In a large field study of prairie plants, a team of biologists found that the more species a plot had, the more "biomass," or plant material, it produced and the better it retained nitrogen, its most crucial nutrient.
Increasingly, conservationists have turned to arguments that biodiversity has value because a greater wealth of species might be useful making ecosystems more productive and stable.
But until now, rigorous scientific evidence to support this idea has been lacking. Researchers say the new study shows not only that biodiversity can increase an ecosystem's productivity but that biodiversity can aid long-term growth as well by increasing an ecosystem's ability to retain key nutrients. Even though the new study was carried out in a grassland, some scientists predict that the result will apply to ecosystems as different as forests and agricultural fields.
"It's a brilliant experiment, and it's a first," said Dr. Samuel McNaughton, an ecosystem ecologist at Syracuse University. "Nobody's ever done what they've done here. It shows that biological diversity does matter to the function and sustainability of ecosystems, and that has a lot of implications for agriculture and human life."
Dr. Peter Kareiva, an ecologist at the University of Washington in Seattle, said: "This is really the first experiment that is unequivocal, clear and striking. You can't attribute these results to any particular species -- it was just the fact that you had a diversity of species that made productivity and sustainability greater. It's the first experiment that's specifically tested biodiversity, per se, as part of its design."
More than 100 years ago, Charles Darwin, writing in "The Origin of Species," suggested that more diverse ecosystems would also be more productive; he also drew his conclusions from grasslands. Referring to an unidentified study, he said fields sown with several grasses consistently yielded more than those with a single species.
Since then, the idea that biodiversity might increase productivity has gathered only scattered interest. But more recently, biologists say, as they took heed of warnings of an impending biodiversity crisis, they began to consider seriously once again the role biodiversity might play in the functioning of ecosystems.
The new study, published Feb. 22 in the journal Nature, was led by Dr. David Tilman, an ecologist at the University of Minnesota, and colleagues there and at the University of Toronto. Working with more than 50 people, the researchers burned, plowed, hand-planted and tended 147 plots at the Cedar Creek Natural History Area in Minnesota.
Each of the 100-square-foot plots grew from 1 to 24 native prairie species. Researchers randomly chose which species would fill out a particular plot's designated number of species. By doing so, they were able to test how the number of species, rather than the kinds of species, affected the growth of plants on the plots and their use of nitrogen.
One of the most labor-intensive parts of the project was weeding all 147 plots. The weeders had to distinguish tiny sprouts of the desired seedling plants from the many undesirable plants whose seeds were still in the soil or had blown in.
In the second year, each plot was examined to estimate how much growth there had been -- the total plant biomass -- and how the plot's plants had used the available nitrogen. The more species a plot had, the greater its biomass of plants and the more nitrogen it had taken up in its increased growth. The fewer the species, the sparser the growth and the greater the amount of nitrogen leaching out of upper soil layers. When nitrogen leached down to where plants' roots could not reach it, it made sustained growth in such plots less likely.
While researchers have yet to determine the exact mechanism by which biodiversity leads to greater productivity and sustainability, they suggest that it is likely to be the mechanism suggested by Darwin himself. The idea is that the more species there are on a plot of land, the more efficiently those many species, all slightly different, can utilize the resources present, thereby producing more actual plant mass.
Tilman and his colleagues also examined native prairies with naturally varying numbers of species and found that more species were associated with greater productivity and nitrogen use, just as in the experimental plots.
"To me, that was as major a finding as the experimental finding," Tilman said. "That's very strong evidence, and it directly applies to undisturbed operating ecosystems."
Earlier research from Tilman's group had strongly hinted that biodiversity might have positive effects on an ecosystem. In an experiment originally designed to study the effects of nitrogen on plant competition, researchers found that plots with different levels of nitrogen also had different levels of biodiversity. Then, in 1987 and 1988, the Midwest's most severe drought in half a century struck. In a 1994 Nature paper, the researchers reported that plots with more species returned much more quickly to full productivity than those with fewer species. But while it seemed likely that the greater variety of species had given the plots their resilience, the results remained clouded by the confounding effects of different nitrogen levels in the plots.
In another paper later that same year in Nature, researchers working in England on a set of miniature ecosystems grown in a greenhouse-like setting, known as the Ecotron, also found that biodiversity could increase productivity. But again researchers remained cautious, in this case because of the difficulty of extrapolating from the small, contrived ecosystems to the rough and tumble reality of nature.
Despite the concurring results, researchers say the controversy over the role of biodiversity in natural ecosystems is far from over.
"I think it's a building block," said Dr. Peter Vitousek, an ecosystem ecologist at Stanford University in Palo Alto, Calif., "but it's not the final answer."
Some say they are reserving judgment, at least in part, because the prairie plots are only two years old.
There also remains the unanswered question of just how many species are required for any given set of ecosystem functions. In the past, some researchers have suggested that relatively few species can perform all the necessary functions and that the remaining species are simply redundant.
In the Nature study, productivity was found to increase dramatically with each new species added, but only until there were 10 species. After that, each new species added much less to the productivity of a plot, suggesting that at least for this one function, 10 species might be able to do much of the work.
But Tilman is quick to point out that while the increase in productivity might not be so great with each species added after the first 10, productivity does continue to increase. And Kareiva noted that for an ecosystem to work, it needed to do many things, like resist pathogens or survive cold years.
As more functions are examined, he said, many more than 10 species might well be required to carry out such functions, with each one providing a different kind of insurance to help buffer an ecosystem as conditions change.
It remains unclear exactly how to translate the new results into a plan of action for managers and farmers. While the new study would seem to suggest that multiple agricultural crops or forest species be grown for maximum productivity, researchers say greenhouse experiments often show a single highly productive species to be as productive or more productive. Researchers say they do not know yet which work will speak most directly to the managed conditions of crop and forest ecosystems.
While commending the work, some researchers suggested that ecologists should not be too quickly drawn off into many costly and time-consuming proofs of ecosystem function.
"People already have a reason to value biodiversity: it's our legacy," Kareiva said. "Now what we need are ways, not just to say what its function is, but to effectively maintain it."