Lawrence Berkeley National Laboratory

Methane emissions reduction has become a high priority due to its tractability relative to other greenhouse gases, and the potential for climate and air-quality benefits. However, regional methane sources have been difficult to ascertain, with discrepancies between top-down and bottom-up estimates. We combined mobile surveys with continuous atmospheric measurements at two fixed locations (one urban and one rural) in the southern San Joaquin Valley of California, USA, using ethane:methane (C2:C1) content and stable carbon isotope ratio (δ13CH4) as tracers for source type. In 6483 km of road surveys conducted over five field campaigns, 108 methane source events from 74 unique locations were sampled. C2:C1 was characteristically low for dairies and wastewater facilities, but was also near zero for many oil and gas sources, notably within a northeastern arc of high-density oil and gas fields. Natural gas sources within the Bakersfield urban area fell consistently within 2.3–4.2 % C2:C1. δ13CH4 was effective at separating dairy (−51.6 ± 3.1 ‰), wastewater (−45.4 ± 3.2 ‰), and thermogenic natural gas (−42.9 ± 1.8 ‰) sources. Oil and gas production sources were less consistent, with 50 % of observations within −45.8 to −42.7 ‰ (median: 43.8 ‰), but with the largest spans of δ13CH4 and C2:C1. The urban location experienced frequent methane enhancements attributed to both thermogenic and biogenic origins, whereas methane enhancements at the rural location enhancements were almost exclusively attributed to biogenic sources. Our results demonstrate considerable utility of diagnostic ethane and δ13CH4 measurements in partitioning several methane source types, with the caveat that oil and gas sources may contain wide-ranging C2:C1 and δ13CH4 due to thermal maturities and secondary processes (microbial and oxidation).