Characterization of Algerian Oil and Gas Methane Emission Point Sources from Satellites to Drive Mitigation Actions

Methane (CH₄) emissions from point sources in the oil and gas sector are highly heterogeneous in space and time, with emission patterns ranging from short-lived events to long-lasting, persistent sources. High-resolution satellite observations provide a unique capability to systematically detect, classify, and monitor large emissions at the facility level, and to explore links between emission behaviour, infrastructure characteristics, and potential mitigation outcomes. 

In this contribution, we will present the analysis of methane point source emissions in the Algerian oil and gas sector using satellite observations from UNEP’s International Methane Emissions Observatory (IMEO) Methane Alert and Response System (MARS). The study combines hyperspectral and multispectral high spatial resolution satellite data from January 2024 to December 2025 and analyses more than 150 emission point sources with large emissions during this time frame. The source analysis includes a classification by facility type (e.g. flares, gas disposal facilities, pipelines) and facility age based on the historical satellite imagery and visual inspection of high-resolution RGB data. This facility classification aims to assess potential relationships between infrastructure characteristics and emission behaviour. 

Given the differences in sensitivity, noise, revisit frequency, etc., of the satellites considered in this study, we investigate detection patterns across satellite types and find systematic differences between hyperspectral-only and multispectral detections. Sources detected exclusively by hyperspectral instruments are associated with lower estimated flux rates, sporadic emissions, or environmental and operational conditions that worsen the detection limits of multispectral sensors. In contrast, single-plume detections captured by multispectral satellites tend to correspond to big, short-duration emission events. Regarding emissions from flares, detections with multispectral instruments are limited to cases where flares are unlit and methane is vented, since active flaring and smoke significantly degrade methane retrievals, preventing the detection of emissions from incomplete combustion. 

Another parameter that is specifically analysed is the duration of emissions and monitoring of their status after MARS notification. While all emissions are important, those identified as long-duration (several days of emission) or frequent (e.g., a flare that repeatedly goes out and vents) are targeted for urgent mitigation recommendations by MARS in its engagement process and monitoring of the potential effect of its notifications when the emission cessation happens. 

Overall, this work demonstrates the value of an integrated classification framework that combines facility type, emission persistence, and multi-sensor satellite observations. Such an approach improves interpretation of satellite-derived methane detections and supports prioritisation of mitigation efforts in the oil and gas sector.