Satellite-based model maps natural and planted forests worldwide

Satellite-based model maps natural and planted forests worldwide
by Simon Mansfield
Sydney, Australia (SPX) Sep 16, 2024

While tree planting is often promoted as a solution to climate change, planted forests can sometimes encroach on natural habitats such as forests, wetlands, and grasslands. This can lead to decreased biodiversity and disrupt essential processes like carbon and water cycling. Although global forest cover has been increasing, it has been difficult to determine whether this growth is due to the expansion of natural forests or newly planted trees. New satellite technology and remote sensing techniques could provide the answer.

Currently, detailed maps differentiating planted and natural forests are insufficient. However, a new study published in the 'Journal of Remote Sensing' on August 21, 2024, proposes an innovative method to generate training samples for accurately mapping these forests at a spatial resolution of 30 meters using satellite imagery.

"Accurately mapping the global distribution of natural and planted forests at a fine spatial resolution is a challenge, but it is crucial for understanding and mitigating environmental issues such as carbon sequestration and biodiversity loss," explained Yuelong Xiao, a doctoral student at Tongji University in Shanghai, China. "Traditional methods often lack sufficient training samples, which hampers the accuracy and resolution of global forest maps. Our study presents a novel approach to overcome this limitation by generating extensive training samples through time-series analysis of Landsat images."

Researchers leveraged data from several sources, including preprocessed Landsat images from Google Earth Engine (1985-2021), Sentinel-1 satellite imagery from 2021, and the 2021 European Space Agency's WorldCover2021 land cover maps. They also used data from the ALOS Global Digital Surface Model. To manage the large volume of data, they divided the globe into 57,559 small tiles and generated 70 million training samples for analysis.

The study used a "frequency of disturbance" value to differentiate between natural and planted forests. Natural forests, being more stable, experience fewer changes over time, whereas planted forests undergo disturbances such as reforestation or deforestation. Pixels with a disturbance frequency higher than three were classified as planted forests, while natural forests showed no disturbances. The researchers also accounted for forests planted before 1985 by using additional distinguishing characteristics.

The model successfully demonstrated that mapping natural and planted forests using autogenerated training samples is not only possible but also less labor-intensive. "This method to accurately map natural and planted forests globally at a 30-meter resolution is reliable and the generated map and training samples represent a valuable resource for future research and environmental management, contributing to efforts in combating climate change," Xiao commented.

Researchers plan to continue refining the mapping system. "Next, we will use the generated training samples and method mapping to update and refine the global map of natural and planted forests regularly. Our ultimate goal is to enhance the accuracy and resolution of forest maps worldwide, providing critical data for policymakers and researchers," added Xiao.

Other contributors to the study include Qunming Wang from Tongji University in China and Hankui K. Zhang from South Dakota State University, USA.

Research Report:Global Natural and Planted Forests Mapping at Fine Spatial Resolution of 30 m'