New top-down constraints on methane emissions from animal agriculture based on GEM airborne measurements in the US Upper Midwest

Agriculture and waste emissions make up the largest anthropogenic source of atmospheric methane (CH4) according to current inventories. However, such bottom-up emission estimates contain inherent uncertainties from extrapolating the limited number of in-situ measurements to larger scales. The US Corn Belt and Upper Midwest is a crucial region for the methane budget as one of the most intensive agricultural areas of the world, with a livestock population of >700 million including 19 million cattle and a majority of national swine operations. In this study, we present new airborne methane measurements from the Greenhouse Emissions in the Midwest (GEM) study, and apply the data to better quantify the methane flux from an array of major agriculture/waste point sources. In total the GEM flights targeted 9 of the largest concentrated animal feeding operations (CAFOs) in the region, plus 2 sugar processing plants that are also thought to be major methane sources. Facilities were revisited multiple times during summer (08/2017), winter (01/2018), and spring (05-06/2018) to better understand seasonal changes in emissions from these source categories. The results reveal significant temporal variability in facility-level emissions: top-down flux estimates derived from repeat visits in a single season sometimes varied by a factor of 2 or more. We further compare our top-down quantified fluxes with state-of-science bottom-up estimates using EPA methodology and informed by site-level information on animal population and management practices. Top-down fluxes are consistent with bottom-up estimates for beef CAFOs, but are significantly lower for dairies (by 40% on average) and for the sugar processing plants (by 80% on average). Results for swine facilities were more variable. The temperature dependent Van't Hoff-Arrhenius seasonality of methane emissions from manure management assumed in bottom-up inventories is not supported by the aircraft measurements, which has implications for seasonal source attribution based on atmospheric inverse modeling.