cd /news/machine-learning/methane-hotspot-explorer · home › topics › machine-learning › article
[ARTICLE · art-158993] src=atmosphere.copernicus.eu ↗ pub= topic=machine-learning verified=true sentiment=· neutral

Methane Hotspot Explorer

The Copernicus Atmosphere Monitoring Service (CAMS) released the Methane Hotspot Explorer, a weekly-updated world map of large methane plumes detected from Copernicus Sentinel-5P TROPOMI satellite data at up to 7 x 5.5 km² spatial resolution. The plumes are identified by a machine learning approach trained on methane scenes with known emission plumes, and each detection is tagged with an estimated source category — coal, oil, gas or solid waste — drawn from bottom-up emission inventories. CAMS cautions that the kilometer-scale pixels cannot identify specific facilities as sources and that detected plumes may reflect transient events that have drifted from their origin.

read1 min views1 publishedOct 27, 2026

The CAMS Methane Hotspot Explorer shows large methane plumes from strongly emitting localised sources detected using the operational Copernicus Sentinel-5P TROPOMI methane satellite data. The TROPOMI instrument provides daily global coverage of methane concentrations at up to 7 x 5.5 km<sup>2</sup> spatial resolution. These observations allow us to detect large methane plumes around the world, which are provided here on a weekly basis. The methane plumes are detected using a machine learning approach that has been trained based on methane scenes with known emission plumes.

The world map shows the estimated source origins of detected plumes based on the most upwind TROPOMI pixel included in the plume. These locations come with large uncertainty as the kilometer-scale resolution of the individual TROPOMI observations (pixels) does not allow the identification of specific facilities as the source of the emissions. Additionally, the detected plume may be related to a transient emission event and could therefore have drifted away from the source.

The viewer also provides an estimated source category based on the most prominent emission source category in the area where the plume has been detected based on bottom-up emission inventories. For this classification, only source categories that are regularly leading to large methane plumes are considered: coal, oil, gas, and solid waste.

Detected plumes are uploaded on a weekly basis. The number of detections fluctuates from week to week because of varying emissions, cloud cover, and viewing geometry. The explorer only shows the clearest and most confident plumes. The detections exclude larger-scale spread-out enhancements such as seen over wetland areas.

── more in #machine-learning 4 stories · sorted by recency
── more on @copernicus atmosphere monitoring service 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/methane-hotspot-expl…] indexed:0 read:1min 2026-10-27 · —