Interview: Where does biodiversity come from?

A new study by researchers from McGill University and IIASA provides insight into how environments promote biodiversity. McGill University evolutionary biologist Ben Haller, who led the study, started the work as part of IIASA’s Young Scientists Summer Program in 2010. In this interview he talks about his new study and his continued collaboration with IIASA.

Ben Haller

Ben Haller

Nexus: What was the broad question you were trying to answer in this work?
Ben Haller: We are interested in the origins of biodiversity.  The world contains quite a large number of different species.  There are about 400,000 different known species of beetles alone.  And beyond beetles, of course, the world is full of different kinds of plants and animals and fungi and bacteria and so forth.

Perhaps the central question of evolutionary biology is what led to all this biodiversity.  Although Darwin supplied the biggest piece of the puzzle with his theory of evolution by natural selection, there is still much left that we don’t understand about the origin of species.

More specifically, our question was how variation in the environment might generate biodiversity.  Landscapes vary in temperature, in elevation, in rainfall, and in all sorts of other environmental variables.  Environments vary biotically, too; some forests are dominated by conifers, while other forests are dominated by deciduous trees, for example, and that creates very different environments for all of the birds and mammals and insects and so forth that live in those forests.  We know that this environmental variation promotes biodiversity; organisms in different environments will evolve to “fit” the environment they are in, and this can lead to the development of new species.

Previous theoretical models have simulated this process.  Simulated organisms would be placed into a simulated environment that had some sort of variation in it, and the simulated organisms would evolve divergently and become different species.  But these models only looked at extremely simple, artificial environments.  We wanted to look at more realistic environments, with more random variation in environmental conditions, to see whether a sort of patchy quilt of different environments across a landscape had a different effect on biodiversity than the simple forms of variation used in earlier models.

In your paper you found something called a “refugium effect.” What do you mean by that?
We found that complex environmental variation promotes the creation of new biodiversity, through a phenomenon that we have called the “refugium effect”. A refugium is a place of refuge.

Imagine the sort of simple environment that has been studied before, with a smooth, continuous gradient from environment type A to environment type B.  Maybe a gradient in elevation, for example; our model does not make any assumption about what aspect of the environment is varying, but we can think of it as elevation.  If a species is adapted to a particular elevation in this environment, it can be hard for that species to move around in the landscape.  It lives at its optimal elevation, and if it ventures outside of that zone, it encounters conditions that it is not well-adapted to, and it has trouble establishing new populations and colonizing the landscape.  As a result, that species might never diversify into other species adapted to other elevations.

Now imagine a landscape with a more complex topographic profile, with high points and low points scattered somewhat randomly across the landscape.  That’s the sort of environmental variation that we modeled in this study.  A species adapted to a particular elevation can find spots that have that elevation in many places in the landscape.  These are refugia.  The species can immediately disperse into those refugia and establish new colonies.  Those colonies will be largely isolated from each other, however, due to their geographic separation, so they can follow independent evolutionary trajectories.  And importantly, the refuge provided by a refugium is limited, because the refugium is surrounded by different habitat; land at a different elevation, here.  A population might scrape out a living in a small refugium, but many of its offspring are going to end up in that surrounding environment, and so there will be a strong pressure to become better-adapted to that surrounding environment.  While the refugium provides an initial foothold, in other words, the pressure is ultimately to leave the refugium and become adapted to the surrounding conditions.  And when that happens, a new ecotype, or perhaps a new species, has been created.

You created a model of an environment in order to investigate how complexity can affect biodiversity. What kind of environment does your model represent and what kinds of organisms live there?
Well, it’s really fairly abstract.  While we’re trying to bring our model closer to the real world, I don’t want to give the impression that we are simulating anything like real environments and real organisms.  In our model, the environment varies in just a single characteristic, and the pattern of variation, while relatively complex, is still much simpler than the variation in the real world.  Similarly, the organisms in our model vary in just a single trait, corresponding to the single axis of variation in the environment.

At present our model is also of asexual organisms that disperse at birth; this is similar to many plants, for example, but is less applicable to most animals.  The model is never going to be a sort of hyper-realistic model of real-world environments and organisms; that would not actually be desirable, as it would then be just as impossible to analyze and understand as the real world is!

The power of theoretical work is that you can distill a real-world question down into a very simple, abstract form and get a very simple, abstract answer, uncluttered by all of the complications that exist in the real world.  At the same time, though, the specific things I’ve just mentioned are things we’d like to work on next: looking at landscapes and organisms that vary in more ways than one, and looking at sexual organisms with perhaps more animal-like movement behaviors.

Besides the refugium effect, were there other new findings from the study?
We also showed that environmental variation is a bit like the story of Goldilocks and the Three Bears; for promoting biodiversity, you can have too little variation (which doesn’t promote diversification), or too much variation (which makes it too difficult to diversify), or the variation can be “just right”.  That had been shown before for very simple environments, but we generalized that result to a broader class of landscapes and variation, which is an important result.

