[Paper Review] Global Tracking and Quantification of Oil and Gas Methane Emissions from Recurrent Sentinel-2 Imagery

Thibaud Ehret, Aurélien De Truchis+7 more|arXiv (Cornell University)|Oct 22, 2021

Atmospheric and Environmental Gas Dynamics33 citations

TL;DR

The paper presents a global methane-emission monitoring framework using recurrent Sentinel-2 imagery to detect and quantify over 1,200 oil-and-gas related plumes, linking Sentinel-2 results with Sentinel-5P and airborne data to establish a global power-law emission distribution.

ABSTRACT

Methane (CH4) emissions estimates from top-down studies over oil and gas basins have revealed systematic under-estimation of CH4 emissions in current national inventories. Sparse but extremely large amounts of CH4 from oil and gas production activities have been detected across the globe, resulting in a significant increase of the overall O&G contribution. However, attribution to specific facilities remains a major challenge unless high-resolution images provide the sufficient granularity within O&G basin. In this paper, we monitor known oil-and-gas infrastructures across the globe using recurrent Sentinel-2 imagery to detect and quantify more than 800 CH4 emissions. In combination with emissions estimates from airborne and Sentinel-5P measurements, we demonstrate the robustness of the fit to a power law from 0.1 tCH4/hr to 600 tCH4/hr. We conclude here that the prevalence of ultra-emitters (> 25tCH4/hr) detected globally by Sentinel-5P directly relates to emission occurrences below its detection threshold. Similar power law coefficients arise from several major oil and gas producers but noticeable differences in emissions magnitudes suggest large differences in maintenance practices and infrastructures across countries.

Motivation & Objective

• Motivate and enable global monitoring of methane emissions from oil and gas using recurrent, high-spatial-resolution satellite imagery.
• Develop automatic detection and quantification methods using SWIR bands sensitive to CH4 in Sentinel-2 and Landsat-8.
• Validate the Sentinel-2 based emissions against Sentinel-5P and airborne measurements and explore global emission distributions.

Proposed method

• Apply Beer-Lambert-based atmospheric attenuation modeling to detect methane in SWIR bands (B11, B12) from Sentinel-2 and Landsat-8.
• Compute a background methane-free model via linear regression on a time series of band ratios to detect anomalies.
• Use a two-step background estimation with outlier rejection and optional albedo-based clustering to improve plume SNR.
• Quantify plume methane by optimizing l_leak using band ratios and HITRAN-based A_CH4, then estimate source rate Q with the IME method and ERA5 wind data.
• Integrate Sentinel-2 results with Sentinel-5P and airborne campaigns to build and validate a global power-law emission model.

Experimental results

Research questions

• RQ1 Can recurrent Sentinel-2 imagery detect and quantify methane plumes from oil and gas infrastructure globally?
• RQ2 How do Sentinel-2-based detections compare with Sentinel-5P and airborne measurements for large and smaller emitters?
• RQ3 Is there a robust global power-law relationship for methane emissions that unifies data from Sentinel-2, Sentinel-5P, and airborne campaigns?
• RQ4 What is the recurrence pattern and geographic distribution of detected methane plumes across major oil and gas regions?

Key findings

• - Detected 1,202 methane plumes across 92 sites in Algeria, Turkmenistan, and the United States over 47 months.
• - 58% of plumes are recurrent events, implying ongoing emission activity rather than isolated incidents.
• - A global power-law relationship is supported by combining Sentinel-2, Sentinel-5P, and airborne data, with Sentinel-2 filling the mid-range (0.1–10 t CH4/h) and Sentinel-5P capturing ultra-emitters (>25 t CH4/h).
• - Permian Basin event (summer 2020) analyzed with Sentinel-2, Landsat-8, Sentinel-5P, and airborne data yielded 16,537±7,146 tons CH4, with start date inferred earlier than prior reports and cross-source agreement.
• - A large-scale dataset was compiled: ~7,000 sites, >1,248,621 pixels (tiles) processed over 562,652 km2, with a 10x10 km2 tile framework and per-tile quantifications.
• - The method provides a practical pathway to track global methane emissions and monitor progress via a simple power-law summary over time.