Mapping Technology Speeds Help to Fire-Scarred Land

IT’S wildfire season, the time not only for burned forests but for the erosion that often follows when bare, scorched soil repels rain instead of letting it infiltrate. The damage from such erosion – from mudslides to contamination of reservoirs – can be extensive. So conservation teams typically move in quickly, often before the fire has been extinguished, to assess problems and protect homes and water supplies before rain comes.

By ANNE EISENBERG NY Times

Now, computer technology may help speed up this job. Instead of getting out their rulers and topographic maps and analyzing each slope for its erosion potential, some workers may soon use software to download digital topographical maps and other data, quickly connecting the information to a computer model.

Chris S. Renschler, an assistant professor of geography at the University at Buffalo, part of the SUNY system, has written software to link people confronting forest fires to a wealth of free geographic data collected by the government and available online. His program, called GeoWepp, leads them to information specific to the place where the fire is burning, then helps them plug this data into a program that estimates problems in watersheds.

"For years we have been doing this erosion assessment by hand," said William Elliot, an agricultural engineer and project leader at the Rocky Mountain Research Station of the United States Forest Service in Moscow, Idaho. Some of the required information might be readily available, but items like the calculation of slopes take time, particularly given the large acreage typical of forest fires.

Workers, though, have only a small window of time in which to assess erosion risk. Runoff causes flooding, and the sediment in water causes problems, too, Dr. Elliot said, especially in Western areas where runoff is a main source of drinking water. Endangered species in forest streams are also affected by the sediment. "We’ve wanted to figure out a way to use Internet databases and computer software analysis to speed matters up," he said.

The software interface created by Dr. Renschler permits this, in part by doing many of the calculations needed for the model automatically.

For instance, when users get access to digital records of terrain elevations provided by the United States Geological Survey, slopes can then be calculated automatically. "In seconds, you can tell how steep the hills are and where possible channels are," Dr. Renschler said.

Successfully linking the vast reservoir of online digital information to the watershed model was complicated, he said, primarily because most geological survey data these days is designed for use with a fairly new computer technology usually called G.I.S., for geographical information systems. G.I.S. is software in which attributes of data like elevation are linked directly to their location, making possible the creation of maps that show at a glance, for example, rivers and streams in a watershed.

But the erosion model was developed independent of G.I.S. technology and was not compatible with it. So Dr. Renschler retrofitted the model, a national project to which about 30 scientists have contributed, to make it match the new world of G.I.S.

The need was pressing. "It took me a month when I first tried to prepare the data describing a particular location manually for the model," he said of early attempts in 2000, when he was a postdoctoral researcher in the Agricultural Research Service’s National Soil Erosion Research Laboratory, on the campus of Purdue University.

Three years later, he thinks he has succeeded. "Using the interface, I now have the process down to 15 minutes," he said. To make the interface fast and appealing, even for beginners, Dr. Renschler wrote subprograms that work interactively with users to derive networks of erosion channels that might develop and then display these networks in maps. He led users to data describing local climate and soil conditions. He even arranged for viewers to see measurements in English units if they preferred. "Lots of people want to use acres rather than hectares," he said.

Workers still have to go out into the field after a fire to see how the vegetation cover has changed and to assess burn severity. "But that’s the only variable you’ll have to go out and get yourself," he said.

About half a dozen workshops have been held to introduce people working in erosion prediction to the new tool, Dr. Renschler said, including a session last March at the Bureau of Land Management’s National Science and Technology Center in Denver. Bill Ypsilantis, a soil scientist at the bureau who organized that workshop, said the new interface met an important need for access to G.I.S. data. "We need a handle on predicting accelerated erosion from wildfires and prescribed burns," he said.

The price tag for mitigating erosion, using measures like seeding or creating log barriers to catch sediment, is substantial. The cost to Forest Service areas after the Hayman fire southwest of Denver last year, which burned nearly 140,000 acres, was more than $14 million, Dr. Elliot said.

The new interface has its costs, too, said Dennis Flanagan, who heads the National Soil Erosion Research Laboratory. Users need roughly $2,000 worth of G.I.S. software on their computers, he said.

G.I.S. capability is increasingly common among conservation groups, but not among the broader audience Dr. Flanagan hopes to reach in the future. "It would be good if one day they were able to do comparable types of simulations via a Web site that runs the G.I.S. there," he said. Dr. Elliot, the project leader at the Forest Service’s Rocky Mountain Research Station, said the new interface might be tested in the field later this summer.

If that happens, Dr. Renschler is ready. "Over the winter we will include any comments we get back from people during the fire season" for revisions of the tool, he said.

"Erosion is a huge, complex problem," he said. "You don’t want to waste any time in fighting it."