Calibration and Validation of a Dynamic Soil Acidification Model at a Large Spatial Scale

摘要

The relatively simple soil acidification model, SMART2, is especially developed for the application at national and European scale, however, it still runs at a point support. In order to increase confidence in model predictions at large spatial scales, the model was calibrated and validated for the Netherlands, using a resolution which is feasible for Europe as a whole. In order to bring available point data to the same support as the model results, about 250 point observations on soil solution concentrations in forest soils where up-scaled towards a 5 x 5 km~2 grid map. A comparison of the map with model predictions using nominal parameter values and the map with the up-scaled observations showed that the predicted Al and NO_3 concentration were clearly over estimated by the model. Nonetheless, the nominal model results were in the 95% confidence interval of the up-scaled observations, calibration clearly improved the model predictions and was able to reduce the uncertainty in model input data. Model validation showed that the model error was relatively large for the nominal run, whereas calibration strongly reduced the model error.