Australian bushfire season 2019-2020 – Severity and reasons in context of available data

The Australian bushfire season of 2019-2020 is now the climate topic of the year – the severe bushfire season has caused more than 2000 houses to burn in the state of New South Wales (NSW) alone. At least 34 people have died and likely over 1 billion mammals, birds and reptiles has been lost (1).

According to wikipedia pages for the 2019-2020 bushfire season (2) 18.9 million hectares of land has been burned as of 14h of January. This sounds severe, but how large is the amount of burned land when comparing to the earlier seasons?

Annual burned area in Australia

There are sources to place this bushfire season in the context like the study by Giglio at al 2013 (3). The paper describes a fourth generation Global Fire Emissions Database (GFED4). This data set combines satellite records like the 500m MODIS burned area maps with active fire data from the Tropical Rainfall Measuring Mission (TRMM) Visible and Infrared Scanner (VIRS) and the Along-Track Scanning Radiometer (ATSR) family of sensors. The paper also provides burned area data for Australia and New Zealand (combined) for the years 1997-2011.

Luckily Louis Giglio and his team have continued to work and have created excellent source of all burned area and fire-based emissions datasets. MODIS Collection 6 (C6) MCD64A1 burned area dataset (4) provides satellite-based burned area data for all continents – and also for Australia.

The data and a great analysis tool are available at dataset provides burned area data for the years 1997-2016. It’s possible to select a continent or country and choose several options for the source data such as emission or burned area data.

Let’s start with burned area data for Australia:

Figure 1: Annual burned area in millions of hectares

Figure 1 shows the total burned area for each year between 1997 and 2016 in millions of hectares. Area burned every year was between 18.2 million hectares (2010) and 94.6 million hectares (2001). On average, the area burned during this time period was 52.9 million hectares. Since there is 769 million hectares of land in Australia, the area burned between 1997 and 2016 was 2.4 – 12.3 % of total land area – every year.

These figures seem very high, so let’s see were the bushfires typically happen. Giglio et al 2013 provides a view to that.

Figure 2: Mean annual area burned in Australia, Image source Giglio et al 2013 supplemental materials

Figure 2 provides Mean annual area burned in Australia, expressed as the fraction of each grid cell that burns each year, derived from the July 1996 – August 2012 monthly GFED4 burned area time series. As we can see the majority of fires are happening within the Australian northern and western territories. But overall, the fires can happen everywhere. There are fewer fires in some desert areas: like the Simpson desert. But if there is sufficient fuel load to burn, the fire seems to be likely at some point.

Thus the area burned so far during the bushfire season 2019-2020 can be placed into a context. The burned area as quoted by several sources (~ 18.9 million hectares) is ~36% of average area burned annually in Australia and exceeds the minimum burned area year in the satellite dataset (year 2010). Thus, it is likely that the quoted area is too low, since the fires in many remote areas are not reported. The real burned area during this season will eventually be available through satellite burned area datasets.

Most of the burned land areas are shrublands, woodlands and open forests. Forests fires happen mostly within eucalyptus forests (Australia’s northern and eastern shore).

The above data provides the details of area being burned in total whether it is forest, non-forest and whether the fire was planned (prescriptive burns) or non-planned. But how about the forests specifically?

Forest fires in Australia

There is another source, which provides a lot of details for forest fires specifically. Australia government’s department of Agriculture provides the “Australia’s State of the Forests Report” for every five year period. The latest one has been published in 2018 (5) and covers years 2011-2016.

This report provides details about forest fires in Australia starting with annual forest fires for seasons 2011-2012 to 2015-2016.

Figure: Annual planned and unplanned area of forest fires in Australia – millions of hectares

Unplanned forest fires burned between 8.9 million hectares (season 2013-2014) and 21.2 million hectares (2012-2013). In addition the area burned due the planned (prescriptive) burns was between 6.2 million hectares (season 2013-2014) and 8.2 million hectares (season 2011-2012).  Also we can see that this data correlates well with the satellite burned area dataset.

Earlier versions of these reports provides similar figures; for example the year 2008 version of this report says that the estimated area of forest burnt in the period from 2001 to 2006 was 24.7 million hectares; an estimated 20.0 million hectares was burnt in unplanned fires and 4.7 million hectares was burnt in planned fires. In average 15.7% of Australian forest land burned every year. According to the latest report, the total area of forest in Australia burnt one or more times during the period 2011–12 to 2015–16 was 55 million hectares (41% of Australia’s total forest area) (5). Some forests had at least one fire per year during five different years between 2011 and 2016. Thus, forest was in fire every year.

