CEE Seminar - Sensing and Analytics Approaches for Fugitive Methane Emissions

This talk addresses scientific and technical challenges to identifying and quantifying fugitive methane emissions from the production, transmission and end use of natural gas. We will examine a Bayesian inference approach built around turbulent transport theory applied to stationary and mobile measurements of methane concentration. Additionally, an information-theoretic sensor path-planning algorithm will be demonstrated for scaling the approach from the single well to the region or enterprise. It is based on a principle of maximum expected entropy reduction in the fugitive source distribution, conditioned on wind fields, associated dispersion, and prior sensor measurements.