Models Or Measurements

No abstracts are to be cited without prior reference to the author.

Many simulation models in the field of ecology and ecophysiology are based on an imperfect understanding of the dynamics of the object systems. This results-often in highly uncertain model outcomes because of uncertain model structure, parameters and inputs. Calibration is the attempt to improve the fit between observations (data) and model results, given the body of knowledge incorporated in the model (including parameters). Data validation evaluates how well the calibrated model with new inputs predicts new observations. In the approach discussed in this paper uncertainty ranges or distributions in the observations are user defined, in accordance wi_th subjective or problem-domain dependent acceptable errors. The uncertainty in the model outcomes is the uncertainty in the a posteriori parameter vectors, produced by the calibration. These parameter vectors are used to compute model results with remaining uncertainty.