Using algorithms and artificial intelligence, a research team led by the Universidad Complutense de Madrid (UCM) developed a tool that, in its initial trials, demonstrated its ability to estimate areas with the best access to groundwater for drinking in Africa. Up to 90%.
In particular, papers published in Hydrology and Earth System Science And Geocarto International Describe the hydrogeological mapping carried out by the MLMapper software in the areas of Bamako and Kaulikoro (Mali) and Ouaddaï (Chad), respectively.
“Ensuring access to water and sanitation for all” is the goal of sustainable development.
“This is mainly due to a lack of hydrogeological knowledge, with the practical consequence of millions of euros in humanitarian aid being lost in vain drilling operations,” said Victor Gomez-Escalonilla Canals, a researcher at UCM’s Department of Geodynamics, Stratigraphy and Paleontology.
One of the authors, a geologist, adds that the main contribution of this research is “the use of machine learning methods to create maps of the potential of reservoir resources in remote areas.”
To conduct these studies, researchers have developed two types of sources: first, variables such as water resources database with information on well success, and second, variables such as rock type or terrain characteristics. The existence of successful wells.
Machine learning algorithms look for patterns in these descriptive variables that lead to one result or another. If the verification process produces satisfactory results, the samples will be expanded to areas where there is no information. This allows to assess whether the hydrogeological conditions in such areas are favorable.
“One of the benefits of this type of research is that most of the work can be done in faculty offices, although the results have yet to be verified on Earth.
The University of New Chattel in Switzerland was also involved in the design of the tool, but two research projects were funded by the Swiss Cooperation Agency and the Spanish Academy of Sciences, Innovation and Universities.
“The next step in the research is to try to assess not only the pros and cons of boreholes, but also to determine which areas may be more productive, in other words, to improve performance. This research is already underway, with the participation of researchers at the University of Bamako,” Gomez-Escalonella said.
Hydrology and Earth System Sciences
Preprocessing Processes in Machine-Learning-Based Groundwater Potential Mapping: An Application for the Coulicoro and Bamako Areas
Date of publication of the article
Disclaimer: AAAS and EurekAlert! They are not responsible for the accuracy of news releases posted to EurekAlert! By providing organizations or by using any information through the Eurekalert system.