|
|
Aderemi, B. A., Olwal, T. O., Ndambuki, J. M., & Rwanga, S. S. (2023). Groundwater levels forecasting using machine learning models: A case study of the groundwater region 10 at Karst Belt, South Africa. Systems and Soft Computing, 5, 200049.
Abstract: The crucial role which groundwater resource plays in our environment and the overall well-being of all living things can not be underestimated. Nonetheless, mismanagement of resources, over-exploitation, inadequate supply of surface water and pollution have led to severe drought and an overall drop in groundwater resources’ levels over the past decades. To address this, a groundwater flow model and several mathematical data-driven models have been developed for forecasting groundwater levels. However, there is a problem of unavailability and scarcity of the on-site input data needed by the data-driven models to forecast the groundwater level. Furthermore, as a result of the dynamics and stochastic characteristics of groundwater, there is a need for an appropriate, accurate and reliable forecasting model to solve these challenges. Over the years, the broad application of Machine Learning (ML) and Artificial Intelligence (AI) models are gaining attraction as an alternative solution for forecasting groundwater levels. Against this background, this article provides an overview of forecasting methods for predicting groundwater levels. Also, this article uses ML models such as Regressions Models, Deep Auto-Regressive models, and Nonlinear Autoregressive Neural Networks with External Input (NARX) to forecast groundwater levels using the groundwater region 10 at Karst belt in South Africa as a case study. This was done using Python Mx. Version 1.9.1., and MATLAB R2022a machine learning toolboxes. Moreover, the Coefficient of Determination (R2);, Root Mean Square Error (RMSE), Mutual Information gain, Mean Absolute Percentage Error (MAPE), Mean Squared Error (MSE), Mean Absolute Error (MAE), and the Mean Absolute Scaled Error (MASE)) models were the forecasting statistical performance metrics used to assess the predictive performance of these models. The results obtained showed that NARX and Support Vector Machine (SVM) have higher performance metrics and accuracy compared to other models used in this study.
|
|
|
|
Tamagnone, P., Comino, E., & Rosso, M. (2020). Rainwater harvesting techniques as an adaptation strategy for flood mitigation. Journal of Hydrology, 586, 124880.
Abstract: The development of adaptation and mitigation strategies to tackle anthropic and climate changes impacts is becoming a priority in drought-prone areas. This study examines the capabilities of indigenous rainwater harvesting techniques (RWHT) to be used as a viable solution for flood mitigation. The study analyses the hydraulic performance of the most used micro-catchment RWHT in sub-Saharan regions, in terms of flow peak reduction (FPR) and volume reduction (VR) at the field and basin scale. Parametrized hyetographs were built to replicate the extreme precipitations that strike Sahelian countries during rainy seasons. 2D hydrodynamic simulations showed that half-moons placed with a staggered configuration (S-HM) have the best performances in reducing runoff. At the field scale, S-HM showed a remarkable FPR of 77% and a VR of 70% in case of extreme rainfall. Instead at the basin scale, in which only 5% of the surface was treated, 13% and 8% respectively for FPR and VR were obtained. In addition, the reduction of the runoff coefficient (Rc) between the different configuration was analyzed. The study critically evaluates hydraulic performances of the different techniques and shows how pitting practices cannot guarantee high performance in case of extreme precipitations. These results will enrich the knowledge of the hydraulic behavior of RWHT; aspect marginally investigated in the scientific literature. Moreover, this study presents the first scientific application of HEC-RAS as a rainfall-runoff model. Despite some limitations, this model has the effective feature of using very high-resolution topography as input for hydraulic simulations. The results presented in this study should encourage stakeholders to upscale the use of RWHT in order to lessen the flood hazard and land degradation that oppresses arid and semi-arid areas.
|
|
|
|
Kamruzzaman, M., & Chowdhury, A. (2023). Flash flooding considerations aside: Knowledge brokering by the extension and advisory services to adapt a farming system to flash flooding. Heliyon, 9(9), 19662.
