Deflating a popular notion, two new reports suggest that ecosystems boasting many species aren't necessarily better at withstanding environmental assaults like drought or fire than are species-poor landscapes. Although their authors emphasize that biodiversity is worth saving for its own sake, it may not be the crucial factor in an ecosystem's survival.
The debate about the importance of biodiversity to ecosystem health has been simmering for decades. In a less varied ecosystem, many biologists have reasoned, the loss of even one species can deal a serious blow because there may not be another species to move into its niche. Biodiversity, they have said, is the key to stability--a hypothesis bolstered by field experiments showing that removal of one or more species can make a community less resilient. But other experts haven't been convinced.
Ecologist Mahesh Sankaran and his colleagues at Syracuse University in New York surveyed 72 plots in three different types of savanna grasslands in India's Western Ghat Mountains, varying in biodiversity from as few as 10 plant species to as many as 50. They simulated grazing by clipping the grass, and wildfires by burning it. At the beginning and after a year, they measured the percentage of the soil surface occupied by each species. As they report in this week's Nature, the species-poor communities were the most stable, with fewer shifts in occupation occurring. But by another measure--the number of species that died out--the more diverse communities had the greater stability. From these conflicting results, Sankaran concludes that sometimes an ecosystem's history (and not its biodiversity) determines its stability. Less diverse communities that have evolved under harsh environmental conditions will tend to have species that are more resistant to further environmental insults, he says. "We shouldn't assume that because the diversity is high, that will act as a buffer."
Using a computer to study the same question, a team led by ecologist Anthony Ives of the University of Wisconsin simulated interactions between species and how they cope with unpredictable environmental events. The researchers conclude in today's Science  that stability--as measured by total biomass--depends more on the ability of species in an ecosystem to withstand environmental stress, than on the sheer number of species. Even an ecosystem with 100 species may be less resilient than one with only 10, says Ives, if those 100 species can't adapt to environmental change.
The papers impress Shaheed Naeem, an ecologist at the University of Washington in Seattle, but he's afraid they won't persuade many other ecologists. Experimentalists, for instance, may argue that Sankaran's study doesn't say much about the effects of biodiversity loss in the real world, because the researchers didn't add or remove any species. "Each camp is so strongly cynical, you have to make a very convincing case," says Naeem.