Assessment of the predictive quality of simple indicator approaches for nitrate leaching from agricultural fields
Introduction
Diffuse nitrogen (N) losses from agricultural fields are a major cause of excessive nitrate concentrations in ground and surface waters and have been of concern since decades (e.g. Bach, 1987, Wendland et al., 1993, Ten Berge, 2002, Behrendt et al., 2003, Delgado et al., 2008). Excessive nitrate concentrations in groundwater can have toxic effects when used as drinking water and cause eutrophication in surface waters (e.g. Townsend et al., 2003, Powlson et al., 2008). Gaseous N losses in form of N2O are an important factor in global warming and the destruction of the stratospheric ozone layer (IPCC, 2007), whereas ammonia volatilization contributes to soil acidification and eutrophication (Follett and Delgado, 2002). Moreover, fertilizer and manure N that is not used by growing crops but lost to the environment represents an economic loss.
For management and environmental planning purposes, it is necessary to assess the risk and magnitude of diffuse N losses from agricultural fields and how they are influenced by factors such as management practices, climate and soil etc. (Meisinger and Delgado, 2002, Havlin, 2004). Utilization of experimental methods to determine N loading, such as analysis of leachate water obtained by suction cups (Sieling and Kage, 2006), monolith-lysimeters (Bohne et al., 1997, Knappe et al., 2002), analysis of percolate from tile drains (Kladivko et al., 2004, Tiemeyer et al., 2008), N concentrations in groundwater (De Ruijter et al., 2007), and also measurements of mineral nitrate content in the soil profile (Wehrmann and Scharpf, 1979), is restricted in practice. The main reasons for this are that routine application of such labor-demanding methods is mostly not viable, measurements can be made only afterwards, and the experimental data are often not suited for generalization (because of the effects of different years with varying weather patterns, different management practices, fertilizer application rates, etc.). Moreover, it is often difficult to explain results from direct measurements, since such data are in most cases integrative and do not allow to separate the effects of soil, climate and management (e.g. Bockstaller et al., 2008).
On the other hand, physically based N transport models have been developed since decades (e.g. Cannavo et al., 2008). With such models, it is – at least in principle – possible to quantify N losses for various environmental conditions and agricultural management practices. However, complex models require many input data, contain many parameters whose values are often not sufficiently known, and are often difficult to operate. This restricts their routine use for assessment of Nloss from agricultural fields and also on a larger regional scale.
As an alternative, simplified qualitative or semi-quantitative Nitrogen Loss Indicators (NLI) have been developed and discussed (Follett et al., 1991, Schröder et al., 2004, Pervanchon et al., 2005, Delgado et al., 2008, Bockstaller et al., 2008). NLIs are a subset of (agri-)environmental indicators (Bockstaller et al., 2008, Makowski et al., 2009). The great number of different NLI approaches differ with respect to their complexity, factors of Nloss, data requirements and type of output (e.g. risk classes, quantified amounts of Nloss etc.). For instance Shaffer and Delgado (2002) presented various leaching indices used in the USA; in Canada, the IROWC-N NLI has been developed (De Jong et al., 2009). Several recent, more elaborated NLIs consider also Nloss by denitrification, ammonia volatilization, surface runoff or erosion: e.g. the “NIT-1” NLI has been developed in the USA (Delgado et al., 2008), but also applied in other countries such as Spain (De Paz et al., 2009); or the French “IN” NLI (Pervanchon et al., 2005). Although even these “complex” NLIs are considerably simpler and quicker to use than full-fledged N models, the utilization of very simple NLIs may be warranted in situations with restricted data availability. For instance, N balances are commonly being used as an indicator for N losses at the landscape and field scale (Goodlass et al., 2003, Drury et al., 2007). However, correlations of N balances with measured N losses often proved to be weak (e.g. Schröder et al., 2004, Sieling and Kage, 2006, De Ruijter et al., 2007, Rankinen et al., 2007).