And finally, we showed that the spatial scale of environmental variation is important to diversification.  Imagine a landscape with very fine-grained variation, and a species with a very long dispersal distance.  The species will not be able to adapt to fit different local conditions, because it doesn’t “fit into” any one local patch in the environment.  Now imagine a landscape with broad-scale variation, continent-scale variation perhaps, and a species with a much shorter dispersal distance.  Now whole populations of the species will “fit into” different local patches with different local conditions, and so those populations will diverge and become different ecotypes or different species.

Three views of one run of the model: Left: a complex landscape, generated by a method described in the paper. Center: A population of organisms that have adapted to local conditions in the environment. Right: The evolutionary history of the population depicted in the center panel, with time proceeding from left to right and phenotype shown on the y-axis.

Three views of one run of the model: Left: a complex landscape, generated by a method described in the paper. Center: A population of organisms that have adapted to local conditions in the environment. Right: The evolutionary history of the population depicted in the center panel, with time proceeding from left to right and phenotype shown on the y-axis.

How does this research apply to the real world?
I think it’s important in two ways. First it’s very important that we work to conserve the biodiversity that exists today, and it’s very hard to conserve something that you don’t even understand.  Humans are presently causing a vast wave of extinction around the planet, because of activities from deforestation and climate change to overfishing and industrial agriculture.  We would like to find ways to moderate those negative effects on the biosphere; but to do that, we need to really understand how the biosphere works and how it responds to our actions.  Our study is a very small piece in that puzzle.

The other way is more speculative but perhaps more direct.  Our model is a model of how environmental variation promotes biodiversity.  It is also possible, however, that the factors that we have found are important for maintaining biodiversity, as well as for producing it.  If we homogenize the environment, as humans often do – think of cutting down rainforest to make soybean plantations, or draining wetlands to make cities, for example – we might be removing the environmental variation that maintains existing biodiversity.  It’s not a new idea from an ecological perspective, but I think it’s a fairly new idea from the evolutionary perspective.  Even if a species can persist in an ecological sense – even if it has sufficient habitat left to survive – it might not persist for evolutionary reasons.

We tried hard to connect our research to the real world, to make it concrete and testable.  In particular, although we used simulated landscapes in our model, we measured their characteristics using metrics that could equally well be applied to real-world landscapes.  So when our results indicate that refugia have such-and-such a quantitative effect in promoting biodiversity in a landscape, that result is expressed in a way that field biologists could go out and test, or perhaps laboratory biologists could test using an experimental system like bacteria or yeast.  We’d very much like to see our theoretical results tested empirically; that back-and-forth between theory and experiment is at the heart of science.

You started this work as part of IIASA’s Young Scientists Summer Program. How has that experience influenced your research?
For one thing, it brought me into contact with my collaborators on this research, Ulf Dieckmann and Rupert Mazzucco.  I have learned an enormous amount from them, and their ideas have profoundly influenced my scientific trajectory.  This was particularly important for me because my PhD supervisor is primarily a field biologist, not a theorist.  Working with Dr. Dieckmann and Dr. Mazzucco in the YSSP gave me experience in working in a more theory-focused environment, and that was very important for my development during my PhD.

Another thing is that the YSSP has opened up international collaboration for me in a way that would not otherwise have been likely to happen.  I’m American, and it can be hard to get funding to go to conferences and initiate collaborations with scientists “across the pond”.  The YSSP, and the funding I received for it from the National Academy of Sciences, let me break through that barrier.  Now, in addition to ongoing work with Dr. Dieckmann and Dr. Mazzucco, I am also involved in research with a group in Zurich, and I have made connections with scientists in Sweden, Japan, Finland, France, South Africa, Germany, Hungary, Greece, the Netherlands, and on and on.  The YSSP is wonderfully international, and so it is a great way to break into the international community of science.  That’s an immensely valuable thing, since the cross-fertilization of ideas from different perspectives drive our thinking forward.

It has also advanced my career in very concrete ways.  As a result of my YSSP project, I ended up with a publication in a major journal, a chapter in my PhD thesis, and a possible place for a postdoc down the road, and famous scientists from whom I can get recommendations when I start looking for a tenure-track job.

But most importantly, I would say that participating in the YSSP made me think about my research in a much broader context than I otherwise would have.  In the YSSP, I was surrounded by fellow graduate students who were not evolutionary biologists, but rather were researching issues of land use and politics and global hunger and sustainability and climate change and deforestation and nuclear disarmament and all of the other things that IIASA works on.  That experience really encouraged me to think about how my work in evolutionary biology connects to all of those other things.  It promoted a big-picture perspective, an interdisciplinary outlook that has stayed with me.

Read more about this work on the IIASA Web site and on the Eco-Evo, Evo-Eco Blog.

Note: This article gives the views of the interviewee, and not the position of the Nexus blog, nor of the International Institute for Applied Systems Analysis.