Half of the global area burned by fires every year is a result of deliberately set and managed fires, rather than wildfires caused by lightning, accidents or arson, a study suggests.
If the finding is correct, it means that the increase in the area burned annually by wildfires as the world heats up will be even greater than currently predicted by computer models.
鈥淲e might have more extreme fires than we realised,鈥� says at King鈥檚 College London.
Advertisement
That, in turn, means efforts to reforest areas may not remove as much carbon dioxide from the atmosphere as is currently projected, he says.
Starting in the 1970s, researchers have been using satellite data to track the area of the planet鈥檚 land surface that burns every year in greater and greater resolution. Today, these observations are fed into computer models to work out the extent to which the risk of fires is increasing due to global heating, and thus how much worse it will get in the future.
We have always known that managed fires, as opposed to wildfires, are responsible for part of the burned area, says Perkins. This includes the burning of stubble after harvesting, for instance, or small fires set to reduce the amount of dead vegetation and the risk of bigger fires later on.
But until now, there haven鈥檛 been good ways of distinguishing between managed fires and wildfires. So Perkins and his team created a database of around 500 managed fires and developed a computer model of their typical characteristics. For example, they tend to be small, but often there are many in close proximity that collectively cover a very large area.
They then applied this to the satellite observations to work out that half of the burned area is due to managed fires. Perkins presented the findings at a meeting of the European Geosciences Union in Vienna, Austria, on 15 April.
The area of managed fires doesn鈥檛 change that much from year to year, says Perkins. 鈥淢anaged fires don鈥檛 bounce around as much with the climate.鈥�
So if managed fires are being underestimated, the variability in wildfires is being underestimated too, which means we are also underestimating the influence of climatic factors. 鈥淲e鈥檝e ended up with models that are far too insensitive,鈥� says Perkins.
