G. David Tilman, PhD
G. David Tilman, PhD, is a Professor of Ecology, Evolution,
and Behavior at the University of Minnesota, and is Director
of the Cedar Creek Natural History Area. Tilman studies patterns
in the biological diversity, structure and dynamics of plant
(and sometimes insect) communities, as well as the effects of
biodiversity on the stability and productivity of ecosystems.
Could you define 'keystone species'?
A keystone species is an organism that has an unusually large effect on how an ecosystem operates, given its abundance. So it could be a rare species that has a noticeable effect, or it could be a hugely abundant species that also has a very large effect.
Define ecological niche.
An ecological niche refers to the way that an organism lives; its habitat that it lives in, the way it exploits the habitat. Basically, it's a loosely defined concept, but as organisms have been studied, we've realized that species really do differ in how they live off of their habitat, what they do in the habitat, and these differences are what allow these species to coexist.
Define a functional species group?
A functional species group refers to organisms that are very similar in how they respond to their environment, that are very similar in how they live.
And how about diversity-stability hypothesis?
The diversity stability hypothesis says that ecosystems that contain more species vary less through time in response to various disturbances they might experience. It could be a drought, it could be an outbreak of a disease or an insect, but they vary less. They're more stable through time.
What is the functional significance of a keystone species?
A keystone species in an ecosystem plays a major role in the function of that ecosystem, and there are lots of examples that have been known.
I would mention one that we've studied in Minnesota. I mean it may or may not seem like a keystone species, but deer turn out to be a keystone species. Deer eat a variety of plants, but they preferentially eat legumes, and by eating legumes they keep this kind of plant in low abundance, and that actually keeps the soil from becoming very fertile. When deer are in low densities, legumes become abundant, the soil becomes more fertile, and we've seen almost the doubling of soil fertility in a 20-year period in response to keeping deer out of areas compared to ones where they were present.
So what are the identity factors for such a species?
The most common interpretation of how to measure if something is identified as a keystone species would be to do an experiment in which you would add it or remove it from an ecosystem, and if you add or remove a species and see a very large, normally cascading series of events, then you would say that was a keystone species.
And this did happen, for instance, with sea otters when they were removed by hunting and the whole shift that happened in the kelp and sea urchin ecosystem, the same thing has happened with the removal of a starfish in the inner tidal zone, and removing various animals, either predators or herbivores, in the ecosystems around the world.
Why would one remove keystone species?
We are removing species that we think might be harmful keystone species, and the biggest effect we've seen have come from deer removal and it was a series of cascading effects. Removing deer allowed legumes to increase in abundance, legumes can fix atmospheric nitrogen. That nitrogen makes the soil more fertile. That then changed the interactions among the other plants and animals in the system, so there's a big shift in the structure of these systems in response to the presence or absence of one species.
And that tells you what about the system?
It's a major insight into how ecosystems operate. There are two major philosophies that have been put forward to explain how ecosystems work. One philosophy is called a "bottom up" approach. That the climate in the area, the fertility of the soil, determines the abundance of the plants, and that the plant abundance is determined the abundance of the animals that eat them--the herbivores--and that then influences the predators, which eat those.
What the keystone concept says is that effects also do come from the top down. I mentioned some examples, but other examples you can think about are aquatic ecosystems, lakes. Large predatory fish are keystone species in lakes; they can cause big shifts in the measure of water quality. How much algae grows in water that comes from the bottom is dependent on whether or not there are high nutrients in the lake ecosystem, but also from the top, by whether or not there are predatory fish on the top.
Those fish eat smaller fish, which then eat little microscopic animals, and those eat algae, and you can influence an abundance of all of those by whether or not you have a keystone predatory fish on the top.
We have been studying dynamics of plants and insects in hundreds of plots at Cedar Creek Natural History Area since 1982. And one of our major techniques is just to annually measure and record the abundance of every species in each one of these blocks, and then look at what happens through time. And it was actually doing this that we came upon our discovery of the relation between diversity and stability.
We had a series of plots that differed in the number of species that were in them, and the plots experienced the normal kind of warm summers, cool summers, wet or dry summers, and then we had a major drought that hit. It was the third worst drought in the hundred and fifty years of recorded meteorology from Minnesota, a very severe drought, and during that drought, we saw that some of these communities, which happen to contain a large number of plant species were harmed by it. They had their biomass fall to about half of what it had been beforehand, but other ones had their biomass fall down to a 10th or a 12th of what it had been. There is a six- or so fold difference in how they were impacted by drought, and the best variable at explaining it--and we looked at 20 or so different things that we measured in these plots--the single best variable at explaining it was the number of species. The more species there were in these plots, the more resistant they were to this major disturbance; the more stable they were.
What was the size of the plots?
We did this work in a total of 207 plots that were located in four different fields, with about 50 plots per field. The plots themselves are 4x4 meters in size. To sample the plots, every year we clip a strip of vegetation that is 10 centimeters wide by 3 meters wide, so it's along, narrow strip of vegetation.
