Text Size

Scientists Use Satellites And Museum Collections To Locate Lizards In Madagascar
NASA satellites help find chameleons in Madagascar
Image Left: A male Parson's chameleon (Calumma parsonii) from Madagascar. This is one of 11 chameleons for which museum specimens and satellite data were used to predict species distributions. Click Image For High Resolution Version. Photo: Christopher J. Raxworthy

Chameleons are well known for their ability to hide from predators.

But they weren't able to hide from NASA, which helped researchers discover seven previously unknown species in Madagascar, off the east coast of Africa.

The NASA-funded study -- which also successfully predicted the geographic distribution of 11 known chameleon species in Madagascar -- uses a computer model combined with satellite data from NASA and other agencies, and data from museum collections about where species have been found in the past.

The study, which appears in a mid-December 2003 issue of Nature, offers a method that may be applied to other species around the world. Predicting the distribution of species is one of the most important and tools for effective conservation, as well as ecological and evolutionary research.

Until now, scientists were required to physically visit a locale and painstakingly find and count species. The new approach allows them to focus on places with a high likelihood for a big payoff. This is especially important for poorly explored regions with diverse climates and habitats. It is also important for those who work with limited budgets, or when there is an urgent need to include species in conservation planning.

The lead author of the study is Christopher Raxworthy, a biologist at the American Museum of Natural History. He's visited Madagascar to study chameleons for the last 20 years. He used computer software that allows users to predict where species might be found, entering information about chameleons in that area.

Raxworthy entered data from museum collections about where these chameleons had been found in the past. Some of this information dated back to the 1800s. He also added the latest satellite images of Madagascar and more data about the island's land cover (as viewed from space), rainfall, cloud cover, average and seasonal temperatures, and topographic data, including elevations.

Environmental records came from several NASA satellites, a Space Shuttle radar mapping mission, and U.S. Geological Survey and National Oceanic and Atmospheric Administration data. The computer then provided maps of the most likely places where these chameleons might be found.

Moderate Resolution Imaging Spectroradiometer (MODIS) image of Madagascar
Image Right: The world's fourth largest island, Madagascar, is featured in this Moderate Resolution Imaging Spectroradiometer (MODIS) image taken by the Terra satellite on September 10, 2003. The narrow strip of green along the east coast of the island is a rain forest. The west coast is lined with baobabs, a desert tree with a fat trunk, and thorny forest. Click Image For High Resolution Version. Photo Credit: NASA/MODIS Rapid Response Team

The model was tested with museum records from different time periods to test its accuracy. Historical data collected prior to 1978 were separated from more modern data that came from specimens found since 1988.

The model showed that the historical data alone was quite accurate (74.7 percent accuracy), the combination of modern and historical data was even better (82.8 percent accuracy), while the modern data alone proved to be the best predictor of where the 11 chameleon species tested might live (85.1 percent accuracy). The older records may have reflected a landscape that has now been altered, and the modern records may fit better with the current satellite data.

Also, in a finding that was completely unexpected, the computer model highlighted three areas where scientists had never found these chameleons. No museum records from these areas had been entered into the model. Raxworthy and colleagues initially thought the new pinpointed regions were due to some sort of computer error. But in visiting two of these areas, they instead found habitats that housed seven new closely-related species previously unrecorded by western science.

Madagascar and its surrounding area is one of 25 biodiversity hotspots -- areas identified by scientists where diverse local plant and animal life appears to be thriving, in spite of human influences nearby. Unfortunately, these hotspots are few and far between. Consider that the 25 hotspots contain 44 percent of all plant species and 35 percent of all land vertebrate species in only 1.4 percent of the planet's land area.

Madagascar itself has lost forests at an alarming rate. Only about ten percent of the original natural habitat remains.

With so much land being changed, nobody knows exactly which species of plants and animals have become extinct, and which species may still be protected. Conservationists around the world have been struggling against time to identify biologically rich areas to protect before they are lost.

Hopefully, the new technique will offer conservationists and scientists a fast-acting tool in their race against time. With tropical forests and other highly bio-diverse areas disappearing around the world, conservationists may now be able to identify and protect critical habitat before it is changed and lost. And many new species may be discovered in the process.

This work is consistent with a new element of the NASA Earth Science applications program focusing on ecological forecasting. The program uses Earth observation data and models to forecast species distributions and how environmental change might affect them.

Krishna Ramanujan
Goddard Space Flight Center, Greenbelt, Md.