Towards a limits of acceptability approach to the calibration of hydrological models: Extending observation error

Within the GLUE methodology, there are a number of advantages of taking a limits of acceptability approach to model evaluation for non-ideal applications where the strong assumptions of statistical identification might be difficult to justify. However, there is a question of how the limits of acceptability might be specified in a way that reflects the different sources of uncertainty in the modeling process. Here, a novel method for identifying behavioural models in an extended GLUE methodology is developed and applied to an application of Dynamic TOPMODEL to the Attert catchment in Luxemburg with semi-distributed inputs to nested sub-catchments. The results raise some important issues about testing model structures as hypotheses of catchment responses.