Equipped with infra-red technology, a powerful computer and a set of mathematical equations, a Monash scientist is seeking to understand the complex forces at work in a bushfire.

By David Bruce

Armed with a box of matches and an infra-red camera, meteorologist Dr Michael Reeder last year arrived at a disused airfield in the Northern Territory and proceeded to set fire to it. Recording his observations on the camera from a cherry picker perched over the blaze, he returned to Monash with enough data to keep his supercomputer chugging away for months.

Having spent most of his working life making sense of how clouds and storms form and disperse in an infinite array of patterns and with a great range of force and menace, Dr Reeder, a mathematician in the Faculty of Science, is applying this knowledge to an equally unpredictable element bushfire.

Despite the long history of bushfires on the Australian and international landscape, Dr Reeder is one of the few people in the world trying to gain a better understanding of how bushfires behave by building a complete mathematical model. Bushfires are unpredictable because of their ability to fundamentally change their size, speed and intensity as they move over varied topography and are influenced by changes in winds and movements in the air temperature. Real bushfires are difficult to study – at the height of a blaze firefighters do not look too kindly on scientists going about collecting data.

The exercise in the Northern Territory was a controlled burn-out on a flat, lightly vegetated field. But the data gathered by the infra-red camera has been digitised and placed into a three-dimensional computer model that can introduce new factors such as a hillside, a strong northerly wind, or mid-summer day’s air temperature.

Researcher Dr Michael Reeder uses modelling to track the 'personality' of a bushfire.

Scientists now have an effective tool to track the ‘personality’ of a bushfire. Without the combination of infra-red technology and a powerful supercomputer, bushfire modelling would not have been possible.

"The data we have drawn from these fires is so complex that analysing only one portion of it can tie up the supercomputer for several days. It is slow, grinding work," said Dr Reeder, who buys time on the supercomputer at the Bureau of Meteorology in Melbourne.

Despite the latest advances with the impressive technology, Dr Reeder said there was still a long way to go to really understanding the mind of a bushfire. "We really are at the point where weather forecasting was in the 1950s," he said. "When computers finally arrived, it immediately changed the accuracy of forecasting to the point where they're pretty good at it these days. We are now at the same sort of early point with bushfire forecasting."

Like the fishermen armed with a detailed weather forecast as he sets out to sea, the plan ultimately is to provide firefighters with detailed predictions on how a bushfire will behave on a given piece of their ground.

The research may not tame one of nature's most ferocious manifestations but when the early February northerlies are gusting over a parched stretch of Australian bushland, the men and women who scan the horizon for the first puff of smoke would be grateful for a measure of predicability in their onerous task.

In the second half of 1999, Dr Reeder is a visiting professor at the State University of New York in the Department of Atmospheric Science. He will also be working closely with Dr Terry Clark at the National Centre for Atmospheric Research in Colorado on their joint research on bushfires.