Using Satellites to Unearth Fossils | |||
Each year the winds and harsh winters of the Gobi peel away additional layers
of the red sandstone at Ukhaa Tolgod, exposing more fossils. Every summer since
1993 researchers from the museum have traveled back to the site, and they've
continued to find remarkable specimens. Despite their success, the
paleontologists still believe that there may be undiscovered sites just as good
or even better in other parts of the desert. So when Novacek and his team make
their yearly pilgrimage, they also take time to search for new fossil beds. |
|||
In general, the fossil-bearing red sandstone can be found in bluffs and
outcrops of eroded sedimentary rock. After the scientists locate a potential site
either by chance, by map, or by word of mouth, they go to the area and poke
around until they see enough evidence of good fossils to start digging. "But
locating these sites is often very difficult," says Novacek. They are dispersed
across the vast desert and topographic and geologic maps of the Gobi aren't
always accurate. Novacek explains that driving around the Gobi looking for sites
that may or may not be there isnt always the best idea. Hundred-degree
temperatures and sandstorms are common in the Gobi. There are no gas stations or
super markets along the way, so the scientists have to bring large, unwieldy supply
and gas trucks with them. Once, while heading for a site they knew existed in the
western Gobi, the team had to go off road. They got mired in a large expanse of
sand and had to dig out the supply trucks and gas tanker twenty-one times in one
day. "I want to know exactly where we are going before I take a column down
there," said Novacek. |
|||
In order to find new sites and pinpoint ones they know exist, Novacek and the
other fossil hunters at the American Museum of Natural History started using
images of the Gobi Desert retrieved from the Landsat 5 satellite. This satellite,
launched by NASA in 1984, moves in a near-circular orbit very nearly from
pole-to-pole around the Earth and scans strip after strip of our spinning planet.
A sensor array onboard known as the Thematic Mapper has seven different types of
detectors, which acquire images of different wavelengths of reflected
sunlight or emitted thermal radiation from the surface of the Earth. One light
detector records only the blue light coming off the Earth (band 1), another
observes all the yellow-green light (band 2), and still another picks up on
thermal radiation (band 7). These data are then beamed back to the planet's
surface where they can be made into detailed images of the Earth's surface.
Scientists can mix and match the various bands into a single image to highlight
various aspects of a section of land. For instance, vegetation in a region can
be highlighted by combining the blue, near-infrared, and red bands of the
Thematic Mapper data. (View a comparison between true color and infrared false
color Landsat data.) |
|||
Unfortunately, there is no simple combination of bands that will immediately pinpoint the fossil sites in the Gobi. Novacek explains that the best they can do is combine blue, near-infrared, and thermal radiation bands on the images of the desert. Together these bands show the rock formations of the Gobi in the most detail. "We then look for the outcrops with the right features and contours," he says. After years of dealing with the geology of the Gobi, Novacek and his team have developed a sense for what types of rock formations will yield fossils and how to spot them on an image. Though there are no hard and fast rules for identifying fossil-bearing outcrops, he explains they are usually found at the base of the larger mountains that run through the Gobi. These outcrops also contain layers of sedimentary rock, and they are well-weathered, low-lying, and sparsely vegetated. All of these aspects can be detected in the satellite images. So far, Novacek says they have had some success using these Landsat images. Last year, in fact, they tracked down one such site northeast of Ukhaa Tolgod and visited it on their yearly campaign. While it was nothing like Ukhaa Tolgod, he says they did find a few well-preserved, fossilized mammal skulls as well as some partial fossils of dinosaurs. The find gives him confidence that in the future the maps will be useful in tracking down both new sites as well as confirming sites shown on the questionable older maps of the Gobi. "And more than anything they will save us many days of driving across the desert," he says. References
Norell, M. A., J. M. Clark, L. M. Chiappe, and D. Dashzeveg, 1995: A Nesting Dinosaur, Nature, 378, 21/28 December, pp. 774-776. Webster, D., 1997: Dinosaurs of the Gobi: Unearthing a fossil trove, National Geographic, July 1996, pp. 73-89. |