Year: 2025
Type: article
Abstract: Global land cover mapping is essential for environmental monitoring and earth system modeling, but the high production costs lead to considerable carbon dioxide emissions that are rarely considered. We develop a carbon dioxide emission method for land cover mapping, which takes into account both direct and indirect carbon dioxide emissions associated with the mapping process. The results show that completing annual global land cover mapping at a 30 m resolution generate an estimated 6,880.7 to 10,126.6 tonnes of carbon dioxide emissions. Based on mapping information extracted from publications published between 2014 and 2024, the cumulative emissions from land cover mapping during this period were approximately 3.89 million tonnes. Projections indicate that by 2050, annual emissions could exceed 184.09 million tonnes. To address the carbon effect, we propose three emission reduction strategies: artificial intelligence-assisted collaborative sample labeling, pre-training models with unlabeled data, and model compression techniques, achieving up to a 35% reduction in carbon dioxide emissions. Our results reveal the potential for carbon dioxide emission reduction in land cover mapping, providing a pathway toward sustainable, low-carbon mapping frameworks. The global land cover mapping contributes to carbon dioxide emissions projected to increase by 2050, and strategies exist to reduce emissions by 35 percent, according to an analysis that uses a process-based estimation model integrated with large-scale literature and mapping data.
Language: English
Subfield: Environmental Engineering
Field: Environmental Science
Domain: Physical Sciences
SDGs: Climate action, Life on land
Open Access status: gold
Estimated APC paid: $3,490
Award: G8536649004
OpenAlex ID: W4417464383