Off-Site Flux Estimates of Volatile Organic Compounds from Oil and Gas Production Facilities Using Fast-Response Instrumentation

Abstract

Flux estimates of volatile organic compounds (VOCs) from oil and gas (O&G) production facilities are fundamental in understanding hazardous air pollutant concentrations and ozone formation. Previous off-site emission estimates derive fluxes by ratioing VOCs measured in canisters to methane fluxes measured in the field. This study uses the Environmental Protection Agency’s Other Test Method 33A (OTM 33A) and a fast-response proton transfer reaction mass spectrometer to make direct measurements of VOC emissions from O&G facilities in the Upper Green River Basin, Wyoming. We report the first off-site direct flux estimates of benzene, toluene, ethylbenzene, and xylenes from upstream O&G production facilities and find that these estimates can vary significantly from flux estimates derived using both the canister ratio technique and from the emission inventory. The 32 OTM 33A flux estimates had arithmetic mean (and 95% CI) as follows: benzene 17.83 (0.22, 98.05) g/h, toluene 34.43 (1.01, 126.76) g/h, C8 aromatics 37.38 (1.06, 225.34) g/h, and methane 2.3 (1.7, 3.1) kg/h. A total of 20% of facilities measured accounted for ∼67% of total BTEX emissions. While this heavy tail is less dramatic than previous observations of methane in other basins, it is more prominent than that predicted by the emission inventory.

Mobile Measurement System for the Rapid and Cost-Effective Surveillance of Methane and Volatile Organic Compound Emissions from Oil and Gas Production Sites

“…29 In contrast, the arithmetic mean (95% CI) of BTEX emission is 0.4 (0.1−0.6) kg/h, which is much higher than the arithmetic mean of 0.09 (0.003−0.38) kg/h found in UGRB. 13 We hypothesize that the great discrepancy is partly caused by the presence of several large BTEX emitters found in this study since we focused on the top ∼50% emitters. Previous work has found that the mean methane emissions (both absolute and productionnormalized) were higher in the Eagle Ford Basin than that in the UGRB.…”

“…This is slightly more skewed than the observed distribution in the UGRB, where the top 20% of sites were responsible for 67% of BTEX emissions. 13 Again, the fact that this study focused on the top ∼50% emitters would naturally reduce the skewness as compared to an unbiased sampling. More importantly, an overlap was found between large emitters, such that the top 20% well pads ranked according to methane emissions were responsible for 79% of total VOCs (C4−C12), 78% BTEX, and 60% methane emissions.…”

“…29 Such a paired approach relies on favorable wind conditions and is most suited to situations where the methane and VOCs are emitted from the same source, which may not always apply to O&G production wells. 12,13 This study addresses the development and demonstration of a ground-based mobile measurement system that can perform rapid emission surveillance of both methane and VOCs from O&G well pads. Real-time measurements of methane and total nonspeciated VOCs were made by an advanced methane analyzer and a photoionization detector (PID), respectively.…”

“…In 2019, the U.S. EPA rescinded the methane requirement in the policy amendment, arguing that regulations for VOCs can simultaneously reduce methane emissions . Some studies have found that methane and VOCs may originate from different sources onsite a O&G production well pad (e.g., separator and storage tank). , Therefore, the effectiveness of achieving methane and VOC emission reductions by only regulating VOC emissions remains to be examined from additional field measurements. Also, only regulating VOC emissions will be ineffective to reduce methane emissions from dry gas wells.…”

Advanced Leak Detection and Quantification of Methane Emissions Using sUAS

“…where C p is the peak concentration from the Gaussian fit, U m is the mean wind speed, σ z and σ y are the vertical and lateral plume dispersion that can be determined from the meteorological conditions, such as the Pasquill-Gifford stability classification curves [128] (see Figure 5). The accuracy of the OTM33A method is explored in [129,130].…”