Abstract
Effective groundwater resource management requires managers, planners, and stakeholders to have access to timely, relevant, and trustworthy information. Researchers have undertaken several attempts to offer groundwater monitoring, prediction, risk analysis, and adaptation. Nevertheless, most of these technologies have excessively expensive license charges, and it is difficult to find a single solution that covers both aquifer monitoring and predictive analysis. This work presents a data-driven decision support tool that monitors and predicts groundwater table depths using an IoT-enabled monitoring system and machine learning-based techniques. The system was designed with open source and low-cost components, which helped to keep the overall cost low. This groundwater decision support tool (GDST) gathers, processes, and analyzes data in order to estimate future changes in groundwater table depths. The predictive analysis incorporates a 95 percent interpretability uncertainty analysis. The program can only generate solid short-term projections due to the inadequate data. Web and mobile apps are used to communicate information to policymakers, managers, decision-makers, and other stakeholders. Despite the short-term forecasts, the review discovered that this open, user-friendly, and efficient technology has the potential to contribute in the sustainable use, governance, and management of groundwater in Sub-Saharan Africa.
Cite this
Omar Haji Kombo; Santhi Kumaran; Emannuel Ndashimye; Omar Kombo, (2023), A Decision Support Tool Based on WSN and ML Models for Adaptation and Management of Groundwater Resources in Sub-Saharan Africa, 1-5