Where The Elephants Roamed

By S. de Silva

This slider shows areas containing habitat for elephants (yellow) in the year 1700 vs. 2000.

Where Did The Elephant Habitats Go?

The bulk of conservation efforts center around protected areas as the primary means of safeguarding wildlife and wild places. This is in response to recognizing that human activities are altering the face of the earth at an alarming pace, leading to the loss and fragmentation of habitat for numerous species. Often, our attention is on proximate threats and especially on particular biomes, such as forests. But over what sort of timescale and what sorts of ecosystems have these changes actually been taking place? In order to protect the biodiversity we have today, we have to really understand the processes that maintain them and how we got to where we are.

Asian elephants provide a good perspective on the problem. I’d heard some conservation organizations state that elephants had lost as much as 90% of their historic range, but I couldn’t find a single scientific reference that showed this. Was it true? When I first saw their distribution on a map, I got curious – though classified as a single species, Asian elephant populations occupy many different types of landscapes ranging from grasslands to rainforests that are now cut-off from another. At one time, these disparate populations must have been joined together and if we could reconstruct what happened to these habitats over time, we would have part of the answer. So, I enlisted a team of collaborators to find out exactly when and where elephant habitats started disappearing. In a way, you can think of elephants as being ambassadors for these ecosystems (a reason they’re referred to as “flagship” species). But we couldn’t know exactly where elephants or their habitats had been simply by looking at maps or records directly, so we needed a different trick.

Asian elephants live in 13 different countries. The supposed post-glacial range of elephants, shown in orange, covers a lot of mountain and desert where they probably never lived. The red dots are locations we used to train the model based on where we know elephants are found today, with purple areas showing this known range. Image from Figure 1 of de Silva et al. 2023.

Modelling The Past

We used an approach called “ecological niche modelling” (aka “species distribution modelling”). Basically, we take locations where we know there are wild elephant populations today and relate them to ecological characteristics of those locations. The important thing is that the locations should represent places where elephants could conceivably thrive, not those where they’re likely to conflict with people (such as agricultural plantations). As the first step, we therefore compiled locations that represented as many of the different ecosystem types from as many countries as we could verify that wild elephants actually occurred, from locations that still contained good habitat.

The second step was to find a dataset that provided the ecological features of these locations. We could think of a lot of features that might matter, but since I was interested in the history of the landscapes, the tricky part was that these same variables had to be available going back in time. Lucky for me, a group from the University of Maryland had just released a new version of something they called the Landuse Harmonization Dataset (LUH2) which did just that. It modelled different land-use and land-cover categories (including not only forests and pastures but also primary non-forested lands, various types of agriculture, and wood harvesting rates) all the way back to the year 850 CE!

With these two types of data, we could construct our model. First we used a machine-learning algorithm called MAXENT (which stands for Maximum Entropy) to relate the locations of elephants to 20 of the LUH2 variables for the year 2000, which we used as our base year. We then took this trained model and re-used it again and again to generate maps for prior years back to 850. We also made maps up to the year 2015, where the LUH2 historical dataset ends. The resulting maps give us a sort of index of “suitability” with 0 being areas that elephants might be less likely to occur and 1 being areas they’d love. In order to calculate how much habitat was available for elephants and how it fragmented over time, we had to convert this continuous scale into a simpler binary classification of unsuitable vs. suitable, which we did based on a statistical cut-off threshold.

Mapping Changes Over the Centuries

Since our dataset didn’t go all the way back to the end of the Pleistocene, I can’t really tell you if that map up there is justifiable. But we did see that there was plenty of habitat for Asian elephants all throughout the continent way back in 850, and it pretty much stayed that way until the 17th or 18th centuries. But then the amount of available habitat took a nosedive, and by 2015 there was a loss of 64%. The video above, from our paper, shows this progression with yellow being “suitable” habitat.

Back in the 1700s, if you were an elephant, you could have more or less wandered around over 40% of the continental habitat without interruption – this doesn’t mean it was all a single type of habitat, because remember, elephants are generalists; these may have been forests, grasslands, or lots of things in-between, but they had plenty of good vegetation. This included landscapes within 100 kilometers of where elephants live today. But by 2015 the largest patch of habitat constituted barely 7% of the total available and less than half of their current range was classified as being suitable.

