ANN ARBOR—As tropical forests and other ecologically sensitive areas throughout the world rapidly disappear, conservationists often find themselves scrambling to identify and protect critical habitat before it is destroyed.

But knowing where to focus their efforts isn’t easy, and sometimes decisions must be made before detailed surveys of an area’s plants and animals can be completed. Using museum collection data and satellite images, a team that includes two researchers from the University of Michigan Museum of Zoology has developed a way of predicting where particular species are most likely to be found—information that can help insure that conservation efforts have the greatest payoff. The researchers describe their approach and a test of the method—involving reptile species in Madagascar—in the Dec. 18 issue of the journal Nature.

In the test, the researchers were able to predict with 75 to 85 percent success where 11 species of chameleons would be found in Madagascar. The method also pinpointed three areas where subsequent surveys uncovered at least seven previously undescribed species. “That information is significant not just because of chameleons in Madagascar, but because the method can be used for any species, anywhere in the world—in the Amazon, for instance, where there are even fewer people doing surveys relative to the amount of forest that’s there and the rate at which it’s disappearing,” said Ronald Nussbaum, professor of ecology and evolutionary biology and a curator at the U-M museum. “This is a way to get ahead of the problem of disappearing rainforests.”

The method uses a computer model that combines information from museum specimens—the species and precise location where a specimen was collected, for example—with satellite remote sensing data, such as land cover, climate, topography and water distribution patterns. The results, which can be displayed as tables or species distribution maps, predict where particular species are currently most likely to be found.

In the test, the method worked best when it used recently collected museum specimen data, but it also yielded good results with older data. That’s important, said Nussbaum, because recent data may be unavailable or incomplete for some species.

The project underscores the importance of natural history museums, said Greg Schneider, coordinator of the museum’s reptile and amphibian collection. Museum visitors usually see only the exhibits in display cases, but behind the scenes are vast research collections that can provide a wealth of information. “We have more than just the animals—the specimens,” said Schneider. “We have all of the data associated with them, and that’s all archived in special ways. Now, with high-speed computers, we can do all kinds of data manipulations to produce the kind of results we got in this study. It’s a fantastic way to use natural history collections that have been accumulating for 100 or 200 years, for conservation efforts.”

Nussbaum and Schneider collaborated on the project with Christopher Raxworthy of the American Museum of Natural History (who is also a research associate of the U-M Museum of Zoology), Enrique Martinez-Meyer of the Universidad Nacional Autonoma de Mexico in Mexico City and Miguel Ortega-Huerta and Townsend Peterson of the University of Kansas. The research was supported by NASA, the American Museum of Natural History, Conservation International, Earthwatch, the National Geographic Society, the National Science Foundation and the World Wide Fund for Nature.

For more information: Ronald Nussbaum—http://www.eeb.lsa.umich.edu/eebfacultydetails.asp?ID=57 University of Michigan Museum of Zoology—http://www.ummz.lsa.umich.edu American Museum of Natural History—http://www.amnh.org/

http://www.eeb.lsa.umich.edu/eebfacultydetails.asp?ID=57http://www.ummz.lsa.umich.edu