Reliable quantification of available water resources is an important prerequisite for sustainable management and development. However, the data required for adequate resource estimation are usually not available due to inadequate and shrinking measurement networks. Hydrologic models are the standard tool used to generate continuous estimates of stream flow and regionalisation has been the traditional approach to model application in poorly gauged basins. The classical approach to regionalization is to calibrate against naturalized observed data, identify hydrologically similar basins, and then transfer calibrated parameter sets to the ungauged basins. One of the problems related to this approach is the relatively few basins for calibration in southern Africa (due to data quality and availability problems), resulting in huge uncertainties in the prediction process. This study proposes an a priori process to directly estimate parameters of the Pitman monthly rainfall runoff model based on physical basin characteristics. The approach is tested on selected basins in southern African.