Improved modelling of biogenic emissions in human-disturbed forest edges and urban areas

作者：Yanli Zhang, Haofan Ran, Alex Guenther, Qiang Zhang, C. George, Abdelwahid Mellouki, Guoying Sheng, Ping’an Peng, Xinming Wang · 发表于：Nature Communications · 年份：2025 · DOI：10.1038/s41467-025-63437-8 · 被引用次数：4 · 研究领域：Atmospheric chemistry and aerosols、Atmospheric and Environmental Gas Dynamics、Plant responses to elevated CO2

Biogenic volatile organic compounds (BVOCs) are critical to biosphere-atmosphere interactions, profoundly influencing atmospheric chemistry, air quality and climate, yet accurately estimating their emissions across diverse ecosystems remains challenging. Here we introduce GEE-MEGAN, a cloud-native extension of the widely used MEGAN2.1 model, integrating dynamic satellite-derived land cover and vegetation within Google Earth Engine to produce near-real-time BVOC emissions at 10-30 m resolution, enabling fine-scale tracking of emissions in rapidly changing environments. GEE-MEGAN reduces BVOC emission estimates by 31% and decreases root mean square errors by up to 48.6% relative to MEGAN2.1 in human-disturbed forest edges, and reveals summertime BVOC emissions up to 25‑fold higher than previous estimates in urban areas such as London, Los Angeles, Paris, and Beijing. By capturing fine-scale landscape heterogeneity and human-driven dynamics, GEE-MEGAN significantly improves BVOC emission estimates, providing crucial insights to the complex interactions among BVOCs, climate, and air quality across both natural and human-modified environments. GEE-MEGAN, a cloud-native MEGAN2.1 extension on Google Earth Engine, uses multisource satellite data to generate near-real-time BVOC emission estimates at 10–30 m resolution, refining emission characterization in urban and forest-edge regions.