News & Media

Tim Wells on Canadian fire behaviour

By: Leith Hillard

Category: Operational Information

  3.52 PM 13 October, 2014


Location: General

Views: 1706

With a lot of high humidity days and unburnt fuel within control lines, the Canadian fires recently attended by about 80 Australians were primarily slope rather than weather-driven.

Two major breaches of control lines on the Mt McAllister fire were largely due to strong overnight katabatic winds.

Fire Behaviour Analyst (FBAN) Tim Wells was deployed to the Mt McAllister fire as a member of a pre-formed team. He loaded up with technology from home – GPS, a weather kit and laptop – but struggled to connect to local information technology systems.

“We’d call back to the local forest service staff and they were massively helpful,” said Tim. “It was all hands on deck.

“They use different fire behaviour models and I had to get my head around the wind flowing the opposite way around high and low pressure systems. In Australia we mostly look at mean sea level pressure on our weather maps, but in Canada that is uncommon: they mostly look at thickness lines which show the temperature of the lowest five kilometres of the atmosphere. The bigger mountains in Canada mean sea level pressure isn’t so helpful for weather forecasting.

“Our fire was burning between 600 and 2000 metres elevation, similar to the Victorian Alps, but we had 3000-metre-high mountains nearby. We were at a latitude of about 55 degrees north. The southern hemisphere equivalent would be a little south of Macquarie Island more than 1000 kilometres south of Tasmania. That led to some icy cold mornings with frost on the ground.

“A weather pattern would last for a few days but, because the fire was in undulating terrain, the fireline conditions would be different around the whole fire. The temperature had been in the 40s before we arrived but was mostly in the high 20s to low 30s with 20 to 50 per cent relative humidity [RH] when we were there. Those conditions were enough to see continual flare-ups and sustained surface fire. The Canadians describe rates of spread in metres per minute as opposed to our kilometres per hour.

“There were fire service remote meteorologists who gave great advice and support, but the Canadian forest service weather information lacked quite a bit of the detail we have access to in Australia.

“They don’t have dedicated ground observers but I spent half my time out in the field. Lots of the terrain wasn’t accessible by vehicles so we used helicopters most days for checking fuels and verifying fire behaviour.

“The fuel structure was varied and most of it didn’t match the standard Canadian fuel models. There was a really deep layer of duff which is organic decomposing material that can dry out and burn. It was similar to our peat fires that can burn underground for weeks, even after lots of rain.

“There was lots of variation in fire behaviour among the different trees. The fire tended to stop spreading when it reached deciduous trees, while some spruce trees would flare and even sustain crown fire under relatively mild conditions. The fires spread more slowly with much less spotting than we see in eucalypt forest. The steep slopes, which were measured in per cent, not degrees, increased rates of spread.”

As in Australia, Canadian fire behaviour models are based on experimental burns and previous wildfires.

“We train in a universal way to read fire behaviour,” continued Tim, “so my Australian training translated well. I applied their models and they worked but it took me a few weeks to be comfortable with the model outputs predicting rates of spread and intensity. I’d feed that fire behaviour and weather information to planning and operations.”

And finally, a fuel hazard that we don’t face in Australia: squirrel caches. Apparently fire can smoulder under snow throughout winter in squirrel caches and then come to the surface and spread again in spring.

“Some of the differences and conversions had my head spinning,” said Tim, “but it also gave me heaps of confidence. “

Last Updated: 13 October 2014