Some 40 scientists from around the world are studying the fire, from a variety of perspectives. At issue are the causes, spread patterns and other scientific characteristics of the more overwhelming and costly kinds of forest fires.

Frank Albini, a leading theoretical fire behavior specialist from the University of Montana, is credited by the Canadian Forest Service for bringing the international group of scientists together.

Besides South African and Canadian scientific community representatives, there are Russian, German, and American participants.

Two contracted NASA representatives are also on hand to participate in the experiment. Russia and Germany could not get involved in the actual burn, although all data retrieved from this portion is expected to be given to the absent contingents.

Key players have been developing the project since 1993. Each university or company is financing its own portion of the experiment. This year, up to five of 10 targeted plots will be burned. The group will return next year to study the remaining plots.

Brian Stock, a senior research scientist with the Canadian Forest Service, said the project is a unique opportunity for scientists from all over the world to participate in experiments ranging from fire burn studies to environmental impact of fires in the boreal forest ecosystem.

The Canadian government spends a reported $500 million a year on fire protection and suppression while it spend just $300,000 a year on research.

"We needed to get the most bang for our buck. Each participant's research is just a small piece of the puzzle," Stock said.

Albini's goal is to create a scientifically accepted theory for crown fire initiation, spread and consumption.

"We want to make measurements of real life fires," he said, pointing out that currently their only knowledge comes from previous fires.

"Eventually we could look at a fire and say, 'Is this a fire we should let burn? Or is this a fire we should spend money and resources on to extinguish?'" Albini said.

But Stock said he believes the findings could have greater implications. Community protection and fuel management strategies would also benefit from the results.

During a test fire on July 1, a 50-meter fire break used to keep fire from burning other test blocks was breached by the intense flame. The 75-by-75-metre test plot burned in just over 90 seconds. Fifty-metre fire breaks are found surrounding most NWT communities.

Jim Roth, of Storm King Mountain Technologies out of Los Angeles, Calif., is testing the effectiveness of fire shelters and protective clothing.

Roth said he was personally motivated to participate after his brother's death three years ago, when the fire shelter deployed by his brother failed to protect him from the flames.

Roth's organization is volunteer-run. He designed the protection equipment himself and paid his own expenses so that he could test his equipment.

Mike Weber, scientific editor of the xxxInternational Journal of Wildland Fires and team leader of fire ecology research with the Canadian Forest Service, is focusing on the effects of fire on the regrowth process in boreal forests.

"Fire is as natural an ecological process as the wind and the rain," he said in an interview.

"Fire plays an astonishing role," he added, estimating a half metre of growth will be seen in five years.

Neither Weber or Stock would say whether global climate changes were related to the increase in the number and intensity of forest fires reported in the last 20 years.

Both scientists did say that co-operation of GNWT fire management and various fire research people has been excellent. They also pointed out how the community has been informed and consulted at every decision making process.

"This has been an interesting test case. Essentially there are two governments to deal with -- the GNWT welcomed the opportunity to be involved and the aboriginal community has had input from the ground up," Weber said. He credited the communication effort of research personnel for a successful project.

Detailed briefing and more detailed information on the experiments can be found at the International Crown Fire Modelling web site at www.nofc.forestry.ca/fire/fmn/nwt.