Unlocking the potential: multi-task deep learning for spaceborne quantitative monitoring of fugitive methane plumes
- 1The Hong Kong Polytechnic University
- 2Chinese Academy of Sciences
- 3Technical University of Munich
Abstract
Quantifying methane point-source fugitive emissions is vital for climate mitigation, yet established methods face significant challenges: matched filter algorithms (e.g. Mag1c) are noise-prone, plume segmentation often relies on subjective manual delineation, and flux rate estimation using the Integrated Mass Enhancement (IME) method is constrained by meteorological data requirements. While fully coupled physical models are currently less feasible due to the complex entanglement of multiple atmospheric processes, we propose a pragmatic alternative: a physical simulation-driven deep learning framework that directly maps hyperspectral imagery to quantitative methane metrics. We built a large-scale, physics-aware training dataset by simulating methane plumes through Large Eddy Simulation (LES) and the Radiative Transfer Equation (RTE). Based on these simulated data, we refined and trained three core deep networks: a U-Net for concentration inversion, a Mask R-CNN for plume segmentation, and a ResNet-50 for emission rate estimation. Each model demonstrated superior precision and robustness compared to its conventional counterpart, showing excellent generalization on real PRISMA data. By leveraging the intrinsic synergy between these tasks, we further developed two Multi-task Learning (MTL) models, MTL-01 and MTL-02, which consistently outperformed the single-task models. This study not only provides a novel and powerful tool for automated methane monitoring, but also highlights the efficacy of MTL in advancing geospatial information science for environmental protection.
Related
- GEDI spaceborne lidar for protected areas monitoring
- Unlocking wheat yield potential through evaluation of high-performing hybrid lines in semiarid conditions of northern Pakistan
- Unlocking Blue Carbon's Potential Requires Targeted Uncertainty Reductions
- IAMAP: Unlocking Deep Learning in QGIS for non-coders and limited computing resources
- Real-time determination of swine manure liquid compost maturity by oxidation-reduction potential and pH monitoring.
BibTeX
@article{si2026unlocking,
title = {{Unlocking the potential: multi-task deep learning for spaceborne quantitative monitoring of fugitive methane plumes}},
author = {Guoxin Si and Shiliang Fu and Wei Yao},
journal = {Geo-spatial Information Science},
year = {2026},
month = oct,
doi = {10.1080/10095020.2026.2741747},
url = {https://doi.org/10.1080/10095020.2026.2741747},
note = {CC BY},
}
RIS
TY - JOUR
TI - Unlocking the potential: multi-task deep learning for spaceborne quantitative monitoring of fugitive methane plumes
AU - Guoxin Si
AU - Shiliang Fu
AU - Wei Yao
PY - 2026
DA - 2026/10/27
JO - Geo-spatial Information Science
DO - 10.1080/10095020.2026.2741747
UR - https://doi.org/10.1080/10095020.2026.2741747
AB - Quantifying methane point-source fugitive emissions is vital for climate mitigation, yet established methods face significant challenges: matched filter algorithms (e.g. Mag1c) are noise-prone, plume segmentation often relies on subjective manual delineation, and flux rate estimation using the Integrated Mass Enhancement (IME) method is constrained by meteorological data requirements. While fully coupled physical models are currently less feasible due to the complex entanglement of multiple atmospheric processes, we propose a pragmatic alternative: a physical simulation-driven deep learning framework that directly maps hyperspectral imagery to quantitative methane metrics. We built a large-scale, physics-aware training dataset by simulating methane plumes through Large Eddy Simulation (LES) and the Radiative Transfer Equation (RTE). Based on these simulated data, we refined and trained three core deep networks: a U-Net for concentration inversion, a Mask R-CNN for plume segmentation, and a ResNet-50 for emission rate estimation. Each model demonstrated superior precision and robustness compared to its conventional counterpart, showing excellent generalization on real PRISMA data. By leveraging the intrinsic synergy between these tasks, we further developed two Multi-task Learning (MTL) models, MTL-01 and MTL-02, which consistently outperformed the single-task models. This study not only provides a novel and powerful tool for automated methane monitoring, but also highlights the efficacy of MTL in advancing geospatial information science for environmental protection.
SN - 1009-5020
SN - 1993-5153
ER -