Modeling methane emissions from US natural gas operations: national gathering station emission factor development and facility/regional-scale top-down to bottom-up reconciliations
Date
2017
Authors
Vaughn, Timothy L., author
Marchese, Anthony J., advisor
Yalin, Azer P., advisor
Olsen, Daniel B., committee member
Opsomer, Jean D., committee member
Journal Title
Journal ISSN
Volume Title
Abstract
United States natural gas dry production increased by 47% between 2005 and 2015 due to the widespread use of horizontal drilling and hydraulic fracturing to extract gas from shale and other tight formations. Natural gas production and consumption is projected to continue to increase for the foreseeable future. In 2016, the natural gas supply chain delivered 29% of the energy used in the U.S., and natural gas surpassed coal as the leading electricity generating source for the first time in U.S. history. When combusted, natural gas produces less CO2 per unit energy released compared to coal or petroleum. However, uncombusted methane (the primary component of natural gas) has a global warming potential 30 times higher than CO2 on a 100 year time horizon (including oxidation to CO2, but excluding climate-carbon feedbacks). Therefore, the net greenhouse gas impacts resulting from displacement of coal and petroleum by natural gas depend on the emission rate of uncombusted natural gas. Short term climate benefits resulting from coal substitution, for example, are lost if the net rate of methane (CH4) emission from the natural gas supply chain exceeds 3—4% . Three studies were conducted to quantify CH4 emissions from the natural gas industry. In particular, these studies focused on quantifying emissions from the gathering and processing sector and reconciling emissions estimates developed using top-down (tracer flux and aircraft) vs. bottom-up (on-site component-level) measurement approaches. In the first study, facility-level CH4 emissions measurements were made at 114 natural gas gathering facilities and 16 processing plants in 13 U.S. states during a 20-week field campaign conducted from October 2013 through April 2014. Measurement results were combined with facility counts obtained from state air permit databases and national inventories in a Monte Carlo simulation to estimate CH4 emissions from U.S. natural gas gathering and processing operations. Annual CH4 emissions from normal operations at gathering facilities totaled 1699 Gg (95% CI=1539—1863 Gg), while normal operations at processing plants totaled 505 Gg (95% CI=459—548 Gg). CH4 emissions from abnormal operations at gathering facilities were estimated in a separate Monte Carlo simulation based on field observations and a sub-set of field measurements. These emissions totaled 169 Gg (+426%/-96%). In the second study, coordinated dual-tracer, aircraft-based, and direct component-level measurements were made at midstream natural gas gathering and boosting stations in the Fayetteville shale in Arkansas, USA. On-site component-level measurements were combined with engineering estimates to generate comprehensive facility-level CH4 emission rate estimates ("study on-site estimates (SOE)") comparable to tracer and aircraft measurements. Concurrent measurements at 14 normally-operating facilities showed a strong correlation between tracer and SOE, but indicated that tracer measurements estimated lower emissions (regression of tracer to SOE=0.91 (95% CI=0.83—0.99, R2=0.89). Tracer and SOE 95% confidence intervals overlapped at 11/14 facilities. Contemporaneous measurements at six facilities suggested that aircraft measurements estimated higher emissions than SOE. Aircraft and study on-site estimate 95% confidence intervals overlapped at 3/6 facilities. In the third study, a detailed spatiotemporal inventory model was developed and used to reconcile top down and bottom-up CH4 emission estimates from natural gas infrastructure and other sources in the Fayetteville shale on two consecutive days. On Thursday October 1, 2015 13:00—15:00 CDT top-down aircraft mass balance flights estimated 28.7 (20.1—37.3 Mg/h 95% CI) from the study area, while the bottom-up ground level area estimate predicted 23.9 (20.9—27.3 Mg/h 95% CI). On Friday October 2, 2015 14:30—16:30 CDT top-down estimated 36.7 (21.3—52.1 Mg/h 95% CI), while bottom-up estimated 21.1 (18.4—24.2 Mg/h 95% CI). Production and gathering activities were the largest contributors to modeled CH4 emissions. In contrast to prior studies, comparisons on two consecutive days indicated overlapping confidence intervals between top-down aircraft estimates and bottom-up inventory-driven estimates. Operator participation and extensive activity data proved critical in understanding emissions as observed by aircraft. In particular, the agreement obtained was possible only because bottom-up models included the variability in production maintenance activities, which showed substantially higher emissions during daytime hours when aircraft-based measurements were performed. Results indicated that that poor activity estimates (counts and timing) for large episodic events likely drives divergence in CH4 emission estimates from production basins, and that even more precise activity data would be required to improve agreement between these two approaches.
Description
Rights Access
Subject
modeling methane emissions