The soil acidification model SMART2 has been developed for application on a national to a continental scale. In this study SMART2 is applied at the European scale, which means that SMART2 was applied to the Netherlands with data that are available at the European scale. In order to calibrate SMART2, for the country as a whole, calibration data and model output had to be attuned to the same support. We assume that for application at the European scale all required data are available at a gridsize of 5* 5 km2. Therefore point measurements at 250 forest stands were upscaled to a 5 * 5 km2 block support, using regression models, derived from ancillary data at a national scale. After aggregation to a 5 * 5 km2 block support, the regression results were subsequently refined by the spatial correlated part of the residuals. The calibration of the model SMART2 resulted in lower predicted concentrations of nitrate and aluminium than the nominal model run that overestimated nitrate and aluminium concentrations. The calibration of the mineralization rate constant affected the nitrate concentrations mostly, whereas aluminium concentrations were mostly affected by the exchange constants.
|Place of Publication||Wageningen|
|Number of pages||48|
|Publication status||Published - 1999|