Fluxes of ammonia over oilseed rape: Overview of the EXAMINE experiment

An integrated field experiment was organized as part of the EC EXAMINE project to quantify surface–atmosphere fluxes of ammonia (NH3). Fluxes of NH3, NH4+ aerosol, HNO3 and HCl were measured using the gradient method with several continuous and batch sampling systems. Within-canopy NH3 was determined to quantify the contribution of leaf cuticles, sub-stomatal apoplastic (intercellular) fluids and other sources and sinks to net fluxes. The campaign included the first field measurements of apoplastic pH and [NH4+], providing independent estimates of the stomatal compensation point (χs) for comparison with micrometeorological results. The latter also compared fluxes before and after cutting. Under the clean conditions of the experiment, [NH3][HNO3] and [NH3][HCl] were much less than values required for aerosol formation, while the NH4+ aerosol size distribution indicated that aerosol evaporation would be much slower than the time-scale of turbulent exchange. Nevertheless, the measurements suggest that aerosol production/growth as well as formation of HCl may have occurred within the canopy. Apoplastic [NH4+] and pH of leaves showed no diurnal patterns, with the main control on χs being the temperature dependence of the solubility equilibria. Fluxes of NH3 were bi-directional (−200 to 620 ng m−2 s−1), with deposition generally occurring when the canopy was wet and emission when it was dry, particularly during the day. Nocturnal emissions indicated a non-stomatal NH3 source, while daytime emissions were larger than indicated by the apoplastic estimates of χs. The within-canopy data, together with an inverse Lagrangian source–sink analysis, showed decomposing litter to be a significant NH3 source, explaining nocturnal NH3 emissions and larger emissions following cutting. Daytime net fluxes were controlled by the top part of the canopy, due to χs for siliques apparently being larger than for leaves. The measurements have been used to develop multi-layer resistance models of NH3 exchange. A 3-layer ‘foliage–litter–silique model’ provides a detailed mechanistic treatment of the component fluxes, while a 2-layer ‘foliage–litter model’ is better suited to generalization in atmospheric transport models. Application of the new models should help improve estimates of regional atmospheric ammonia budgets.