Paradoxically
as some scientists have accepted wildfire as normal parts of many ecosystems, others are looking at high-tech ways to detect these fires automatically, using geostationary (defined) satellites.

Although the technology would help fire-fighters get to the scene sooner, when the fires are easier to fight, the immediate goal is to monitor fire and get some hard data on its occurrence, says Elaine Prins, assistant researcher at the Cooperative Institute for Meteorological Satellite Studies (CIMSS), housed at the University of Wisconsin-Madison.

The launching of the GOES-8 satellite has brought the goal of a hands-off, automated fire detector much closer to reality. The satellite orbits 22,300 miles above the equator and looks down at North and South America. In 1995, while studying Brazil, CIMSS researchers used a computer to analyze images and crank out maps and coordinates (defined) of fire locations. The results were automatically placed on the World Wide Web. "One hour after the satellite's image was received on Earth, it was on the Web, and scientists in South America could read it and then monitor the fires by airplane," Prins notes. (But don't bother surfing for the fire listing, they've been electro-erased.)

The key advantage of GOES-8 over previous satellites is its finer vision. Each infrared pixel covers a square of Earth 4 kilometers on a side. Its predecessor, in comparison, covered almost 14 kilometers squared per pixel.

Nonetheless, the detector can detect a fire only 1 to 2 acres in size. How? Isn't that like catching a minnow with a net designed for a whale?

Not exactly, says Prins. "The key is the satellite's ability to monitor the earth in more than one infrared frequency at once." Then, using a computer program that uses information from both channels, fires far smaller than the pixel size can be detected.

Eye in the sky?
One reason for using Brazil as a test bed is because it's closer to the equator than the United States or Canada, so the satellite has a more direct view of the surface. So the next challenge is to see how well GOES-8 does in the Northern Hemisphere. By next summer, the CIMSS researchers want to have that computer program working smoothly enough that there will be "no need to touch the [computer] code," Prins says. Then, by the summer of 1998, she says an automated detector could be in service. "Where the U.S. Forest Service can't afford to put lookouts anymore, we'd be able to find fires every 15 minutes," she says (the satellite downloads data that often), if the fires cover at least an acre or two.

This image (above) shows the basic science of the GOES-8 Automated Biomass Burning Algorithm (ABBA) for automated fire detection. Each panel represents a region 780 by 800 kilometers centered in central Brazil at 1:45 p.m. local time on August 24, 1995.

The milky regions in the "visible" panel show smoke plumes primarily along the boundary between the forest to the west and grassland to the east (winds are from the east).

Both the 4 and 11 micron panels show the contrast between the cooler tropical forest in the west and the warmer grassland in the east. The 4 micron panel also shows several hotter regions associated with fire along the boundary between forest and grassland and along a road to another new farming area to the west. The 4-11 micron panel shows the satellite-observed temperature differences between the 4 and 11 micron channels. Typically the difference is about 5 degrees Kelvin for non-fire pixels, but for fires these differences are much higher.

For more on this technique, see "Trends in South American...". For a description of the new GOES-8 fire detection capabilities, see "Monitoring fire activity".

Satellite observation gives a new ability to obtain a planetary view of natural history, allowing scientists to replace surmises and educated guesses with hard data. One surprising result of the study of South America, Prins says, was that despite the concern about burning in the rain forest, 80 to 90 percent of the fires occurred in grassland bordering the rain forest, where farmers burn land to clear it for planting.

Still, despite the relatively small proportion of burning in the rain forest, the damage done to biological diversity by burning is much greater than it is in the grassland, where there is less diversity to preserve. More important, fire has been a natural part of the grassland ecosystem for many years.

Want to "see" some images of fires in the Great Plains also taken by GOES-8?

Satellites are also rather handy for observing lightning...

So why do ecologists think fires can be a good thing?

There are 1 2 3 4 5 6 7 8 9 10 11 documents. (Glossary | Bibliography)