Just think about that for a minute.

This means that perhaps half of the elephant populations are living in suboptimal habitat – areas that aren’t really so good for them. As we discussed in an earlier post about another study, elephant populations can hang on for quite a while even when things are not so great, because they are long-lived (unless hunted) and breed slowly, so it might be difficult to notice when there is a slow decline. But poor habitat means they will breed even more poorly and calves may not survive, eventually causing the population to go extinct – unless they move to better habitat. This can create its own challenges, as elephants might pass through places that are unused to having elephants, with people who don’t know how to react.

Going country by country, the largest losses happened in India (which lost >80%) and China (>90%). These two countries had the largest land areas and hence the largest amounts of absolute habitat to begin with. Today, India still has the largest share of the elephant population, but China has a very small share. In these and other countries, the loss of habitat not just in the short term but these very long timescales may underlie so-called “human-elephant conflict” as elephants and people adjust to the new normal: lots of elephants trying to make a living on inadequate habitat, and lots of agrarian communities trying to make a living next to elephants.

Why did this happen and what should we do? Read on to find out.

A paddy field by day contains lots of human activity but by night the elephants are out and about.

Two Revolutions

The changes in Asia reflect events that swept the globe during that time across all continents. Colonialism, which was already happening, had already made inroads into the natural resources of many countries. But the industrial revolution really allowed these resources to get exploited at a pace and scale that was unprecedented. These two together dramatically changed the way lands were managed – taking them out of the hands of local communities and placing them in the hands of governing bodies or enterprises that were pre-cursors to modern-day multi-nationals. The conversion of these elephant ecosystems, as I call them, reflects this massive geopolitical and economic shift, which was most pronounced in South Asia as well as the far East.

But looking carefully at the timing, another observation really stands out. Something else is going on in Southeast Asia. Notably, the biggest and brightest patch of habitat, which was located in what is now central Thailand, was around much longer. It doesn’t disappear until the middle of the last century, right around the time of a different kind of revolution – the so-called Green Revolution that brought about industrial agriculture. Large-scale plantations such as tea, rubber, coffee, timber and the like had already been put in place during colonial times, but now there were fresh land conversions for crops such as rice, fruit and vegetables. Of course, there are many other human activities that have ramped up too – mining, road building, urbanization. But all of this really represents this pair of revolutions that continue to play out today as our globalized economy gobbles up resources at an alarming pace.

What Next?

So what can we do? Should we redouble our efforts to increase the amount of land that is strictly protected from human activity?

Well, not exactly.

As we discussed in a post about our last study, elephants can’t be contained in protected areas because usually these areas are just too small. There’s no way any park or sanctuary in Asia can be large enough to be fully sufficient without also displacing a lot of people, which has been the way many protected areas were created in the first place.

But a more profound observation, based on mounting evidence from other studies, is that people might have been an important component of these very ecosystems. The vast majority of ecosystems – be they grasslands, forests, or something in-between – have been managed by people for hundreds if not thousands of years. Humans (yes humans!) may be vital to maintaining the kinds of habitats that species like elephants and other wildlife actually thrive in. Practices such as setting managed fires, seasonal and shifting cultivation, creation of water sources, and lots of other things that people historically did in these landscapes long before they were colonized and claimed for other purposes.

This was a difficult thing for me to wrap my head around, because as a biologist in many ways I had been trained to think about how “wild” nature operates independently of us. Non-interference with what is happening is fundamental to the study of animal behavior, for instance. It wasn’t just my scientific training, I later recognized, but it’s the separation of humans from nature that arises from a very particular social-cultural worldview that now predominates our society. This view has largely buried our actual history – one in which we were respectful stewards of the land.

As a species, we most certainly do not practice non-interference: when we see an elephant (or any other animal) in trouble, some of us go out of our way to help it. This is something I love about our species, and it gives me hope.

The world today is a very different place. As we head towards a planet of 8 billion people with climate change looming, is it even possible to share space with wildlife like we did before? Elephants force us to reckon with our place in nature if we are to continue to live together.

Read the paper: https://www.nature.com/articles/s41598-023-30650-8 

Related Posts:

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s