Delineation of groundwater potential zones using multi-criteria analysis (AHP), frequency ratios (RF), remote sensing and GIS: a case study of the Batcham municipality (west Cameroon)

Abstract

Groundwater resource management and development should be done sustainably, which calls for accurate assessment based on new methods and scientific principles. There is an increasing demand for potable water supplies, to meet the woefully inadequate domestic water needs in the Batcham municipality (west of Cameroun). The dearth of knowledge on sustainable management of water resources is to blame for this predicament. As integrated methodologies are not used in groundwater exploration programs, the current problem has gotten worse. Innovative scientific principles of groundwater resource exploration are required to effectively counteract the non-rationalization of water and the problem of unacceptable failure rates in drilled water borehole development. This will improve the appropriate and sustainable use of these resources. Therefore, the goal of this research is to map groundwater potential zones by utilizing readily available datasets and remote sensing (RS), GIS, Frequency Ratios (FR), and multi-criteria analysis (AHP) techniques. The database used is made up of geological data, hydroclimatic data, well data, and satellite images. Each of the methods used is based on the study of the relationships between the factors controlling the groundwater potential. Among these factors, ten (10) were retained based on previous work; these are slope, elevation, land use, land cover, lineament density, drainage density, topographic wetness index, curvature, geology, rainfall, and the static water level. The AHP allowed, according to the user's expertise, to assign weights to each factor controlling the water potential. The FR made it possible to study the relationships linking each variable with high flow rates. The thematic layers obtained were integrated into a GIS environment to generate the groundwater potential map. Four classes of groundwater potential have been identified: bad, moderate, good, and excellent. The bad classes represent 8.32% and 11.21%; moderate 36.57% and 28.70%; good 43.55% and 38.93%; and excellent 11.56% and 21.16%, for the AHP and FR models, respectively. The overlying of wells with high flow rates on areas of excellent groundwater potential shows perfect concordance, which validated the final map. Likewise, the resulting map was validated by sensitivity curve analysis, and the receiver operating characteristic (ROC) curve from which result accuracy was obtained is 70.63% and 80.28% for the AHP method and FR method, respectively. Finally, the FR model outperformed the AHP model in terms of performance and it is possible to use these dependable models to describe groundwater potential zones in the other regions of Cameroon.