Possible malfunction in widely used methane sampler deserves attention but poses limited implications for supply chain emission estimates

Abstract

Estimates of methane emissions from natural gas production sites in the United States based on recent studies have been questioned due to possible malfunction of the Bacharach Hi Flow ® Sampler (BHFS), the primary measurement instrument used for two out of five source types examined in those studies (equipment leaks and chemical injection pumps). Without assessing whether the BHFS malfunction occurred in those studies, we constrain the possible underestimation of emissions associated with the BHFS-based results by excluding potentially affected measurements. Assuming leak emission rates are similar for sites with low and high methane content, U.S. methane emissions from equipment leaks and chemical injection pumps in recent studies could be underestimated by up to 40-80% due to a malfunctioning BHFS. We discuss uncertainties associated with this estimate. While a 40-80% underestimation is important when characterizing individual source categories, the potential implications are attenuated when aggregating emissions across the five sources examined in the recent studies (<12-24%), across all sources in the natural gas production segment (<7-14%), or across the entire supply chain (<2-5%). Therefore, potential errors caused by BHFS malfunction in recent studies would not substantially alter estimates of methane emissions from the U.S. natural gas supply chain. The possible malfunction of such a widely-used instrument deserves further attention to ensure that its use in quantifying emissions from individual supply chain components is not compromised.

Retracted: Methane emissions from the Marcellus Shale in southwestern Pennsylvania and northern West Virginia based on airborne measurements

“…The validity of the study by Allen et al . [] is still an ongoing point of contention [ Alvarez et al ., ]. Given the current low cost of natural gas, a purely economic driver to reduce the atmospheric release of CH 4 by leakage does not exist.…”

“…For example, Howard [2015a] and Howarth [2015] have argued that Allen et al [2013] might have underestimated CH 4 emissions and hence obtained an erroneously low CH 4 leakage rate. The validity of the study by Allen et al [2013] is still an ongoing point of contention [Alvarez et al, 2016]. Given the current low cost of natural gas, a purely economic driver to reduce the atmospheric release of CH 4 by leakage does not exist.…”

Temporal Variations in Methane Emissions from an Unconventional Well Site

“…22 However, a recent examination by Alvarez et al suggested that sensor issues may be a part of discrepancies, but that operator practices and general lack of complete data sets could also lead to discrepancies. 31 Therefore, we used data from Allen et al to compare with our emissions. 8…”

Methane Emissions from the Marcellus Shale in Southwestern Pennsylvania and Northern West Virginia Based on Airborne Measurements

“…For example, Howard (2015a) and Howarth (2015) have argued that Allen et al (2013) might have underestimated CH 4 emissions and hence obtained an erroneously low CH 4 emission rate. The validity of the study by Allen et al (2013) is still an ongoing point of contention (Alvarez et al, 2016(Alvarez et al, , 2018Guglielmi, 2018). It is noteworthy that there is considerable temporal variability in CH 4 emission rates in shale gas production regions, which could lead to variability in these estimates (Lavoie et al, 2017).…”