Hydrological modelling, process understanding and uncertainty in a southern African context: lessons from the northern hemisphere

Abstract

AbstractDuring the four decades of Keith Beven's career there have been many developments in the science of hydrological modelling. Some have focused on the links between hydrological process understanding and the structure and complexity of hydrological models, others on the related issues of modelling uncertainty. The southern Africa region continues to be generally less well endowed with the resources required to contribute to these research developments, but they are critical for successful water resources management decision‐making in data scarce areas, and go beyond academic interest. Consequently, the focus in the region has been on adding a local context to northern hemisphere research as well as trying to put it into practice. The challenge in southern Africa has always been to extrapolate from published research ideas and decide how they can be effectively used in larger scale practical applications in data‐poor areas. The paper examines the issues of model complexity, links with process understanding and the broad topic of model uncertainty estimation in the context of data scarce areas and how the science questions relate to improvements in water resources decision making. The conclusions suggest that the southern African region has benefited a great deal from several decades of northern hemisphere research (including those by Beven) and that some values have been added through the focus on practical implementation. The region should also embrace the opportunities presented by the need to link realistic uncertainty estimates with risk‐based water resources decision‐making, thereby contributing to the international debate on this important topic. Copyright © 2015 John Wiley & Sons, Ltd.