A Bayesian hierarchical model for methane emission source apportionment

Abstract

Reducing methane emissions from the oil and gas sector is a key component of short-term climate action. Emission reduction efforts are often conducted at the individual site-level, where being able to apportion emissions between a finite number of potentially emitting equipment is necessary for leak detection and repair as well as regulatory reporting of annualized emissions. We present a hierarchical Bayesian model, referred to as the multisource detection, localization, and quantification (MDLQ) model, for performing source apportionment on oil and gas sites using methane measurements from point sensor networks. The MDLQ model accounts for autocorrelation in the sensor data and enforces sparsity in the emission rate estimates via a spike-and-slab prior, as oil and gas equipment often emit intermittently. We use the MDLQ model to apportion methane emissions on an experimental oil and gas site designed to release methane in known quantities, providing a means of model evaluation. Data from this experiment are unique in their size (i.e., the number of controlled releases) and in their close approximation of emission characteristics on real oil and gas sites. As such, this study provides a baseline level of apportionment accuracy that can be expected when using point sensor networks on operational sites.

Figures (7)

• Figure 1: Aerial view of the METEC facility located in Fort Collins, Colorado. The five potential emission sources are identified with colored squares, and the locations of the 10 methane sensors are marked with pins. Pins with a gray interior show sensors that measure wind speed and direction in addition to methane concentrations.
• Figure 2: Example data from February 12, 2024. (a) True release data, with color corresponding to the emission source. (b) Minute-by-minute maximum of the concentration observations across the 10 CMS sensors. (c) Minute-by-minute median of the wind speed observations across the 8 sensors with anemometers. (d) Minute-by-minute circular median of the wind direction observations across the 8 sensors with anemometers.
• Figure 3: Methane concentration observations and concentration predictions from the Gaussian puff atmospheric dispersion model during one of the METEC controlled releases.
• Figure 4: Hierarchical structure and conditional independencies of the MDLQ model. Each circle represents a parameter, with fixed hyperparameters in gray and random variables in white. Arrows indicate when one parameter is a function of other parameters.
• Figure 5: Output from the MDLQ model during 5 days of the METEC controlled releases. (a) shows the true emission state and (b) shows the estimated emission state from the MDLQ model. Color corresponds to the emitting equipment, and the black line shows the true site-level total emission rate over time.