Deposition monitoring networks: what monitoring is required to give reasonable estimates of ammonia/ammonium?

Deposition is one of the main loss terms for ammonia and ammonium from the atmosphere. It is also the input for ecosystems that can lead to drastic changes and effects. Deposition networks are needed to evaluate the need and the effect of policies to reduce nitrogen emissions, but also for studying deposition parameters and for developing deposition models. As with ambient concentrations of ammonia, deposition, especially dry deposition, varies strongly in space and in time. Furthermore, the bi-directional surface-atmosphere exchange of ammonia makes the combination of ambient concentration measurements with inferential models inadequate. Developing deposition monitoring networks with reasonable accuracy and representativeness is therefore not straightforward. In Europe several projects have addressed deposition monitoring. From these results it is concluded that a monitoring strategy should consist of a network with a limited amount of super sites combined with a larger number of sites where low cost methods are applied, together with models for generalisation.     

Field intercomparison of precipitation measurements performed within the framework of the Pan European Intensive Monitoring Program of EU/ICP Forest

A 6-month field intercomparison study on precipitation measurements was performed at Schagerbrug near the west coast in the Netherlands. Twenty bulk sampling systems and two wet-only samplers were evaluated on accuracy, sampling strategy and performance under field conditions. Bulk precipitation fluxes of NO3-, NH4+, H+ and Kjeldahl-N generally could be determined with a greater accuracy than bulk precipitation fluxes of SO4(2-), Na+, Cl-, Mg2+, Ca2+, Alkalinity and H+. Bulk precipitation fluxes of K+ generally had the lowest accuracy. Only 20% of the sampling systems differed less than 10% from the best estimate, whereas most systems (60%) differed more than 20% from the best estimate. The inaccuracy induced by the sampling system appeared to be much larger than that resulting from the analysis of the samples by different laboratories as determined by ring-tests.