The present paper examines the validity of the application of a distributed rainfall-runoff-erosion model on a regional basis in a semi-arid region of Brazil. The model tested is known as WESP (Water Erosion Simulation Program) developed by Lopes (1987). The model simulates the hydrograph and the sediment graph for individual events and thus, the model when properly calibrated can serve as a predictive tool for runoff and soil erosion on an individual or a continuous basis. The possibility of the existence of regionally applicable values for the parameters of the model WESP would be very interesting from the point of view of predictability of runoff and soil erosion from ungauged basins in the region. For this purpose, data collected in several experimental units in the experimental basin of Sumé as well as in the experimental basin of So Joo de Cariri have been used. The model was calibrated and validated in each of the experimental units (consisting of erosion plots and micro-basins) and the average value from all of the units for each of the parameters was considered to be the regional value. With these average values, all the events in all the experimental units were simulated. The results show that the runoff and erosion values simulated with this single parameter set were very good in all the units, being well within the acceptable deviations. The present paper examines the validity of the application of a distributed rainfall-runoff-erosion model on a regional basis in a semi-arid region of Brazil. The model was known as WESP (Water Erosion Simulation Program) developed by Lopes (1987) The model simulates the hydrograph and the sediment graph for individual events and thus, the model when properly calibrated can serve as a predictive tool for runoff and soil erosion on an individual or a continuous basis. The possibility of the existence of regionally applicable values ​​for for the parameters of the model WESP would be very interesting from the point of view of predictability of runoff and soil erosion from unguged basins in the region. For this purpose, data collected in several experimental units in the experimental basin of Sumé as well as in the experimental basin of So Joo de Cariri have been used. The model was calibrated and validated in each of the experimental units (consisting of erosion plots and micro- With these average values, all the events in all the experimental units were simulated. The results show that the runoff and erosion values ​​simulated with this single parameter set were very good in all the units, being well within the acceptable deviations.