Skip to content
BY-NC-ND 4.0 license Open Access Published by De Gruyter Open Access September 7, 2018

Assessment of runoff nitrogen load reduction measures for agricultural catchments

Marta Martínková , Tomáš Hejduk , Petr Fučík , Jan Vymazal and Martin Hanel EMAIL logo
From the journal Open Geosciences


Water quality in rural catchments is influenced by many societal and bio-physical factors (e.g. different pollution sources, land use and land cover changes). Good ecological status and surface water quality are currently challenged mainly due to different poorly identified pollution sources. The main objective of this study is to estimate the potential of different measures (land use changes and/or reduction in point sources) and their combinations in decreasing the nitrate-nitrogen load from Jankovský stream catchment. The eco-hydrological model SWIM, which simulates dynamics of nutrients in a catchment was used in the study. The simulations for scenario measures showed that nitrate-nitrogen loads at the outlet can be decreased more by reduction of municipal nitrate-nitrogen sources rather than by agricultural land-use change. Overall, the modeling results demonstrated that the most effective scenario was the combination of total reduction of municipal nitrate-nitrogen sources and agricultural land-use change.

1 Introduction

The issue of water quality in rural catchments is of global importance [1, 2, 3]. Especially, the significance of groundwater for safety of water supply is increasing with regard to global climate change and rapid development in rural areas [4, 5]. Notably, rural catchments are heavily influenced by direct anthropogenic impacts from point and non-point nutrient sources. The European Water Framework Directive (WFD) focuses on integrated water management in river catchments to achieve, enhance or maintain a good status of fresh waters [6]. Achieving good ecological status is essential aspect of this. Good ecological status and surface water quality are currently challenged mainly due to different (often poorly identified) pollution sources. Nevertheless, agriculture is traditionally supposed to be the most important source of nutrients emitted to waters [7].

Nitrate pollution is a particular issue in the rural catchments globally: nitrogen as a major plant nutrient is often applied in large amounts at agriculture land to obtain optimal yields. Moreover, the municipal sources from waste water treatment plants (WWTP), which are monitored, and from scattered sources (which are not monitored - mainly households without connection to municipal sewerage networks) contribute importantly to the overall nitrogen load. Furthermore, atmospheric deposition can also influence the leaching of nitrogen (N) significantly [8].

Constructed wetlands (CWs) are supposed to reduce nitrogen load from municipal scattered and point N sources; change in use of agriculture land is supposed to reduce nitrogen load from non-point sources. While there is a number of studies quantifying the land use change effects on water quality, the combined assessment of the effects of land use change and constructed wetlands has (to our knowledge) not been published before.

CWs represent promising alternative to traditional waste water treatment: they are relatively inexpensive to construct and operate; they have significant environmental and aesthetic benefits and are perceived recently as a reliable wastewater treatment technology [9]. Moreover, their important advantage regarding the climate variability and change impacts is the relative robustness against flooding. The CWs are undergoing rapid development recently and also the intensified CW designs and methods removing more “exotic” pollutants have been presented recently [10].

The estimation of contribution of different sources to overall load can be carried out using numerical models. Additionally, the numerical models allow estimation of the efficiency of measures to decrease the contaminant load. A broad spectrum of diverse approaches for modeling of water quality is available: from simple models based on statistical analysis and conceptual water quality models to physically-based models describing the processes in the catchment in a different level of detail.

The territory of the Czech Republic is an representative example of development of different nitrates sources. The impact of agriculture as one of the major sources of nitrates in the Czech Republic declined: the consumption of nitrogen fertilizers sharply decreased in the beginning of 1990s (approximately 50% reduction in N fertilization) [11]. A decrease in nitrate concentrations in surface water in some regions of the Czech Republic was observed and it was partly attributed to lower nitrogen fertilizer doses as well as due to grassing of ploughland especially in submontane or areas less suitable for intense agriculture [12]. On the other hand, the importance of non-agricultural sources such as scattered municipalities is still significant in the Czech Republic. Such sources constitute considerable part of the total nitrogen load, as was confirmed by the assessment of surface water quality [13]. The possible reasons for this situation are mainly construction and operational costs for wastewater treatment and/or construction of unified (combined) sewerage systems without connection to WWTP.

The motivation for this study was to assess the potential of feasible protective measures using simple yet robust mathematical modeling. Consequently, our results allow to formulate recommendations to improve the status of surface water in the catchment, which is important for drinking water supply. The novelty of this study is to address the mature issue of searching for a solution of N-NO3 loads from semi-gauged catchments, propose feasible measures and simultaneously test the potential of these measures. The results will be transferable for catchments with analogical properties. Moreover, such objective is in accordance with integrated approach of WFD and indirectly contributes to its target, i.e. good ecological status of surface water.

