Is Optical Gas Imaging Effective for Detecting Fugitive Methane Emissions? - A Technological and Policy Perspective

Mitigating fugitive methane emissions from the oil and gas industry has become an important concern for both businesses and regulators. While recent studies have improved our understanding of emissions from all sectors of the natural gas supply chain, cost-effectively identifying leaks over expansive natural gas infrastructure remains a significant challenge. Recently, the Environmental Protection Agency (EPA) has recommended the use of optical gas imaging (OGI) technologies to be used in industry-wide leak detection and repair (LDAR) programs. However, there has been little to no systematic study of the effectiveness of infrared-camera-based OGI technology for leak detection applications. Here, we develop a physics-based model that simulates a passive infrared camera imaging a methane leak against varying background and ambient conditions. We verify the simulation tool through a series of large-volume controlled release field experiments wherein known quantities of methane were released and imaged from a range of distances. After simulator verification, we analyze the effects of environmental conditions like temperature, wind, and imaging background on the amount of methane detected from a statistically representative survey program. We also examine the effects of LDAR design parameters like imaging distance, leak size distribution, and gas composition. We show that imaging distance strongly affects leak detection - EPA's expectation of a 60% reduction in fugitive emissions based on a semi-annual LDAR survey will be realized only if leaks are imaged at a distance less than 10 m from the source under ideal environmental conditions. Local wind speed is also shown to be important. We show that minimum detection limits are 3 to 4 times higher for wet-gas compositions that contain a significant fraction of ethane and propane, resulting a significantly large leakage rate. We also explore the importance of `super-emitters' on the performance of an OGI-based leak detection program, and show that OGI technology can be used as an approximate leak-quantification method to selectively target the biggest leaks. Finally, we also provide recommendations and best-practices guidelines for achieving expected methane mitigation.