<p>Reactive nitrogen (N<sub>r</sub>) compounds comprise essential nutrients for plants. However, a large supply of nitrogen by fertilization through atmospheric deposition may be harmful for ecosystems such as peatlands and may lead to a loss of biodiversity, soil acidification and eutrophication. In addition, nitrogen compounds may cause adverse human health impacts. Large parts of N<sub>r</sub> emissions originate from anthropogenic activities. &#160;Emission hotspots of &#931;N<sub>r</sub>, i.e. the sum of all N<sub>r</sub> compounds, are related to crop production and livestock farming (mainly through ammonia, NH<sub>3</sub>) and fossil fuel combustion by transport and industry (mainly through nitrogen oxides, NO<sub>2 </sub>and NO). Such additional amount of N<sub>r</sub> will enhance its biosphere-atmosphere exchange, affect plant health and can influence its photosynthetic capacity. Therefore, it is necessary to thoroughly estimate the nitrogen exchange between biosphere and atmosphere.</p><p>For measuring the nitrogen mixing ratios a converter for reactive nitrogen (TRANC: Total Reactive Atmospheric Nitrogen Converter) was used. The TRANC converts all reactive nitrogen compounds, except for nitrous oxide (N<sub>2</sub>O), to nitric oxide (NO) and is coupled to a fast-response chemiluminescence detector (CLD). Due to a low detection limit and a response time of about 0.3s the TRANC-CLD system can be used for flux calculation based on the eddy covariance (EC) technique. Flux losses, which are related to the experimental setup, different response characteristics and the general high reactivity of most N gases and aerosols, occur in the high frequency range. We estimated damping factors of approximately 20% with an empirical cospectral approach.</p><p>For getting a reliable prediction of &#931;N<sub>r</sub> fluxes through deposition models, long-term flux measurements offer the possibility to verify the nitrogen uptake capacity and to investigate exchange characteristics of &#931;N<sub>r </sub>in different ecosystems.</p><p>In this study, we compare modelled dry deposition fluxes using the deposition module DEPAC (DEPosition of Acidifying Compounds) within the chemical transport model LOTOS-EUROS (LOng Term Ozone Simulation &#8211; EURopean Operational Smog) against &#931;N<sub>r</sub> flux measurements of the TRANC-CLD for a remote mixed forest site with hardly any local anthropogenic emission sources. This procedure allows to determine the background load and the natural exchange characteristics of nitrogen under low atmospheric concentrations. Therefore, the broad-scale dry deposition predicted directly by LOTOS-EUROS was compared to site-specific modelling results obtained using measured meteorological input data as well as the directly measured &#931;N<sub>r</sub> fluxes. In addition, the influence of land-use weighting in LOTOS-EUROS was examined. We further compare our results to &#931;N<sub>r</sub> deposition estimates obtained with canopy budget techniques. Measured &#931;N<sub>r</sub> dry deposition at the site was 4.5 kg N ha<sup>-</sup><sup>1</sup> yr<sup>-</sup><sup>1</sup>, in close agreement with modelled estimates using DEPAC with measured drivers (5.2 kg N ha<sup>-</sup><sup>1</sup> yr<sup>-</sup><sup>1</sup>) and as integrated in the chemical transport model LOTOS-EUROS (5.2 kg N ha<sup>-</sup><sup>1</sup> yr<sup>-</sup><sup>1</sup> to 6.9 kg N ha<sup>-</sup><sup>1</sup> yr<sup>-</sup><sup>1</sup> depending on the weighting of land-use classes).</p><p>Our study is the first one presenting 2.5 years flux measurements of &#931;N<sub>r</sub> above a remote mixed forest. Further verifications of long-term flux measurements against deposition models are useful to improve them and result in better understanding of exchange processes of &#931;N<sub>r</sub>.</p>
Spatial representativeness and uncertainty of eddy covariance carbon flux measurements for upscaling net ecosystem productivity to the grid scale
