Main authors: Oene Oenema, Meindert Commelin, Piet Groenendijk, John Williams, Susanne Klages, Isobel Wright, Morten Graversgaard, Irina Calciu, António Ferreira, Tommy Dalgaard, Nicolas Surdyk, Marina Pintar, Christophoros Christophoridis, Peter Schipper, Donnacha Doody
FAIRWAYiS Editor: Jane Brandt
Source document: »Oenema, O. et al. 2018. Review of measures to decrease nitrate pollution of drinking water sources. FAIRWAY Project Deliverable 4.1, 125 pp


Agriculture is a main source of nitrate pollution of the aquatic system. That is related to the facts that

  1. agricultural land covers roughly 40% of the total land area of EU-28, equivalent to 174 million ha in 2013,
  2. agriculture is a large user of nitrogen (N) for producing food and feed, and
  3. on average only 50 to 60% of the applied N is taken up by the crop and withdrawn in harvested crop yield.

The remainder is lost to the atmosphere via ammonia volatilization and denitrification or to lost water bodies via leaching and surface runoff. The loss of nitrate-N from agriculture to groundwater and surface waters depends on farming system, management, soil type and geomorphology, and climate. These factors define both

  • the sources of nitrate pollution and
  • the loss pathways (e.g., downward leaching to groundwater or overland flow (surface run-off), erosion, and subsoil lateral leaching to surface waters (Leip et al., 2011)).

This article provides a brief overview of the sources of nitrate pollution in agriculture. First, a summary of farming systems and of management in EU-28 is presented, as these define the input and the utilization of nitrogen in agriculture. Secondly, an brief overview is presented of the N input-output balance, as the balance is an indicator of the potential for N pollution of water resources (Klages et al., 2018).

Contents table
1. Farming systems
2. Characterization of management

1. Farming systems

About 60% of the utilized agricultural area in the EU-28 in 2013 was classified as arable land, 34% as grassland and 6% as permanent cropland (orchards, vineyards). These areas are managed by some 10 million farms, which are mostly family farms. Basically, each farm is managed in a unique manner (Eurostat, 2015).

There is a huge variation in farming systems, because of differences in their resource basis, enterprise pattern, crops, animals, management and also the use of nitrogen. A first characterization is commonly made between

  1. specialized crop production systems,
  2. specialized animal production systems, and
  3. mixed production systems.

Eurostat (2015a) distinguishes 8 main farm types (Table 3) , which reflect the aforementioned three categories, and three main classes of land use.

Table 3. Agricultural holdings by farm type in EU-28 in 2013 (Eurostat, 2015a)

Code Farm type Number of holdings in EU-28 (millions) Number of holdings in EU-28 (%)
1 Specialist field crops 3.20 29.6
2 Specialist horticulture 0.21 1.9
3 Specialist permanent crops 1.89 17.4
4 Specialist grazing livestock 1.86 17.1
5 Specialist granivores1 1.02 9.4
6 Mixed livestock 0.48 4.4
6 Mixed cropping 0.52 4.8
7 Mix crop-livestock2 1.50 13.8
8 Other 0.16 1.5
  Total 10.84 100.0

1) Granivorous literally means ‘feeding on grains and seeds’. In practices it means farms with monogastric animals, mainly pigs and poultry, where often a significant fraction of the feed is imported.
2) Mixed crop-livestock holding have neither livestock nor crop production as dominant activity; an activity is called dominant if it provides at least two-thirds of the production of an agricultural holding.

Anderson et al (2016) developed a farm typology for EU agriculture on the basis of:

  • Specialisation: Measured as the output value from the main activity; 10 farm specialization types.
  • Size: Measured as the economic size of the farms; 3 classes: <16; 16-40; >40 ESU
  • Intensity: Measured as the total output in Euro per ha; 3 classes: <500; 500-3000; >3000 euro/ha
  • Land use: Measured as the proportion of the agricultural area covered by specific types of crops; 9 different land use types were distinguished.

The farm typology of Anderson (2006) is a useful framework for characterizing farm types, as farm size, intensity, specialization and land use are all important determinants for N use. The farm typology does however not address the level of externalization of feed use in animal production farms in sufficient detail. A large fraction of animal farms do purchase animal feed from elsewhere, which affects N inputs, N output and N surplus of the farm. The level of externalization can be defined as the percentage of the feed (in dry weight) used on the farm that is imported from elsewhere (Table 4).

