Emission and short range re-deposition of ammonia on grassland determined by an inverse dispersion method

Methods to measure gaseous emissions from spatially confined sources (e.g. from plots with slurry application or grazing excreta) often rely on atmospheric dispersion models to relate downwind concentration or flux measurements to the respective emission rate. Since ammonia (NH₃) is very reactive and water soluble, it adsorbs readily to surfaces leading to a significant loss due to re-deposition, which affects concentration and surface-atmosphere exchange measurements even at short distances from the source.

In the present study, a backward Lagrangian stochastic (bLS) model was used, which has been enhanced to incorporate dry re-deposition. Controlled release experiments were carried out at a grassland site with a parallel release of ammonia (NH₃) and methane (CH₄) at a defined release rate through a source grid with 36 critical orifices at ground level. NH₃ and CH₄ concentrations and vertical fluxes (by eddy covariance) were measured downwind of the source and the bLS model was used to infer the emission fluxes from the measured quantities, with and without considering dry re-deposition. When disregarding deposition, the median recovery fractions (ratios between the observed emission and the known release rate) were close to 100% for CH₄ but considerably lower (between 32% and 72%) for NH₃. The difference in the recovered fraction of NH₃ compared to CH₄ could be attributed to dry deposition loss between the source and the concentration measurements. In a second step,  surface deposition velocities were derived to match recovered fractions of NH₃ and CH₄. This resulted in median NH₃ deposition velocities (related to 2 m height) between 0.6 and 1.7 cm s^-1. Compared to literature values this is in the expected range of deposition velocities for grassland sites.