Scientists Unveil High-Resolution Dataset to Combat Land Degradation

An international team of researchers has developed the world's first 30-meter resolution dataset to track changes in land productivity—a critical dataset for fighting global land degradation. Published in Scientific Data, this innovation promises to help nations track progress toward Sustainable Development Goal (SDG) 15.3, which aims to achieve a "Land Degradation Neutrality" (LDN) world by 2030.

Land degradation—the decline in soil health, vegetation, and biodiversity—threatens food security, climate resilience, and ecosystems. To combat this, the UN 2030 Agenda for Sustainable Development includes SDG target 15.3, which encourages countries to monitor land degradation and take actions to achieve LDN. A key indicator for this goal is Land Productivity Dynamics (LPD), which measures changes in vegetation health over time. Until now, global LPD datasets were limited to a coarse 250-meter resolution, making it difficult to pinpoint small-scale degradation, such as overgrazed pastures or deforested patches.

Led by Prof. LI Xiaosong from the Aerospace Information Research Institute (AIR) of the Chinese Academy of Sciences (CAS), and in collaboration with UN Food and Agriculture Organization (FAO), Beijing Normal University, as well as other global institutions, the joint team combined satellite imagery from Landsat-8 (30-meter resolution) and MODIS sensors to create a sharper, more reliable Normalized Difference Vegetation Index (NDVI)—a measure of plant health. Using Google Earth Engine, a powerful cloud-based platform for processing planetary-scale data, they analyzed a decade of observations (2013–2022) to generate the first 30-meter global LPD map. This resolution allows scientists to detect subtle changes in farmland, forests, and grasslands that were previously invisible.

The dataset was validated using multiple datasets, achieving over 80% accuracy in identifying areas with declining land productivity. This precision is vital for locating “degradation hotspots,” enabling governments to prioritize restoration efforts.

The United Nations Convention to Combat Desertification (UNCCD) has adopted the dataset for Small Island Developing States (SIDS)—nations highly vulnerable to land loss—as the default dataset for SDG 15.3.1 reporting. Additionally, the algorithm and product have been adopted and recommended in the Good Practice Guidance for SDG 15.3.1 reporting.

The team plans to expand the dataset's applications, from guiding reforestation projects to monitoring drought impacts. As Prof. Li notes, "This isn't just about better maps—it's about giving nations the tools to heal their land." With the 2030 SDG deadline approaching, this high-resolution lens could be a game-changer in the race to protect Earth's lifelines.

Global 30-meter LPD result. (Image by AIR)

Comparison of LPD results for Small Island Developing States (SIDS). (Image by AIR)