Abstract

A detailed water balance and conceptual flow model was calculated and developed for the Sandspruit catchment for the period 1990 to 2010 on a winter rainfall water-year (1 April - 31 March) basis. The Sandspruit catchment (quaternary catchment G10J) is located in the Western Cape Province of South Africa and is a tributary of the Berg River. It contributes significantly to the salinisation of the mid- to lower-reaches of the Berg River and thus the hydrological drivers need to be quantified and conceptualised in order to develop salinity management strategies. Various components of the water balance, i.e. precipitation, evaporation, streamflow, recharge, etc., were monitored and quantified. In addition, stable environmental isotopes and water balance modelling were used to perform hydrograph separation as well as to quantify components of the water balance. Annual streamflow in the catchment during the period of observation was variable, ranging between 0.026 mm·a-1 and 75.401 mm·a-1. Streamflow volumes also exhibit high variability between water years. Catchment annual rainfall varied between 351 and 655 mm·a-1, averaging at 473 mm·a-1. On average, 6.5% of rainfall was converted to streamflow during the period of observation. Evapotranspiration was found to be the dominant component of the water balance, as it comprises, on average, 94% of precipitation in the catchment. Groundwater recharge was calculated to average at 29 mm·a-1. The water balance model (J2000) performed well during the simulation period with all measures of performance exhibiting acceptable values. Simulation results indicate that streamflow is driven by interflow from the soil horizon (94.68% of streamflow), followed by overland flow (4.92% of streamflow). These results, together with the physiographic conditions evident in the catchment, were used to develop a conceptual flow model. Streamflow is interpreted to be driven by quickflow, i.e. overland flow and interflow, with minimal contribution from groundwater, and is also more dependent on the rainfall distribution in time rather than on the annual volume. The correlation between average annual streamflow and average rainfall was observed to be poor, suggesting that alternative factors, e.g. the spatial distribution of winter wheat, the temporal distribution of rainfall, climatic variables (temperature), etc., exert a greater influence on streamflow. The water balance and conceptual flow model will form the basis for the application of distributed hydrological modelling in the Sandspruit catchment and the development of salinity management strategies. Results from this investigation, e.g. ET estimates, methods to quantify groundwater recharge, hydrograph separation, etc., could potentially be extrapolated to other semi-arid areas.

Highlights

  • The quantification of a catchment’s water yield is a fundamental problem in hydrology, the volume of water available at the catchment outlet over a fixed time period (Poncea and Shetty, 1995)

  • This is lower than the average of 8.6% for South Africa suggested by Alexander (1985)

  • The results indicate that groundwater sampled at Zwavelberg exhibits a depleted signature, providing evidence that this is the source/ recharge area in the catchment and that recharge is immediate

Read more

Summary

Introduction

The quantification of a catchment’s water yield is a fundamental problem in hydrology, the volume of water available at the catchment outlet over a fixed time period (Poncea and Shetty, 1995). Conceptual models do not take into account the detailed geometry and small-scale variability of catchments, but rather consider the catchment as an ensemble of interconnected conceptual storages (Jothityangkoon et al, 2001). They allow for the identification or quantification of the principal factors (meteorological, plant, soil, etc.) which control the processes of water loss. They allow for the generation of synthetic sequences of hydrological data for various purposes, including water resources design and management (Xu, 1999). A lack of capacity and inadequate infrastructure does, mean that detailed information pertaining to all the terms of the water balance equation is rarely available to catchment hydrologists (Everson, 2001)

Methods
Results
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call