Objective To assess the 'cumulative' uncertainty attached to Spatial Decision Support System (SDSS) for managing groundwater quality at European level with special emphasis on nitrates and aluminium contamination as caused by agricultural activities and atmospheric deposition; and to determine the magnitude of error caused by individual sources in order to place the balance of error in a risk assessment perspective. The potential strategic implications on uncertainty linked to SDSS will also be assessed.Analytical models and Geographic Information System (GIS), the main building blocks of SDSS, carry with them uncertainties which will, in a cumulative process, be propagated and reflected in the system output. Limits of confidence and accuracy of SDSS are, currently, not available in a formalised and comprehensive way. Research is required in order to assess the cumulative uncertainty associated with sampling, analysis, data collections, approximation and constraints in modelling, model calibration and spatial resolution. The preliminary phase of the research project will be focused on the adaption of the physically-based models on the national and European scale. MIKE-SHE - hydrological modelling system - coupled with the crop-soil model DAISY will be used for modelling nitrate contamination of groundwater from agricultural sources while SMART 2 - acidification model - will be used for studying the leaching of aluminium and nitrate from natural systems. A general interface enabling ARC/INFO to be read by the MIKE SHE and SMART2 models and vice versa will be made in order to facilitate an efficient transfer of data from standard data- bases available on European level. The preliminary phase will also include the gathering of geo-referenced data. The first phase will aim at determining the overall error in the model output. By means of validation procedures in which model predictions are compared with real values, accuracy of the models will be assessed. Effects of scale change on model performance output will also be evaluated. Variability of the spatial attributes as their statistical properties are size dependent has important implications on the accuracy of the model output when used at various scales.Therefore the error linked to input data, mainly the geo-referenced data, will be quantified in a further step. Both parametric (kriging method) and non-parametric method (Bayesian Markov geostatiscal method) will be used.Uncertainty in the inputs to the environmental models will propagate to the output.A sophisticated MONTE CARLO simulation, which is quite suitable for a parallel computing approach, will be used to analyze this error propagation. The stochastic simulation module will then be built into the interface between GIS and the dynamic models. During the second phase of the research project particular attention will be paid to the balance of error and the minimisation of uncertainty in the light of sampling strategy and risk assessment. The contribution of individual error sources can be obtained by using the partitioning property, which states that the output variance is approximately equal to a sum of contributions, each of which is attributable to the error of an individual output. Minimisation of uncertainty will be studied in the light of a cost-efficiency analysis. Finally the focus of the research will be on the analysis of how a better understanding of the distribution of uncertainties linked to SDSS can contribute to the optimisation of cost-effectiveness in land-use planning and other regulatory efforts to protect groundwater resources at national level. Fields of science natural sciencesearth and related environmental sciencesphysical geographycartographygeographic information systemsnatural scienceschemical sciencesinorganic chemistrypost-transition metalsnatural sciencescomputer and information sciencessoftwaresoftware applicationssimulation software Programme(s) FP4-ENV 2C - Specific programme of research and technological development in the field of environment and climate, 1994-1998 Topic(s) 01010401 - European water resources Call for proposal Data not available Funding Scheme CSC - Cost-sharing contracts Coordinator Dansk Hydraulisk Institut EU contribution No data Address 5,Agern Allé 5 2970 Hørsholm Denmark See on map Total cost No data Participants (5) Sort alphabetically Sort by EU Contribution Expand all Collapse all CHALMERS UNIVERSITY OF TECHNOLOGY Sweden EU contribution No data Address 8,Sven Hultins Gata 8 412 96 GOETEBORG See on map Total cost No data Dienst Landbouwkundig Onderzoek Netherlands EU contribution No data Address 11,Marijkeweg 6708 PE Wageningen See on map Total cost No data Geographic Information Management SA Luxembourg EU contribution No data Address 9,Rue Jean-Pierre Sauvage 2514 Luxembourg See on map Total cost No data THE ROYAL VETERINARY AND AGRICULTURAL UNIVERSITY Denmark EU contribution No data Address 10,Agrovej 10 2630 TAASTRUP See on map Total cost No data Universiteit van Amsterdam Netherlands EU contribution No data Address 130,Nieuwe Prinsengracht 1018 VZ Amsterdam See on map Total cost No data