Review of machine learning algorithms used in groundwater availability studies in Africa - analysis of geological and climate input variables

Aug 3, 2026

Upload by Liam

Groundwater is vital for potable water supply, agriculture, and economic development across Africa, yet the region faces persistent scarcity driven by population growth, climate change, and over-exploitation. A review summarizes how machine learning has been applied over the past decade to groundwater availability research, focusing on groundwater level prediction and groundwater potential mapping studies in Africa. The work compiles relevant papers, categorizes the machine learning algorithms, and analyzes the geological and climate inputs by frequency of use. Results indicate fuzzy-based approaches are most common, hydrology/hydrogeology variables dominate level prediction, geology is most studied for mapping, and precipitation receives the greatest climate attention. The review concludes that expanded research is needed and highlights machine learning’s value for water-resource management and decision-making.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages20

Reading Time50 min

AccessFree to download

Abstract

Introduction

Groundwater importance and demand

Land cover and recharge impacts

Climate change and variability

Water security context

Review Review of machine learning algorithms used in groundwater availability studies in Africa: analysis of geological and climate input variables Haoulata Touré1 · Cyril D. Boateng2,7 · Solomon S. R. Gidigasu1 · David D. Wemegah2 · Vera Mensah1 · Jeffrey N. A. Aryee3 · Marian A. Osei3,4,5 · Jesse Gilbert3 · Samuel K. Afful6 Received: 12 February 2024 / Accepted: 17 July 2024 © The Author(s) 2024 OPEN Abstract Groundwater is crucial for Africa’s potable water supply, agriculture, and economic development. However, the continent faces challenges with groundwater scarcity due to factors like population growth, climate change, and over-exploitation. Over the past ten years, machine learning has been increasingly and successfully used in groundwater availability studies across the world. This review paper explores the application of machine learning techniques in groundwater availability studies including groundwater level prediction and groundwater potential mapping studies by focusing on some of the studies conducted in Africa. The methodology involved downloading relevant papers, identifying and categorizing the machine learning algorithms employed, and quantifying their use. Geological and climatic variables were also identified, analyzed, and categorized to measure their usage frequency. The different algorithms and input variables extracted from each paper are graphically represented in this document highlighting the most employed ones. The findings suggest that more research needs to be conducted on the use of machine learning algorithms on this topic in Africa. In the reviewed papers Fuzzy-based algorithms are commonly used. The groundwater level prediction studies primarily focus on input variables related to hydrology/hydrogeology, while for potential mapping, geological aspects are the most investigated variables. In terms of climate, precipitation receives the most attention in the reviewed studies. The study highlights the potential of machine learning in improving water resource management and decision-making in the region. Keywords Groundwater level prediction · Groundwater potential mapping · Machine learning · Africa 1 Introduction Groundwater, which is the water occupying all the voids within a geological stratum is the main source of potable water for the majority of Africans [1]. According to the United Nations of Environment Programme (UNEP), 75% of Africans, mainly in Northern and Southern Africa, rely on groundwater as their primary drinking water supply [2] . It has been estimated that around 2 billion of the global population relies on groundwater as their primary water source [3]. This includes * Cyril D. Boateng, [cyrilboat@knust.edu.gh |](cyrilboat@knust.edu.gh |1Department of Geological)[1](cyrilboat@knust.edu.gh |1Department of Geological)[Department of Geological](cyrilboat@knust.edu.gh |1Department of Geological) Engineering, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana. 2Department of Physics, College of Science, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana. 3Department of Meteorology and Climate Science, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana. 4Centre for Ecology and Hydrology, Crowmarsh Gifford, Oxfordshire, UK. 5School of Earth and Environment, University of Leeds, Leeds, UK. 6Department of Computer Science, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana. 7Caburu Company Ltd, Madina, P.O. BOX MD 2046, Accra, Ghana. Discover Water (2024) 4:109 | [https://doi.org/10.1007/s43832-024-00109-6](https://doi.org/10.1007/s43832-024-00109-6) drinking water as well as various domestic and industrial purposes. The global population is increasing at a rate of 80 million people each year. This necessitates the identification of methods to augment the global water supply by an estimated 64 billion cubic meters annually [4]. Due to the increase in population growth and urbanization, the dependenc

FAQs

What key groundwater research tasks are covered by the review?

The review focuses on machine learning applications for groundwater level prediction and groundwater potential mapping studies in Africa.

Which types of input variables receive the most attention in the reviewed studies?

For groundwater level prediction, hydrology/hydrogeology-related variables are primarily used. For potential mapping, geological aspects are most investigated, and precipitation is the most emphasized climate variable.

What do the findings suggest about algorithm usage and future research needs?

Fuzzy-based algorithms are commonly used in the reviewed papers. The review also indicates more research is needed on machine learning algorithm use in groundwater availability studies in Africa.

Aug 3, 2026

Upload by Liam