Species diversity: It's not who they are, it's what they do

Contacts:
David Tilman, (612) 625-5743 or (612) 434-5131 or (612) 922-5781
Deane Morrison, University News Service, (612) 624-2346, [email protected]

Species diversity: It's not who they are, it's what they do

What makes a healthy ecosystem? Ecologists long suspected that a diversity of species was important, and recent work by University of Minnesota ecologist David Tilman and others has demonstrated that. Now, Tilman and several colleagues have discovered why a diversity of species controls ecosystem sustainability. Working with plots of prairie plants, the researchers found that what counts most is not the number of species per se, but the number of different ways species perform a variety of functions such as growing, cycling nutrients or producing seed. The work will be published in the Aug. 29 issue of Science.

Natural ecosystems are akin to towns, which require maintenance workers, teachers, barbers, doctors, etc. in order to thrive, said Tilman. For example, a town may have a great diversity of people, but if all the teachers packed up and left, the town would suffer. Similarly, natural ecosystems suffer most if people alter them by removing all the species that perform a certain vital function, even if the total number of species remains the same. Or, as the researchers put it, what matters even more than a diversity of species is a diversity of functions, or functional diversity.

"This work was motivated by a huge issue--humans dominate all ecosystems of the world," Tilman said. "Society is managing ecosystems without realizing it. We are directly and indirectly changing which species and how many species live in both managed and natural ecosystems. This has potentially serious implications for how these ecosystems function and the services that they can provide to society.

"If we remove one tree species from an area and replace it with another, we have to be concerned with which species are being exchanged, because the action could dramatically affect how the ecosystem functions. Another example would be agriculture. We know that monocultures, whether planted with corn, white pine or Douglas fir, are not sustainable in the long term. Soils lose fertility, disease and pests become problems, and productivity declines unless there are great inputs of fertilizers, other chemicals and energy. This work shows that the wisest way to manage is by increasing the functional diversity of such ecosystems."

The researchers examined 289 plots, each about 198 square yards, that were planted and weeded to have either zero, 1, 2, 4, 8, 16 or 32 species of prairie plants. The plants fell into five "functional groups": legumes, which convert nitrogen from the air into usable nitrates; cool- and warm-season grasses, which grow best during those respective seasons; woody plants, namely oak trees; and forbs, broadleaf plants that produce large amounts of seed. Each plot contained plants representing between zero and five of these functional groups. Because a random draw decided which species went to which plot, some plots had lots of species but only a few functional groups, while others had lots of both or few of both.

The researchers looked at how well each plot performed in terms of several variables, including the amount of vegetation, soil fertility, light penetration and how much nitrogen was contained in the plants' tissues. Then, using sophisticated statistical techniques, they determined how the number of species and the number of functional groups separately affected the performance of each plot. When they looked at the number of plant species per plot or the number of functional groups per plot, they found it was more beneficial to have a diversity of functional groups (that is, more of them) than a diversity of species per se. However, when they compared the plots according to the number of functional groups and the number of species within each group, both mattered. Thus, the performance of these experimental ecosystems was being controlled by both the diversity of functional groups and the number of species (species diversity) within each group. In other words, if a plot of land!

doesn't have any, say, woody plants, it would be good to add one species but better to add two.

Species diversity seems to help ecosystems to the extent that it entails a diversity of functions. Invoking the town analogy, a town may have a large diversity of peoples and cultures, but if everybody's a butcher or a baker, the place will soon be crying for a candlestick maker--not to mention tinkers, tailors, soldiers and even spies (who could go out and find people with new occupations). And while adding occupations to a town will generally help it, so would adding diversity within occupations; better to have more than one doctor, police officer or, of course, teacher. The study results, said Tilman, show that any process that disrupts the types of species in an ecosystem could be as damaging as a process that affects the number of species.

"Every year about 6,000 exotic plant species are accidentally introduced to new areas somewhere in the world," Tilman said. "Some of them have a huge impact on the species composition of the new area. Other human activities that change species composition include the killing off of predators, disturbance and fragmentation of habitats, species extinctions, air pollution and increases in greenhouse gases. About half of the planet's land surface is directly controlled for human benefit. We have to think hard about how much effort we're willing to put into doing it right, because, like it or not, we are now the managers of the Earth."

Tilman's colleagues were Johannes Knops and Evan Siemann, department of ecology, evolution and behavior, University of Minnesota; Peter Reich, department of forest resources, University of Minnesota; David Wedin, botany department, University of Toronto; and Mark Ritchie, department of fisheries and wildlife, Utah State University. The work was performed at the University of Minnesota's Long Term Ecological Research facility in Cedar Creek, Minn.

News releases also on WWW at http://www.umn.edu/urelate/news.html