The Soil and Water Integrated Model (SWIM) has been already successfully implemented for various types of catchments from meso- to macro-scale [8, 14].In this study, we implemented SWIM for estimation of the effect of protective measures in the catchment of the Jankovský stream (Czech Republic, Figure 1). It is a rural catchment with agricultural land and considerable anthropological N-NO3 sources, which are not routinely monitored. It is situated within the catchment area of the Švihov Reservoir, which is an important source of drinking water for the capital of the Czech Republic (Prague).

Figure 1 The Jankovský stream catchment.
Figure 1

The Jankovský stream catchment.

The main objective of this study is to estimate the potential of different measures and their combinations in decreasing the nitrate-nitrogen (N-NO3) from a rural semi-gauged catchment. Specifically, we tested the potential of constructed wetlands and change in use of agriculture land on total N-NO3 load from Jankovský stream catchment.

2 Materials and Methods

2.1 SWIM

SWIM is a dynamic process-based eco-hydrological numerical model developed on the basis of the models SWAT [15] and MATSALU [16]. Up until now, SWIM has been intensively tested, validated and applied for simulating water discharge and water quality including hydrological extremes, nutrient dynamics and erosion [14].

SWIM uses the following climate variables in daily time step: maximum, minimum and mean temperature, precipitation, solar radiation and air humidity.

In this study we used the version of SWIM, where all processes are simulated by disaggregating a catchment into subcatchments and hydrotopes. The hydrotopes are sets of elementary units in a subcatchment with homogeneous soil and land use types. It is assumed that a hydrotope behaves uniformly regarding hydrological processes and nutrient cycles. The simulation of water and nutrient cycling starts from hydrotopes, then water and nutrient fluxes are aggregated for subcatchments and routed via river network to the catchment outlet.

Each soil type is described by numerical ID, number of soil layers, thickness, geophysical (clay, silt and sand content, bulk density, porosity, available water capacity, field capacity, saturated conductivity) and geochemical properties (organic C and N content) of soil layers. Soil types recognized for Jankovský stream catchment have 4 layers.

The Muskingum flow routing method is used in the SWIM. Two parameters are defined: the storage time constant for the reach (alternatively an average travel time for a flood wave) and dimensionless weighting factor in river reach routing that indicates the relative importance of the input and outflow in determining the storage in a reach. Deposition and remobilization in the streams are two components in the sediment routing model; they operate simultaneously. N-NO3 is considered as a conservative material in the model for the time of an runoff model. It is routed by adding contributions from all subcatchments to determine the total load for the catchment.

The hydrological system is represented by four slots in the SWIM: the soil surface, the root zone, the shallow aquifer and the deep aquifer. Processes considered for the root zone are surface runoff, infiltration, evapotranspiration, percolation and interflow. Hydrological processes in the aquifer zone are: groundwater recharge, capillary rise to the root zone, lateral flow and percolation to the deep aquifer.

The nutrient modules of SWIM contain pools of active and stable phases, inorganic and organic phases and nutrients in the plant residue for nitrogen and phosphorus. The represented processes are as follows: mineral and organic fertilization, input with precipitation, mineralization, denitrification, plant uptake, leaching to groundwater, and losses with surface runoff, interflow and erosion. The nitrogen mineralization considers two sources: (a) fresh organic pool associated with crop residue and (b) active organic pool associated with the soil humus. Organic nutrient flows between the active and stable organic pools are governed by the equilibrium equations and depend on the C:N ratio, soil temperature, and soil-water content.

Denitrification in soil appears only in the conditions of oxygen deficit, which usually occurs when soil is wet. Denitrification rate is estimated as a function of soil water factor, soil temperature, organic matter, and mineral nitrogen content. The soil water factor is an exponential function of soil moisture with an increasing trend when soil becomes wet. Crop uptake of nitrogen is estimated using a supply and demand approach. The optimal nitrogen concentration for a given crop is calculated as functions of the growth stage. The daily crop demand of nitrogen is estimated as the product of biomass growth and optimal concentration in plants. Actual nitrogen uptake is the minimum of supply and demand. Uptake starts at the upper soil layer and proceeds downward until the daily demand is met or until all nitrogen content has been depleted. Amounts of soluble nitrogen in surface flow, lateral subsurface flow and percolation are estimated as the products of the volume of water and the average concentration. While passing the soil and groundwater by flowing to the river system within surface flow, interflow and base flow nitrogen is subject to retention and decomposition processes [17].

