Modeling and regulating hydrosalinity dynamics in the Sandspruit river catchment (Western Cape)

[摘要] ENGLISH ABSTRACT: Bugan, R.D.H. Modelling and regulating hydrosalinity dynamics in the Sandspruit Rivercatchment (Western Cape). PhD dissertation, Stellenbosch University.The presence and impacts of dryland salinity are increasingly become evident in the semi-aridWestern Cape. This may have serious consequences for a region which has already beenclassified as water scarce. This dissertation is a first attempt at providing a methodology forregulating the hydrosalinity dynamics in a catchment affected by dryland salinity, i.e. theSandspruit catchment, through the use of a distributed hydrological model. It documents theentire hydrological modelling process, i.e. the progression from data collection to modelapplication. A review of previous work has revealed that salinisation is a result of land usechange from perennial indigenous deep rooted vegetation to annual shallow rooted croppingsystems. This has altered the water and salinity dynamics in the catchment resulting in themobilisation of stored salts and subsequently the salinisation of land and water resources. Theidentification of dryland salinity mitigation measures requires thorough knowledge of the waterand salinity dynamics of the study area. A detailed water balance and conceptual flow model wascalculated and developed for the Sandspruit catchment. The annual streamflow and precipitationranged between 0.026 mm a-1 - 75.401 mm a-1 and 351 and 655 mm a-1 (averaging at 473 mm a-1), respectively. Evapotranspiration was found to be the dominant component of the waterbalance, as it comprises, on average, 94% of precipitation. Streamflow is interpreted to be drivenby quickflow, i.e. overland flow and interflow, with minimal contribution from groundwater.Quantification of the catchment scale salinity fluxes indicated the Sandspruit catchment is in astate of salt depletion, i.e. salt output exceeds salt input. The total salt input to and output fromthe Sandspruit catchment ranged between 2 261 - 3 684 t Catchment-1 and 12 671 t a-1 - 21 409 ta-1, respectively. Knowledge of the spatial distribution of salt storage is essential for identifyingtarget areas to implement mitigation measures. A correlation between the salinity of sedimentsamples collected during borehole drilling and the groundwater EC (r2 = 0.75) allowed for thepoint data of salt storage to be interpolated. Interpolated salt storage ranged between 3 t ha-1 and674 t ha-1, exhibiting generally increasing storage with decreasing ground elevation. Thequantified water and salinity fluxes formed the basis for the application of the JAMS/J2000-NaClhydrological model in the Sandspruit catchment. The model was able to adequately simulate thehydrology of the catchment, exhibiting a daily Nash-Sutcliffe Efficiency of 0.61. The simulatedand observed salt outputs exhibited discrepancies at daily scale but were comparable at an annualscale. Recharge control, through the introduction of deep rooted perennial species, has beenidentified as the dominant measure to mitigate the impacts of dryland salinity. The effect of various land use change scenarios on the catchment hydrosalinity balance was evaluated with theJAMS/J2000-NaCl model. The simulated hydrosalinity balance exhibited sensitivity to land usechange, with rooting depth being the main factor, and the spatial distribution of vegetation. Revegetationwith Mixed forests, Evergreen forests and Range Brush were most effective inreducing salt leaching, when the 'salinity hotspots were targeted for re-vegetation (Scenario 3).This re-vegetation strategy resulted in an almost 50% reduction in catchment salt output. Overall,the results of the scenario simulations provided evidence for the consideration of re-vegetationstrategies as a dryland salinity mitigation measure in the Sandspruit catchment. The importanceof a targeted approach was also highlighted, i.e. mitigation measures should be implemented inareas which exhibit a high salt storage.

[发布机构] Stellenbosch University