Large CH4 leak rates have been observed in the Uintah Basin of eastern Utah, an area with over 10,000 active and producing natural gas and oil wells. In this paper, we model CH4 concentrations at four sites in the Uintah Basin and compare the simulated results to in situ observations at these sites during two spring time periods in 2015 and 2016. These sites include a baseline location (Fruitland), two sites near oil wells (Roosevelt and Castlepeak), and a site near natural gas wells (Horsepool). To interpret these measurements and relate observed CH4 variations to emissions, we carried out atmospheric simulations using the Stochastic Time‐Inverted Lagrangian Transport model driven by meteorological fields simulated by the Weather Research and Forecasting and High Resolution Rapid Refresh models. These simulations were combined with two different emission inventories: (1) aircraft‐derived basin‐wide emissions allocated spatially using oil and gas well locations, from the National Oceanic and Atmospheric Administration (NOAA), and (2) a bottom‐up inventory for the entire U.S., from the Environmental Protection Agency (EPA). At both Horsepool and Castlepeak, the diurnal cycle of modeled CH4 concentrations was captured using NOAA emission estimates but was underestimated using the EPA inventory. These findings corroborate emission estimates from the NOAA inventory, based on daytime mass balance estimates, and provide additional support for a suggested leak rate from the Uintah Basin that is higher than most other regions with natural gas and oil development.