A Data-Driven Modeling Approach for Estimating Methane Emission Sources from Satellite, Environmental, and Oil/Gas Well Data

There has been an increased commitment to capping GHG emissions from oil/gas (O/G) operations. However, reliable monitoring and quantification of methane emissions from particular sources are difficult for several reasons: (1) high-resolution sampling data over an extensive geographical area is rare; (2) identifying emission sources from large-scale monitoring satellite data is extremely challenging, often yielding inconsistent and highly varying results.
Satellite retrievals of near-surface concentrations of atmospheric methane, together with other environmental (weather), and oil/gas operational/production data, can be used to infer potential emission sources. This project aims to develop practical workflows for quantifying methane emission sources from O/G operations by (1) building machine-learning models for correlating production data and well activity to methane emission rates, and (2) understanding the relationships between O/G activities and emission source characteristics. The outcomes would be useful for identifying better monitoring opportunities.

Jeffrey Bian
Faculty Supervisor: 
Juliana Leung
Partner University: