We studied atmospheric methane observations from November 2016 to October 2017 from one rural and two urban towers in the Baltimore-Washington region (BWR). Methane observations at these three towers display distinct seasonal and diurnal cycles with maxima at night and in the early morning, reflecting local emissions and boundary layer dynamics. Peaks in winter concentrations and vertical gradients indicate strong local anthropogenic wintertime methane sources in urban regions. In contrast, our analysis shows larger local emissions in summer at the rural site, suggesting a dominant influence of wetland emissions. We compared observed enhancements (mole fractions above the 5th percentile) to simulated methane enhancements using the WRF-STILT model driven by two EDGAR inventories. When run with EDGAR 5.0, the low bias of modeled versus measured methane was greater (ratio of 1.9) than the bias found when using the EDGAR 4.2 emission inventory (ratio of 1.3). However, the correlation of modeled versus measured methane was stronger (~1.2 times higher) for EDGAR 5.0 compared to results found using EDGAR 4.2. In winter, the inclusion of wetland emissions using WETCHARTs had little impact on the mean bias, but during summer, the low bias for all hours using EDGAR 5.0 improved by from 63 to 23 nanomoles per mole of dry air or parts per billion (ppb) at the rural site. We conclude that both versions of EDGAR underestimate the regional anthropogenic emissions of methane, but version 5.0 has a more accurate spatial representation.