SWIM usually simulates snowmelt and meltwater out-flow by simple degree-day method at the subcatchment scale. Here, to be able to incorporate the spring snowmelt, we used the improved version: the relevant parameters are calibrated (threshold temperature for snow fall and threshold temperature for snow melt, snow correction factor) [8]. We used the version of SWIM that allows to incorporate the municipal nitrogen point sources [8]. We took into account the atmospheric deposition by using a global parameter for the whole catchment, representing the average concentration of nitrate in precipitation (0.8 mgl, [18]).

2.2 Study area

Jankovský stream catchment is situated in Bohemian-Moravian Higlands, Czech Republic. Drainage area of the Jankovský stream is 129.44 km2. The altitude of the catchment is between 445 and 767 m above mean sea level. The mean annual total precipitation is 670 mm and the mean annual air temperature is 7.1°C. Soil conditions are heterogeneous; the major soil types are acid Cambisols (arenic, modal and gleyed forms) of a sandy to sandy-loam texture. Further, in sloped and alluvial positions, Pseudogleys and Gleysols are located, with a loam to clay-loam texture.

Geomorphologically, the area consists of an undulated mosaic of upland peneplains and shallow to deep valleys. Major parent rock is paragneiss, other encountered ones are granite, orthogneiss and quartzite; sporadically sandy and loamy eluvium. There is characteristic shallow aquifer which is fixed to the quaternary porous sediments or to the weathering zones of the crystalline rocks or to zone of shallow disjunction of crevices. Therefore, catchment hydrogeology manifests two different patterns; permanent aquifers exist in the places where the regolith is thicker (in local depressions and at saddle-like sites), in the fractured and faulted zones of solid rocks and in alluvial deposits in narrow valleys, i.e. in discharge zones. The recharge zones are mainly located in the uppermost parts of the catchment, close to the catchment divide, peaks and ridges. The deep, permanent groundwater table in the recharge zones usually lies at the depth of several metres, whereas the perched, shallow groundwater table of recharge and transient zones occurs already at around 0.8-1.2 meters. Generally, the turnover of groundwater is quick [19].

The catchment land use is as follows: arable land (48.5%), forest (31.5%), grassland (meadows and pastures, 15.8%), built-up areas (3.2%) and water bodies (1%). There are 37 settlements, predominantly small villages and dwellings. The Jankovský stream has a length of 18.3 km and has five main tributaries. In the catchment, there are about 200 fishponds, from tiny (0.1 ha) to larger ones (30 ha). Agricultural land in the catchment is partly tile-drained with systematic subsurface drainage. Drainage spreads on 10.5% of catchment area; 16% of agricultural land is drained. It has been shown that the tile drainage runoff pattern in crystalline catchments (as our study area) is complicated [20].

To built the model of a given catchment with SWIM, it has to be described by four input raster maps representing its most relevant properties regarding the nutrient dynamics: land-use, soils, digital elevation model and sub-catchments [21]. For Jankovský stream catchment, the study area was described by maps prepared in 10 × 10 m resolution: the land-use based on Corine 2006 corrected by aerial images processing and analyses; soil map based on [22]; digital elevation model (DEM) and sub-catchment map based on DEM (17 subcatchments in Jankovský stream catchment).

In total, five types of soils are recognized in the Jankovský stream catchment for modeling purposes: two cambisol, two gleysol and pseudogley type.

Importantly, Jankovský stream catchment is located in the catchment area of the Czech largest drinking water reservoir: Švihov reservoir on the Želivka river. The location of the catchment within the Czech Republic and the Želivka river basin, land-use and main watercourses and monitoring sites are shown in Figure 1. Both the agricultural and municipal sources of nitrogen contribute to the Jankovský stream, however, their actual shares have not been evaluated yet.

2.3 Materials

The gridded climate data (precipitation, temperature and humidity) were provided by the Czech Hydrometeorological Institute. The time series of each climate variable requested by SWIM were interpolated in the centroids of subcatchments by the inverse distance weighting method. For calibration purposes time series of daily discharge at the catchment outlet were used.

To analyze the municipal nitrogen sources we used the equivalent-adjusted inhabitant-pollution production approach and wastewater treatment type [23]. We obtained annual data on number of inhabitants and seasonal vacationers for 2004–2014 and the information on wastewater treatment (WWT) for all 37 settlements in order to estimate the nitrogen pollution released to surface water for each WWT. Finally, we aggregated the estimated daily nitrogen loads from the settlements to the subcatchments.