And the advantage of sampling in that way is that a long narrow strip will average over the patchiness that you see within a plot. So 4x4 meters, that's sort of maybe the size of a living room. That kind of vegetation in nature can be quite patchy, what plants occur where, but by taking a long narrow strip we average across that, and that allows us to get a quantitative measure of abundance of the species with a minimal amount of work. Minimal is maybe not the right word. It takes about 12 to 15 people helping you do this, for about two weeks of our time in a summer, to clip and then sort all this vegetation into species, pick them up, look at them, identify them--and there's sometimes a lot of argument as to what some little plant is--identify them by species, and you have to dry them to get rid of the water that's in them, and weigh them, and then you have to enter the data in the computers, and everything else, and analyze and look at what's happening, and do this year after year after year in several hundred plots. And if you're lucky, something like a drought or an insect outbreak comes, and you can try to understand how the system works, because now you have data on all these species. You can see now what happens when one is hurt by something. What we saw happening was, when one was harmed by a drought, other ones increased. They sort of increased using the resources left unconsumed by the species that was harmed by the drought.
What's the role of your statistical measures?
In the work we're doing, we are using a variety of statistical techniques, but we're not actually trying to observe the abundance of a given species and find out whether a rare species is threatened or endangered.
To do that would require sampling in a very different scale than what we'd been doing. We've been trying to understand how ecosystems operate, and our ecosystems contain some relatively rare species that we sample enough that we can record their abundances. Other ones that are really rare, we don't find them enough in our samples ever to know what's going on. You take a totally different sampling scheme to deal with them.
But in our analyses, the kinds of data we analyze now--there weren't computers capable of analyzing the data that we play with now 20 years ago when we started, so it's a very large dataset, there'd be hundreds and hundreds of species in hundreds of plots sampled every year for the last 21 years, and with that kind of dataset it was too large, and the kinds of analyses we needed to do were too complex ever do be done until the modern era, if you will, with computers and software that exist now.
Is there a very big math component?
I've always loved mathematics and I just played with numbers as a kid, and loved mathematics, took a lot of math in college. And mathematics is becoming a very important part of the discipline of ecology. E ecology is a science that has a strong experimental component, a strong observational component, but also an increasingly strong component for mathematical theory, trying to integrate the ideas that we observe in nature and try to understand their implications.
A large part of what I do is to develop mathematical theories, theories that may help explain how ecosystems function, how their functioning depends upon the number of species that are in them, what the species are that are in an ecosystem and the various disturbances these ecosystems experience.
So what do your post-docs do in winter?
Most of us-faculty, graduate students, post-docs-spend a lot of our time in the winter months, when we're not out doing field work, analyzing data. It's a very important aspect of our job. We've gathered an immense amount of data every summer. We have to make sure these data are verified, and put into databases in appropriate ways so we don't have mistakes in them, and then we have to try to understand what these data mean and it's a process that goes on.
And I always tell my students that, if you don't know what happened last year, there's no reason to go out and do something again the next year, so it really is a very critical part of the process. You go out into the field with ideas, with hypotheses, one year you gather the data that year and then the fun part, frankly, is coming back and seeing which hypotheses worked and which ones you have to reject. And frankly, sometimes it's a lot more fun to reject an idea than just find support for it, because that gives you an insight.
Do you see biodiversity as a natural insurance policy?
In our studies of biological diversity, one thing that we've learned is that, because species are not identical to each other, that one species can in some manner compensate for what another species is not doing. Probably the single variable about an ecosystem that an ecologist would say is important is the productivity of that ecosystem, how much living biomass is produced by the plants in that ecosystem.
Well, plants differ from each other, and some plants in our systems have shallow roots, and they can grow well when the soil is moist, but when the soil dries out they basically dry out and wait there in dormancy. Other ones have deeper roots, they can grow when it's dry. Some of them have the capability of growing very well when it's cool; other ones grow best when it is warm.
And these differences among species allow them to compensate for each other, the one takes the place of the other. If you have this kind of diversity in an ecosystem, it means that when that ecosystem is disturbed, this diversity is almost an insurance policy, because when one does poorly-the nitrogen or the water that it doesn't take up-something else can take it up and grow with it. And that kind of compensation basically ensures that the function of that system will not be harmed as much as it would have been from a disturbance if there had been fewer species in the system.
How did the drought provide you with an example of this compensation?
In 1988, we had a major drought, and the drought caused different losses in productivity depending upon the number of species. One of the things we saw, when we looked at individual species and how they responded, was that there were several species of prairie grasses that were more abundant in the drought, and a year after the drought, than they ever had been before and that they ever have been since. These are the species that were very drought resistant and were able to not only continue growing as they had before the drought, but during the drought they were able to take up the nutrients the plants that were harmed by the drought couldn't take up and these drought-resistant plants used them and grew and by doing that, compensated for much of the effect of the drought on the ecosystem.
Could you define biomass?
Biomass refers to the weight of all of the living organisms in an ecosystem. And normally we will sort it by major types, major functions, such as biomass of plants or biomass of herbivores or biomass of carnivores. And when I'm saying biomass, I mainly mean biomass of plants, which is what we focus most of our work on.
How is the biomass weighed and measured?
Normally, you measure the dry weight of organisms, and so you cut plants in the field and put them in a big drying oven, dry them out, and weigh them once they're dry, because there's so much variation in the moisture content depending upon the species and so on, and if you dry them out like that, the biomass is almost equivalent to measuring something like the caloric content, the energy content of the food.
And the greater biomass results in what?
