A new hydrological and soil erosion model has been developed and tested: LISEM, the Limburg soil erosion model. The model uses physically based equations to describe interception, infiltration and soil water transport, storage in surface depressions, splash and flow detachment, transport capacity and overland and channel flow. From the validation results it is clear that, although the model has several advantages over other models, the results of LISEM 1.0 are far from perfect. Based on the sensitivity analysis and field observations, the main reasons for these differences seems to be the spatial and temporal variability of the soil hydraulic conductivity and the initial pressure head at the basin scale. Another reason for the differences between measured and simulated results is our lack or understanding of the theory of hydrological and soil erosion processes. KEY WORDS soil erosion; catchment hydrology; simulated models; validation; sensitivity analysis; calibration; scenarios THE LISEM MODEL Within the scope of the erosion project described by De Roo (1996), a new physically based hydrological and soil erosion model has been developed, which can be used for planning and conservation purposes: the Limburg soil erosion model (LISEM) (De Roo et al., 1996). The development and structure of the LISEM model is based on experiences with the ANSWERS model (Beasley et al., 1980; De Roo et al., 1989; De Roo, 1993) and SWATRE (Belmans et al., 1983), but the process descriptions are changed totally. The processes incorporated into the model are rainfall, interception, surface storage in micro-depressions, infiltration, vertical movement of water in the soil, overland flow, channel flow, detachment by rainfall and throughfall, detachment by overland flow and transport capacity of the flow. In addition, the influence of tractor wheelings and small paved roads (smaller than the pixel size) on the hydrological and soil erosion processes is taken into account. Vertical movement of water in the soil is simulated using the Richard's equation. Storage of surface water in micro-depressions is modelled using the concept of random roughness. For the distributed flow routing, a four-point finite-difference solution of the kinematic wave is used together with Manning's equation.
ADVANTAGES AND DISADVANTAGES OF THE LISEM MODEL
Validating a model entails not only a comparison of the measured and simulated outputs, but also a 'mathematical model validation ' (Howes and Anderson, 1988). In this stage it is verified whether the assumptions made about the real system by the model are reasonable and whether the model adequately reflects the essential features and behaviour of the real system that are relevant to the application in mind.