Abstract: The development of agriculture sector and livelihood in Bangladesh are threatened by various climatic stressors, including flash flooding. Therefore, Extension and advisory services (EAS) need to navigate the knowledge landscape effectively to connect various farm actors and help secure the optimum benefits of knowledge and information for making rational decisions. However, little is known how EAS can perform this task to combat various effects of climate change. This study investigates the means of brokering knowledge by the EAS to help the farming sector adapt to flash flooding. The research was conducted in the north-eastern part of Bangladesh with 73 staff of the Department of Agricultural Extension (DAE), the largest public EAS in Bangladesh. The results showed that DAE primarily dealt with crop production-related information. However, EAS did not navigate knowledge and information about flash flooding, such as weather forecasting and crop-saving-embankments updates, among the farming actors. Moreover, they missed the broad utilization of internet-based-communication channels to rapidly navigate information and knowledge about possible flash flooding and its adaptation strategies. This article provides some policy implications to effectively support the adaptation of farming system to flash flooding through EAS.
|
|
|
|
Priestley, S. C., Payne, T. E., Harrison, J. J., Post, V. E. A., Shand, P., Love, A. J., et al. (2018). Use of U-isotopes in exploring groundwater flow and inter-aquifer leakage in the south-western margin of the Great Artesian Basin and Arckaringa Basin, central Australia. Applied Geochemistry, 98, 331–344.
Abstract: The distribution of uranium isotopes (238U and 234U) in groundwaters of the south-western margin of the Great Artesian Basin (GAB), Australia, and underlying Arckaringa Basin were examined using groundwater samples and a sequential extraction of aquifer sediments. Rock weathering, the geochemical environment and α-recoil of daughter products control the 238U and 234U isotope distributions giving rise to large spatial variations. Generally, the shallowest aquifer (J aquifer) contains groundwater with higher 238U activity concentrations and 234U/238U activity ratios close to secular equilibrium. However, the source input of uranium is spatially variable as intermittent recharge from ephemeral rivers passes through rocks that have already undergone extensive weathering and contain low 238U activity concentrations. Other locations in the J aquifer that receive little or no recharge contain higher 238U activity concentrations because uranium from localised uranium-rich rocks have been leached into solution and the geochemical environment allows the uranium to be kept in solution. The geochemical conditions of the deeper aquifers generally result in lower 238U activity concentrations in the groundwater accompanied by higher 234U/238U activity ratios. The sequential extraction of aquifer sediments showed that α-recoil of 234U from the solid mineral phases into the groundwater, rather than dissolution of, or exchange with the groundwater accessible minerals in the aquifer, caused enrichment of groundwater 234U/238U activity ratios in the Boorthanna Formation. Decay of 238U in uranium-rich coatings on J aquifer sediments caused resistant phase 234U/238U activity ratio enrichment. The groundwater 234U/238U activity ratio is dependent on groundwater residence time or flow rate, depending on the flow path trajectory. Thus, uranium isotope variations confirmed earlier groundwater flow interpretations based on other tracers; however, spatial heterogeneity, and the lack of clear regional correlations, made it difficult to identify recharge and inter-aquifer leakage.
|
|
|
|
Zhang, H., Gao, J., Xu, L., & Zhang, X. (2022). Case studies of radioactivity of drilling mud for in situ leaching uranium mining in China. Journal of Environmental Radioactivity, 251-252, 106982.
Abstract: The drilling mud from in situ leaching uranium mining is a type of low-radioactivity waste that contains natural nuclides and other harmful substances. In order to determine whether the drilling mud can meet the requirements of radioactive exemption management standards, field investigations and data simulations were conducted in this study. Two typical uranium mines were selected for onsite investigations. Drilling mud from different layers (i.e., the upper covering layer and ore-bearing layer) and from different stages (e.g., logging stage mud, drilling expansion stage mud, and mixed mud) was sampled. For each sample, the 238U and 226Ra concentrations of the solid components and the U and 226Ra concentrations of the supernatant were analyzed. The results revealed that the highest 238U and 226Ra concentrations of the solid components were 4122 Bq/kg and 4077 Bq/kg, while the 238U and 226Ra concentrations of the mixed drilling mud were all less than 300 Bq/kg. A radioactivity estimation model was established for scenario analysis. Exemption management screening lines of waste drilling mud, which can be used to classify and treat the drilling project according to the deposit’s grade and conditions, were proposed for in situ leaching drilling projects.
|
|