Finally, the fertilization data – N in kg, type of fertilizer (organic/mineral) and day of application, typical for the Jankovský stream catchment – were also implemented in the model. Data on livestock numbers and farming were obtained from the Ministry of Agriculture of the Czech Republic. We incorporated these data into the model using livestock-unit-pollution production approach.

For proper calibration, the observational data representing the outlet of given catchment are needed. For Jankovský stream catchment outlet we used the water flows in daily time step and time series of N-NO3 concentrations, which were available with sampling frequency of 1-2 per month.

2.4 Water quality improving scenarios

Two strategies for improvement of water quality were considered: land-use change and additional treatment of municipal pollution by constructed wetlands. For a land-use change scenario we considered conversion of all catchment arable land, situated on coarse-textured, shallow and leaching-prone soils to grassland. This leads to 25% reduction of cropland (Figure 2).

Figure 2 Current (left) and scenario (right) land use. The converted areas are marked in red.
Figure 2

Current (left) and scenario (right) land use. The converted areas are marked in red.

Constructed wetlands were considered as elements able to reduce the municipal (point) sources under the 37 settlements in the catchment. To account for the effect of constructed wetlands, the estimated municipal pollution was reduced by 50% or 100%. Note that the latter value is unrealistic, however, we included it to evaluate the limits of nutrient reduction (Figure 3).

Figure 3 Placement of the constructed wetlands in the catchment area (red dots and the name of the settlements). Violet numbers indicate the SWIM subbasins.
Figure 3

Placement of the constructed wetlands in the catchment area (red dots and the name of the settlements). Violet numbers indicate the SWIM subbasins.

Scenarios that are assessed in our study are based on reduction of municipal N-NO3 sources (by constructed wetlands), reduction of agricultural sources (by land-use change) and the combination of both:

  1. Scenario 1: No municipal N-NO3 sources (assuming perfect treatment).

  2. Scenario 2: Municipal N-NO3 sources reduced to 50%.

  3. Scenario 3: Land-use change.

  4. Scenario 4: Combination of no municipal N-NO3 sources and land-use change.

  5. Scenario 5: Combination of municipal N-NO3 sources reduced to 50% and land-use change.

3 Results

3.1 Modeling with SWIM

The calibration of SWIM is processed in two steps: the water discharge is calibrated first, and after that, the N-NO3 is calibrated. For calibration of discharge, we used the non-dimensional efficiency criterion NSE [24]. NSE is a measure describing the squared differences between the observed and the simulated values on a daily time step. We did the calibration manually and by using the calibration system PEST ( The ranges of parameters were taken from literature [25]. We calibrated the N-NO3 loads on average annual value. As objective function, the relative deviation was used.

The following parameters were used for hydrological calibration: curve numbers, alpha factor for groundwater (characterizes the rate at which groundwater flow is returned to the stream), threshold temperature for snow fall and threshold temperature for snow melt, snow correction factor and two routing coefficients. For N-NO3 load, the residence time and the decomposition rates in surface, subsurface and groundwater were used as calibration parameters.

The studied period was 2004–2008. The comparison of observed and simulated daily discharge is presented in Figure 4. The model represents the discharge at the outlet quite reasonably. However, the high flows are overestimated and low flows are underestimated in the model simulation. To reduce the effect of this bias we limit our further discussions on the effects of protective measures to annual water balance and annual mean N-NO3 loads. Annual water balance (Table 1) is represented adequately by the model.

Figure 4 The comparison of observed and simulated daily discharge.
Figure 4

The comparison of observed and simulated daily discharge.

Table 1

Observed and simulated annual water balance. Units are m3 × 1000.

yearsimulatedobservedrel. diff.

We carried out calibration of N-NO3 for the outlet of the catchment while taking into account estimated municipal N-NO3 sources, atmospheric deposition and agricultural sources. The nutrient transformation processes on the croplands and in the underlying soil are to a large extent affected by soil conditions (texture, structure, soil water dynamics and temperature) and organic matter content (organic nitrogen and carbon). The quantity of N-NO3 that enters the river system and the catchment outlet is also controlled by characteristics of vegetation and the type of crops.

