The integration of vision transformers and SAM for automated methane super-emitter detection using TROPOMI data

Mohammad Marjani; Masoud Mahdianpari; Daniel J Varon; Fariba Mohammadimanesh

doi:10.1016/j.jenvman.2025.127034

The integration of vision transformers and SAM for automated methane super-emitter detection using TROPOMI data

J Environ Manage. 2025 Oct:393:127034. doi: 10.1016/j.jenvman.2025.127034. Epub 2025 Aug 24.

Authors

Mohammad Marjani¹, Masoud Mahdianpari², Daniel J Varon³, Fariba Mohammadimanesh⁴

Affiliations

¹ Department of Electrical and Computer Engineering, Memorial University of Newfoundland, St. John's, NL, A1B 3X5, Canada.
² Department of Electrical and Computer Engineering, Memorial University of Newfoundland, St. John's, NL, A1B 3X5, Canada; C-CORE, St. John's, NL, A1B 3X5, Canada. Electronic address: m.mahdianpari@mun.ca.
³ School of Engineering and Applied Sciences, Harvard University, Cambridge, United States.
⁴ Canada Centre for Remote Sensing, Natural Resources Canada, 580 Booth Street, Ottawa, ON, Canada.

PMID: 40854284
DOI: 10.1016/j.jenvman.2025.127034

Abstract

Methane (CH₄) significantly contributes to global warming, with a global warming potential approximately 84 times greater than carbon dioxide (CO₂) over 20 years. Numerous studies have shown that a small number of high-emitting point sources, known as super-emitters, account for a disproportionately large share of total anthropogenic CH₄ emissions, underscoring the urgency of targeted detection strategies. Recently, a growing focus has been on using remote sensing technology for CH₄ monitoring across various emission sources. As such, this study introduces an automated solution for identifying CH₄ super-emitters using Sentinel-5P (S5P) satellite data. Specifically, a deep learning (DL) framework that integrates a Vision Transformer (ViT) and the Segment Anything Model (SAM) for CH₄ plume detection is proposed. The ViT model is trained using CH₄ plume locations reported by the Netherlands Institute for Space Research (SRON) to classify the presence or absence of CH₄ plumes within image patches, achieving an overall accuracy (OA) of 0.92. Subsequently, SAM extracts plume boundaries from patches identified as plumes by the ViT. Integrated Mass Enhancement (IME) is then used to quantify emission rates based on the SAM-generated masks. This approach is applied to various known emission regions, including Turkmenistan, Spain, Algeria, Argentina, China, Iran, the United States, India, and Morocco, identifying significant CH₄ plumes with emission rates up to 92 t/h. The reported rates align closely with SRON values for the same dates and locations. While the ViT model requires supervised training, the SAM component operates without mask-specific training data, enabling automated and generalizable mask extraction. This study demonstrates the potential of combining advanced AI models with satellite data for effective CH₄ monitoring and environmental assessment.

Keywords: Deep learning (DL); Methane (CH(4)); Segment anything model (SAM); Sentinel-5P (S5P); Super-emitters; Vision transformer (ViT).

MeSH terms

Deep Learning
Environmental Monitoring* / methods
Global Warming
Methane* / analysis
Remote Sensing Technology

Substances

Methane