Methane Anomaly Detector

The Environmental Protection Agency (EPA) estimates that methane is causing 25 times more damage to the atmosphere than carbon dioxide. The detection of large and frequent methane (CH4) leaks linked to oil and gas production has raised concerns in the ability of natural gas to effectively reduce greenhouse gas (GHG) emissions as a substitute to coal. Knowing when methane leaks occur is essential to reducing methane emissions, but collecting frequent, precise, on-ground methane measurements is expensive. Methane Anomaly Detector (MAD) identifies potential methane emission anomalies in California by applying deep learning techniques on data collected by Sentinel 5P, a satellite launched by the European Space Agency. Our web application showcases near real-time data that is easily accessible, offers personalized location search to view anomalies in specific areas, and enables insight-driven decisions from interactive visualizations. MAD enables local regulators and methane-emitting facility owners to identify abnormal methane levels in their region and kick off more in-depth investigations that aid in the reduction of harmful effects of methane.