The prevailing crops in the Jankovský stream catchment are winter wheat, fodder plants, spring barley and oil seed rape. The nitrogen calibration was done assuming winter wheat on all agricultural areas. This is not fully realistic, however, it is acceptable for the water quality modeling with SWIM and fully sufficient for purposes of this study (see also [8, 26]).

The comparison of observed and simulated N-NO3 loads is given in Tab. 2. The model reproduce the N-NO3 loads reasonably well. The N-NO3 are significantly overestimated for 2007. The total relative difference between observed and simulated N-NO3 load was 2%. As expected, the N-NO3 loads are higher in wet (2005) and lower in dry (2008) years.

Table 2

The comparison between observed and simulated N-NO3 loads. Units are kg.

yearsimulatedobservedrel. diff.

3.2 Scenarios of protective measures

The results of impact of different measures and their combinations on N-NO3 load are presented in Figure 5. The total reduction in municipal N-NO3 sources (Scenario 1) leads to a larger decrease in N-NO3 loads than the considered agricultural land-use change only (Scenario 3). However, when more realistic effect of constructed wetlands is considered (i.e. Scenario 2), the land use change scenario (Scenario 3), clearly reduces the nitrogen loads more.

Figure 5 The results of simulation with different scenarios for N-NO3 load.
Figure 5

The results of simulation with different scenarios for N-NO3 load.

The largest decrease in N-NO3 was simulated for Scenario 4, i.e. the combination of measures has larger effect than only the reduction of municipal sources or only the agricultural land-use (as expected).

Regarding the discharge, modeling for scenarios provided contrasting results. The reduction in municipal N-NO3 has understandably no impact on discharge since only the reduction of concentration (not runoff) is considered. On the other hand, the scenarios with land-use change led to slight increase in mean annual discharge at the outlet (not shown).

4 Discussion

According to our results (Figure 5), the contribution of municipal sources to the total N-NO3 loads is considerable and reduction of these sources can thus improve the nutrient balance in the catchment outlet even more than a “realistic” (i.e. made on a relatively small fraction of arable land) land-use change (and consequent changes in fertilization). This observation is in contrast with conclusions of Hesse et al. [26] for an agricultural catchment. The possible explanation is different catchment conditions and soils properties.

In Scenario 1 we run the model with no municipal N-NO3 sources. This scenario might seem to be unrealistic, yet it provides valuable information on what is the maximum potential influence of reduction of N-NO3 from municipal sources. Scenario 2 (50% reduction in municipal sources) provides estimate on change in N-NO3 loads in the case that the constructed wetlands (CW) would be implemented in the catchment: reduction of 50% N-NO3 corresponds to capacity of CW to reduce the nitrogen compounds. It is worth noting that reported efficiency of CWEs in reduction of N-NO3 is even higher. For instance, CWs are able to decrease concentration of N-NO3 up to 70% and NH3 up to 60% [27]. Also denitrification bioreactors are very effective in reducing N-NO3 concentration in waste water (even up to 74% [28]).

There are several fishponds in the Jankovský stream catchment (Figure 1). Fishponds may affect both water quantity and quality at the catchment outlet. Their presence in the catchment might be partial explanation for overestimation of the high flows and underestimation of low flows by the model (Figure 4). They are potentially relevant source of N-NO3, in principal similar to municipal sources, that we have not implemented in the modeling due to data unavailability. For instance, according to Kopp et al. 2012 a fishpond on stream may cause increase in N-NO3 concentration downstream [29]. On the other hand, it is possible to use reeds in shallow lakes or fishponds for treatment of wastewater [30].

The scenario with agricultural land-use change produced a decrease in N-NO3 load around 5%. The agricultural land-use change can be seen as an analogy to reduction of nitrogen fertilization [31]. Our results are in accord with results of Whitehead et al. [32]. According to them, 50% reduction in fertilizers (due to corresponding land-use change) resulted in 27% decrease of N-NO3 concentration in 20 years. Additionally, Constantin et al. [31] reported small but significant impact of fertilization reduction on leaching of N-NO3.

In our simulation the agricultural land-use change was less effective than the reduction of municipal N-NO3 sources. The possible explanation for this can be that nitrogen export was more related to mineralization of soil organic nitrogen pools due to drainage and tillage than the external nitrogen sources [11]. This hypothesis is supported by a study in similar catchment in the Czech Republic, which concludes that soil water regime and mineralization of soil organic matter is dominant source of nitrogen and therefore decrease in fertilization does not always result in decrease of nitrogen loads in stream water [12]. In addition, somewhat contrasting results were obtained for poorer soils where the SWIM simulations indicate, that the fertilization decreasing measures are more effective than the reduction in municipal N-NO3 sources [26]. Finally, it is worth noting that the scenario land-use change affected only ca 25% of the catchment area. The effects would be stronger when more extensive land-use changes were considered.

