Automatic detection of methane emissions in multispectral satellite imagery using a vision transformer

Автоматическое обнаружение выбросов метана на мультиспектральных спутниковых снимках с использованием трансформера для компьютерного зрения
Claudia Hulbert, Bertrand Rouet‐Leduc
2024-05-14

Sentinel-2methane emission detectionmethane point sourcesmultispectral satellite imageryvision transformer
Abstract Curbing methane emissions is among the most effective actions that can be taken to slow down global warming. However, monitoring emissions remains challenging, as detection methods have a limited quantification completeness due to trade-offs that have to be made between coverage, resolution, and detection accuracy. Here we show that deep learning can overcome the trade-off in terms of spectral resolution that comes with multi-spectral satellite data, resulting in a methane detection tool with global coverage and high temporal and spatial resolution. We compare our detections with airborne methane measurement campaigns, which suggests that our method can detect methane point sources in Sentinel-2 data down to plumes of 0.01 km 2 , corresponding to 200 to 300 kg CH 4 h −1 sources. Our model shows an order of magnitude improvement over the state-of-the-art, providing a significant step towards the automated, high resolution detection of methane emissions at a global scale, every few days.
1
A vision-transformer deep-learning method detects methane point sources in multispectral Sentinel-2 satellite imagery with global coverage and high temporal and spatial resolution.
2
Comparisons with airborne methane measurements indicate detection of plumes as small as 0.01 km², corresponding to sources emitting approximately 200–300 kg CH₄ h⁻¹.
3
The approach enables automated, high-resolution global monitoring of methane emissions approximately every few days. तर
4
The method addresses the spectral-resolution trade-off inherent in multispectral satellite data, improving methane detection capability without sacrificing broad coverage.
5
The model achieves an order-of-magnitude improvement over the state of the art for automated methane-emission detection.

Methane point-source emissions captured in Sentinel-2 multispectral satellite imagery

Automatic high-resolution detection and quantification sensitivity of methane plumes using deep learning, including coverage, spatial and temporal resolution

Publication Details
Publication Date
2024-05-14
Journal
Publisher
ISSN
Cited by
109
Access Type
Author Information
Authors
Claudia Hulbert
Bertrand Rouet‐Leduc
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%