Using satellite technology, the world's first global flood observatory pinpoints and archives floods, one of nature's deadliest phenomena.
When the floods that devastated the upper Mississippi River in 1993 receded, they left behind more than just muddy debris. They also planted the seed of an idea that Bob Brakenridge hoped could help prevent similarly unexpected disasters in the future.
The July 1993 flood - the most destructive in U.S. history - affected nine states, especially Iowa, Missouri, and the Dakotas. Floodwaters submerged seventeen thousand square miles of land, killing nearly fifty people and causing the evacuation of twenty-six thousand more. More than four hundred counties were declared disaster areas. Meteorologists described the flood as a once-in-a-century event, while homeowners and policymakers wondered why no one had predicted the possibility of such a calamity.
"The water was so high, you might only see the tops of grain silos sticking up here and there and maybe the roof of one of the taller buildings," says Brakenridge, who visited the soggy region that summer with the support of a special grant from the National Geographic Society. From a small plane nine hundred feet above the Mississippi, he toured the destruction from Wisconsin to Missouri. "Even from up in the air, you could look out almost to the horizon and there was nothing but water."
Brakenridge's aerial tour of the region was more than just sight-seeing. As a fluvial geomorphologist, he was interested in creating the first truly accurate maps of large-scale flooding by using new radar satellites that rely on reflected microwaves rather than visible light to capture an image. Previous attempts to map floods were limited, because optical satellites could not see through clouds; the newer radar satellites, however, could. And so, up in the single-engine plane, Brakenridge compared the actual flooding below him to images taken by radar satellites.
The radar satellite images were remarkably effective in capturing the extent of flooding, and the science of flood remote sensing was born. "This was a whole new field," said Brakenridge, who came to Dartmouth in 1988.
A decade later, Brakenridge and his colleagues, research associate Sebastien Caquard and research assistant Elaine Anderson at the Dartmouth Flood Observatory (DFO) have taken this idea even further, using both radar and optical satellites to detect, map, measure, and analyze major floods worldwide. The DFO receives grant support from the NASA Earth Science Enterprise and uses data collected by NASA satellites and sensors. Among the other agencies the DFO partners with are NASA's Jet Propulsion Laboratory, the U.S. Army Corps of Engineers, the National Oceanic and Atmospheric Administration, and the U.S. Geological Survey.
One of the DFO's major projects has been the creation of the first global flood archive, which uses information from news, governmental, instrumental, and satellite sources to document the extent and variety over time of flood inundation. In the long run, the archive could provide information vital to climatologists investigating global warming, as well as to urban planners and engineers responsible for new development.
More immediately, however, the maps created by the DFO are helping save lives. International relief agencies such as the United Nations World Food Program use the images to target emergency aid to remote areas. In regions lacking a sophisticated communications infrastructure and mass media, DFO maps are, in some cases, the only reliable way for agencies to determine where devastation is greatest.
Floods are the most common natural disasters on Earth, comprising forty percent of natural calamities. Flooding is also extremely expensive: Not including hurricanes and tropical storms, twelve of the fifty-four most expensive weather disasters in the U.S. since 1980 have been floods, according to the National Climatic Data Center. If flooding as a result of hurricanes and tropical storms is included in that total, the number rises to twenty-six. The median amount paid out each year by the National Flood Insurance Program since 1980 is about 368 million dollars.
Given how common floods are overall, the DFO tracking is limited to major flood events and those that result in fatalities or significant damage. Determining which floods fit this description is somewhat subjective. Hydrologists often use the frequency of recurrence as one criterion. Areas that flood on a regular basis typically are not included.
"Deciding what is a major flood actually turns out to be quite a big problem," Brakenridge said. "As the population increases in certain areas of the world, like China, Bangladesh, Africa, and Southeast Asia, for example, people are being forced to live in areas where it's normal to see flooding every other year. So flood events that aren't really unusual are becoming major relief problems because of all the people who live in these areas."
Statistics gathered by the DFO reinforce this observation. In 2002, for example, Asia experienced thirty-seven percent of the world's major floods, but suffered sixty-six percent of the total number of flood deaths and eighty-nine percent of the displaced persons (see chart). In northeastern India and Bangladesh alone, more than 500 people were killed and another fifteen million had to be evacuated because of monsoon rains that ravaged the area.
The DFO's newest project, monitoring large seasonal wetlands, could help prevent similar disasters in the future. By looking closely at the gradual day-to-day changes of the wetlands, with their distinct wet and dry seasons, the researchers can watch the progression of monsoons. Wetlands are also critical to river systems, providing natural reservoirs for excess rainfall. Many experts believe the 1993 floods along the Mississippi were exacerbated by the fact that almost eighty percent of the river's original wetlands have been drained since the 1940s.
Ultimately, Brakenridge would like to develop a way to better predict where dangerous flooding will occur, but to do so will require several decades worth of information. "As the Earth's climate changes, the geographic patterns and frequency of extreme meteorological events will also change," says Brakendridge. "Using sustained multi-sensor Earth observations from space to observe and measure extreme floods will ultimately help us to examine, analyze, and even predict these devastating events."