Above: Mount Kenya’s nipple, viewed from Ol’Pejeta Conservancy during an early morning trapping routine. Photo credit: Anne-Marie Hodge 2014.

In 2014, I boarded a plane to Nairobi to volunteer as a field assistant for a Ph.D. student (Anne-Marie Hodge – check out her website here). I had just finished my B.Sc. at the time, and the trip was a sort of “volun-cation” before starting graduate school for myself. On the menu: trapping little predators to sample their tufts of fur. Here’s why.

A mesocarnivore is a mid-sized predator, such as a white-tailed mongoose (Ichneumia albicauda) or a black-backed jackal (Canis mesomelas) – two species that we frequently found in the tomahawk traps. Mesocarnivores eat an assortment of rodents, invertebrates and eggs. Since they aren’t exactly at the top of the food chain, their populations are regulated both by food availability (bottom-up control) and predation (top-down control) from larger carnivores. What we don’t know is how these forces interact to result in the observed population distributions of these species. In a world with a rapidly changing landscape, knowing why the distributions are the way they are will help us choose the best conservation approaches in the future.

A black-backed jackal (Canis mesomelas). Photo credit: Anne-Marie Hodge 2014.
A black-backed jackal (Canis mesomelas). Photo credit: Anne-Marie Hodge 2014.

As it turns out, the bottom-up and top-down forces are difficult to quantify directly. For example, to know the level of top-down control on jackals, ideally we would want to find out how many jackals are killed as a result of predation. However, dead animals don’t tend to last long in the savannah, and the evidence of the kill would be quickly consumed. Predation could also be estimated by monitoring the numbers of predators in the area, but it turns out that collaring big cats with radio transmitters is a lot of work. And so we resort to an approach that is so often used in scientific research: using a less ideal, easy measurement as a proxy for the more ideal, but less feasible one. I want to make it clear that this is not an improper practice; it means that we need to consider extra caveats, but using this approach means that we can answer questions which may be otherwise impossible to answer or that may require methods which are outside a typical lab’s operating budget.

Human habitation is a reasonably good proxy for the abundance of large carnivores – where humans live, people tend to control the numbers of top predators because if they don’t, villagers and livestock could be attacked. Where humans don’t live, carnivores are free to roam undisturbed. What’s more, since mesocarnivores are not a threat to people, their populations aren’t necessarily different whether there are people around or not. This creates just the conditions we need for a good experiment: using this proxy, we can easily predict where there will be more or less top-down pressure on an otherwise constant mesocarnivore populations.

The second element – bottom-up control – is a little more complicated. Rainfall is a great proxy for plant biomass, but mesocarnivores eat mostly small animals rather than plants. There is also some competition between the mesocarnivores for similar food sources, making it less straight-forward to predict who has access to more food. Thankfully, we have a technique that can profile an animal’s diet based on something you might not expect: the chemistry of its fur.

Mass spectrometry (MS) is the tool for the job. Here are the fundamentals: MS can detect minute mass differences in compounds, and conveniently, depending on where an organism gets its food, its proteins can be made up of compounds with slightly different masses. This is because of varying carbon and nitrogen isotope distributions in different living things (an isotope is a different version of the same element that has a different atomic mass). For example, the carbon and nitrogen isotopes that are by far most abundant are 12C and 14N, but 14C and 15N are also stable isotopes (i.e. non-radioactive), albeit far less abundant. In other words, stable forms of carbon can have 6 protons and 6 neutrons ( = 12) or 6 protons and 8 neutrons ( = 14). Similarly, nitrogen can have 7 protons and 7 neutrons, or 7 protons and 8 neutrons. The kicker is that different organisms have different relative amounts of these isotopes, and animals that are higher up on the food chain will adopt the cumulative isotope signature of everything they eat. So, since MS can measure the proportions of these isotopes, we can use it to assign a dietary signature to a species from something as small as a fur sample.

Image from http://www.legacyamc.com/

Whiskers, though, are the real gold mine for information. Whiskers grow much more slowly than fur, often remaining on the same animal for years at a time. As it grows, the newest part of the whisker gets the isotope signature that reflects the animal’s diet, but once its diet changes, so does its signature. This means that a few months later, when the whisker grows out and the wet season brings a new buffet of things to eat, the isotope signature at the new base of the whisker will reflect this. The cycle repeats until the whisker can be read like a diet signature timeline from tip (old) to follicle (new).

All this is why I found myself staring down a snarling jackal as I plucked out a few whiskers through the bars of his tomahawk trap. He didn’t realize it, but he was handing over the history of his diet with those small samples. This is an ongoing research project, so we don’t yet know exactly how the bottom-up and top-down forces are interacting, but I’ll be sure to update this post when the research is published. Regardless of what the answer will be, it was a pretty good way to spend a vacation.

Acknowledgement: Anne-Marie Hodge for her input on this article.