Generally, the drainage of water from fields is important source of N-NO3 to surface water [19, 33]. To manage it, several relatively simple and effective measures are available. Besides grassing of ploughland, these measures are known as drainage water management, e.g. widening the tile drainage spacing [35], performing controlled drainage, proper adjustment of timing and rate of nitrogen fertilization, quantification of soil organic matter mineralization and using appropriate yield goals [36] or, where applicable, elimination of land drainage [37]. Very effective measure to reduce leaching of N-NO3 in drainage water are the catch crops (reducing the N-NO3 concentration in drained water in the order of 34-52% [31]. Finally, the edge-of-field treatment wetlands (CWs focused on treatment of the cropland drainage water) can remove up to 68% of N-NO3 [38].

The scenarios with agricultural land-use change resulted in slightly larger and more balanced discharge. The possible explanation is the reduction of transpiration especially when comparing the grassland with corn or oil seed rape [39]. The relative change in discharge is most pronounced in the driest year 2008.

5 Conclusions

We used the eco-hydrological model SWIM for assessment of the effectiveness of different protective measures on decreasing the N-NO3 in surface water runoff from a rural catchment. Based on the analysis of N-NO3 sources in the Jankovský stream catchment, we implemented two main scenarios: reduction of municipal N-NO3 sources and agricultural land-use change and their combinations. Since the information on municipal N-NO3 sources was not available, we carried out their estimation using the equivalent-adjusted inhabitant-pollution production approach and wastewater treatment type.

Results of modeling with SWIM for these scenarios showed that the reduction of N-NO3 loads at the outlet by elimination of municipal sources is comparable to land-use change at 25% of the catchment area. Overall, the most effective scenario was the combination of total reduction of municipal N-NO3 sources and agricultural land-use change.

Importantly, the agricultural land-use change influenced positively the flow at the outlet of the catchment. It was pronounced significantly in dry year. Such measures have high potential not only regarding the water quality and ecological status of surface water, but they can contribute to solution of drought as an important impact of climate change.

Our results highlighted the potential of relatively inexpensive and robust protective measures to improve the quality of surface waters and their ecological status.


The research leading to these results has received funding from the Norwegian Financial Mechanism 2009–2014 under Project Contract no. 7F14341 “Assessing water quality improvement options concerning nutrient and pharmaceutical contaminants in rural watersheds“.


[1] de Souza, A.L., Fonseca D.G., Libório, R.A., Tanaka M. O., Influence of riparian vegetation and forest structure on the water quality of rural low-order streams in SE Brazil, Forest Ecol. Manag., 2013, 298, 12–1810.1016/j.foreco.2013.02.022Search in Google Scholar

[2] Srikanth R., Challenges of sustainable water quality management in rural India, Current Science, 2009, 317–325Search in Google Scholar

[3] Li P., Tian R., Xue C., Wu J., Progress, opportunities, and key fields for groundwater quality research under the impacts of human activities in China with a special focus on Western China, Environ. Sci. Pollut. Res. In., 2017, 24, 13224–1323410.1007/s11356-017-8753-7Search in Google Scholar PubMed

[4] Wu J., Sun Z., Evaluation of shallow groundwater contamination and associated human health risk in an alluvial plain impacted by agricultural and industrial activities, mid20 west China, Exposure and Health, 2016, 8, 311– 32910.1007/s12403-015-0170-xSearch in Google Scholar

[5] Li P., He S., He X., Tian R., Seasonal hydrochemical characterization and groundwater quality delineation based on matter element extension analysis in a paper wastewater irrigation area, Northwest China, Exposure and Health, 2017, 1–1810.1007/s12403-017-0258-6Search in Google Scholar

[6] Voulvoulis, N., Arpon K.D., and Giakoumis T., The EU Water Framework Directive: From great expectations to problems with implementation, Sci. Total Environ., 2017, 575, 358–36610.1016/j.scitotenv.2016.09.228Search in Google Scholar PubMed

[7] de Wit M., Behrendt H., Bendoricchio G., Bleuten W., van Gaans P., The contribution of agriculture to nutrient pollution in three European rivers, with reference to the European Nitrates Directive, EWMO, 2002, 19Search in Google Scholar

