Liang Boming, Hao Yuanyuan, Tang Zhuangsheng, He Nianpeng, Li Mingxu
Key Laboratory of Grassland Ecosystem of the Ministry of Education, College of Grassland Science, Gansu Agricultural University, Lanzhou, 730070, China.
Key Laboratory of Grassland Ecosystem of the Ministry of Education, College of Grassland Science, Gansu Agricultural University, Lanzhou, 730070, China.
Environ Res. 2025 Mar 1;268:120773. doi: 10.1016/j.envres.2025.120773. Epub 2025 Jan 4.
Accurate quantifying of methane (CH) emissions is a critical aspect of current research on regional carbon budgets. However, due to limitations in observational data, research methodologies, and an incomplete understanding of process mechanisms, significant uncertainties persist in the assessment of wetland CH fluxes in China. In this study, we developed a machine learning model by integrating measured CH fluxes with related environmental data to produce a high-resolution (1 km) dataset of CH fluxes from China's wetlands for the period 2000-2020. Our results estimate that the wetland CH flux in China is approximately 1.54 ± 0.03 mg CH m h, with total annual emissions of 10.85 ± 0.26 Tg CH yr. Yangtze River Basin (6.01 Tg CH yr), Northeastern China (1.65 Tg CH yr), and the Qinghai-Tibetan Plateau (1.34 Tg CH yr⁻) were identified as the primary contributing regions. Notably, total CH emissions from China's wetlands exhibited a significant declining trend from 2000 to 2020, primarily driven by a substantial decrease in emissions from the Yangtze River Basin and Southern China, where paddy field wetlands are predominant. In contrast, an increasing trend was observed in Northeastern China and the Tibetan Plateau, characterized by natural wetlands. Further analysis revealed that the spatial and temporal dynamics of CH emissions from China's wetlands are closely linked to vegetation activity. This study highlights the spatial and temporal patterns of wetland CH fluxes in China and investigates their potential driving mechanisms, offering valuable data support and a theoretical foundation for national CH emission reduction strategies and wetland management programs.
准确量化甲烷（CH）排放是当前区域碳预算研究的一个关键方面。然而，由于观测数据、研究方法的局限性以及对过程机制的不完全理解，中国湿地CH通量评估中仍存在重大不确定性。在本研究中，我们通过将实测CH通量与相关环境数据相结合，开发了一个机器学习模型，以生成2000 - 2020年期间中国湿地CH通量的高分辨率（1公里）数据集。我们的结果估计，中国湿地CH通量约为1.54±0.03毫克CH·平方米·小时，年总排放量为10.85±0.26太克CH/年。长江流域（6.01太克CH/年）、中国东北（1.65太克CH/年）和青藏高原（1.34太克CH/年）被确定为主要贡献区域。值得注意的是，中国湿地的CH总排放量在2000年至2020年期间呈现出显著下降趋势，主要是由于以稻田湿地为主的长江流域和中国南方排放量大幅减少。相比之下，以天然湿地为主的中国东北和青藏高原则呈现出上升趋势。进一步分析表明，中国湿地CH排放的时空动态与植被活动密切相关。本研究突出了中国湿地CH通量的时空格局，并探讨了其潜在驱动机制，为国家CH减排战略和湿地管理计划提供了有价值的数据支持和理论基础。
