Comparison of two micrometeorological and three enclosure methods for measuring ammonia emission after slurry application in two field experiments

Samenvatting

Ammonia emission following field application of animal slurry is a significant problem for the environment and human health. Accurate emission measurements are crucial for inventories, research, and mitigation. However, there may be large differences between results obtained with different methods. In this study measurement methods were compared in two field experiments: in Denmark (I-AU, trailing hose application, summer, arable land) and the Netherlands (II-WUR, slurry shallow injection, autumn, grassland) over 7 days each. Two micrometeorological methods (Integrated Horizontal Flux (IHF) and backward Lagrangian stochastic (bLS)) and three enclosure methods (Dräger tube method (DTM), wind tunnels (WT), and dynamic flux chambers (FC)) were included in one or both. Measuring in parallel eliminated effects of local factors influencing emission. Relative systematic error in micrometeorological methods (bLS variants and IHF) was estimated from measurements as about 25 % as a standard deviation among methods based on random-effects models. DTM emission measurements were lower than other methods by as much as 34 % of applied TAN compared to bLS. The emission rate measured by IHF followed the same pattern as the other methods soon after slurry application, but total emission was lower (5 % of applied TAN lower than bLS). Different concentration measurement methods used with bLS showed differences of 1–13 % of applied TAN. FC emission was 9–15 % of applied TAN higher than IHF and bLS, but 13 % lower than WT. WT emissions were high and depended on the air exchange rate. Overall relative uncertainty in total emission measured with micrometeorological methods was estimated at 24 and 31 % of the measured value (standard deviation), implying a 95 % confidence interval of about 60 %-160 % of emission measured in a single plot using a micrometeorological method.