How much heat was released from the 2019-2020 Australian bush fires? And is that likely to have a measurable impact on global atmospheric heat content?

Maybe as a secondary question: How much impact will the CO2 released by the fires have? And (maybe a bit to hard to answer) how much will regrowth draw down?

  • 2
    $\begingroup$ It should be possible to get an estimate by doing something like (energy per mass of wood) x (mass of wood per hectare) x (burnt area). An example here: large.stanford.edu/courses/2017/ph240/timcheck1 $\endgroup$ Commented Feb 12, 2020 at 9:08
  • $\begingroup$ You don't know the mass of wood burned per hectare, not all the wood in the burned area was consumed. Any "estimate" would be no more than a guess. $\endgroup$ Commented Feb 12, 2020 at 9:20
  • 8
    $\begingroup$ That's how scientists make estimates all the time. For instance, when University of Sydney announced one billion animals killed in bushfires, they just did (average animal density) x (cleared area): sydney.edu.au/news-opinion/news/2020/01/08/… Of course it does not mean that exactly one billion animals died, but it tells us that it's not 100 millions, and it's not 10 billions. It gives us an order of magnitude. You can do the same with the energy released. $\endgroup$ Commented Feb 12, 2020 at 9:39
  • 1
    $\begingroup$ This Washington Post article gives a partial answer: 400 million tonnes of CO2 emitted by January 24th ... but it is not a total. The impact of the ash, some of which may have made it to Antarctica, could be greater than the impact of the CO2 emitted (by changing albedo) -- but that is just my speculation. $\endgroup$
    – M Juckes
    Commented Feb 13, 2020 at 23:42

2 Answers 2


Right. We can make some estimates of the scale of the problem, but they will come with a healthy margin of error.

  • If we assume that wood has a calorific value of 18.5 GJ/t (from the phyllis2 database)
  • The area burned is 18.6 Mha (from Wikipedia here)
  • The standing volume of material is circa 1500 m3/ha (an educated guess based on Eucalyptus values in Forest Yield)
  • The density of timber is circa 0.5t per dry m3 (from Forest mensuration handbook)
  • and only 90% of that material actually burns (a guess)


There's around 675 tonnes of material burned per ha releasing 12,150 GJ of energy which is equal to 225,990,000 TJ for the total area or 3,587,143 Hiroshima bombs (at 63 TJ each)...

This is a (very) rough calculation, but it gives a scale of the amount of heat released. This is about 10 day's worth of solar constant radiation over the same area, so it might appear in local temperature readings, but bear in mind that 1) there was already unusually high temperatures in the area, and 2) fires like this produce a lot of smoke which will have had an effect on the local radiation budget.

In terms of all that CO2, you can expect it to be reabsorbed by the surrounding biosphere which is described in the impulse response function described in the IPCC AR5 WG1 CH8 technical supplement (8.SM.11.1 Equations for the Global Warming Potential). You can also expect the burned areas to regrow - which will absorb CO2 as well. Depending on rates of growth, I would expect this to take several decades. - but there is a lot of very complex processes, so this is open to a lot of discussion!

Hope this helps.

  • 3
    $\begingroup$ +1 for linking back to the sources of your estimates! $\endgroup$ Commented Feb 12, 2020 at 10:48
  • 2
    $\begingroup$ I feel like the most important part of this answer is buried in the middle paragraph: "This is about 10 day's worth of solar constant radiation over the same area..." So the heat generated is basically negligible in terms of global climate. I'm fairly sure that the CO2 released will have a bigger effect before it is reabsorbed, but I don't know the numbers for that. $\endgroup$
    – craq
    Commented Feb 12, 2020 at 19:08
  • 1
    $\begingroup$ Query regarding unit in first bullet point, should the unit GJt be GJ/t? $\endgroup$
    – Fred
    Commented Feb 12, 2020 at 20:25
  • 2
    $\begingroup$ There is an easier way: just put a dollar sign to enter equation mode, then underscore _ or circumflex ^ for subscript or superscript, then the number, and finally another dollar sign to leave equation mode. If you want several numbers in a row, just put them between curly braces. It's all explained here: earthscience.meta.stackexchange.com/q/371/18081 $\endgroup$ Commented Feb 12, 2020 at 21:09
  • 2
    $\begingroup$ Given the estimates/uncertainties in the calculation saying 225,990,000 TJ is not correct. Better would be 226 EJ. $\endgroup$
    – Jan Doggen
    Commented Feb 13, 2020 at 8:23

EDIT: As pointed out by Jean-Marie Prival and klanomath in the comments, I originally misread the specific heat of wood as per tonne, rather than per kilogram, and also used the wrong conversion between TWh and TJ.

Here are my rough calculations, based on CO2-release estimates from NASA:

  • 306 MtCO2 (wikipedia)
  • Wood spec energy = 4.50 kWh/kg (engineeringtoolbox.com)
  • Wood spec CO2 = 1.83 kg CO2/kg fuel (engineeringtoolbox.com)
  • $\Rightarrow$ wood burnt = 167.2 Mt = $1.67\times10^{11}$ kg
  • $\Rightarrow$ energy released $\approx 7.525\times10^{11}$ kWh
    • 1KWh = $3.6\times 10^{-6}$ TJ
    • = $2.708\times10^{6}$ TJ
  • Hiroshima = 54-75 TJ (wikipedia)
  • $\Rightarrow$ Aus Bushfires $\approx$ 36k-50k Hiroshima bombs worth of heat.

According to the New York post, the ocean (which accounts for 90% of the earth's warming budget) is warming at a rate of around 5 Hiroshimas a second, which would suggest that the total heat from the bushfires would be eclipsed by global warming forcing in around 3 hours.

  • $\begingroup$ Oooh that's interesting! Do you want to swap and check each other's working? $\endgroup$
    – Will
    Commented Feb 13, 2020 at 8:15
  • $\begingroup$ With other words: The part of the total fire loading you are expecting to be burned is ~9 t/ha (wood/other organic material) and Will estimates 675 t/ha (wood/other organic material). (BTW: compare this with Germany, which has an average "timber stock" of ~160t/ha and Sweden one of ~80t/ha (this is per total area not only forests). $\endgroup$
    – klanomath
    Commented Feb 17, 2020 at 16:53
  • $\begingroup$ Thanks @klanomath, fixed again. Still has a 2-order-of-magnitude difference from Will's answer. If there are more mistakes, this should probably become a community wiki ;) $\endgroup$
    – naught101
    Commented Feb 18, 2020 at 1:20
  • 1
    $\begingroup$ @naught101 I'm reasonably sure I've pitched it a bit high - The fuel volume is based on a mature eucalyptus stand with a generous fudge for branch-wood and under-storey vegetation (so misses out scrub woodland or immature stands). I've also assumed that 90% of it will be burned (which is probably too generous) So mine is probably a extreme high-end estimate $\endgroup$
    – Will
    Commented Feb 18, 2020 at 8:48
  • 1
    $\begingroup$ @Will Yep. Still, the end result of "somewhere between 2.7e6 (mine) and 2.3e8 TJ (yours)" actually seems like fairly reasonable constraints, given the complexity of the problem :) $\endgroup$
    – naught101
    Commented Feb 19, 2020 at 2:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.