COLLEGE PARK, MARYLAND--Carefully selecting certain species to protect may not be the most efficient way to preserve biodiversity. An analysis of species distributions in North America, unveiled here on 18 June at the annual meeting of the Society for Conservation Biology, suggests that randomly picking imperiled species--rather than focusing on particular ones--protects as many life-forms in less land.
Scientists have argued that protecting the habitat of key species--such as large carnivores that need expanses of land, or organisms such as prairie dogs on which many others depend--would shelter additional species, thus offering a bigger bang for the conservation buck. Some conservation organizations, such as the Wildlands Project, actively pursue such schemes. Skeptical, ecologists William Fagan of Arizona State University in Tempe and Sandy Andelman of the University of California, Santa Barbara, set out to see whether this strategy is likely to pay off.
The team examined biodiversity databases for three regions of varying scales: the United States except Hawaii, the Pacific Northwest's Columbia Plateau, and coastal sage scrub in southern California. Each database tracked the presence or absence of organisms either listed as threatened or endangered by state or federal authorities, perilously close to being listed, or species of concern to The Nature Conservancy, which maintains biodiversity databases nationwide. The researchers identified several criteria for choosing umbrella species--such as big carnivores, most long-lived, habitat specialists, and ones on the verge of extinction--and analyzed how many imperiled species might be saved by protecting the habitat of that type of umbrella species, compared to species saved by randomly picking a set of species to serve as the umbrella. They found that a random selection worked better in all three regions: That strategy protected comparable numbers of imperiled species while requiring fewer sites to do so. Fagan says, "Our analysis suggests that extensive reliance upon umbrella schemes represents a poor allocation of scarce conservation resources."
Curtis Flather, a landscape ecologist with the U.S. Forest Service in Fort Collins, Colorado, says he and his colleagues have "wondered whether random approaches wouldn't work just as well as ecological criteria." Such research is important, he says, because resource managers use the umbrella concepts as if they have been tested, "but the degree of testing that is going on is minuscule."