A fast growing economy and a more demanding population in Europe is leading to increasing socio-economic pressures, and large-scale land use and landscape changes are taking place. An increasing pressure is put on water resources and conflicts are arising for water distribution among different stakeholders, and degradation of water quality is exacerbating the problem. Although nutrients are essential for plant and animal populations, high concentrations can degrade water and soil quality. Excess nutrients have adverse effects on the environment such as acid rain caused by ammonia volatilisation, greenhouse emission due to denitrification, and impairment of drinking water supplies due to nitrate leaching. Both nitrogen and phosphorus are responsible for the eutrophication of many lakes and ponds. To control and reduce pollution coming from nutrients, the EC has been setting stringent regulations. To reduce and prevent water pollution caused or induced by nitrate from agricultural sources, the Nitrates Directive (91/676/EEC, 1991) requires Member States to designate Nitrate Vulnerable Zones (NVZ), and imposes to implement various action programs for reducing water pollution. In addition, the Water Framework Directive (2000/60/EC, 2000) required Member States to perform an analysis of pressure and impact on the surface and subsurface water resources by the end of 2004, where diffuse sources of pollution had to be identified as well as their impacts on the ecological status of surface and subsurface waters.
In Western Europe, agriculture is the main source of nitrogen loading to water bodies while agriculture and households contribute the most to phosphorus loading. The contribution of agriculture to nutrient loading into surface waters is highly variable ranging from more than 80% in Denmark to less than 30% in Finland for nitrogen and from more than 40% for Germany and Finland to about 17% for Belgium and United Kingdom for phosphorus (OECD, 2001). As the total load of nutrient coming from point sources has severely dropped, emphasis has been put on controlling diffuse sources. Combating diffuse pollution from agriculture is complicated due to the temporal and spatial lag between the management actions taken at the farm level and the environmental response (Schröder et al., 2004). Beside the correct identification and quantification of sources, cost-effective nutrient mitigation requires the delineation of critical source areas, which contribute disproportionate amounts of nutrients to receiving waters. According to Dickinson et al. (1990), targeting and prioritizing diffuse pollution control has the potential to triple pollutant reduction, is financially attractive, and minimizes the extent of area affected negatively by restrictive land practices.
To combat high river nutrient loads, the scientific community has been asked to provide reliable modelling tools to evaluate nutrient sources contribution to water pollution, propose sustainable alternative management practices to alleviate such pressure on water resources, and develop appropriate sampling strategies for monitoring the impact of implementation of best management practices (Wasson et al., 2003). Modelling is essential to the implementation of cost-effective and environmental friendly management strategies to optimize nutrient use and reduce their losses in terrestrial ecosystems. Several modelling methodologies are available to assess the fate of agrochemicals. These methodologies vary in complexity and are tailored to specific problems and specific scale (Schouman and Silgram, 2003). Several dilemma are involved in the selection of the appropriate models as the scale of different environmental problems linked to agriculture, and in particular those associated with nitrate excess, varies from the local to regional, country, continental, and global scales (Carton and Jarvis., 2001). In addition, the mismatch between spatial and temporal scale of remediation actions and their environmental effects (Schröeder et al, 2004) makes the model selection procedure more difficult. Scale is a thus a critical issue in model selection. The scale of the model is often determined by the dominant processes to be simulated, neglecting lesser important ones (Heuvelink and Pebesma, 1999). In addition, to simulate the impact of Best Management Practices (BMPs), various scales have to be taken into consideration as their implementation can be done at local scale, or can be part of larger strategies elaborated at regional or country level.
This paper presents a tiered approach for addressing nutrient fate at various scales that makes best use of readily available data at EU level taking into account the policy requirements of various Directives such as that of the Water Framework Directive (2000/60/EC, 2000) and the Nitrates Directive (91/676/EEC, 1991). Firstly (tier 1) a statistical approach (Grizzetti et al., 2005a) is used at large-river basin scale as a screening tool to identify catchments where nutrient losses are the highest and/or can cause a threat to water bodies. In a subsequent step (tier 2), the semi-distributed, physically-based SWAT (Arnold et al., 1998) model is used to identify within those areas, the major processes and pathways controlling and contributing to nutrient losses. The final step (tier 3) involves the use of the farm-scale model EPIC (William, 1995) to elaborate appropriate farming practices that could reduce pollution load
without endangering the farm economic sustainability. The tier 1 approach has no forecasting ability, while the tools used in tier 2 and 3 aim at analysing the actual situation, and forecast the impact of potential environmental and/or economical conditions. This paper will detail initially the various methodologies used, the data sets available at EU level followed by an application on the Loire/Vilaine (north western France) basin to illustrate the implemented tiered approach.