Global Map Reveals One-third of World's Forests Disturbed in Two Decades
31 Mar 2026
A new global map is offering the clearest picture yet of how and why the world's forests are changing.
In a study published in Earth System Science Data, an international research team has created the first high-resolution global dataset of forest disturbance types, —the Global Forest Disturbance Type Dataset (GFD), showing that 31% of the world's forest area, approximately 1.247 billion hectares, experienced disturbances between 2000 and 2020.
Built using 30-meter-resolution Landsat program imagery, combined with machine learning and spatial analysis, the dataset distinguishes 11 different types of forest disturbance, ranging from fires and logging to shifting cultivation and urban expansion. The result is a detailed and globally consistent view of forest change that goes beyond simply detecting tree loss by identifying its underlying causes.
The research, led by WANG Li of the Aerospace Information Research Institute of the Chinese Academy of Sciences (AIRCAS), reveals that human-driven activities dominate global forest disturbances. Plantation renewal accounts for 44% of disturbed areas, followed by shifting cultivation at 24% and forest fires at 11%, and infrastructure development accounted for the remaining 21%.
Together, plantation management and shifting cultivation account for more than two-thirds of all disturbances worldwide, underscoring the profound role of land-use practices in reshaping forest ecosystems.
"At a global scale, we have long known that forests are changing rapidly, but understanding the specific drivers has remained a major challenge," said Prof. Wang. "By distinguishing different types of disturbances, our dataset helps reveal the underlying mechanisms of forest change and provides a more solid scientific basis for targeted conservation and management."
At the same time, the study highlights signs of recovery. Newly established forests, resulting from afforestation and reforestation, account for about 3% of total disturbed areas, with notable concentrations in China, India, and Brazil.
To produce the dataset, the team integrated multi-source satellite observations, land cover data, and 18 environmental indicators within the Google Earth Engine platform. This approach enabled classification of disturbance types with an accuracy exceeding 85%, representing a substantial improvement over previous coarse-resolution products.
By identifying not only where forests are changing but also why these changes occur, the dataset fills a critical gap in global forest monitoring. It provides a powerful new tool for scientists, policymakers, and conservationists seeking to better understand deforestation, land-use dynamics, and ecosystem recovery.
The researchers note that the dataset could support more targeted forest management strategies, improve carbon accounting, and strengthen efforts to mitigate climate change.
The study was conducted in collaboration with scientists from the University of Copenhagen, Tsinghua University, and China University of Geosciences, and was supported by multiple funding programs, including China's National Key Research and Development Program and the National Natural Science Foundation of China.
Research News
Global Map Reveals One-third of World's Forests Disturbed in Two Decades
A new global map is offering the clearest picture yet of how and why the world's forests are changing.
In a study published in Earth System Science Data, an international research team has created the first high-resolution global dataset of forest disturbance types, —the Global Forest Disturbance Type Dataset (GFD), showing that 31% of the world's forest area, approximately 1.247 billion hectares, experienced disturbances between 2000 and 2020.
Built using 30-meter-resolution Landsat program imagery, combined with machine learning and spatial analysis, the dataset distinguishes 11 different types of forest disturbance, ranging from fires and logging to shifting cultivation and urban expansion. The result is a detailed and globally consistent view of forest change that goes beyond simply detecting tree loss by identifying its underlying causes.
The research, led by WANG Li of the Aerospace Information Research Institute of the Chinese Academy of Sciences (AIRCAS), reveals that human-driven activities dominate global forest disturbances. Plantation renewal accounts for 44% of disturbed areas, followed by shifting cultivation at 24% and forest fires at 11%, and infrastructure development accounted for the remaining 21%.
Together, plantation management and shifting cultivation account for more than two-thirds of all disturbances worldwide, underscoring the profound role of land-use practices in reshaping forest ecosystems.
"At a global scale, we have long known that forests are changing rapidly, but understanding the specific drivers has remained a major challenge," said Prof. Wang. "By distinguishing different types of disturbances, our dataset helps reveal the underlying mechanisms of forest change and provides a more solid scientific basis for targeted conservation and management."
At the same time, the study highlights signs of recovery. Newly established forests, resulting from afforestation and reforestation, account for about 3% of total disturbed areas, with notable concentrations in China, India, and Brazil.
To produce the dataset, the team integrated multi-source satellite observations, land cover data, and 18 environmental indicators within the Google Earth Engine platform. This approach enabled classification of disturbance types with an accuracy exceeding 85%, representing a substantial improvement over previous coarse-resolution products.
By identifying not only where forests are changing but also why these changes occur, the dataset fills a critical gap in global forest monitoring. It provides a powerful new tool for scientists, policymakers, and conservationists seeking to better understand deforestation, land-use dynamics, and ecosystem recovery.
The researchers note that the dataset could support more targeted forest management strategies, improve carbon accounting, and strengthen efforts to mitigate climate change.
The study was conducted in collaboration with scientists from the University of Copenhagen, Tsinghua University, and China University of Geosciences, and was supported by multiple funding programs, including China's National Key Research and Development Program and the National Natural Science Foundation of China.