Modelling Across Scales: The Medalus Family of Models

Abstract

Soil erosion forecasting is relevant at a wide range of time and space scales; from the field scale up to national or global scales. Changes of time and space scales generally go together, because finer time resolution requires better understanding of hydrological and sediment transport processes, and generally implies a more detailed spatial scale. At different scales, different groups of processes tend to become dominant, so that the effective focus of the models also changes. At the scale of the single erosion plot up to the hillslope catena, the timing and volume of overland flow hydrographs is critical, together with its distribution across rill and inter-rill areas. At the scale of the catchment, topography, soil and vegetation patterns become more important, and it is essential to consider periods from single storm events up to several decades, over which these patterns may change. At the coarser national to global scales, climate and lithology become the critical variables, with associated time spans from a few years up to the hundreds or thousands of years over which significant climatic change naturally occurs.