- The fires in Greece were a tragedy and newsworthy one, as no one likes tourists areas to be hit or ancient olive groves to go up in flames. But in the global scheme of things, they were a rounding error.
- Africa matters. It seems half the fires are in Africa. Is this because it has the most tropical and sub tropical forests?
- Europe's figures are surprisingly large. Typically more than USA+Canada, despite this being a smaller continent. Where are these all? Are remote Russian forests regularly going up in flames and no one notices?
- Will the data separate out what I think are the three types of wildfires:
1. Unintended destructive fires. These of course are the ones that make the headlines.
2. Illegal (or unethical) land clearing, whether government sanctioned or not.
3. Land management fires, often to clear scrub around fire resistant trees, of the type widely used in Australia over the last 10,000 years.
There is a fourth limitation to this data, which is the most important. "Area burnt" doesn't say anything about the severity of the fire. For example, in North America, the evidence would suggest that for thousands of years substantially larger areas burned (predominantly do to indigenous managed fire). However, these were relatively low intensity "healthy fires"; they would clear underbrush and small trees, but tend not to kill larger trees. They tended to be smaller fires, resulting in patchwork environments. In the present day, due to the combined impacts of forestry practices, fire suppression (increasing fuel loads), and climate change, the recent fire in North America have tended to be a lot more severe & destructive. My understanding is the same is true of Australia, and surely many other geographies.
Sometimes, your data can obscure more than it reveals.
Here is one graph that used to be available on the USDA site but some years ago was truncated to ~1930 et seq and the large years in the teens and 20s removed.
limitations perhaps include the fact that once a forest has burnt down its not going to be available to burn for many decades [if at all given the trajectory of the earth's climate].
Even for the Amazon the accumulative area burnt is appreciable now and for a less forested place - say a Greek island - a big fire can remove much of what could be burnt in the future, therefore it may look as though things are improving when no more, or even less is burnt this year say, but the truth is worse. [No forest - no burn this year - Oh things are great !]
I do not know if its possible to account for this fact in how data are presented?
Many wildfires in Australia are the result of burnoffs for fire hazard reduction gone out of control. Hazard reduction reduces the humidity in natural areas making fires more likley and act as bellows when fires do occur. It is difficult to separate climate change from other human causes of fires. Increase in the number of transmission lines from wind farms is another risk for wildfires. Battery storages are another as lithium is highly volatile.
Anything above 1000 hectares my mind just registers as 'a lot'. Wouldn't it be useful for people (news outlets) to switch to sq km for slightly easier visualisation? In UK we have an intuitive unit 'area the size of Wales' - which is about 2m hectares.
Please keep up this important work. 🙏🏼
Thank you Hannah, very insightful work.
Some deductions:
- The fires in Greece were a tragedy and newsworthy one, as no one likes tourists areas to be hit or ancient olive groves to go up in flames. But in the global scheme of things, they were a rounding error.
- Africa matters. It seems half the fires are in Africa. Is this because it has the most tropical and sub tropical forests?
- Europe's figures are surprisingly large. Typically more than USA+Canada, despite this being a smaller continent. Where are these all? Are remote Russian forests regularly going up in flames and no one notices?
- Will the data separate out what I think are the three types of wildfires:
1. Unintended destructive fires. These of course are the ones that make the headlines.
2. Illegal (or unethical) land clearing, whether government sanctioned or not.
3. Land management fires, often to clear scrub around fire resistant trees, of the type widely used in Australia over the last 10,000 years.
There is a fourth limitation to this data, which is the most important. "Area burnt" doesn't say anything about the severity of the fire. For example, in North America, the evidence would suggest that for thousands of years substantially larger areas burned (predominantly do to indigenous managed fire). However, these were relatively low intensity "healthy fires"; they would clear underbrush and small trees, but tend not to kill larger trees. They tended to be smaller fires, resulting in patchwork environments. In the present day, due to the combined impacts of forestry practices, fire suppression (increasing fuel loads), and climate change, the recent fire in North America have tended to be a lot more severe & destructive. My understanding is the same is true of Australia, and surely many other geographies.
Sometimes, your data can obscure more than it reveals.
Here is one graph that used to be available on the USDA site but some years ago was truncated to ~1930 et seq and the large years in the teens and 20s removed.
https://fee.org/articles/forest-fires-aren-t-at-historic-highs-in-the-united-states-not-even-close/
Scroll down to the X excerpt for graph.
Interesting and helpful, thanks.
limitations perhaps include the fact that once a forest has burnt down its not going to be available to burn for many decades [if at all given the trajectory of the earth's climate].
Even for the Amazon the accumulative area burnt is appreciable now and for a less forested place - say a Greek island - a big fire can remove much of what could be burnt in the future, therefore it may look as though things are improving when no more, or even less is burnt this year say, but the truth is worse. [No forest - no burn this year - Oh things are great !]
I do not know if its possible to account for this fact in how data are presented?
Great data insights, thank's a lot!
How much is deforestation taken into consideration in this data? An analysis relative to the evolution of sq km would be interesting :-)
Many wildfires in Australia are the result of burnoffs for fire hazard reduction gone out of control. Hazard reduction reduces the humidity in natural areas making fires more likley and act as bellows when fires do occur. It is difficult to separate climate change from other human causes of fires. Increase in the number of transmission lines from wind farms is another risk for wildfires. Battery storages are another as lithium is highly volatile.
Anything above 1000 hectares my mind just registers as 'a lot'. Wouldn't it be useful for people (news outlets) to switch to sq km for slightly easier visualisation? In UK we have an intuitive unit 'area the size of Wales' - which is about 2m hectares.
Thanks for doing this important work. I hope you will advertise it on X/Twitter also.