Holistic approach to groundwater recharge assessment in the Upper Crocodile River Basin, Johannesburg, South Africa
Date
2019-07
Authors
Leketa, Khahliso Clifford
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
In the highly urbanised and water stressed Upper Crocodile River Basin (UCRB), Johannesburg, South Africa, which has wastewater induced surface water resources and a complex land use and geological setting, a holistic assessment of groundwater recharge was undertaken to understand the groundwater provenance and to address the prevailing water security issues. This work emanated from a project entitled, “Understanding Groundwater Recharge in the Limpopo River Basin (GRECHLIM)”.
Samples from Johannesburg rainfall, surface water and groundwater were collected and analysed for δ18O, δ2H and 3H, while selected groundwater sources were additionally sampled and analysed for δ13C and 14C. Albert Farm spring was monitored monthly for stable isotopes and yield over a period of 14 months. The Thornthwaite Monthly Water Balance Model (TMWBM), Stream Segment Water Budget, Reservoir Water Budget, Baseflow Separation and Water Table Fluctuation methods were applied to assess groundwater recharge, while Precipitation Runoff Modelling System (PRMS) was applied to assess the impacts of changes in climate variables on the hydrology of the UCRB.
The temperature and amount effects were observed in rainfall and the Johannesburg Local Meteoric Water Line (JLMWL) was constructed as ��2��=6.7��18��+10‰, while the temperature-δ18O regression line was ���� 18=0.552��−14.1‰. A high stable isotope variability was observed in the Johannesburg rainfall with the most depleted sample having δ2H and δ18O values of -131.6‰ and -17.82‰, respectively, and the most enriched rainfall having 51.4‰ and 7.03‰, respectively. A high variability was also observed in the monthly stable isotope samples of the Albert Farm spring with the signature ranging from -6.01‰ to -2.28‰ for δ18O and from -21.6‰ to -10.48‰ for δ2H with d-excess ranging from 3.23‰ to 18.7‰ for the review period. The majority of the groundwater samples across the UCRB plot close to the JLMWL and the most depleted samples were those from the dolomitic aquifer, while the most enriched were the boreholes positioned downstream of the Hartbeespoort Dam. The samples from the dolomite aquifer were enriched in stable carbon isotope (δ13C), while those from the granites, shales and quartzites were depleted, with intermediate values observed at the margins of granites and dolomites. The 14C results indicated that groundwater in the UCRB has an MRT between 0 (present) and 3856 years, deducing high aquifer heterogeneity. The estimated daily air temperature at the time of recharge is similar to the current daily air temperature during rainfall, while the estimated annual air temperatures are lower than the present annual temperatures.
An average potential recharge estimate of 4.3% was obtained in the UCRB using the TMWBM with low amounts estimated in the warmer low elevation northern part of the UCRB and higher amount in the cooler elevated southern portion. This indicated the influence of temperature on recharge. The Stream Segment and Reservoir Water Budget methods indicated spatial and temporal net gain and net loss across the UCRB and the Reservoir Water Budget indicated that annually, the Hartbeespoort Dam experiences a net loss of 7% of its inflow, which is referred to as focused recharge amounting to 2 084 131 m3. The focused recharge is deduced to occur through the Brits fault lines and the bedding planes in the Magaliesberg quartzite at the rate of 202 m/year deduced from tritium analysis. The application of the Baseflow Separation method in different quaternary catchments of the UCRB gave an area weighted average recharge estimate of 6.7% for the UCRB, while the Water Table Fluctuation method gave an average recharge estimate of 11.4% in the dolomitic aquifers. The hydrograph and stable isotope analysis of the piezometers indicated the influence of interflow contribution, deducing that the baseflow estimate in the basement granites is inclusive of interflow.
It is concluded that change in seasons and moisture sources are the causes of the high variability in rainfall isotopes. The high temporal variability in stable isotopes and d-excess values from the same groundwater source further confirms that the UCRB receives rainfall from different moisture sources. This indicates that in regions that receive rainfall of different isotopic signature, the use of a single sample to characterise recharge may not be a good practice as this may erroneously yield biased interpretations. Effective recharge in the UCRB predominantly occurs through diffuse mechanisms, while focused recharge is mainly observed near the surface water bodies. The depleted nature (in δ18O and δ2H) of groundwater samples in the dolomitic aquifers is deduced to be a result of altitude effect or possibilities of an exchange of oxygen between the dissolved CO2 and groundwater. However, the possibilities of oxygen exchange have been recommended for further investigation. The stable carbon isotopes (δ13C) indicated closed system carbon evolution in the dolomitic aquifer and open system in the granites, shales and quartzites. From the signature of stable isotope of carbon, oxygen and hydrogen, it is further concluded that there is mixing of the two water types at the periphery of the dolomitic and granitic aquifers.
Combined interpretation of 14C age and the estimated annual air temperatures indicates that recharge of older groundwater occurred at lower temperatures, while recharge of more recent groundwater occurred at warmer temperatures showing evidence of increase in annual air temperatures with time. The observed similarity between the calculated and the current daily temperatures on one side, and the observed variability between the calculated and current annual air temperatures on another side, could be related to the fact that regardless of the changes in global temperatures, the physical in-cloud condensation conditions, which dictate the isotopic signature in daily rainfall are not changing.
In general, the study indicates that the stable isotope effects in rainfall can successfully be used to assess the air temperature on the day of the rainfall that generated recharge and to determine the dependency of recharge amount on rainfall amount if the aquifer is recharged by rainfall without undergoing extreme evaporation prior to recharge.
The climate simulations indicate that as baseflow decreases due to changes in climate variables and wastewater increases due to growth in urbanisation, the amount of wastewater shall dominate baseflow in the UCRB leading to more polluted surface water. Therefore, better treatment and constraints on pollution are recommended for the future.
Description
A thesis submitted to the Faculty of Science, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the degree of Doctor of Philosophy.
Keywords
Citation
Leketa, Khahliso Clifford, Holistic approach to groundwater recharge assessment in the Upper Crocodile River Basin, Johannesburg, South Africa, University of the Witwatersrand, Johannesburg, <http://hdl.handle.net/10539/29632>