Table 4. Characterization of farms in EU-28

Nr Characteristics Unit of characterisation
1 Specialisation

Specialization type, and output derived from the main activity, in %; The 10 dominant specialized farm types are: (i)arable farms, (ii) horticultural farms, (iii) permanent crops, (iv) dairy farms, (v) beef farms, (vi) pig farms, (vii) poultry farms, (viii) sheep and goat, (ix) mixed livestock, (x) mixed farms

2 Land use Crop rotation and crop types, in %
3 Size Value of output, in European Size Units (ESU), and UAA, in ha
4 Intensity Value of output, in Euro per ha
5 Externalization Purchased feed, in % of total feed

2. Characterization of management

The importance of individual farmer decisions on nitrogen flows and balances are large; much depends upon the skill and precision with which farmers decide on the acceptable level of risk associated with each farm operation to determine nutrient application/management regimes (Jarvis et al., 2011). Farmers have multiple roles: they are managers and risk takers. And their skills determine the level of risk they are prepared to take to achieve financial gain and/or environmental benefit. However, the majority of farmers are businessmen and women, and many are entrepreneurs, whose primary aim is to optimize their production system to the benefit of themselves and perhaps of society as well. As a result, there is a wide variation in N input and N utilization (Jarvis et al., 2011; Stoumann Jensen et al., 2011).

Management is often considered to be the most important factor for the performance of the farm and of the utilization and losses of N. Management is usually defined as ‘a set of activities to achieve objectives’. It includes a sequence (cycle) of

  1. analysis of the current situation and of possible options,
  2. decision making,
  3. planning of the activities,
  4. execution,
  5. monitoring, and
  6. verification and control of achievements.

These management activities relate to different components of the farm.

Crop management includes:

  1. crop rotation aspects, i.e. crop sequence, use of cover crops and under growth, use of legumes, use of buffer zones. Crop rotations define both N input and N output in harvested crop. The crop statistics of Eurostat distinguishes 17 categories for cereals and 29 for other main crops, 40 categories for vegetables, 41 for permanent crops.
  2. soil cultivation aspects, i.e., conventional (mouldboard) ploughing or minimum tillage or zero tillage. Soil cultivation affects the amounts of N that are released through net soil mineralization.
  3. nutrient management, i.e., use of soil fertility analyses, organic farming, use of animal manures without low emission techniques, use of animal manures with low-emission techniques, use of fertilizers, use of GPS controlled fertilizer application. All these factors influence N input as well as N utilization at farm level
  4. pest management, i.e., use of chemical control and/or biological control measures. This factor greatly influences crop yield and thereby the N output and the overall N balance
  5. irrigation and drainage aspects, i.e., no irrigation, sprinkler irrigation, flood irrigation, drip irrigation and/or fertigation. These factors influence both crop yield and N output as well as the nitrate leaching losses.

Livestock management includes:

  1. Animal categories, i.e., Dairy cattle – beef cattle – pigs – poultry – sheep – goats. These categories greatly differ in protein-N requirements, N retention and N excretion.
  2. Herd related aspects, i.e. number of dairy cattle, replacement heifers, calves for replacement, number of fattening and suckling cattle, number of sows and fattening pigs, number of broilers and laying hens; The ratio between productive and supporting animals influence greatly the N utilization efficiency per unit of animal product produced
  3. Feeding management, i.e., number of grazing days per year, kg of concentrate per dairy cow, percent protein in animal feed. This influences the N utilization efficiency per unit of animal product
  4. Animal performance, i.e., milk production per cow per year (kg), calving interval (days), number of piglets per sow, feed conversion (kg feed per kg pork; kg feed per kg broiler; kg feed per kg egg); Again, this influences the N utilization efficiency per unit of animal product.
  5. Animal health management, i.e., veterinary cost, in % of total costs. This again influences the N utilization efficiency per unit of animal product
  6. Manure management, i.e., solid manure or slurry, covered manure storages, manure export; m3 per year, low-emission manure application. This affects the N losses from manure and manure storages and the effectiveness of manure N as N fertilizer. Animal manure is a main source of N in EU-28, which is used to fertilize cropland and grassland but att he same time is a main source of nitrate pollution of groundwater and surface waters (Oenema et al., 2007).