To support a decision as to which NLI method could be suited for a given site, management option, and data availability, output validation of different NLIs using experimental data would be useful. Such studies are relatively scarce: for instance, in the Netherlands several source-based NLIs were tested and compared (Ten Berge, 2002, De Ruijter et al., 2007), in France, output validations of simple source-based NLIs using the “Receiver Operating Characteristic” methodology were conducted (Bockstaller et al., 2008, Makowski et al., 2009).
The objectives of this study were to test some simple NLIs using field data of Nloss and to compare the results obtained with different NLI methods. We restrict the testing procedure here to a relatively small number of some basic NLIs. To simplify the analysis, only leaching through the soil profile was considered. It is generally accepted that leaching through the soil profile is the dominant pathway of Nloss in many situations, and the impact by leaching losses of N on ground and surface waters is more direct than the environmental impact by gaseous N losses and surface erosion losses. We used experimental field data of Nloss both from Northeast Germany (Miegel and Zachow, 2006, Tiemeyer et al., 2008) and published data from different geographical regions, with different crop types and methods of Nloss measurement.
Section snippets
Field sites/experimental data
We utilized N leaching data from two field sites in Mecklenburg-Vorpommern (NE Germany). We extended the database with published data from field experiments in Germany, Scotland, and North America (Table 1, Table 2). The field experiments encompass different measurement methods of Nloss and different overall duration, a wide range of crop types, N input rates (Table 1), soil textures, water holding capacity of the root zone at field capacity (WHC(rz)), seepage rates, rates of N leaching,
N balance
For all datasets pooled, the correlation between Nbal and Nloss is rather low when single-year data are considered (although significant at the 0.01 level, Table 3, Fig. 2). Fig. 2 shows a large scatter, some datasets even show large negative Nbal values (<−150 kg N ha−1 y−1) with high Nloss. The correlation is distinctly higher when data averaged over several years are considered (Table 4). On the other hand, correlations between Nbal and Nconc in the leachate were not significant, neither for
N balance
N balances in the simplified form used here proved to be a poor indicator for possible N leaching losses for the time scale of a single year and a slightly better indicator for longterm data. This is in accordance with many previous studies (e.g. Oenema et al., 2003, Sieling and Kage, 2006, De Ruijter et al., 2007). The relatively high correlation for Central Iowa and Wittkoppenberg may be explained by the fact that at these sites, several different N fertilizer rates were applied and compared
Conclusions
In this study, four relatively simple NLIs were compared: Nbal, EF, PNCL and a composite NLI calculated as the product of Nbal and EF (BEP). Experimental data of Nloss comprising several datasets from field sites in Germany, Scotland, and North America were used for testing the predictions of these NLI approaches (output validation).
Calculated NLI values were compared with measured Nloss and Nconc using linear correlation analysis, mean errors and root mean square errors. For single-year data,
References (61)
- et al.
The movement of nitrate fertiliser from the soil surface to drainage waters by preferential flow in weakly structured soils, Slapton, S. Devon
Agric. Ecosyst. Environ.
(1985) - et al.
Risk of water contamination by nitrogen in Canada as estimated by the IROWC-N model
J. Environ. Manage.
(2009) - et al.
An index approach to assess nitrogen losses to the environment
Ecol. Eng.
(2008) - et al.
Input output accounting systems in the European community – an appraisal of their usefulness in raising awareness of environmental problems
Eur. J. Agron.
(2003) - et al.
Catchment and drainage field nitrogen balances and nitrogen loss in three agriculturally influenced Latvian watersheds
Eur. J. Agron.
(2003) Ground water nitrates in other developed countries (Europe) – relationships to land use patterns
- et al.
Measuring the accuracy of agro-environmental indicators
J. Environ. Manage.
(2009) - et al.
Approaches and uncertainties in nutrient budgets: implications for nutrient management and environmental policies
Eur. J. Agron.
(2003) - et al.
Effects of lowering nitrogen and phosphorus surpluses in agriculture on the quality of groundwater and surface water in the Netherlands
J. Hydrol.
(2005) - et al.
A novel indicator of environmental risks due to nitrogen management on grasslands
Agric. Ecosyst. Environ.
