Résumé
Runoff simulation in highly anthropized catchments is complex, but essential for water management, especially in poorly gauged and data-scarce hydrosystems of the West African Sahel. This study aims to evaluate the effect of different calibration schemes on runoff simulation. The physically-based and semi-distributed hydrological SWAT (Soil and Water Assessment Tool) model is used to simulate daily runoff in the Nakanbé catchment at Wayen station in Burkina Faso (in the West African Sahel) over the period 2006-2012. Four (4) hydrometric stations (Dombré, Rambo, Ramsa and Wayen) gauging 4 nested catchments (ranging from 1060 to 21 178 km.sup.2 in size) are considered. The added value of the consideration of nested catchments is assessed through the following 3 calibration schemes: one-site (OS) at the entire catchment outlet (Wayen); multi-sites with nested sub-catchments (MS1); and multi-sites without considering nested sub-catchments (MS2). The results indicate that OS and MS2 schemes perform well (KGE 0.7, |PBIAS|3 %), with MS2 scheme being superior (KGE, PBIAS). However, the MS1 scheme (KGE = 0.68; PBIAS = -22.9 %) performed worse in comparison to the traditional OS scheme. The comparison of the three modelling schemes provides evidence that accounting for nested sub-catchments does not necessarily improve the quality of rainfall-runoff simulations. Yet, multi-site calibration should be favoured when catchments are not nested.