News Release

New high-resolution map shows wildfires caused one third of global forest loss between 2001 and 2019

Peer-Reviewed Publication

Frontiers

Global forest loss due to fire

image: Global disaggregation of 2001–2019 Hansen et al. (2013) 30meter forest loss map into forest loss due to fire vs. other direct drivers of loss view more 

Credit: Alexandra Tyukavina, Peter Potapov, Matthew C. Hansen, Amy H. Pickens, Stephen Stehman, Svetlana Turubanova, Diana Parker, Viviana Zalles, André Lima, Indrani Kommareddy, Xiao-Peng Song, Lei WANG, and Nancy Harris

Researchers have produced the first ever 30m (30 meter pixel) resolution global map of fire-induced forest loss. The study shows that one third of global forest loss between 2001 and 2019 was due to fires, which is higher than previous estimates. The area of fire-related forest loss is increasing globally, particularly within the tropical primary forests of Africa and Latin America. The researchers hope that their map helps improve modeling of future forest fire loss rates under various climate crisis scenarios.

The severe wildfires in Australia, California, and Brazil in recent years have attracted the world’s attention to the threat of forest fires. However, there is a lack of clear and consistent global data on the types and drivers of forest fires. 

“In the case of the Amazonian fires of 2019, there was no clear information on what exactly was burning: forests or previously deforested areas that had been converted to pasture and cropland,” explained Dr Alexandra Tyukavina, of the University of Maryland.  

“Even I, as a geographer, was terrified reading all the headlines that seemed to imply that the last patch of the Amazonian rainforest was on fire, which was not true.”

Tyukavina and her colleagues attempted to close the information gap surrounding forest fires by producing the first ever 30m resolution (a 30 meter pixel represents a square patch of the land with a 30m side) global map of forest loss due to wildfires between 2001 and 2019. The study was published in Frontiers in Remote Sensing.

Mapping forest loss

Forest loss was defined as the removal of woody vegetation over 5 meters in height. To accurately map forest loss types and drivers, the researchers distinguished between forest loss due to fires and forest loss due to other factors such as agriculture (mechanical clearing of forests), floods, and hurricanes. 

“We have been producing the global forest loss map and updating it annually since 2013, but we did not attribute drivers of forest loss. This means that we did not know what exactly happened to forests (fire or mechanical removal) and we didn't know whether this forest loss was temporary (for example, due to slash-and-burn farming) or long-term (deforestation),” said Tyukavina.

“These distinctions of different kinds of forest loss are important from standpoints of carbon accounting and land management.”

The new map showed that the proportion of global forest loss due to fires between 2001 and 2019 is 26% to 29%, higher than previously estimated. It showed near-consistent increases in wildfires across the globe, with boreal forests having the highest proportion of forest loss (69%-73%), followed by subtropical forests (19%-22%), temperate forests (17%-21%), and tropical forests (6%-9%).

The map showed a worrying increase in fires in primary tropical forests of Latin America and Africa. Primary forests are forests in their ‘final form’; fully grown, dense, and bursting with biodiversity. They are extremely important for the environment. Typically, fires in primary tropical forests are very rare.

The map as baseline

“The map is annual, and not near-real-time, so it won't help monitoring forest fires in real time but will be useful as a historic baseline regarding the rates of forest loss due to fire,” said Tyukavina.

The researchers propose to use the map as a tool for forest management, policy and conservation program development, and climate modeling.
“Our new map is a good indicator of where the fires resulting in the loss of forest canopy have happened in the last two decades, and could help guide future finer-detail national, regional and local-scale studies, or tune coarser-scale global models,” concluded Tyukavina.


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.