The management of crop and livestock farms can be captured by the N balance. Table 5 presents the input and output items for the farm N balance. These data allow to be estimated N use efficiency (NUE) and N surplus at farm level, for basically all farm types. Input and output items have to be reported only once on the balance. In the case that animals are imported to the farm and other animals are exported, only the net results should be presented, i.e., on the right-hand side of the balance . Similarly, in the case that animal manure is imported to the farm and other manure exported, only the net manure N input should be reported, as input (Table 5). Hence, manure is seen as an input (and not as a harvested output). Reporting the inputs and outputs on the proper side of the balance is important, as it allows a better comparison between farms.

Table 5. Input and output items considered for the farm N balance

Nitrogen input items  Nitrogen output items 
Mineral fertilizers I1 Crop products O1
Feed and fodder (net) I2 Animals (net) O2
Biological nitrogen fixation I3 Animal products (milk, egg, wool) O3
Atmospheric N deposition I4 (orchard) Trees (net) O4
Compost and sewage sludge I5    
Seed and planting material I6    
Bedding material (straw, saw dust) I7    
Animal manure (net) I8    
Irrigation water I I9 Surplus ∑I-∑O
Total sum Total sum

The soil is a main store of N, especially the top soil (plough layer). A small percentage of the total amount of N in soil (2000 to 10000 kg ha-1) is in the form of ammonium and nitrate and directly available to plants. Most of the N is stored in soil organic matter and not directly available to plants. Changes in soil organic N are common following changes in crop rotation and especially following the conversion of permanent grassland to arable land and vice versa. Changes are also common following changes in manuring and fertilization, changes in soil cultivation practices, and changes in weather conditions (mean temperature, rainfall). These changes can have a large effect on nitrate leaching losses.

D41 fig08
Figure 8

The main inputs to the farm are via mineral fertiliser, imported animal manure, fixation of atmospheric nitrogen by leguminous crops (beans, pulses, clover, alfalfa), N deposition from the atmosphere, and import of livestock feed. Inputs in seed and bedding used for animals are generally minor inputs, although the latter can be significant for some traditional animal husbandry systems. The main outputs from the farm are in crop and animal products. The main N flows within mixed crop-livestock farms are the consumption of feed by livestock, the return of nitrogen to the field in the excreta of grazing animals, and the removal of manure from manure storages to the field.

Farms differ greatly in the relationship between total N input, N output and the resulting NUE and N surplus. For intensively managed grassland-based dairy farms N surpluses may range from 80 to 300 kg per ha per year (Figure 9). For arable farms, N surpluses are usually in the range 0 to 100 kg per ha year. This wide variation is related to differences in farming systems and management. It is also related to interactions between crop, soil and climate, which affect N demand (because of differences in crop yield) and soil N supply (because of differences in net soil N mineralisation).

D41 fig09
Figure 9
D41 fig10
Figure 10

Figure 10 shows response curves of wheat and barley yields to N applications for different sites and years in the UK. There were huge differences in economic optimal fertilizer application rates, which are almost impossible to assess by the farmer at the start of the growing season. As farmers benefit more from high yields than from low yields, there is a tendency that farmers fertilize for high yields. This is one of the reasons that NUE is relatively low in years when yields are low and that N losses are relatively high in these years. Table 6 shows the average and the ranges of the fertilizer N use for main crops in EU-27. Evidently, the minimum and maximum N input differ by a factor or 3 to 4 between farms.  

Table 6. Average annual, minimum, maximum fertilizer N-use for main European crops in EU-27 (Stoumann Jensen and Schjoerring, 2011)

Crop Average (kg/ha) Range (min-max) kg/ha Crop area million ha
Oilseed rape 148 50-195 6.1
Sugar beet 123 50-160 1.9
Wheat 113 25-200 25.9
Grain Maize 106 26-200 9.0
Potato 98 40-185 2.2
Barley 88 15-145 13.9
Grassland 69 10-170 30.5
Silage Maize 65 10-126 4.7
Rye, triticale, oats, rice 64 10-110 8.7

In summary, total N use at farm level mainly depends on farming system and management. The potential for N losses depend on the difference (N surplus) between total N input and total N output, both of which greatly vary between farms and across EU-28 (Figure 11). The surplus of N in agriculture is highest in western Europe. Main N sources are N fertilizers and animal manures, while soils may also act as a N source following ploughing-up grassland and changes in soil cultivation (not shown in Figure 10). In addition, there are N inputs via biological N2 fixation and atmospheric deposition. The estimations shown in Figure 11 have not been checked and corrected by estimations at national scales by experts from Member States.

D41 fig11
Figure 11


Note: For full references to papers quoted in this article see

» References


Go To Top