Analyzing methane emissions from the San Joaquin Valley, California using combined airborne and tower measurements

California is committed to reduce methane (CH₄) emissions to 1990 levels by 2020 through the California Global Warming Act of 2006 (AB-32). Because of this, the California Air Resources Board (CARB) compiles State-wide emission inventories of greenhouse gases, including CH₄ and carbon dioxide (CO₂); the two most important greenhouse gases. In California the major anthropogenic CH₄ emission sources are ruminant livestock, landfills, wastewater treatment, losses during oil and gas extraction and transmission, and combustion. Emission inventories and ecosystem models provide valuable estimates of spatial and temporal distributions of CH₄. However, uncertainty remains because many factors controlling emissions are poorly quantified. Atmospheric measurements and inverse modeling provide an independent method to quantify local and regional CH₄ emission estimates. In order to assess the spatial variation of CH₄ and CO₂ emissions and to identify local “hot spots,” airborne measurements of CH₄ and CO₂ were made by the Alpha Jet Atmospheric eXperiment (AJAX) over the San Francisco Bay Area (SFBA) and San Joaquin Valley (SJV) in January and February 2013 during the Discover-AQ-CA study. High mixing ratios of CH₄ (> 2 ppb) were observed in-flight with a high degree of spatial variability. CH₄ and CO₂ measured from tall towers are analyzed providing information on the temporal variability and comparisons to airborne data. To provide an independent method to quantify CH₄ emissions, the GEOS-Chem (v9-01-03) chemical transport model (CTM) is used to simulate SFBA/SJV CH4 measurements. The North America nested-grid version of GEOS-Chem utilizes CH₄ emissions from EDGAR 4.2, GFEDv3, and a wetland emission parameterization. A comparison between observations and emissions from GEOS-Chem and CALGEM emission maps are made along with a discussion of CH₄ and CO₂ source apportionment and the use of GEOS-Chem-Adjoint to perform inversion modeling.