Determining Hydrological Variability Using a Multi-Catchment Model Approach for the Western Cape, South Africa

Abstract

Understanding the impacts of climate change requires the development of hydrological modelling tools. However, data scarcity hinders model application, performance, process simulation and uncertainty, especially for Sub-Saharan Africa. In this study, a multi-catchment approach was used to assess hydrological process variability in the Western Cape (WC) of South Africa using the JAMS/J2000 rainfall–runoff model and a Monte Carlo analysis (MCA). Due to much steeper slopes and lower evapotranspiration, the models suggest that WC is dominated by surface runoff from mountainous regions and regional groundwater flow. The results highlight the impact of the catchment size, availability and position of hydroclimatic and anthropogenic factors and the frequency of the signal-to-noise ratio (water balance). For large catchments (>5000 km2), the calibration was able to achieve a Nash–Sutcliffe efficiency (NSE) of 0.61 to 0.88. For small catchments (<2000 km2), NSE was between 0.23 to 0.39. The large catchments had an overall surface runoff, interflow and baseflow contribution of 44, 19 and 37%, respectively, and lower overall uncertainty. The simulated flow components for the small catchments were variable and these results are less certain. The use of a multi-catchment approach allows for identifying the specific factors impacting parameter sensitivities and in turn provides a means to improve hydrological process simulation.

Between the Cape Fold Mountains and the deep blue sea: Comparative phylogeography of selected codistributed ectotherms reveals asynchronous cladogenesis

“…The CFM may have also played a causal role in other factors which influence the phylogeographic structure of taxa, such as drainage systems, climatic patterns and vegetation types. The CFM syntaxis zone represents the interface between western and southern flowing drainage systems, serving to delineate the Eerste and Berg River catchment areas in the west, and the Breede and Bot River catchments to the southeast (Watson et al, 2021 ). Price et al ( 2010 ) highlighted the significant impact which watersheds and catchments can play in cladogenesis even in the absence of geographic features as dramatic as the CFM, and high mountain streams have been frequently shown to demonstrate high levels of species richness in invertebrates, often with distinct biota in discrete mountain catchments (Daniels, 2003 ; Gouws et al, 2004 , 2010 ; Stewart, 1992 ; Wishart & Hughes, 2001 ).…”

Towards the development of an isotope‐enabled rainfall‐runoff model: Improving the ability to capture hydrological and anthropogenic change

“…Bias correction can be difficult in environments where extreme isotope signatures are the most important and drive the hydrological response, such as for the Western Cape (van Rooyen et al, 2021;Watson, Eilers, & Miller, 2020), but various quantile mapping approaches (Gudmundsson et al, 2012) are available and can be used under different climatic conditions. As the JAMS/J2000 has a modular component structure these new developments could support a wider use of isotope-enabled models, given that the JAMS/J2000 has already been applied across many environments and climates (Fink et al, 2007(Fink et al, , 2017Künne et al, 2019;Meinhardt et al, 2018;Nepal et al, 2021;Watson et al, 2021).…”

“…Conceptual rainfall‐runoff models can use the streamflow signal to separate the hydrograph into proportions of flow from surface runoff, interflow and baseflow (Jakeman & Hornberger, 1993; Krause, 2001; Willems, 2009). However, as most conceptual model calibrations are usually solely dependent on the streamflow signal, background noise, such as reservoir release and irrigation withdrawals amongst others, can influence model hydrograph separation (Payan et al, 2008; Watson et al, 2021; Xue et al, 2014). Understanding the proportion of surface runoff, interflow and baseflow, which provides insight into crucial hydrological processes such as infiltration, evapotranspiration/root water uptake and groundwater recharge/deep percolation, is important to constrain and predict hydrological change.…”

“…The Berg River catchment has recently been subject to an intense drought (2015)(2016)(2017)(2018), and future climate adaptation measures hinge on the ability to accurately simulate current and future hydrological flows. The approach investigated the potential of reducing the model parameter uncertainty of previous studies (Watson et al, 2021;Watson, Kralisch, et al, 2020;Watson, Miller, Künne, & Kralisch, 2022) and whether more robust model parameters could be identified for modelling under enhanced climate variability outlined by Watson, Midgley, Ray, et al, 2022. Furthermore, the catchment has shown recent changes in average hydrological flows (Watson, Miller, Künne, & Kralisch, 2022), as a response to changes in the upper Berg River, and the J2000iso was used to validate these changes.…”

How Climate Extremes Influence Conceptual Rainfall-Runoff Model Performance and Uncertainty

“…In order to understand the relative sensitivities of the parameters θ and ones influenced by climatic disturbances, θ dry values were compared with θ wet from the NSGA-II calibration. To reduce the overall number of parameters and to include parameters with the most notable solution direction differences, parameters identified as the most sensitive in a previous study were selected for the MCA and DYNIA (Watson et al, 2021b). These parameters likewise were presumed to have the largest impact (most sensitive) on the simulated streamflow time of concentration.…”