Greater biomass normally indicates greater productivity. In a grassland that's true. It's not true in all ecosystems. The relationships are complex, and it doesn't need to indicate greater stability. Stability just refers to the year-to-year variation in that biomass.
How would you describe the portfolio effect?
The portfolio effect is a phrase that we use to describe why it is that more diverse systems are more stable, and there are a lot of analogies between the human economy and how ecosystems work. In the economy, there are businesses competing with each other, and in nature there are species competing with each other. There are consumers, if you will, in our economy and in nature there are consumers of various resources, and these consumer resource interactions are a major way that both natural economies and the human economy are structured.
And in looking at our experiments and seeing the effects of diversity on stability, that more diverse systems were more stable, we were wondering why this happened, because at that time the mathematical theory did not predict that that should happen.
And we basically borrowed some ideas from the mathematics of economics and many people know that if you invest in a more broadly diversified portfolio of stock that it varies less from year to year in its value than if you just choose one or two kinds of stock. And that comes about because you're averaging things that tend to vary somewhat at random, and when you average more and more things together, it tends to vary less.
And so one thing that happens, because nature always has good and bad years for different species, is that if you have many species growing together and you average their responses, the average of that more diverse portfolio in nature is more stable. It doesn't vary as much. That's one reason why diversity can affect stability.
The other reason has to do with compensatory interactions: that when one species goes down, something else is freed of competition-it can go up-and that's another mechanism in addition to the portfolio effect that causes natural systems to be more stable when they're more diverse.
What are some examples of the portfolio effect?
The idea of the portfolio effect really came from a controversy that arose after we published our first paper about the drought, and the result that greater diversity was leading to more stability, because at that time there weren't theoretical explanations in ecology that predicted that more diversity should lead to greater stability. In fact, the predictions really went the other direction.
So we had our result, and that result we had analyzed in a large number of ways. It was a robust result. We couldn't find any way that we could reject the hypothesis that more diverse systems in nature in our ecosystem were more stable. So we published, it but we really didn't have anything other than a few hand-waving explanations for why it would be so.
And new results in science lead to discussion, and to controversy, and this led to some controversy, and some very insightful scientists started writing papers, and one paper they came out with on this suggested an alternative explanation: that this might be what they called a statistical averaging effect; that you basically you're averaging variation in abundance of the species, and that it might not have to do with biological interactions at all, but just that you had these randomly varying species and they're going up and down at random, and the more they go up and down, the less their average changes. It was a very neat idea, a very important idea.
The paper led us to explore models we'd never played with before. We played with them some more, and came up in those models with a combination of that idea which, with the effect of these complementary interactions among species where the species aren't just going up and down at random. They're going up and down, but in fact, something is going down because it's harmed by some condition, and when it goes down, it frees resources, which lets something else eat them, consume them, and go up.
And so what we said, well, it's not just random noise. In fact they're going up and down in a very specific way. They are competing with each other, and these interactions are causing them to go up and down, and that added effect is what we call this compensation, or complementarity effect, that has to do with different niches that the species have, and how those allow one to compensate for another one when it's harmed.
And so it's really one of these situations in science--which science is full of--where a new idea comes, it leads to controversy, the controversy leads to new questions, new insights and in this case they were explored by other researchers and by us, and we came up with these two new ideas that could explain why greater diversity will lead to greater stability.
How can you differentiate between the effects of diversity on stability vs. the effects of diversity on productivity?
To talk about diversity's effects on ecosystems, we have to go back about ten or so years. And ten or so years ago there were a few people-Edward Wilson, Paul Ehrlich, and a few others-who were suggesting that diversity would affect how ecosystems operated, but there were no experimental results, and there was no underlying theory except sort of the gut instinct of some very smart people that said that diversity should matter.
And so at about the same time, within a few months of each other, two papers appeared in the journal Nature. One was a paper we published on diversity and stability [Tilman et al. (1996) Productivity and sustainability influenced by biodiversity in grassland ecosystems. Nature 379, 718-720.] ; the other is a paper that John Lawton and others from England published on diversity and productivity. And these two papers were the first two experimental demonstrations that diversity might be having surprisingly strong significant effects on the functioning of ecosystem.
What has turned out to be the usefulness of this hypothesis?
The experiments in theory that have developed in the last ten years suggest that there are many strong links between what had been very disparate approaches and ideas in ecology-ideas that were more sort of evolutionary ecological questions about how species evolved and differentiated, ideas about the interactions among the few species, and ideas about how whole ecosystems operate. Those ideas had never really been linked together, and yet there was an increasing awareness that biological diversity mattered, and for biological diversity to matter, there had to be specific kinds of differences in the traits of species. This really led to a series of insights into how, if you will, nature operates.
In particular, for biodiversity to matter, species have to differ in their traits; in other words, one species that does a better job at one thing is worse at something else. If one species were good at everything, according to theoretical experiments that have been done, it should be the only species on earth. There should be one plant that's best at living in deserts, and best at living in wet spots, and in cold temperatures, warm temperatures, on soils with nitrogen or phosphorous and so on. That's not the case. Species have tradeoffs. When they're better at doing one thing, they're worse at something else. And with that realization came the understanding that you cannot explain why the world is so diverse, why there are so many species, unless you look at the tradeoffs species have. That was a big insight to explaining diversity.