[8] Martínková M., Hesse C., Krysanova V., Vetter T., Hanel M., Potential impact of climate change on nitrate load from the Jizera catchment (Czech Republic), Phys. Chem. Earth, Parts A/B/C, 2011, 36, 673–68310.1016/j.pce.2011.08.013Search in Google Scholar

[9] Vymazal J., Constructed wetlands for wastewater treatment: five decades of experience, Environ. Sci. Technol., 2010, 45, 61–6910.1021/es101403qSearch in Google Scholar PubMed

[10] Carvalho P., Arias C., Brix H., Constructed wetlands for water treatment: New developments, Water, 2017, 397, 910.3390/w9060397Search in Google Scholar

[11] Kopáček J., Hejzlar J., Posch M., Factors controlling the export of nitrogen from agricultural land in a large central European catchment during 1900–2010, Environ. Sci. Technol., 2013, 47, 6400–640710.1021/es400181mSearch in Google Scholar PubMed

[12] Kvítek T., Žlábek P., Bystřický V., Fučík P., Lexa M., Gergel J., Novák P., Ondr P., Changes of nitrate concentrations in surface waters influenced by land use in the crystalline complex of the Czech Republic, Phys. Chem. Earth, Parts A/B/C, 2009, 34, 541–55110.1016/j.pce.2008.07.003Search in Google Scholar

[13] Judová P., Jánský B., Water quality in rural areas of the Czech Republic: Key study Šlapanka River catchment, Limnologica, 2005, 35, 160–16810.1016/j.limno.2005.06.003Search in Google Scholar

[14] Krysanova V., Hattermann F., Huang S., Hesse C., Vetter T., Liersch S., Koch H., Kundzewicz Z.W., Modelling climate and land use change impacts with SWIM: lessons learnt from multiple applications, Hydrolog. Sci. J., 2014, 60, 606-63510.1080/02626667.2014.925560Search in Google Scholar

[15] Arnold J., Soil G., SWAT (Soil and Water Assessment Tool). Grassland, Soil and Water Research Laboratory, USDA, Agricultural Research Service, 1994Search in Google Scholar

[16] Krysanova V., Meiner A., Roosaare J., Vasilyev A., Simulation modelling of the coastal waters pollution from 70 agricultural watershed, Ecological modelling, 1989, 49, 7–2910.1016/0304-3800(89)90041-0Search in Google Scholar

[17] Hattermann F., Krysanova V., Habeck V., Bronstert A., Integrating wetlands and riparian zones in river basin modelling, Ecological Modelling, 2006, 199, 379–39210.1016/j.ecolmodel.2005.06.012Search in Google Scholar

[18] Hrkal Z., Prchalová H., Fottová D., Trends in impact of acidification on groundwater bodies in the Czech Republic; an estimation of atmospheric deposition at the horizon 2015, J. Atmos. Chem., 2006, 53, 1–1210.1007/s10874-006-0911-0Search in Google Scholar

[19] Fučík P., Zajíček A., Duffková R., Kvítek T., Water quality of agricultural drainage systems in the Czech Republic– options for its improvement. In: Lee T. (Ed.), Research and Practices in Water Quality, InTech, 2015, 239–26210.5772/59298Search in Google Scholar

[20] Zajíček A., Pomije T., Kvítek T., Event water detection in tile drainage runoff using stable isotopes and a water temperature in small agricultural catchment in Bohemian-Moravian Highlands, Czech Republic, Environ. Earth Sci., 2016, 75, 1–1310.1007/s12665-016-5561-1Search in Google Scholar

[21] Krysanova V., Wechsung V., SWIM (Soil and Water Integrated Model). PIK, Potsdam, 2000.Search in Google Scholar

[22] Němečková S., Slámová R., Šípek V., Climate change impact assessment on various components of the hydrological regime of the Malše River basin, Journal of Hydrology and Hydromechanics, 2011, 59, 131–14310.2478/v10098-011-0011-ySearch in Google Scholar

[23] Hejzlar J., Metodika bilanční analýzy zdroju živin v povodí, Studie, Biology Centre, Czech Academy of Science, Institute of Hydrobiology, 2016.Search in Google Scholar

[24] Nash J.E., Sutcliffe J.V., River flow forecasting through conceptual models part I–A discussion of principles, Journal of Hydrology, 1970, 10, 282–29010.1016/0022-1694(70)90255-6Search in Google Scholar

[25] Huang S., Hesse C., Krysanova V., Hattermann F., From meso- to macro-scale dynamic water quality modelling for the assessment of land use change scenarios, Ecological Modelling, 2009, 220, 2543–255810.1016/j.ecolmodel.2009.06.043Search in Google Scholar