(2005)
Nutrient management regulations in The Netherlands
Geoderma
The effects of nutrient losses from agriculture on ground and surface water quality: the position of science in developing indicators for regulation
Environ. Sci. Policy
Permissible manure and fertilizer use in dairy farming systems on sandy soils in The Netherlands to comply with the nitrates directive target
Eur. J. Agron.
N balance as an indicator of N leaching in an oilseed rape – winter wheat – winter barley rotation
Agric. Ecosyst. Environ.
Analysing nitrate losses from an artificially drained lowland catchment (north-eastern Germany) with a mixing model
Agric. Ecosyst. Environ.
The monitoring of nitrogen surpluses from agriculture
Environ. Pollut.
A comparative study of nitrate leaching from soils of differing textures under similar climatic and cropping conditions
J. Hydrol.
Statistical methods for evaluating a crop nitrogen simulation model, N_ABLE
Agric. Syst.
Nitrate-nitrogen in tile drainage as affected by fertilization
J. Environ. Qual.
Nutrient Emissions into River Basins of Germany on the Basis of a Harmonized Procedure. UBA-Texte 82
Agri-environmental indicators to assess cropping and farming systems. A review
Agron. Sustain. Dev.
Simulation von Stickstoff-Austrägen aus Lysimetern
Z. Pflanz. Bodenk.
Tillage effects on hydraulic properties and macroporosity in silty and sandy soils
Soil. Sci. Soc. Am. J.
Modeling N dynamics to assess environmental impacts of cropped soils
Adv. Agron.
The indicator of risk of water contamination by nitrate-nitrogen
Can. J. Soil Sci.
Sequential NLEAP simulations to examine effect of early and late planted winter cover crops on nitrogen dynamics
J. Soil Water Conserv.
Use of a new GIS nitrogen index assessment tool for evaluation of nitrate leaching across a Mediterranean region
J. Hydrol.
Nitrate in upper groundwater on farms under tillage as affected by fertilizer use, soil type and groundwater table
Nutr. Cycl. Agroecosyst.
Residual soil nitrogen indicator for agricultural land in Canada
Can. J. Soil Sci.
Cited by (53)
Paddy water managements for diffuse nitrogen and phosphorus pollution control in China: A comprehensive review and emerging prospects
2023, Agricultural Water ManagementModeling groundwater nitrate concentrations using spatial and non-spatial regression models in a semi-arid environment
2022, Water Science and EngineeringCitation Excerpt :Biophysically-based process models have been applied to solving the nitrogen cycle processes on field-to-catchment scales (Buczko et al., 2010; Cannavo et al., 2008). However, use of these models may be limited due to data requirements, model parameterizations, and initial assumptions (Buczko et al., 2010; Cannavo et al., 2008). Many statistical models, including linear, nonlinear, and nonparametric approaches, have been applied in environmental studies (Amini et al., 2008; Banerjee et al., 2011; Cho et al., 2011; Marques et al., 2006; Nolan et al., 2014; Rahmati et al., 2019; Shiri et al., 2013).
Suitability of operational N direct field emissions models to represent contrasting agricultural situations in agricultural LCA: Review and prospectus
2022, Science of the Total EnvironmentReuse in the agro-industrial: Irrigation with treated slaughterhouse effluent in grass
2020, Journal of Cleaner ProductionDriving factors of nitrate leaching in arable organic cropping systems in Northern France
2019, Agriculture, Ecosystems and EnvironmentCitation Excerpt :These values appear to be as variable but lower than those reported in other organic systems, ranging from 5 to 143 kg N ha−1 yr−1 (Küstermann et al., 2010; Sapkota et al., 2012; Anglade et al., 2015b). We did not find any relationship between the N surplus and N leaching, confirming previous observations made in conventional or organic systems with moderate values of N surplus (Buczko et al., 2010; Constantin et al., 2010; Pugesgaard et al., 2017). Wick et al. (2012) also found that N surplus explained little variance of nitrate concentration in groundwater despite a positive regression between the two variables.