Simple SummaryDefined criteria for the application of the CO2 balance method in a naturally ventilated barn provided reliable data. This specification enabled the acquisition and quantification of CH4 and NH3 in a naturally ventilated dairy barn, as well as detecting decreasing NH3 emissions affected by supplementing an Acacia mearnsii condensed tannin extract to a dairy cattle ration. Moreover, long-term measurements were possible and can be used to examine feed-related mitigation strategies at a barn level in the future.AbstractExtensive experimentation on individual animals in respiration chambers has already been carried out to evaluate the potential of dietary changes and opportunities to mitigate CH4 emissions from ruminants. Although it is difficult to determine the air exchange rate of open barn spaces, measurements at the herd level should provide similarly reliable and robust results. The primary objective of this study was (1) to define a validity range (data classification criteria (DCC)) for the variables of wind velocity and wind direction during long-term measurements at barn level; and (2) to apply this validity range to a feeding trial in a naturally cross-flow ventilated dairy barn. The application of the DCC permitted quantification of CH4 and NH3 emissions during a feeding trial consisting of four periods. Differences between the control group (no supplement) and the experimental group fed a ration supplemented with condensed Acacia mearnsii tannins (CT) became apparent. Notably, CT concentrations of 1% and 3% of ration dry matter did not reduce CH4 emissions. In contrast, NH3 emissions decreased 34.5% when 3% CT was supplemented. The data confirm that quantification of trace gases in a naturally ventilated barn at the herd level is possible.

The overall measurement of farm level greenhouse gas (GHG) emissions in dairy production is not feasible, from either an engineering or administrative point of view. Instead, computational model systems are used to generate emission inventories, demanding a validation by measurement data. This paper tests the GHG calculation of the dairy farm-level optimization model DAIRYDYN, including methane (CH₄) from enteric fermentation and managed manure. The model involves four emission calculation procedures (indicators), differing in the aggregation level of relevant input variables. The corresponding emission factors used by the indicators range from default per cow (activity level) emissions up to emission factors based on feed intake, manure amount, and milk production intensity. For validation of the CH₄ accounting of the model, 1-year CH₄ measurements of an experimental free-stall dairy farm in Germany are compared to model simulation results. An advantage of this interdisciplinary study is given by the correspondence of the model parameterization and simulation horizon with the experimental farm's characteristics and measurement period. The results clarify that modeled emission inventories (2,898, 4,637, 4,247, and 3,600 kg CO₂-eq. cow(-1) year(-1)) lead to more or less good approximations of online measurements (average 3,845 kg CO₂-eq. cow(-1) year(-1) (±275 owing to manure management)) depending on the indicator utilized. The more farm-specific characteristics are used by the GHG indicator; the lower is the bias of the modeled emissions. Results underline that an accurate emission calculation procedure should capture differences in energy intake, owing to milk production intensity as well as manure storage time. Despite the differences between indicator estimates, the deviation of modeled GHGs using detailed indicators in DAIRYDYN from on-farm measurements is relatively low (between -6.4% and 10.5%), compared with findings from the literature.