But the same concept, the same underlying models of species differentiation that predict that higher diversity should lead to greater productivity, that higher diversity should lead to greater ecosystems stability, that higher diversity should lead to more efficient utilization of the limiting resources in that ecosystem-these things are all linked.
In fact, it is these mechanisms that allow species to coexist that are the very same mechanisms that cause diversity to influence the productivity and stability of an ecosystem.
What sort of feedback have you seen from this work?
It's interesting, doing this work on diversity. When we started it, most people thought that the productivity of an ecosystem influenced its diversity. They thought that the stability of an ecosystem influences diversity. No one thought that it went the other way with diversity influencing productivity, or diversity influencing stability. And as this work has progressed it has become increasingly clear to people in ecology that the link is probably going both ways; for example, take a system that is unproductive, but then has some species invade it. That invasive species will increase the productivity. But at some point, it's likely to happen that, as more species come into the ecosystem and it becomes more and more productive, the productivity of the system will prevent the further invasion by more species. But the mathematics of this has not been worked out well. There haven't been good experimental tests, but the ideas are there which suggests that there's this interplay, in a longer term perspective, between how diverse a system is and how stable and how productive it is, and that the cause and effect goes in both directions.
Can you define the term 'biocomplexity'?
Biocomplexity is a phrase that's just sort of come into the vocabulary of scientists in the last, oh, five, maybe ten years at the most, and it recognizes something that maybe was obvious long ago, but wasn't studied or appreciated, and that is that nature is phenomenally complex.
If you look at a single acre of prairie, you'll find about 200 species of flowering and other vascular plants in it. If you look in the soil, you'll find at least 5,000 different species of bacteria, probably a thousand or so species of fungi. We've looked at insects in an acre of prairie and seen probably 500 to 700 different species of insects, and then there's probably another 40 or so species of vertebrates that you can find there-birds, mammals of various sorts, lizards, and so on. These are incredibly complex ecosystems.
With that number of species all interacting with each other, it's very hard to know how these systems operate and what the cause and effect is, and what the effect of the lost or gain of one or few species might be.
We talked about keystone species earlier. It could be that a viral disease, an incredibly rare organism, you could probably not measure its biomass with all the viruses that cause disease in a given ecosystem. It would be trivial. You couldn't weigh them. It could be that that, more than anything else, could determine how that ecosystem functions.
I think that's what's happening right now as the West Nile virus is sweeping across North America. It's causing massive death in some bird species and not in others. Well, some bird species eat some insects, and other bird species specialize in feeding on other insects. If the bird species die that feed on one kind of insect, then that insect will become much more abundant. Whatever that insect feeds on will become much less abundant. We have no ideas for that. The addition of one virus in an ecosystem, how is it going to affect the functioning of North American ecosystems?
How important is the control of such microbes?
We focus a lot of our effort on plants, and, of all organisms, for some reason people study plants more than other organisms. I tend to like processes, but still, when studying processes I'm more likely to look at plants. I first started studying algae in the Great Lakes and then I switched to prairie plants because somehow plants attracted me.
But in the soil, there are thousands and thousands of different kinds of bacteria and fungi interacting with each other, and they're interacting in the same kind of food webs that happen above ground. They are consuming various resources, they're being consumed by other species, they have their predators, their parasites, their diseases going on in the soil, and that soil ecosystem is just as important in determining the functioning of the whole prairie as are the plants aboveground. They really are linked. The soil organisms, as far as the plants are concerned, are important for providing nutrients to the plant, but the soil organisms don't do this deliberately. They're living their own lives and interacting with each other.
The links between microorganisms and their biology, and plants, and vertebrates, and so on, insects, are not at all well studied, but it is clear that there is an incredible amount of very important ecology going on among these organisms, and that they merit the same kind of in-depth study that we've been giving to the more visible, larger organisms.
Let's shift back to your experiments. How do you physically control the experimental plots?
The biodiversity experiment, when we set it up in 1993, was an unusually large and bold, ambitious, maybe foolish experiment for ecology, because we took 20 acres of land, plowed it to get rid of the vegetation that was there, and then divided it up into 342 plots, and then sprinkled seed of the appropriate species onto each plot.
Well, as everybody with a garden knows, not only did the species we planted grow there, but so did many, many different species of weedy plants. They're sitting on the seed bank just like they do in a garden or in your backyard. And each plot, when we started first started, was 13x13 meters--was about 40x40 feet. Now they're down to 9x9 meters, because we couldn't take care of them when they were that big. We had to make them smaller.
So we planted the species, and the first year what happened was they were just basically waist-high weeds. Annual plants came up from the soil. Most of the annuals were just the same kind of weed you find in a garden plot, and living down in the shade of these annuals were little tiny plants that grow tall, the prairie plants. Prairie plants grow very slowly. Prairie plants have a lot of their biomass in the roots, because the prairie is a system that is water- and nitrogen-limited, and they make a lot of roots to be able to get at those resources. But because they put so much energy into their roots, they don't grow very quickly, and so the first year they were tiny, and we just let them grow and let the weeds be there and so on.
The next year, the prairie plants were a little taller, but whatever you see aboveground, there's about three or four times more biomass than that belowground, and they were starting to inhibit the weedy plants, and that year we went through and tried to pull out at least the biggest, most obvious weeds that were in some plots, tried to let the prairie plants grow a bit faster.
