Water resource protection is central to sustainable water supply management for human wellbeing and for the ecological ecosystem to flourish. This review paper focuses on highlighting the role of groundwater and surface water protection practice to improve their sustainable utilization in South Africa. Using an integrative approach, this paper initially reflects on the history of water resources utilization, and it examines what is understood by the term “water resources protection”. This review paper then continues by providing an analysis of the current practice at global and local levels. The study found evidence of water resource utilization in the ancient times with limited challenges despite unavailability of regulation mechanisms. However, in recent times water resource availability challenges linked to water availability and water quality deterioration are evident globally despite having policies and regulation in place. Based on the findings of the study, a novel conceptual results-oriented policy monitoring framework is proposed, and it was distinctively designed to address challenges identified in practice in the South African context. The framework is configured on (1) the vision of the catchment protection and sustainable use, (2) water resources protection practices, (3) the state of water resources, and (4) policy evaluation and review processes as the key elements to drive success in policy implementation and monitoring practice. Therefore, this paper provides a foundation for results-oriented policy monitoring for water resources protection to improve sustainable water resources utilization in the country. The proposed framework can be considered as a reference guide that can be used to monitor policy practice for water resources protection by following a result-oriented monitoring approach

Empirical evidence continues to show that climate change remains a threat to the stability of the hydrologic system. As the climate system interacts with the hydrologic cycle, one significant repercussion of global warming includes changes in water availability at both regional and local scales. Climate change adaptation is intrinsically difficult to attain due to the dynamic earth system and lack of a comprehensive understanding of future climate and its associated uncertainties. Mostly in developing countries, climate adaptation is hampered by scarcity of good quality and adequate hydro-meteorological data. This article provides a synopsis of the modelling chain applied to investigate the response of the hydrologic system under changing climate, which includes choosing the appropriate global climate models, downscaling techniques, emission scenarios, and the approach to be used in hydrologic modelling. The conventional criteria for choosing a suitable hydrological model are discussed. The advancement of emission scenarios including the latest Shared Socioeconomic Pathways and their role in climate modelling, impact assessment, and adaptation, are also highlighted. This paper also discusses the uncertainties associated with modelling the hydrological impacts of climate change and the plausible approaches for reducing such uncertainties. Among the outcomes of this review include highlights of studies on the commonly used hydrological models for assessing the impact of climate change particularly in the sub-Saharan Africa region and some specific reviews in southern Africa. Further, the reviews show that as human systems keep on dominating within the earth system in several ways, effective modelling should involve coupling earth and human systems models as these may truly represent the bidirectional feedback experienced in the modern world. The paper concludes that adequate hydro-meteorological data is key to having a robust model and effective climate adaptation measures, hence in poorly gauged basins use of artificial neural networks and satellite datasets have shown to be successful tools, including for model calibration and validation.

Groundwater models are valuable tools to quantify the response of groundwater level to hydrological stresses induced by climate variability and groundwater extraction. These models strive for sustainable groundwater management by balancing recharge, discharge, and natural processes, with groundwater level serving as a critical response variable. While traditional numerical models are labour-intensive, machine learning and deep learning offer a data-driven alternative, learning from historical data to predict groundwater level variations. The groundwater level in wells is typically recorded as continuous groundwater level time series data and is essential for implementing managed aquifer recharge within a particular region. Machine learning and deep learning are essential tools to generate a data-driven approach to modeling an area, and there is a need to understand if they are the most suitable tools to improve model prediction. To address this objective, the study evaluates two machine learning algorithms - Random Forest (RF) and Support Vector Machine (SVM); and two deep learning algorithms - Simple Recurrent Neural Network (SimpleRNN) and Long Short-Term Memory (LSTM) for modeling groundwater level changes in the West Coast Aquifer System of South Africa. Analysis of regression error metrics on the test dataset revealed that SVM outperformed the other models in terms of the root mean square error, whereas random forest had the best performance in terms of the MAE. In the accuracy analysis of predicted groundwater levels, SVM achieved the highest accuracy with an MAE of 0.356 m and an RMSE of 0.372 m. The study concludes that machine learning and deep learning are effective tools for improved modeling and prediction of groundwater level. Further research can incorporate more detailed geologic information of the study area for enhanced interpretation.

