High nitrogen load threats forest stand stability and affects subsequent function of the mountain forest over long time. The Orlicke hory Mts. are one of the most heavily nitrogen-polluted areas in the Czech Republic (HRTAKA et al. 2003; FOTTOVA 2003). The high deposition of nitrogen contributes to its accumulation in the forest ecosystems, which influences environment negatively (LUNDBORG 1997; FENN et al. 1998; PETERJOHN et al. 1999). The area of interest is threatened by nitrogen load particularly in the 5th, 6th and 7th forest vegetation zones (fir-beech, spruce-beech and beech-spruce). Nitrogen load poses also a risk for water and soil since the Orlicke hory Mts. have been declared a Protected Area of Natural Accumulation of Water (Czech acronym CHOPAV) for 39 years (see Czech Government Decree No. 40/1978 Coll.). Increased nitrogen load has been detected since 1990s. Assessing previously published studies, both national and from abroad ones, we hypothesized a way to reduce nitrogen load impact in the area of interest. One can summarize that substantially different views of mitigation of nitrogen load impact on the forest have been published (DE VRIES et al. 2012; KREUTZER et al. 2009). Despite missing signs of acute impacts on the Orlicke hory Mts. forests, increased nitrogen load poses a risk of further soil acidification and development of conditions supporting fungi infestation, and abiotic agents (frost, snow and wind) damage due to excessive growth and postponed maturing of new tissues. The primary objective of our study is to learn whether trend of nitrogen load (concentration, deposition) has been or has not been changed over long time. The secondary objectives are (1) nitrogen balance study within the UDL watershed and (2) nitrate content in surface waters of the area of interest. Air concentrations of NOx and atmospheric depositions of N-NO3, N-NH4 and total N were evaluated over long term. Time series began in 1995 (air concentrations) and 1993 (depositions). Data are from automated monitoring stations operated by Czech Hydrometeorological Institute. Atmospheric NOx concentrations were the highest in winter and the lowest in summer, amounting 10 mu g.m(-3) and 6 mu g.m(-3), respectively, at moderate decrease during the 2014-2016 period against the previous 2011-2013. This trend has been recently confirmed also by OULEHLE et al. (2016) via moderate declining N deposition. The UDL watershed was investigated by the Forestry and Game Management Research Institute - Opocno Research Station in cooperation with the Czech Geological Survey (GEOMON network). UDL deposition data were taken in open area, throughfall under young spruce, mature spruce and mature beech stands. UDL treatments were also used to monitor soil water saturation and nitrate saturation in water leaving the UDL watershed. To investigate nitrate concentrations in surface waters of the area of interest, samples were collected in October 1st 2015, May 11th 2016 and October 20th 2016. In total 23 streams and three springs were sampled (Table 1). Nitrogen (kg/ha) was shown separately according to compound origin, i. e. N-NH4 and N-NO3 deposited in open area, under forest stands and its export from the watershed (Fig. 7). N-NO3 deposition was greater compared to N-NH4. Nitrate N was the lowest alternately in the open area and in the watershed outlet water, higher under mature beech and the highest under mature spruce. 90% of the UDL watershed is a young spruce small-pole stand; this part of the forest cover, therefore substantially affects the N balance (Fig. 8) in the watershed. Exported stream runoff N was subtracted from throughfall N. The result could be considered the N consumed by the young spruces amounting 14.5 kg. ha(-1).year(-1). One can infer that the UDL is N-saturated and N consumption is in accordance with the development of the stand. Remaining N is exported to the stream water. Three times repeated sampling of surface waters did not reveal increased N concentrations (Fig. 9). More N-NO3 was in water from foothills compared to higher altitudes. The UDL watershed outlet water samples showed distinct differences in annual patterns of nitrate export. In dormant season, nitrate consumption by micro- and macro-vegetation component is low, stream water contains much more nitrate compared to summer (Fig. 10, 11). In winter hydrological half-year, a biforked (January, March) increased nitrate export is obvious (Fig. 11). Atmospheric N deposition differed obviously between both spruce-beech and spruce-open area treatments within UDL watershed. The high throughfall N deposition should be reflected in stream export and/or N soil pool. However, we found neither the former nor the latter. As the mature spruce and beech N consumption totals 13 kg.ha(-1) and 17 kg.ha(-1) respectively (see LOCHMAN, KANTOR 1985), we can suppose a great release of N due to denitrification, i. e. microbial reduction of NO3 into N(2)O( )and N-2. This raises another issue since N2O (80% is N) is considered a much more efficient greenhouse gas compared to CO2. UDL conditions seem to be supporting this process (see NUTTER, MORRIS 2004) because the watershed is a partly water-logged, drained site being high in carbon and "fertilized" with high N deposition. WEN et al. (2017) reported greater N2O emission from beech compared to spruce forest stand. From N balance point of view, atmospheric deposition and fixation in biomass and soil are the input processes, while extraction of wood and denitrification represent the export processes from forest ecosystems (VAN BREEMEN et al. 2002). LUNDBORG (1997) calculated that extraction of logging residues from N-polluted boreal spruce forests would decrease the N-lo ad amounting 1,000 kg.ha(-1) over 80-100 years of the rotation. Based on this, LUNDBORG (1997) proposed also a shorter rotation and low-emission combustion of the logging residues. Above-ground biomass should not be, however, extracted as whole tree biomass including removal of both slash higher in nutrients (KuNES et al. 2013) and forest floor. This practice would lead to depletion of the site over decades with subsequent low production of the forest (MusHiNsiu et al. 2017). The wood left on site becomes an environment where more N is fixed as wood-destroying and mycorrhizal fungi colonize it (RINNE et al. 2017). More N (PALVIAINEN et al. 2010a), P and Mg (PALVIAINEN et al. 2010b) were reported in rotting spruce stumps compared to birch ones. Low N concentrations in surface waters are attributable to restored consumption of N by forest ecosystems recovering from former air-pollution load (KoLidt et al. 2015). Higher export of nitrate in surface water were also related to loss of wood increment (KoLAft et al. 2015; GODEK et al. 2015). It is attributable to former malfunction of the roots. Similarly the intake of nitrate by roots changes in dormant season (ALEXANDER et al. 2007; ZAHORA et al. 2011; BETTEZ et al. 2015); our findings support that as well. Contemporary increased export of nitrate in water due to bark beetle outbreaks reported BEUDERT et al. (2015) and KOPACEK et al. (2017). The loss of roots' function led to three times higher nitrate concentrations in the Plane catchment-lake system; since 2011 the concentrations of nitrate decreased towards the level prior to forest death (KOPACEK et al. 2017). If nitrate concentration does not exceed 1 mg.l(-1), surface water can be considered clean (PITTER 1999). This was exceeded in all samples excepting the two spring water sites (5cest, VP). Despite some pollution, the all samples were relatively clean since the WHO limit for drinking water is 50 mg NO3 (i.e. 11.3 mg N-NO3 per liter).