But by the third year, the prairie plants had really turned the corner. They were full-sized plants. They had lots of root biomass, and it turns out they are much better competitors for nitrogen and water in these soils than are these weedy plants. So by the third year, most of the weedy plant biomass we had there was squeezed out by the superior competitors, by the plants that we planted in our experiment--but not all of it.
And so to try to have this be an experiment in which we have direct control on which species are in each plot, and how many species are in each plot, we have to weed the plots, and we do that by hand and it takes about 15 people two or three weeks every summer in a couple of one-week bouts.
You walk into a plot, you have in front of you a list of all the species that are supposed to be in that plot. You read the names to each other so you all know what's supposed to be there and then you walk and look at each plant and say, "Are you such and such? Are you one of the plants that are supposed to be here?" And you look at it, and you think about that, and if the answer is no, before you pull it out you say to somebody else, "Am I right? Is this really the wrong species?" And everybody says, "Yeah, pull it out," and you sort of have this group consensus as you walk through a plot and pull out the ones that don't belong.
And typically now these plots are 9x9 meters, so 30x30 feet in size, we will pull out about a half of a grocery bag of weeds from a plot in the summer. It's not that much compared to how much biomass is in these pretty tall plants in these plots. But if we didn't do that, some of these plants might be able to invade and to start influencing how the systems function.
Of course what you have to do in a experiment is control your variables, and nobody else had ever had an experiment in ecology in which they had controlled diversity, controlled the number of species in an experimental manner, had enough replication where they could actually detect whether or not there were effects of diversity on how those ecosystems function.
So we had to, as part of our experimental protocol, we had to keep out other things that weren't part of the experimental plan. Luckily, you know, prairie plants do that pretty well on their own, and if that weren't so, it would have been impossible to weed them thoroughly enough, and there would have been so much disturbance from weeding them that it could have influenced our results. It turns out to be a pretty minor part of what's going on in the plots.
How did you realize that you were able to control the prairie as an experimental environment?
Let me sort of tell you how our experiment came into being. We had this drought. The drought happened in 1988. We saw some responses after the drought that we thought were interesting. About 1991, we started analyzing the data. Our first analysis suggested a really big effect on diversity. None of us believed it. We thought it was totally bogus. We thought it would be explained by something else.
We actually put the data aside for a while, and then I had a visitor come to me on a sabbatical as a visiting professor, John Downing, from Iowa State University, and I would show John some of these results. I'd say, well, here's a really funny result. It says, more diverse systems are more stable, but you know, it's really probably caused by some other variable.
"Well," John said, "you know, we can analyze for that. Why don't we do that?" So we started playing a game where each of us tried to show that diversity didn't matter, and we did this for about three months-did hundreds and hundreds of different analyses, and we could never reject the hypothesis-so at that point in time we suddenly changed our minds, and we thought, "Well, maybe diversity really does matter," and as soon as you look at it that way you start doing something differently. We started saying, "Well, if diversity matters, why might it matter? How might it matter?
That led us to realize that the initial results in the diversity-stability experiment suggested that we had to do a real experiment where we had direct control on which species we were going to plot and how many.
And so in 1993, which is a year before our first paper on this was even published, we started setting up this experiment. So before we'd even gotten a paper accepted and published in a scientific journal, we'd had the insight that diversity would matter, but it was clear to us we had to control diversity experimentally to do it right.
Anyone reading our first paper on this could say, well, sure. You happen to choose plots at different diversity for some reason when you started the experiment and you also treated them in different ways which caused the virtue to change, and the end result of that was that you saw what would look like a diversity stability effect, but maybe is really caused by something else that you didn't happen to measure.
You overcome all of those criticisms if you directly control diversity in a whole series of replicated plots. We have 40 plots that are planted to one species, 40-some plots planted at two, 40 more planted to four, and then 40 more to eight, and 40 more to 16 species, and each plot contains a random draw of species from this whole group of prairie species that we use. And so you're able to look at the average response of all of these randomly chosen monocultures with the randomly chosen two species plots, and the fours, the eights, and the sixteens and you can find out from those analyses if there really is a diversity effect. And their plots aren't just randomly chosen for what's in them. They're randomly spread across this whole 20 acres, so there's not some little corner over here where all the monocultures are planted, which would mean the difference might lie in the soil. They're spread all around this area all randomly.
So you use the rigorous methods of experimental design to eliminate all these possible confounding variables, such that you have an experiment that directly lets you test, "does diversity matter?"
So why had no one had done it before us? The reason was that no one had an insight that diversity really would matter. There's an old idea that had been rejected, and it wasn't really a part of active ecology, so no one had tried it because it just didn't seem interesting.
How were you able to build on this model?
We did start this experiment in 1993 before our results for diversity-stability had been published, but the only reason we were able to start the experiment is that we've been fortunate to be funded as a site for long-term ecological research. This is a grant from the National Science Foundation that gives us the flexibility to have a unique new idea, and go out and set up in the field and try it.
We had people who were working with us in the summer--our field assistants, our grad students and post docs--who all chipped in and got the experiment off the ground and running very, very quickly from when we first had the idea. And we knew when we did this, when we dedicated part of our long-term funding that we'd received to maintaining and sampling this experiment, we knew we could run it for a long enough period of time that we could see whatever the real results of diversity were.
