DRY DEPOSITION OF REDUCED NITROGEN

The paper reviews the results of measurements of ammonia (NH3) exchange and considers the application of this information in approaches to modelling atmospheric NH3 inputs to ecosystems. Ammonia dry deposition is seen to be highly dependent on land management and the magnitude of agricultural nitrogen inputs. Efficient dry deposition is observed to many semi-natural and forest ecosystems, with typical deposition velocities (V(d)) in the range 5-50 mm s-1. By contrast, intensively grazed pastures and fertilized arable croplands frequently show net NH3 emission, with fluxes dependent on the form of N input, environmental conditions (temperature, wetness) and plant growth stage. Other factors affecting the exchange of NH3 with ecosystems include the existence of an NH3 ''compensation point'' within plant tissues and interactions with acid gases such as SO2. Available information on NH3 exchange over the sea and aquatic water bodies suggests that both emission and deposition can occur depending on NH3 air concentrations and water compensation points. Approaches to mapping NH3 dry deposition include inferring deposition from monitored air concentrations and values of V(d) (''inferential technique''), and atmospheric transport models. The former may be more site specific, but is limited by lack of NH3 concentration monitoring, while the latter provides good regional coverage, but poor estimates of actual inputs to specific ecosystems (grids often > 50 km x 50 km). A possible advance is to apply the air concentration field outputs of large scale transport models, with locally derived V(d), though this fails to treat the significant local variability in NH3 emissions and air concentrations. An improvement would be the further development of local scale atmospheric transport models (e.g., 5 km x 5 km grids) incorporating land use information and ecosystem specific V(d) on a fine spatial scale.