars, but our research and that of others shows invasive techniques have long-term negative effects, and they’re also costly.
Paradoxically one approach might be right under our noses….footprints! But wait, isn’t that just going back to the old ‘pugmark’ approach? Absolutely not! At WildTrack (wildtrack.org) we have developed a rigorous, scientific footprint identification technique (FIT) that can identify individual tigers, and also determine their sex with > 90% accuracy. It can even identify which foot is represented by the print.
FIT works like this. A person with a digital camera, or smartphone, takes images of footprints along a trail. They use a ruler to provide a scale, and a GPS to get a geotag for the image. These trackers need a good eye too, but that’s a natural for foresters and anyone who has grown up in tiger range areas.
After the images are captured they are processed in JMP data visualisation software from the SAS Institute (jmp.com) and a new statistical model developed by WildTrack will tell if the track belongs to a known animal, or a new animal. Tigers can then be mapped, strategies implemented and anti-poaching patrols deployed.
Furthermore, FIT could fill in the gaps where camera-traps fear to tread. It’s cheap, accurate, and very engaging for local communities. If groups of local people in low-density tiger range areas carried simple digital cameras or smartphones with them, they could snap a picture every time they came across footprints, and be paid for those leading to an identification. What a fantastic citizen-science resource! Local anti-poaching patrols could also collect images while on patrol. FIT would then contribute to tiger conservation at three levels – direct monitoring, community involvement and law-enforcement.
We’re currently working with renowned conservationist Stuart Pimm at Duke University and with JMP data visualisation software to help introduce FIT for monitoring endangered species.
Of course no single technique can provide perfect monitoring in all areas. Different approaches should work together, preferably in a non-invasive ‘toolbox’ to provide strengths in different areas. For example, FIT can work very effectively with camera-traps.
Meanwhile, our colleagues in China are already using FIT to monitor the Amur tiger – in the most challenging snow conditions! Take a look at our video :
Which path will the tiger finally take? India has invested hugely in tiger conservation and has no shortage of expertise. Money is no longer the main limiting factor. Isn’t it time to step boldly into the tigers‘ footprints?