[26] Hesse C., Krysanova V., Päzolt J., Hattermann F.F., Ecohydrological modelling in a highly regulated lowland catchment to find measures for improving water quality, Ecological Modelling, 2008, 218, 135–14810.1016/j.ecolmodel.2008.06.035Search in Google Scholar

[27] Vymazal J., The use of hybrid constructed wetlands for wastewater treatment with special attention to nitrogen removal: a review of a recent development, Water Research, 2013, 47, 4795–481110.1016/j.watres.2013.05.029Search in Google Scholar PubMed

[28] Gottschall N., Edwards M., Craiovan E., Frey S., Sunohara M., Ball B., Zoski E., Topp E., Khan I., Clark I., et al., Amending woodchip bioreactors with water treatment plant residuals to treat nitrogen, phosphorus, and veterinary antibiotic compounds in tile drainage, Ecological Engineering, 2016, 95, 852–86410.1016/j.ecoleng.2016.06.011Search in Google Scholar

[29] Kopp R., Vítek T., Štastnỳ J., Sukop I., Brabec T., Ziková A., Spurnỳ P., Mareš J., Water quality and biotic community of a highland stream under the influence of a eutrophic fishpond, Int. Rev. Hydrobiol., 2012, 97, 26–4010.1002/iroh.201111426Search in Google Scholar

[30] Dinka M., Kiss A., Magyar N., Ágoston-Szabó E., Effects of the introduction of pre-treated wastewater in a shallow lake reed stand, Open Geosciences, 2016, 8, 62–7710.1515/geo-2016-0008Search in Google Scholar

[31] Constantin J., Mary B., Laurent F., Aubrion G., Fontaine A., Kerveillant P., Beaudoin N., Effects of catch crops, no till and reduced nitrogen fertilization on nitrogen leaching and balance in three long-term experiments, Agric. Ecosyst. Environ., 2010, 135, 268–27810.1016/j.agee.2009.10.005Search in Google Scholar

[32] Whitehead P., Wilby R., Butterfield D., Wade A., Impacts of climate change on in-stream nitrogen in a lowland chalk stream: an appraisal of adaptation strategies, Sci. Total Environ., 2006, 365, 260–27310.1016/j.scitotenv.2006.02.040Search in Google Scholar PubMed

[33] Hirt U., Hammann T., Meyer B. C., Mesoscalic estimation of nitrogen discharge via drainage systems, Limnologica, 2005, 35, 206–21910.1016/j.limno.2005.06.005Search in Google Scholar

[34] Williams M., King K., Fausey N., Drainage water management effects on tile discharge and water quality, Agric. Water Manag., 2015, 148, 43–5110.1016/j.agwat.2014.09.017Search in Google Scholar

[35] Kladivko E., Frankenberger J., Jaynes D., Meek D., Jenkinson B., Fausey N., Nitrate leaching to subsurface drains as affected by drain spacing and changes in crop production system, J. Environ. Qual., 2004, 33, 1803–181310.2134/jeq2004.1803Search in Google Scholar PubMed

[36] Dinnes D.L., Karlen D.L., Jaynes D.B., Kaspar T.C., Hatfield J.L., Colvin T.S., Cambardella C.A., Nitrogen management strategies to reduce nitrate leaching in tile-drained Midwestern soils, Agronomy Journal, 2002, 94, 153–17110.2134/agronj2002.1530Search in Google Scholar

[37] Kulhavý Z., Fuč P., Adaptation options for land drainage systems towards sustainable agriculture and the environment: a Czech perspective, Pol. J. Environ. Stud., 2015, 24, 2015.10.15244/pjoes/34963Search in Google Scholar

[38] Lenhart C., Gordon B., Gamble J., Current D., Ross N., Herring L., Nieber J., Peterson H., Design and hydrologic performance of a tile drainage treatment wetland in Minnesota, USA, Water, 2016, 8, 54910.3390/w8120549Search in Google Scholar

[39] Duffková, Zajíček A., Nováková E., et al., Actual evapotranspiration from partially tile-drained fields as influenced by soil properties, terrain and crop, Soil Water Res., 2011, 6, 131–1410.17221/20/2010-SWRSearch in Google Scholar

Received: 2017-06-18
Accepted: 2018-06-07
Published Online: 2018-09-07

© 2018 M. Martínková et al., published by De Gruyter.

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.

Downloaded on 9.12.2022 from
Scroll Up Arrow