The environment is not the only key factor in determining which species will populate a given habitat -- random, stochastic processes may also play a significant role, according to a study published online today (May 27) on the Science Express website. Study author linkurl:Jonathan Chase,;http://www.biology.wustl.edu/faculty/chase/opening_page.htm a community ecologist at Washington University in Saint Louis, talked with The Scientist about why randomness is so important to species composition,
By Jef Akst | May 27, 2010
The environment is not the only key factor in determining which species will populate a given habitat -- random, stochastic processes may also play a significant role, according to a study published online today (May 27) on the Science Express website. Study author linkurl:Jonathan Chase,;http://www.biology.wustl.edu/faculty/chase/opening_page.htm a community ecologist at Washington University in Saint Louis, talked with The Scientist about why randomness is so important to species composition, and what the findings mean for conserving and restoring the world's ecosystems.
The Scientist: What was your motivation for this experiment?
Students and other collaborators sampling diversity at each of Chase's experimental ponds Image: Jonathan M. Chase
Jonathan Chase: There are environmental factors that create [variations in diversity] -- higher productivity, lower productivity, or disturbed environments and undisturbed environments. Those "deterministic" reasons why species compositions shift from one place to the other [are] kind of what the worldview of much of ecology is about -- how does the environment determine which species live in which places? But there's an alternative perspective that's really been emerging lately is that randomness can really play a really strong role on occasion in determining which species live where, and that sort of flies in the face of our want to create these rules in ecology.
In my own research, I had been creating these replicated communities, what we call mesocosms [constructed in] 300 gallon cattle tanks. We'd be very careful about replicating things, but there was always a lot of unexplained variance that would occur, even among ponds that are sitting in the same field and filled with the same water and so on. So I was inspired to basically ask why you see that variation.
TS: What did you do?
JC: I essentially put these ponds out and put the same water in and the same soil and, in fact, the same species pool. All I did was vary which species went into the habitats first, just to simulate a little bit of randomness early on in the experiment -- the kind of randomness that we often see in natural ecosystems. I wanted to see how important this initial variation was, what we call priority effects. When you see priority effects, essentially the species that gets in first is able to grow to high numbers and on some occasions it can actually change the dynamics of that ecosystem and preclude the invasion of the future species. If more deterministic processes play a stronger role, we then see that whichever the best species is for that particular environment will win regardless of when it invades that community.
The only [other] thing I did manipulate in this experiment was nutrients, which stimulates productivity. Then I just let the experiment run. And then at the end of the 7 years, I was able to go back and ask how important were those initial differences in those first two years in determining what the community looked like over the following several years.
TS: What did you expect to find?
JC: [Previously] we found that higher productivity ponds were more different from one another than lower productivity ponds. So in the low productivity ponds, we hypothesized that there would be a strong filter on the pool of species that could live there [because] only a few types of species [can] persist in these really nutrient-poor ecosystems, whereas pretty much anybody can live in the more nutrient-rich ecosystems, and species interactions and randomness can play a stronger role.
And that's exactly what we found. Over the last 4 years of the experiment, those higher productivity ponds where we dumped nutrients in were extremely variable. You go from pond to pond to pond and it's just a completely different group of species -- different plants, different dragonflies, different zooplankton -- even though they're only 10 meters apart from one another and all arrayed in an old field. [In] the low productivity ponds, you go from one low productivity pond to the next to the next, and it's all the same stuff. In those more nutrient rich ponds, it seemed like history was able to play a stronger role in determining what the final community looked like.
TS: How can these results and concepts be applied to conservation ecology efforts?
JC: If we want to understand how to conserve or maybe even restore biodiversity, it's pretty important to know what creates it in the first place. And it turns out that despite our best efforts, we're still a long way from really understanding the patterns and the mechanisms that create diversity.
More importantly, we tend to perceive diversity by counting the numbers of individuals that we see in a given locality, and we don't necessarily perceive diversity as the differences among communities from site to site -- this "Β-diversity." In this case, history seems to play an important role in creating some of that variation. In the context of restoration, we often forget about this randomness -- this historical perspective -- and in doing so, restoration ecology might actually be minimizing the influence of history and not necessarily recreating diversity at this larger spatial scale.
TS: What the next step in your research?
JC: These pond communities are great for doing long-term experiments, but to me it's not satisfying to think about this as only relevant to ponds. I want this process to be relevant across communities and ecosystems. So I'm working with a group at the National Center for Ecological Analysis and Synthesis in Santa Barbara to explore the generality of the influence of productivity on Β-diversity across larger biogeographic gradients, like from Arctic systems to the tropics, and [how] other factors potentially influence Β-diversity, such as predators or habitat degradation caused by human influences.
J.M. Chase, "Stochastic community assembly causes higher biodiversity in more productive environments," Sciencexpress, 10.1126/science.1187820, 2010. **__Related stories:__***linkurl:Bombs and biodiversity go hand in hand;http://www.the-scientist.com/article/display/43708/ [February 2007]*linkurl:Bacteria may have endless diversity;http://www.the-scientist.com/article/display/22780/ [28th September 2005]*linkurl:The Future of Biodiversity;http://www.the-scientist.com/article/display/13070/ [27th May 2002]
The ecological variables of manipulating the ponds include a. the order in which species are added and b. the nutrient-addition permutations:\n\na. varying the order in which species are added subtly changes the details of species subsequent interactions which induce additional variation in the way in which subsequently added species interact. What you're really doing here is manipulating the successionary dynamics of (~primary) ecological succession.\n\nThe resulting emergent variation is likely disproportionately due to the founder effect (traditionally associated with evolutionary process but more recently in theoretical ecological succession as the "network founder effect" *). Additional variables are inadvertently introduced through the variation among the individuals added, e.g., the specimens easiest to catch for transporting to the study site are subtly different from those left behind and introduced into other ponds.\n\nb. These inadvertent 'deterministic' variations could then be magnified through the mechanism of nutrient manipulation due to original or successionary variation.\n\nBTW, the same phenomena likely occur in organism development; it likely provides an important and non-DNA source of variation among conspecifics (perhaps personality predisposition and orientation as well).\n\nDennis@ecocosm.com\n\n*Hollenberg 2007 "On the evolution and dynamics of biological networks" _Revista di Biologia/Biology Forum_ 100(1) 93-118 (see page 102ff)\n -- please e-mail me for a copy.