TOPKAPI: a model for the representation of the rainfall‐runoff process at different scales

Abstract The paper introduces a new distributed rainfall‐runoff model derived upon the assumption that the horizontal flow at a point in the soil and over the surface can be approximated by means of a kinematic wave model. The point assumption is then integrated up to a finite pixel dimension, thus converting the original differential equation into a non‐linear reservoir equation based upon physically meaningful parameters, the solution of which can be found numerically. The catchment behaviour is finally obtained by aggregating the non‐linear reservoirs into three cascades, representing the soil, the surface and the drainage network, following the topographic and geomorphologic elements of the catchment. The main advantage of this approach lies in its capability of being applied at increasing spatial scales without losing the physical interpretation of the model and parameters. Other advantages of the TOPKAPI approach can be found in the full use of the digital elevation model, soil maps, land‐use maps, etc., for the derivation of the non‐linear reservoir cascade and for the estimate of the model parameters. This allows for the extension of the model to ungauged catchments and as a promising tool for the impact assessment of climatic as well as land‐use changes. Three case‐studies are presented, showing the model application to catchments that are different in nature and size (from a few to over a thousand square kilometres) and described by means of widely different elementary grid cells (pixels). The results of these applications are interesting, and they essentially fulfil the general purpose. Although the work to be done is still great, the model appears to be a promising tool worthy of further investigation. Copyright © 2002 John Wiley & Sons, Ltd.