Identifying changes in volcanic unrest and tracking eruptive activity are fundamental for volcanic surveillance and monitoring. Magmatic gases, particularly sulphur dioxide (SO2), play a crucial role in influencing eruptive styles, making the monitoring of SO2 emissions essential. Recent advancements in satellite remote sensing technology, including higher spatial resolution and sensitivity, have enhanced our ability to detect SO2 emissions from volcanoes worldwide. However, traditional fixed-threshold algorithms struggle to automatically distinguish volcanic SO2 emissions from non-volcanic sources. Additionally, accurately quantifying SO2 emissions is challenging due to their dependence on plume height, particularly when reaching high altitudes. To address these challenges, we developed an Artificial Intelligence (AI) algorithm that detects and quantifies volcanic SO2 emissions in near real-time. Our approach utilizes a Random Forest (RF) model, a supervised Machine Learning (ML) algorithm, to identify volcanic SO2 emissions and integrates Cloud Top Height (CTH) data to enhance the accuracy of SO2 mass quantification during intense volcanic eruptions. This AI algorithm, fully implemented in Google Earth Engine (GEE), leverages data from the TROPOspheric Monitoring Instrument (TROPOMI) aboard the Copernicus Sentinel-5 Precursor (S5P) satellite to automatically retrieve daily volcanic SO2 plumes and CTH. We validated the model's performance against the Radius classifier, a state-of-the-art tool, and generalized its application across various volcanoes (Etna, Villarrica, Fuego, Pacaya, and Cumbre Vieja) with differing degassing activities, SO2 emission rates, and plume geometries. Our findings demonstrate that the proposed AI approach effectively identifies and quantifies SO2 plumes emitted by different volcanoes, enabling the investigation of SO2 emission time series that reflect magma dynamics.

中文翻译：

---

使用 Sentinel-5 TROPOMI 和人工智能监测地球大气层：量化火山 SO2 排放


识别火山动荡的变化并跟踪喷发活动是火山监测和监测的基础。岩浆气体，尤其是二氧化硫 （SO2），在影响喷发方式方面起着至关重要的作用，因此监测 SO2 排放至关重要。卫星遥感技术的最新进展，包括更高的空间分辨率和灵敏度，增强了我们检测全球火山 SO2 排放的能力。然而，传统的固定阈值算法难以自动区分火山 SO2 排放和非火山源。此外，由于 SO2 排放依赖于羽流高度，因此准确量化 SO2 排放具有挑战性，尤其是在到达高海拔地区时。为了应对这些挑战，我们开发了一种人工智能 （AI） 算法，可以近乎实时地检测和量化火山 SO2 排放。我们的方法利用随机森林 （RF） 模型，一种监督机器学习 （ML） 算法来识别火山 SO2 排放，并整合云顶高度 （CTH） 数据，以提高强烈火山喷发期间 SO2 质量量化的准确性。这种 AI 算法完全在 Google Earth Engine （GEE） 中实现，利用来自哥白尼哨兵 5 号前体 （S5P） 卫星上的 TROPOspheric Monitoring Instrument （TROPOMI） 的数据来自动检索每日火山 SO2 羽流和 CTH。我们针对最先进的工具 Radius 分类器验证了该模型的性能，并将其应用推广到具有不同脱气活动、SO2 排放率和羽流几何形状的各种火山（埃特纳火山、比利亚里卡火山、富埃戈火山、帕卡亚火山和 Cumbre Vieja）。 我们的研究结果表明，所提出的 AI 方法有效地识别和量化了不同火山排放的 SO2 羽流，从而能够研究反映岩浆动力学的 SO2 排放时间序列。