Quantifying and contextualising cyclone-driven, extreme flood magnitudes in bedrock-influenced dryland rivers. (January 2019)

• Record Type: Journal Article
• Authors: Heritage, George; Entwistle, Neil; Milan, David; Tooth, Stephen
• Is Part Of: Advances in water resources. Volume 123(2019)
• Page Start: 145
• Page End: 159
• Publication Date: 2019-01
• Subjects: Dryland river -- 2D hydraulic modelling -- Extreme flood -- Flood estimation -- Palaeoflood -- Sabie river -- Olifants river; Hydrology -- Periodicals; Hydrodynamics -- Periodicals; Hydraulic engineering -- Periodicals; 551.48
• DOI: 10.1016/j.advwatres.2018.11.006
• Languages: English
• ISSNs: 0309-1708

Abstract

Abstract : Novel 2D hydrodynamic modelling is used to estimate high magnitude floods on the Sabie and Olifants rivers, South Africa. We estimate extreme peak discharges in the range of 4470–5630 m 3 s −1 for the Sabie river (upstream catchment area 5715 km 2 ) and 14, 407–16, 772 m 3 s −1 for the Olifants river (upstream catchment area 53, 820 km 2 ). The Olifants 2012 peak discharge ranks among the largest recorded or estimated for any southern African river in the last couple of hundred years. On the KNP rivers and some other dryland rivers globally, increases in extreme flood magnitude and/or frequency potentially could trigger state changes from current mixed bedrock-alluvial systems characterised by high geomorphic and ecological diversity towards more dominantly bedrock channels with lower diversity. Abstract: In many drylands worldwide, rivers are subjected to episodic, extreme flood events and associated sediment stripping. These events may trigger transformations from mixed bedrock-alluvial channels characterised by high geomorphic and ecological diversity towards more dominantly bedrock channels with lower diversity. To date, hydrological and hydraulic data has tended to be limited for these bedrock-influenced dryland rivers, but recent advances in high-resolution data capture are enabling greater integration of different investigative approaches, which is helping to inform assessment of river response to changing hydroclimatic extremes. Here, we use field and remotely sensed data along with a novel 2D hydrodynamic modelling approach to estimate, for the first time, peak discharges that occurred during cyclone-driven floods in the Kruger National Park, eastern South Africa, in January 2012. We estimate peak discharges in the range of 4470–5630 m 3 s −1 for the Sabie River (upstream catchment area 5715 km 2 ) and 14, 407–16, 772 m 3 s −1 for the Olifants River (upstream catchment area 53, 820 km 2 ). These estimates place both floods in the extreme category for each river, with the Olifants peak discharge ranking among the largest recorded or estimated for any southern African river in the last couple of hundred years. On both rivers, the floods resulted in significant changes to dryland river morphology, sediment flux and vegetation communities. Our modelling approach may be transferable to other sparsely gauged or ungauged rivers, and to sites where palaeoflood evidence is preserved. Against a backdrop of mounting evidence for global increases in hydroclimatic extremes, additional studies will help to refine our understanding of the relative and synergistic impacts of high-magnitude flood events on dryland river development.