Temporal variability in emissions from oil and gas supply chains depends on the spatial scale at which emissions are aggregated. This work demonstrates a framework for simulating temporally and spatially resolved emission inventories that can be broadly applied in oil and gas production regions. Emissions of methane, ethane, volatile organic compounds (VOCs), and nitrogen oxides (NOxs) from oil and gas facilities in the Marcellus production region were estimated at a one-hour time resolution for the calendar year 2023 and were aggregated at the grid cell (4 km by 4 km), county, and basin level. Maximum to average emission rate ratios decreased as the scale of spatial aggregation increased and differed by pollutant. At the grid cell level, ratios of maximum to average emission rates exceeded 100 in some grid cells for VOCs. In contrast, basin level maximum to average ratios for NOx emission rates were less than 1.1. The sources driving temporal variability in hydrocarbon emissions were well completions and liquid unloadings, while the sources driving temporal variability in NOx emissions were preproduction activities such as drilling and hydraulic fracturing. Temporally and spatially resolved inventories can inform pollutant- and region-specific measurement campaigns and mitigation strategies. Reconciliation between inventories and observations must consider event frequency, duration, and persistence, along with the spatial scale and timing of measurements.
Temporal variability in emissions from oil and gas supply chains depends on the spatial scale at which emissions are aggregated. Emissions of methane, ethane, volatile organic compounds (VOCs) and nitrogen oxides (NOx) from oil and gas facilities in the Marcellus production region were estimated at a one-hour time resolution for calendar year 2023. Emissions from more than 200,000 well sites were included in the analyses and hourly emissions were aggregated at the grid cell (4 km by 4 km), county and basin level. Ratios of maximum to annual average hourly predicted emission rates were calculated at the grid cell, county and basin level for each pollutant. Maximum to average emission rate ratios decreased as the scale of spatial aggregation increased. At the grid cell level, ratios of maximum to average emission rates exceeded 100 in some grid cells for hydrocarbon emissions such as VOCs. In contrast, basin level maximum to average ratios for NOx emission rates were less than 1.1. The sources driving temporal variability for hydrocarbon emissions were well completions and liquid unloadings, while the sources driving temporal variability in NOx emissions were pre-production activities such as drilling and hydraulic fracturing. Accurately reconciling predicted emissions with measurements requires accounting for the timing, duration, and frequency of intermittent events, and depends on the type of measurements being made.
Emission inventories at fine spatial and temporal scales were developed for light alkanes, volatile organic compounds (VOCs), and nitrogen oxides (NOx) from upstream and midstream oil and gas operations in the Permian Basin oil and gas production region for 2022–2024. The inventories were spatially aggregated at basin, county, and 12 km by 12 km grid cell levels, and temporally resolved at hourly resolution, with underlying methods capable of generating inventories at other spatial and temporal scales. Spatial and temporal variability in emissions in the Permian were compared at various spatial scales with inventories for the Marcellus oil and gas production region, developed using the same methods. Emission sources that drive spatial and temporal variability differ by regional production characteristics, the level of spatial aggregation, and emitted species. Temporal variability in emissions decreases as the scale of spatial aggregation increases. Among counties with at least 10 active producing wells, maximum-to-annual-average hourly emission rate ratios reached 2.5 for methane, 2.8 for VOCs, and 2.3 for NOx. At the 12 km by 12 km grid cell level, the corresponding maximum ratios were 33.7, 26.5, and 13.9. These ratios illustrate the magnitude of short-term emission variability and the extent to which peak hourly emissions can exceed annual average estimates, with potential implications for episodic air-quality impact assessment. Compared with the gas-dominated Marcellus Basin, the oil-dominated Permian Basin shows lower temporal variability in hydrocarbon emissions due to fewer episodic gas production related sources (e.g., liquid unloadings) and a greater contribution from near-continuous oil production related sources (e.g., associated gas venting and tank flash). In contrast, NOx emissions exhibit higher temporal variability in the Permian due to more frequent preproduction activities associated with new well development. The spatially and temporally resolved emission inventories by source category and chemical species can be further combined with chemical transport modeling and air quality modeling to support assessment of regional air quality events, such as localized and episodic ozone formation.