Have Ecologists Oversold Biodiversity?

Some scientists question experiments on how numerous species help ecosystems

By LILA GUTERMAN

If environmentalists were to write down their Ten Commandments, one of the sacred principles would surely be, "Honor thy species." The green movement takes as a truism that ecosystems are healthier when they contain many species of plants

ALSO SEE: Colloquy: Join a discussion on the issues raised in this article

and animals. Ecological scientists have even coined a term for this riot of life: biodiversity.

Though the idea now seems natural to environmentalists, scientists have long wondered whether an abundance of species truly improves the health of ecosystems and the way they work. Experiments to test that link were not completed until the mid-1990's, when some large-scale, much-heralded studies seemed to provide a positive answer. Last year, the Ecological Society of America enshrined the importance of biodiversity to ecosystems in a report intended for educators and policymakers. The report concluded that, because ecosystems are vital to human welfare, we must "adopt the prudent strategy of preserving biodiversity in order to safeguard ecosystem processes vital to society."

It sounded harmless enough. But the publication of the article touched off a fire- storm of debate that had been smoldering within ecology.

A group of scientists charged that the society's report ignored a different viewpoint held by many. The studies cited by the report, the scientists said, were flawed and didn't justify the conservation recommendation. Diversity is worth saving for moral, aesthetic, and even economic reasons, the critics said, but it might not make ecosystems healthier or more efficient.

The altercation went public when, in a letter in the July issue of the Bulletin of the Ecological Society of America, eight ecologists bluntly charged that the report was "biased" and "little more than a propaganda document"; made "indefensible statements"; and set a "dangerous precedent" for scientific societies by presenting only one side of the debate, even though the report seemed to represent the entire 7,600-member society.

They wrote, "Our concern is that unjustifiable actions are being made to protect this single rationale for biodiversity conservation, and that scientific objectivity is being compromised as a result."

Today, the controversy encompasses issues beyond scientific disagreement.

Some scientists claim that the eminent researchers who lead the movement to link biodiversity to ecosystem health, including John H. Lawton at the Imperial College of Science, Technology, and Medicine's Silwood Park campus, in England, and David Tilman at the University of Minnesota-Twin Cities, have exerted so much influence in the field that the major journals are silencing the critics.

The backing of the environmental movement amplifies the message of those renowned researchers, and may even distort the conclusions they draw from the data, the skeptics charge. "Ecological scientists need to be very, very careful to clearly separate the results of experiments from feelings about what should be," says William K. Lauenroth, a professor of rangeland-ecosystem science at Colorado State University, and one of the letter's authors.

"What these guys are saying is, a system works better if it's got 20 species in it rather than five," says Phil Grime, director of the department of comparative plant ecology at the University of Sheffield, in England, and an author of the letter. "Of course the conservation lobby want to hear this."

The controversy began quietly, as many scientific debates do. In the mid-90's, researchers published the results of field and laboratory experiments to show that greater species diversity meant improved stability or productivity in a plant community, both taken as signs of greater ecosystem health. The papers concluded that extinctions may be threatening ecosystems, which are fundamental to life on earth.

In 1994, Mr. Lawton, his colleague Shahid Naeem, and several other ecologists at Imperial College published a paper in Nature on model ecosystems they had established in indoor chambers. The researchers set up some plots with few species and others to which they added species. They found that the plants in chambers with more species tended to produce more biomass, the sum total of living matter in the plants. An accompanying commentary in the journal called the paper "the first unambiguous documentation of the effects of biodiversity on ecosystem processes."

But critics asserted that flaws had tainted the experiments and biased the results. Some quick library research on the plants in the British experiment pointed to a major problem, says Michael A. Huston, an ecologist at Oak Ridge National Laboratory and the most outspoken critic of the biodiversity experiments. "As they increased diversity, they were adding larger plants. All their results were inevitable," since larger plants produce more biomass.

It wasn't the only time that skeptics would find a flaw in a study. That same year, in the same journal, Mr. Tilman's research group described how the worst drought in 50 years affected an experiment on the Minnesota prairie. The plots with higher species diversity had resisted the drought better than those with fewer species. But the more-diverse plots also had received less fertilizer in an earlier experiment, so it wasn't clear which factor produced the result.

In 1996, Mr. Tilman reported in Nature on a study that eliminated that confounding factor. He seeded 147 plots, each about 100 feet square, with between one and 24 species, randomly chosen from plants native to the prairie. After two years, he found that the more-diverse plots produced more vegetation.

Again Nature published a commentary praising the work, and again critics chafed, pointing out that the more-diverse plots had a greater likelihood of containing large plants. That "sampling effect" is nothing but an artifact of the experimental design of randomly choosing species, Mr. Huston and other scientists say. The experiments don't model reality well, they charge, because natural ecosystems do not contain random assemblages of species, nor are extinctions random. David A. Wardle, an ecologist with Landcare Research, a government research institute in Lincoln, New Zealand, goes even further in his critique: "Use of a random-effects model is, to be blunt, simply a nonsense."

Mr. Huston wrote a rebuttal, but Nature rejected it. He sent a copy to Mr. Tilman and later published it in the less-prominent journal Oecologia. "I didn't pay that much attention to his initial complaints, thinking that they were misplaced, frankly," Mr. Tilman says. But Mr. Huston has kept copies of correspondence in which Mr. Tilman advised him to "calm down, put aside your obvious disdain for me, re-read your [rebuttal] paper, and throw it away."