And we were very fortunate that way, that we had that kind of long-term funding and that the National Science Foundation makes that kind of funding available for ecological research.
It is critical that this kind of work be done at other places around the world. It requires that other governments or funding agencies be able to and willing to support other researchers to do this kind of work.
There was a wonderful experiment done in eight different European countries, supported by the European Union. It was just like our experiments-it started about four years later-but they did the work not just at one site like we did. They did it in Greece; they did it in Ireland; in England, northern England; in Spain and France; and I think in some Scandinavian countries. It was across Europe, and by doing it across Europe you could draw much broader inferences, if you can show that there's a result. We showed a result that happened in one field in Minnesota. The work done in Europe showed that the same result, almost an identical result, just an amazingly identical result happened on average across all of these countries in Europe, which says it happened all across the European continent, that whole large mass, and it happened that they used different species of plants--whatever were common at each local site. It happened with all different kinds of plants from the grasslands, from across all those whole eight countries of Europe. One can draw a much broader inference, it was a wonderful experiment.
They were funded for three years, and they were not given renewal for three more years because the science committee for the European Union said it was time to move on and ask some other question.
Unfortunately, at the end of three years, these experiments are barely established. I told you about our experiments. At the end of two years, we just finally had reasonable plants. We got rid of the weeds and our plants were finally big enough to start showing some responses.
Our most highly significant results in terms of how clear the pattern is didn't happen until the 5th or 6th year, and the pattern's becoming clearer and clearer through time as these plants have longer and longer to interact. We're in the 9th year now, and we have our clearest results ever from the 9th year. We see very distinct, clear effects of diversity on the function of these systems. That European experiment, which frankly was in my mind the best one like this ever set up, died before it ever had a chance for us to find out what its long-term implications were.
What are your hopes for your research and the field of ecology?
One of the problems with ecology as a science is that the way it's been funded-it has always been funded very poorly. There is a minor amount of money coming into the science compared to the money that comes into some other disciplines like physics, which need large accelerators and even they're having trouble, or astronomy, which needs a very large radio telescopes or light gathering telescopes or the Hubble telescope and so on. The sum spent on those things are huge compared to the money that's ever spent in ecology.
And what this has meant is that, in ecology, if a result is found in one place in the world, once, we'll believe it probably applies to all places in the world forever, because we don't have the money to go out and do it again. And if you're just building up academic knowledge, and it's only of interest to academics and it has no practical application to humans, then maybe you can accept that. It's sort of a low-budget method, and it's all you can do, so we live with it.
But what is happening in ecology in the last 10 or 20 years is that what we find out is directly relevant for the quality of human life on earth, is directly relevant for public policy. The public policy debates about loss of biological diversity and the Endangered Species Act, the function of ecosystems, how should agricultural lands be managed, how should other lands be managed, wetland laws-all these things depend on scientific knowledge, and the scientific knowledge that we've been able to gather so far in ecology comes from a few very specific examples here and there.
I would assert that it's very important for a society to support experimental and observational long-term research in a variety of countries around the world, because we have to know general principles that we can use to really guide long-term national and international policy. And until we've really done it in many places, we don't know whether the results of Cedar Creek are unique to Cedar Creek. From the three-year experiment in Europe, they don't seem to be. But I wouldn't want to have to make major international decisions based upon only one or two examples of some phenomenon.
What would you most want teachers to understand?
I guess the take-home lesson for me from this-there are two kinds. One is the scientific take home lesson, and that is that we now have discovered the tradeoffs among species, the differences among species, which allow species to coexist; and we've discovered that those same differences that have allowed two or three different million species to evolve and coexist on earth, those same forces also mean that the number of species in an ecosystem influences how it functions, and that systems that are more diverse are more stable, more productive, they're more efficient users of resources, they're more efficient providers to society of services that humans need from natural ecosystems. To me, that's sort of Lesson One.
And Lesson Two, which I also think is very important, is that this kind of knowledge is very poorly understood by the general public, is very poorly understood by students and the public. We don't really realize, because of how we earn our livings, that so much of what humans need to live on this earth comes from other organisms, and it ultimately goes back to the natural and managed ecosystems that are on the earth and how we manage these. Humans now, without trying to, we weren't deliberately trying to manage the whole earth, but we've in fact owned and managed all the lands of the earth and we almost manage all the oceans of the earth by how we harvest fish. And we're doing these in ways right now that, as you learn more about them, are probably not very wise. We're not leaving the ability for future generations to live as well as we're currently living.
And I guess that's Lesson Two that I see from this, that I think is a very important one, that we have to gain enough knowledge about how nature works and how we are interdependent with nature to help us find wiser ways to live on earth.
In all of this, what is the role of climate change?
Global change and global climate change have become of increasing interest to society and to politicians and relevant for global international policy, and there are significant reasons to be concerned about climate change. Everything that we know right now suggests that humans, by releasing carbon dioxide and other greenhouse gases, are changing climate at a rate more rapid than it's ever changed over thousands or tens of thousands of years.
What is not often known is that climate is only one of a large number of ways that humans are now dominating global ecosystems, and there are other effects that humans are causing, which I and many ecologists assert are as strong as climate change, and have at least a serious implications for the future suitability and sustainability of earth.
