and Jason Sharples
were kind enough to give some talks on the new research into fire weather last night. Their research covers the high country of SE Australia, where I happen to live. Of course, none of this is part of official NSW RFS doctrine -- yet.
First, the TL;DR
- You can get Foehn Winds in SE Australia. Particularly down on the coasts but the Brindabellas can produce the same effect in my area. These can lead to sudden drops in RH and increases in temperature in the wee hours of the morning, just when it looks like a good time for a back-burn.
- There's a process known as channelling where wind in the lee of a steep slope causes a fire to go sideways to the prevailing wind, with lots of spotting both sideways and ahead of the wind. You can spot this by a leading edge of dark smoke moving across the wind.
- There's an area called the thermal belt with warm, dry air on the sides of slopes, even if the top and bottom areas are cold and moist.
- Fires evolve through small, medium, large, very large and plume-driven phases. The plume-driven phase is where the weather created by the fire feeds back into the fire itself, causing real trouble.
- Fires in rugged landscapes are only suppressed when they exit the rugged landscape.
- There are effects called low-level jets and inversion subsidence that can cause high points to suddenly become hot and dry at night, even though things are stable further down.
- A simplified indicator of when things are really going to go to shit is when the temperature (Celsius) gets close to 40 more than the RH (%).
- The HighFire Risk Project is developing tools to help decisions about these things.
Foehn (Föhn) Winds
Basically, you get a dry wind which comes down from a high area. As it comes down, the pressure increases and the temperature rises. Traditionally, these are caused by air having to rise over a range, dropping its moisture as rain and then descending on the other side. The mechanism in this neck of the woods can also be that the low-level air gets diverted, leading to the high level air dropping down into the lee of the range.
What this means is that you can get really sudden rises/drops in temperature/RH -- less than an hour -- appearing. This will really screw up your back-burn.
Foehn winds are semi-predictable. You can see when this is likely on the synoptic charts. An intense low in the Bight or a trough running NW-SE across Australia are warning signs. Apparently, there are humidity maps that the BOM
produce. If you can find them (I can't, maybe here
) then you can see slots of dry air coming in from the West.http://bushfirecrc.com/publications/downloads/Fire-Note-62-low-res.pdfhttp://www.bushfirecrc.com/posters/B_Sharples1.pdf
Fairly high winds (this effect is pretty much guaranteed at 25kph or so) crest over a steep slope and create an eddy on the lee slope. This can be very strong and explains why, during Black Saturday, the down slopes were often more damaged than the up slopes.http://www.bushfirecrc.com/publications/B_Sharples4.pdf
We kind of get taught about this in fire weather. Now the really interesting bit. Once a fire has established itself on the lee side, the eddies start to spiral along the slope, away from the fire. So you end up with a channel of fire moving across the slope and the wind; in both directions in some cases. But it gets better. The spiralling eddies will pick up lots of embers and create spots both along the channel and in the "normal" direction of the fire. Dave Hubbard mentioned the Lake George escarpment fire from a few years back and said that this effect had happened there.http://www.highfirerisk.com.au/papers/afac07_1poster.pdf
I'm not sure where to put the video Rick showed of a hill in Canberra during the 2003 fires but here will do. You can see a lot of spot fires starting and then you gets what's effectively a landscape flashover; the whole lee slope of the hill lights up at once. Yikes!
You can have cool, moist air at the peak and cool moist air in the vally. And half-way up the slope, it's hot and dry. What happens is that the top of a hill cools off rapidly at night. This cold air runs down the hill, but tends to run down the gullies only; the spurs remain warm. The cool air them pools in the valleys. http://www.bushfirecrc.com/publications/B_Sharples.pdf
Plume-Driven Fires and Rugged Landscapes
I'm not really up on the details of all this. It's generally said that large enough fires generate their own weather. There's more to it than what I have here, but the thing that I noticed was that, once a pyro-culmulus cloud starts generating rain, the heat released from forming the rain drops gets fed back into the fire. http://www.highfirerisk.com.au/papers/afac08_09_poster.pdf
Rugged landscapes make everything difficult. They're also prone to multiple lightning strikes.http://www.bushfirecrc.com/publications/B_Sharples5.pdf
There's a model for fire evolutionhttp://www.highfirerisk.com.au/papers/afac06_3poster.pdf
Low-level Jets and Subsidence Inversions
These both cause sudden rises/drops in temperature/RH at high altitudes, even if everything is stable at lower altitudes. Low level jets are just what they say they are, jets of warm, dry air that have somehow got themselves trapped at a lower altitude. They come in an hit whatever peak happens to be in the way. Subsidence Inversions are high-level temperature inversions that lower throughout the night. Eventually, high peaks will punch through the inversion layer.
Mildish version of these events happen about every 4-5 days in the high country. Strong versions about once every 9 days. So you can http://www.highfirerisk.com.au/papers/afac08_03_poster.pdfhttp://www.highfirerisk.com.au/papers/afac08_04_poster.pdf
Simple Fire Danger Estimation
FMI = 10 - 0.25 (T - H) where FMI=Fuel moisture index/drought factor (0-10), T = temperature (Celsius) and H = relative humidity
FDI = W / FMI where FDI = Fire danger index, W = wind speed (kph)
As T - H approaches 40, the FMI approaches 0 and the FDI takes off into the stratosphere. This measure doesn't seem to produce the same really high peaks as the McArthur meters on very high days, but it's pretty easy to calculate.http://www.bushfirecrc.com/publications/B_Sharples2.pdf
There's a poster of watch-out
rules of thumb for these effects. http://www.bushfirecrc.com/publications/B_Sharples6.pdf