Mr. Tilman and others believe the sampling effect is actually a mechanism by which biodiversity could affect ecosystem function, and say that some environments may indeed be random communities of plants. Even if it weren't, he says, new results on his prairie plots suggest that the sampling effect is not driving the link between diversity and productivity.

What might be happening instead?

The plants may complement each other, he says. When resources such as nutrients and water are limited, if different species exploit them in different ways, adding more species can lead to more efficient exploitation, and thereby greater productivity, he says.

One way to prove that plants are doing that is by looking for a phenomenon called overyielding. If a diverse plot produces more living matter than its single-most-productive component species, grown in monoculture, then the plants must complement each other.

An experiment published last November in Science claimed to have found just that, at least in certain locations. The huge, $1.7-million effort involved plots seeded randomly, like Mr. Tilman's, at eight sites in Europe. Researchers at several of the sites had planted monocultures of all of the species in the diverse plots, and a few of them found overyielding. But Mr. Huston led a group of scientists' rebuttal, published by Science in August, which said that the overyielding occurred only when scientists added a single, important plant -- a legume, which made the nutrient nitrogen available to other plants -- and was not an effect of increasing species diversity per se.

"It's very common in science to have different interpretations of the same data," says Andy Hector, the lead author of the European report and a colleague of Mr. Lawton's at Imperial College. He says he is re-analyzing the data in light of some of the criticisms.

The debate reaches beyond the technical issues to charges of prejudice made by those who question the link between biodiversity and ecosystem function. Mr. Huston sees evidence of that partiality in the Ecological Society of America report that sparked the current controversy. "It's essentially consistent with the conspiracy theory that there's a small group of people that's manipulating the publication and publicity process to push this specific agenda and promote these specific experiments," Mr. Huston says. He says he was not surprised that the report ignored the alternative interpretations of the experiments, since the lead author on the panel of 12 was Mr. Naeem, who is now at the University of Washington, and Mr. Tilman was an author and the editor of the series of reports. "Of the 25 papers that were cited [in the report], 18 of them were by the people who had actually written the article," says Sheffield's Mr. Grime.

"There are quite a lot of experiments that have used different designs to Tilman and Naeem that do not show the results they get," says Mr. Wardle, of Landcare Research. "These studies don't have the same sort of conservation appeal, and therefore do not get the publicity." What's more, he points out, the most productive natural ecosystems are not the most diverse.

Mr. Wardle even accuses major journals of bias toward papers that purport to link biodiversity to ecosystem function. "I'm skeptical about who gets to referee them," he says. "How did they get published?" asks Mr. Grime. "There are senior figures in science who are associated with them."

The skeptics also object to the generalizations made in many of the research papers and in the report, which imply that the results support conserving biodiversity. Mark W. Schwartz, an associate professor of environmental science and policy at the University of California at Davis, says the experiments are not conclusive enough and have been performed on too few ecosystems to translate into general conservation strategies. He says, "If we grab onto the idea before there's actually support for it, we're going to be making mistakes and leaving ourselves vulnerable to people who oppose conservation and say, what's the evidence for that?"

Not surprisingly, Mr. Tilman disagrees. "The least these results suggest is, it would be foolish to lose diversity from ecosystems," he says. Mr. Naeem agrees: "We can't bring back species once they go extinct."

But both say that much more work needs to be done in other ecosystems before a general law linking biodiversity to ecosystem function could be accepted. Most of the experiments so far have been performed on grasslands.

Mr. Tilman expresses surprise at the implication that journals are excluding the opposing view, saying that Mr. Wardle, Mr. Huston, and Mr. Grime have all published papers or letters in Science in recent years.

He and several of the authors of the report for educators and policymakers defend it. "The report was written in a cautious tone to try to reflect what we know and don't know," says Mr. Tilman.

"To tell you the truth, as panel chair, I found the report to be a fairly weak document, given the irreversibility of biodiversity decline," says Mr. Naeem. But he says he argued for inclusion of a diagram, which was later cut, showing different relationships between biodiversity and ecosystem health found in other experiments.

One of the report's authors, David U. Hooper, an assistant professor of biology at Western Washington University, says he was generally pleased with the result but still felt that the alternative viewpoints received too little emphasis. "It was intended as a consensus document. Obviously, we missed. It wasn't," he says. "That was unfortunate, because I don't think it would have taken a lot of tweaking."

In response to the uproar, the Ecological Society of America has changed some of the policies surrounding reports of this type. "We took all the steps we can to minimize any conflicts of interest in the future," says Diana H. Wall, an ecologist at Colorado State University, who was president of the society when the report and letter in response were published.

She also says that future reports will bear a disclaimer saying they are not position statements representing the entire society. Mr. Hooper and Peter M. Vitousek, a professor of population biology at Stanford University, are organizing a panel of scientists with varying viewpoints to draft a position paper to represent the society.

In fact, Mr. Naeem and Michel Loreau, a French ecologist involved in the European research, have organized a meeting in Paris in December to bring the combatants together to discuss the issues. All involved are cautiously optimistic about resolving the items of contention between scientists who have debated in print but rarely, if ever, met face-to-face. "There's an inherent resistance to allow anything to challenge one's own carefully guarded point of view," says Lonnie W. Aarssen, a professor of biology at Queen's University at Kingston, in Ontario, Canada. "There have been a lot of great debates in ecology over the years, and eventually the dust settles and ... people start learning from what happened during the debate. I think that'll happen with this, too."

Easy-to-print version

E-mail this article