VANDPLANLAEGNING OG FORUDSIGELSER Usikkerheder i data, modeller og beslutninger i miljöplanlägning / Predictions in Water Resources Planning Uncertainties in data, models and decisions
The development and application of simulation models in water and environmenral planning and management have expanded greatly over the last decades. Such models are used in, for example, water quality assessments, and it is expected thai models will be applied for analyses and decision making to a greater and greater extent in the future. However, such models are approximate (better or worse) copies of reality, and parameters and input functions are essenually uncertain variables or processes. Consequently, the outpurs of such models are also uncertain variables or processes, which again has an important impact on assessments and decisions. It is, therefore, viral for a proper assessment and solution of environrnental problems that the inherent uncertainties in data and models are quantified and reduced to an acceptable level. The level of acceptance depends on the significance (economically, socially) of the decisions to be taken. The paper discusses an example in water quality modelling, and it is concluded that a proper assessment and documenration ofall significant uncertainty contributions can improve the reliability of decisions.