Humans release as much nitrogen into all land ecosystems as all natural processes combined. That may not sound like it matters. Nitrogen is a good thing. It's a fertilizer, right? But the trouble is that organisms have evolved on Earth over the last three and a half billion years, an Earth where nitrogen is the main limiting factor.
Organisms have a whole array of specializations, which allow them to deal with low levels of nitrogen, and nitrogen varying in space and time, and so on, and this is one of the main ways that species are differentiated. There has been great selection for efficiency in using this major limiting resource.
But what happens when more and more of this just falls out of the sky, after it was used as fertilizer or created by combustion or fossil fuels, is that the organisms that were very efficient at using it get squeezed out by ones that aren't. And we've been doing some long-term nitrogen addition experiments trying to mimic the effects of nitro-deposition. And in native prairie, inside of a nine-square mile reserve where nothing comes in and bothers it, we just doubled the normal rate of nitrogen deposition to some plots over the last 20-some years, and that doubling has caused a 30% loss of biological diversity from those plots, and that kind of doubling is happening all around the world right now, just from nitrogen deposition caused by human activities.
But we can detect this loss of diversity because we actually have permanent plots. We've been counting and measuring everything in them year after year after year for 20-some years. Nobody else does this kind of thing. People aren't out there counting all the species in other ruminant stands.
So you might think, well, you know, humans will have an impact some place, but we'll save the nature here; we'll save a bit of nature there. Well, that's not true when there's climate change, because the climate affects the whole globe. It's not true for nitrogen deposition; it also affects the whole globe. Phosphorous, it doesn't rain out of the sky, but in agriculture we release a lot of phosphorous, which ends up going into rivers, streams, and so on, and into lakes, and the ocean. And in rivers, streams and lakes, phosphorous has the same effect that nitrogen has on land. "Phosphorous utrification," it's called, pollutes these rivers, streams, and lakes. It's why phosphorous has to be removed from sewage, because the phosphorous in it was causing massive changes in lake quality, causing these scummy layers of blue algae to form in the lakes, and making them no longer of use for fishing or recreation.
Well, the same basic biological changes are happening to the earth ecosystems, terrestrial ecosystems, because of nitrogen. We are introducing exotic species. The world has major land masses that are separated from each other, and most organisms on their own couldn't get from North America to South America, much less from Europe to North America, and Australia, and people are moving organisms around willy-nilly. We are homogenizing the globe, and that has big effects on what's going on.
We are releasing pesticides which are moved around the world, and many people don't realize this but pesticides evaporate in the air like water does, and water, when it gets cold, comes out of the air as either rain or snow, and pesticides come out of the air when it's cold too-it takes negative 10, 20, 30 degrees, but it comes out as pesticide snow.
And you can go to pristine lakes 50 or a hundred miles away from the nearest pesticides, up at a high elevation on a mountain, but it's cold at high elevations of the mountains, and you can find deformities in the fish and pesticides in the fish. Pesticides that were used to try to stop malaria in some tropical habitat can end up in lakes in the Yukon Territory.
There are these global cycles. There are many, many ways that humans are affecting the globe other than just climate change, and all of these are areas of I would assert significant concern, and that are at least as important as global climate change and they're probably going to interact with each other in ways that may not be very pleasant, and very few of these are very well understood at this point in time.
Has modern agriculture moved away from monocultures?
If you look at all the lands of the earth and ask who manages those lands, the single largest group of land managers for the world is the farmers, the agriculturalists. We estimate that by the Year 2050-about 50 years from now-that agriculturalists will be managing about half of the usable lands of the world. We're not talking about tundra, we're not talking about desert, we're not talking about mountains, but the land that actually in some ways you consider useable to humans--agriculturalists will be managing about half of that.
How that land is managed is incredibly important for the quality of life on the rest of the earth, because these lands produce lots of services and value to society. And one service, if you will, is production of food. That's a very important service. We need food, and agriculture has done wonderful things in the last 40 years at increasing global food production.
There was significant concern in the 1960s that we might have massive starvation because food production around the world was not increasing as quickly as population. The green revolution had a dramatic and highly positive effect on that, and it has provided food that has prevented starvation, it has allowed people to be able to live and grow and develop their full physical mental capabilities and to contribute to society in meaningful ways. It's greatly increased the quality of life.
But there have been some side effects of that that were not appreciated, that are building up, and that we need to deal with in the long term if we're going to avoid some major problems that could be caused by agriculture.
To produce all of the food that we are producing right now, we now add as much nitrogen in fertilizers as was produced by the system in all various forms on its own naturally, so we sort of doubled the nitrogen cycle of the world. A lot of this nitrogen leaks out of farmlands and into groundwater, where it decreases its quality, and into the lakes, rivers, streams, and oceans and so on, and so it moves through the air and hits other land services. We use pesticides, we use phosphorous, we use irrigation.
And when we do that, we get a crop, a yield of food, but we no longer have these lands, which used to produce clean drinking water, for instance, pristine drinking water. We often farm soils in ways that we use up the nitrogen content and the carbon content of the soil, so we're sort of mining some soils as opposed to farming them sustainably.