Big data analytics (BDA) is a novel concept focusing on leveraging large volumes of heterogeneous data through advanced analytics to drive information discovery. This paper aims to highlight the potential role BDA can play to improve groundwater management in the Southern African Development Community (SADC) region in Africa. Through a review of the literature, this paper defines the concepts of big data, big data sources in groundwater, big data analytics, big data platforms and framework and how they can be used to support groundwater management in the SADC region. BDA may support groundwater management in SADC region by filling in data gaps and transforming these data into useful information. In recent times, machine learning and artificial intelligence have stood out as a novel tool for data-driven modeling. Managing big data from collection to information delivery requires critical application of selected tools, techniques and methods. Hence, in this paper we present a conceptual framework that can be used to manage the implementation of BDA in a groundwater management context. Then, we highlight challenges limiting the application of BDA which included technological constraints and institutional barriers. In conclusion, the paper shows that sufficient big data exist in groundwater domain and that BDA exists to be used in groundwater sciences thereby providing the basis to further explore data-driven sciences in groundwater management.

Machine learning and deep learning have demonstrated usefulness in modelling various groundwater phenomena. However, these techniques require large amounts of data to develop reliable models. In the Southern African Development Community, groundwater datasets are generally poorly developed. Hence, the question arises as to whether machine learning can be a reliable tool to support groundwater management in the data-scarce environments of Southern Africa. This study tests two machine learning algorithms, a gradient-boosted decision tree (GBDT) and a long short-term memory neural network (LSTM-NN), to model groundwater level (GWL) changes in the Shire Valley Alluvial Aquifer. Using data from two boreholes, Ngabu (sample size = 96) and Nsanje (sample size = 45), we model two predictive scenarios: (I) predicting the change in the current month’s groundwater level, and (II) predicting the change in the following month’s groundwater level. For the Ngabu borehole, GBDT achieved R² scores of 0.19 and 0.14, while LSTM achieved R² scores of 0.30 and 0.30, in experiments I and II, respectively. For the Nsanje borehole, GBDT achieved R² of −0.04 and −0.21, while LSTM achieved R² scores of 0.03 and −0.15, in experiments I and II, respectively. The results illustrate that LSTM performs better than the GBDT model, especially regarding slightly greater time series and extreme GWL changes. However, closer inspection reveals that where datasets are relatively small (e.g., Nsanje), the GBDT model may be more efficient, considering the cost required to tune, train, and test the LSTM model. Assessing the full spectrum of results, we concluded that these small sample sizes might not be sufficient to develop generalised and reliable machine learning models.

The devastating socioeconomic impacts of recent droughts have intensified the need for improved drought monitoring in South Africa (SA). This study has shown that not all indices can be universally applicable at all regions worldwide, and there is no single index that can represent all aspects of droughts. The aim of this study was to review the performance and applicability of the Palmer drought severity index (PDSI), surface water supply index (SWSI), vegetation condition index (VCI), standardised precipitation index (SPI), standardised precipitation evapotranspiration index (SPEI), standardised streamflow index (SSI), standardised groundwater index (SGI), and GRACE (Gravity Recovery and Climate Experiment)-based drought indices in SA and provide guidelines for selecting feasible candidates for integrated drought monitoring. The review is based on the ‘2016 World Meteorological Organisation (WMO) Handbook of Drought Indicators and Indices’ guidelines. The PDSI and SWSI are not feasible in SA, mainly because they are relatively complex to compute and interpret and cannot use readily available and accessible data. Combining the SPI, SPEI, VCI, SSI, and SGI using multi-index or hybrid methods is recommended. Hence, with best fitting probability distribution functions (PDFs) used, and with an informed choice between parametric and non-parametric approaches, this combination has the potential for integrated drought monitoring. Due to the scarcity of groundwater data, investigations on the use of GRACE-based groundwater drought indices must be carried out. These findings may contribute to improved drought early warning and monitoring in SA.