Determination of grid scale parameters in catchment modelling by upscaling local scale parameters

A review of scaling approaches and their advantages and limitations is given. Furthermore a framework classifying the different scaling approaches into four groups is proposed. In aggregation the smaller scale equations are applied at the smaller scale and outputs are subsequently aggregated to larger scale units. This allows smaller scale parameters to be assessed directly from field data. In upscaling, a distinction is made between three cases: (i) the smaller scale equations are assumed valid without change, (ii) the smaller scale equations are extended in a theoretical framework to account for the spatial variability of smaller scale parameter variability, and (iii) new equations are developed particularly for the larger scale. Only in case (ii) is it possible to assess the large-scale parameters directly from field data; however generalisation of this approach to catchment scale models is a formidable task. Effective parameters in practise have to be assessed through calibration in most cases. The four scaling approaches as well as various combinations are illustrated by a series of case studies.