And I think that as I look at the world and what the big issues are, one big issue is energy use. The other big issue I would assert is farming. We must try to find ways-and we don't have quick easy answers, this is not an easy question-ways to farm that are going to maximize the net benefits that humans receive from that farming-the food and ecosystem services, the quality of the air, the quality of the water-that maximizes that total return to society.
Well, how can that be done? Can biodiversity contribute to that? And the answer is no and yes, depending upon what kind of agriculture you're looking at.
In one sense, there are certain crops like corn, wheat, and rice that we know no other way to grow them efficiently except to grow them by themselves. Now in those crops, we use genetic diversity within those crops to help us fight pests that attack the plants, whether it's a disease or insects. We use genetic diversity within the crops to help us find a variety of wheat or corn or rice that gives us the best yield on this kind of soil and in this kind of climate, and people in agriculture are working very hard to make sure we don't lose the natural genetic diversity of these crops, and are making sure they can find the diversity and use it to maximize the yields that we do get. So we need to use the diversity of those crops but it's not as if we can plant corn, wheat, and rice and five other grains in the same field and get a reasonable yield. They don't necessarily grow well together. They're hard to harvest, and so that idea of using biological diversity for high production agriculture is unlikely to make it as far as planting many things together.
But if I talk about the land or the earth, of all the land dedicated to agriculture, about one-quarter is dedicated to growing crops intensively, a single corn, wheat or rice crop in an area. The other three-quarters of the land is dedicated to a more pastoral approach, where you have pastures and you're grazing animals on those pastures, and it is in those conditions where biological diversity can be a very important tool to use.
We know from what we've looked at in our grassland ecosystems that greater diversity leads to greater productivity and greater stability of that production. The same is true for forests. Most timber is not produced in monoculture stands, but a lot of it is. A lot of what's happening in forestry in the last 20 or 30 years has been to sort of copy agriculture and try to grow trees as if a tree were corn.
Well, corn grows in about three or four months, maybe five months depending on where you are, and you harvest it. And the conditions it grows in are pretty relatively predictable. When you grow trees, they're gonna grow for 20, 30, 40, 50, 60 years, and there could be outbreaks of insects, climate could change during that period of time, there could be many, many diseases that could come through. So if you have, if you will, all of society's eggs in one basket, if you just convert the whole northwest of the United States into let's say Douglas fir, not that anyone would think of doing that, mind you, but if that ever happened, think about what risk that poses to our society. We need wood to build houses. We need wood-it is a very valuable material for our society.
Same thing is true with Lava Valley pine in the southeast of the U. S. That's about a third of all the lumber comes from there, a third comes from the Pacific Northwest, and frankly most of it does come from Douglas fir or Lava Valley pine. We're putting all of our eggs into two species.
What happens if a pathogen comes into North America that can kill Douglas fir? We lost the chestnut in the eastern United States. It was the dominant tree in the eastern U.S. in the early 1900s. A pathogen was brought into the United States and killed off all the chestnuts. We lost all that. Luckily, these were diverse forests and something else grew up and took its place-oaks and maples and so on took the place of the chestnut.
Well, when you grow things in monoculture, you don't have that ability, plus everything that we had done suggests that monocultures are not in the long term going to be as productive as mixtures of species.
So I would assert that we need to rethink some of our forest practices, to try to find out how can we maximize the total goods and services produced by forests-the same thing for grasslands-and use biological diversity as a tool in optimizing the goods and services humans get from ecosystems.
What is the motivating question that drives you?
I started doing ecology for a variety of reasons. I was interested in nature, as well as interested in mathematics, and I saw a chance to sort of use mathematical and experimental approaches to answer questions that I thought were interesting.
I was also interested, though, in gathering knowledge that was relevant for humans in human life on earth, and human societies, and my view at that time was that human actions were having significant environmental impacts and that we didn't know enough about how ecosystems worked to manage our actions as well as we should be.
And so I am inspired by that. I really am interested in doing research that pushes forward fundamental knowledge in the discipline of ecology, but is simultaneously relevant to the human predicament on earth. There are six billion of us heading toward nine billion, and I think per capita consumption right now is about five times--each person consumes about five times more now than we did a hundred years ago, so we have many more people consuming much, much more. What are the implications of that? How can we actually live sustainably on the earth? How can we come up with a way of living that means that people a hundred years from now will have the same quality of life that we have, or maybe a better quality of life than we have?
To answer that question requires fundamental advances in our knowledge of science and our knowledge of how ecosystems operate. It also requires synthesis of science with economics and with public policy and so on, and I have enjoyed all of those activities. I greatly enjoy collaborating with colleagues in economics and policy and agriculture and so on, even colleagues in ethics. There are major ethical issues. Humans are ethical organisms, which is in its own right is very interesting. You know why are we ethical? What does that imply?
Well, the least that it implies is that we are very social, and ethics suggest that we're willing to give up things that might help us in the short term for the privilege of being part of a longer term, a larger social organization, and maybe we're willing to give up things that help us in the short term because we realize that in the long term, we're better off if we do something else, so there's sort of the short-term gain vs. long-term gains. Those I think are two essences of ethics.
So what keeps me going? I guess I'm innately curious about nature. I'm very curious about why the world is as diverse as it is. We have many theoretical explanations of why that might be so, but we really don't know what's causing it. And I really want to help gather knowledge that helps us find some pathways toward a more sustainable life for humans on earth.