Validation of system models of deep geological disposal of high-level nuclear waste

In research environments related to commercial nuclear power, validation of models has been defined to be concerned with assuring that the model, to be validated, represents reality. Assurance, in this context, has been mostly argued on grounds of consensus of the relevant experts. In short, and much simplified, a model has been defined to be valid, if a sufficient number of experts have said so. In this paper, it is argued, strictly speaking you can never validate a model of the real world, only obtain an increasing trust in the ability of the model to predict reality. The only valid model of reality, is reality itself. As a contribution towards a better understanding of system models of deep geological disposal, the paper deals with validation of such models from a systemic point of view. it is concluded, that correct validation procedures will normally give a correct model