Keli-10/Groundwater-quality

An end-to-end study combining hydrochemistry, GIS interpolation, multivariate statistics and explainable machine learning to assess groundwater quality across Ghana's Central Region. **Groundwater Quality Assessment in the Central Region of Ghana** **Spatial and Statistical Analysis Using Machine Learning Techniques** An end-to-end study combining hydrochemistry, GIS interpolation, multivariate statistics and explainable machine learning to assess groundwater quality across Ghana's Central Region. ________________________________________ **Why this project exists?** Groundwater is an important potable water source in the Central Region. Its quality is influenced by natural processes such as rock weathering, mineral dissolution and coastal salinization, as well as human activities including waste disposal, farming and urbanisation. This study combines Water Quality Index, GIS, multivariate statistics and machine learning to address nonlinear relationships between water quality parameters and prediction at unsampled locations. ________________________________________ **Study Area & Data** Samples 112 boreholes and hand-dug wells Coverage 16 administrative districts · 44 area councils and communities Parameters 17 physicochemical variables Geology Granite-dominated, with coastal sands, gneiss and mixed sedimentary formations Source EDJAH GEOSERVICES Ghana groundwater database ________________________________________ **Method** The analysis consists of four stages: 1. **Water Quality Index** Weighted arithmetic WQI based on Brown et al. (1972), benchmarked against WHO and Ghana EPA drinking-water standards. 2. **Spatial Analysis** Inverse Distance Weighting (IDW) interpolation of key parameters using QGIS. 3. **Multivariate Statistics** Pearson correlation, