Skip to Main content Skip to Navigation

Experimental analysis and modelling of the effects of cropping practices on nonpoint nitrate pollution of a deep aquifer in an arable farming area.

Abstract : Surface and groundwaters must regain good chemical and biological quality before 2015 according to European Directives. Good Agricultural Practices (GAPs) establish a compromise between the risks of pollution and the loss of revenue. The minimum result expected is conformity with the drinking water standard of 50 mgNO3.L-1 in the collected water and the avoidance of transfer of pollution from the hydrosphere into the atmosphere. However their implementation does not guarantee that these objectives will be reached ; that requires a means of quantifying the impact of effective agricultural practices on nitrate pollution. We have tested different methods of quantification by using data from a collaborative experiment on pollution prevention, carried out on the site of Bruyères (02). The question targeted was « what is the impact of GAP, applied regularly, on the scale of a catchment area ? » The research question was « can nitrate pollution be modelled, in a farming situation, on the scale of a catchment area ? » The study site is a plateau of 187 ha which supplies a groundwater aquifer located in the Lutetian geological layer. This aquifer feeds five main springs which have suffered increasing pollution since 1970. The 21 cultivated fields on the plateau were subjected to regular implementation of GAPs by the three farmers since 1990. The farming practices and the hydrogeology of the site were characterised. The flow rates of the springs respond to effective rainfall after a delay of a week. The mean residence time of the tritium molecule in the aquifer is 25 years, because of the thickness of the unsaturated zone. In view of this delay, an intermediate level of evaluation is necessary : the losses under the root zone. The methods of quantification differ according to their degree of dependence on the experimental data : i) the LIXIM mathematical model, associated with all the observed data ; ii) a stochastic model of crop response to the nitrogen rate, initialised each year, iii) the functional dynamic model STICS, which can simulate the losses of the soil/plant/atmosphere system continuously over several years. The predictions of the variables of economic and environmental interest are compared with the observed data on the scale of the station and of the basin. The environmental and economic impacts for different scenarios of pollution prevention are simulated. The mineral nitrogen residues at harvest and at the beginning of winter are similar and stable over time at 41 et 57 kgN.ha-1 respectively. The averaging of the losses, calculated with LIXIM, over the crop rotation, smooths out the crop factor and makes the soil type the principal determinant of the concentration. The mean weighted nitrate concentration in the percolating water is 46 mgNO3.L-1 for the cultivated zone and 37 mgNO3.L-1 for the whole basin. This good result is confirmed qualitatively by the fall observed in the contents at several collection points since the year 2000. The response time of the aquifer would be equal to half of its renewal time. The reduction in the nitrate content of the percolation water permitted by GAPs, compared with a conventional scenario, is between 27 and 39%, depending on the simulation method. The cost of the GAPs is 0.07
Document type :
Complete list of metadatas

Cited literature [204 references]  Display  Hide  Download
Contributor : Ecole Agroparistech <>
Submitted on : Wednesday, July 4, 2007 - 8:00:00 AM
Last modification on : Monday, October 19, 2020 - 11:09:24 AM
Long-term archiving on: : Wednesday, September 8, 2010 - 5:21:11 PM


  • HAL Id : pastel-00002665, version 1



Nicolas Beaudoin. Experimental analysis and modelling of the effects of cropping practices on nonpoint nitrate pollution of a deep aquifer in an arable farming area.. Sciences of the Universe [physics]. INAPG (AgroParisTech), 2006. English. ⟨NNT : 2006INAP0028⟩. ⟨pastel-00002665⟩



Record views


Files downloads