Short-term methane emissions from 2 dairy farms in California estimated by different measurement techniques and US Environmental Protection Agency inventory methodology: A case study.

Reported estimates of CH emissions from ruminants and manure management are up to 2 times higher in atmospheric top-down calculations than in bottom-up (BU) inventories. We explored this discrepancy by estimating CH emissions of 2 dairy facilities in California with US Environmental Protection Agency (US EPA) methodology, which is used for BU inventories, and 3 independent measurement techniques: (1) open-path measurements with inverse dispersion modeling (hereafter open-path), (2) vehicle measurements with tracer flux ratio method, and (3) aircraft measurements with the closed-path method. All 3 techniques were used to estimate whole-facility CH emissions during 3 to 6 d per farm in the summer of 2016. In addition, open-path was used to estimate whole-facility CH emissions over 13 to 14 d per farm in the winter of 2017. Our objectives were to (1) compare whole-facility CH measurements utilizing the different measurement techniques, (2) compare whole-facility CH measurements to US EPA inventory methodology estimates, and (3) compare CH emissions between 2 dairies. Whole-facility CH estimates were similar among measurement techniques. No seasonality was detected for CH emissions from animal housing, but CH emissions from liquid manure storage were 3 to 6 times greater during the summer than during the winter measurement periods. The findings confirm previous studies showing that whole-facility CH emissions need to be measured throughout the year to estimate and evaluate annual inventories. Open-path measurements for liquid manure storage emissions were similar to monthly US EPA estimates during the summer, but not during the winter measurement periods. However, the numerical difference was relatively small considering yearly emission estimates. Manure CH emissions contributed 69 to 79% and 26 to 47% of whole-facility CH emissions during the summer and winter measurement periods, respectively. Methane yields from animal housing were similar between farms (on average 20.9 g of CH/kg of dry matter intake), but CH emissions normalized by volatile solids (VS) loading from liquid manure storage (g of CH per day/kg of VS produced by all cattle per day) at 1 dairy were 1.7 and 3.5 times greater than at the other during the summer (234 vs. 137 g of CH/kg of VS) and winter measurement periods (78 vs. 22 g of CH/kg of VS), respectively. We attributed much of this difference to the proportion of manure stored in liquid (anaerobic) form, and suggest that manure management practices that reduce the amount of manure solids stored in liquid form could significantly reduce dairy CH emissions.