That is a lot of forest fires in one country. You would imagine that after these fires there are no forests left in Australia. But there is and according to the report, the area of forest has even increased slightly between 1990 and 2016. Most of the forested ecosystems in Australia are ecologically adapted to fire and even require it for regeneration.

For example – Eucalyptus trees do not just resist fire, they actively encourage it. Eucalyptus leaves don’t decompose and are highly flammable. Some species for these trees hold their seeds inside small capsules. Fire triggers massive drop of seeds to the ground cleaned by the forest fire (6). Due to the flammable materials generated by Eucalyptus trees, the forest fire in Eucalyptus forest is inevitable sooner or later. Sooner it happens, more controlled the fire is and less harm it will generate to the trees and animals. Avoiding fires too long is clearly not a good idea. Due to this there are a lot of planned (prescriptive burns) in Australia. Prescriptive burns are the only way of managing the volume of burnable biomass in Australian forests.

In summary, the Australian bushfire season 2019-2020 overall – despite of all the harm it has caused to lives – both for humans and animals – has not been exceptional on country level. It has not been one of the worst seasons in any metric e.g. not with the area of burned land or burned forests. But there is something special happening in New South Wales in particular.

Fires in New South Wales

Almost all the publicity regarding the 2019-2020 bushfire season in Australia has been related to the fires in New South Wales. And indeed, according to the MODIS fire count data from there is something extraordinary going in in Southeast Australia – especially in New South Wales, where the number of fires detected is about four times higher than previous records.

Figure: Eastern Australia fire counts (7)

Why the fires are so intense especially in New South Wales?

Positive Indian Ocean Dipole event

Incidentally there is an exceptional natural event going on. An exceptionally positive Indian Ocean Dipole (8) is currently ongoing (9) and has caused severe weather not only in Australia, but in Africa too (10). The event among the strongest in 60 years (12).

Why is this relevant to the extreme fires in South-East Australia? According to the study Cai et al 2009 (11) there is a systematic linkage between positive Indian Dipole events and severe fires in Southeast Australia. Almost half of most severe fires have occurred during pIOD.

Some of the studies have tried to link pIOD to the Climate Change, but so far the climate model’s ability to predict the pIOD has been less than optimal (13).

Lack of sufficient prescribed burning

According to studies, the hazardous level of fuel loads can be reached within 2 to 4 years from the low intensity prescribed burning in South East Australia (14). But the prescribed burning practices are not popular among locals. The smoke from the hazard reduction burns is a nuisance and health issue itself (15).

New South Wales has about 20 million hectares of forests and the current level of prescribed burning is ~ 200000 hectares annually. This level of prescribed burning will do little to reduce the risks of catastrophic bushfires.

But one thing is sure: the debate about the right level of prescribed burning will continue (16).


  • All-in-all the bushfire season in Australia is not abnormal for Australian scale
  • Consider Australia to be a continent of fire.
  • Most ecosystems in Australia are ecologically adapted to the fire and will even require it
  • The only effective way to manage the fire hazards in Australia is to manage the fuel loads
  • Natural Indian Ocean Dipole events (and ENSO events) has and will have the effect on droughts in Australia
  • Hazardous volume of fuel loads together with abnormally positive Indian Ocean dipole and the associated drought are the prime reasons for extreme bushfire season in Southeast Australia and especially in New South Wales during this season

Further reading

Australia’s state of forests report 1998 provides a lot of good background information about the forests and forest fires in Australia in the past.


  • Special credit to Joanne Nova for readability comments to the text and for background-checking all the references.


  3. Giglio, L., J. T. Randerson, and G. R. van der Werf (2013), Analysis of daily, monthly, and annual burned area using thefourth-generation global fire emissions database (GFED4),J. Geophys. Res. Biogeosci.,118, 317–328, doi:10.1002/jgrg.20042.
  4. Giglio, L., Boschetti, L., Roy, D.P., Humber, M.L., Justice, C.O., 2018. The collection 6 MODIS burned area mapping algorithm and product. Remote Sens. Environ. 217,72–85.
  5. Australia’s State of the Forests Report 2018;
  7. 2019-2020 Australian bushfire season; image credit; image and all other images used with
  11. Cai, W., Cowan, T., & Raupach, M. (2009). Positive Indian Ocean dipole events precondition southeast Australia bushfires. Geophysical Research Letters, 36, L19710.
  13. Cai, W., and T. Cowan, 2013: Why is the amplitude of the Indian Ocean dipole overly large in CMIP3 and CMIP5 climate models? Geophys. Res. Lett., 40, 1200–1205,
  14. Morrison et al 1996, Conservation conflicts over burning bush in south-eastern Australia

One thought on “Australian bushfire season 2019-2020 – Severity and reasons in context of available data

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s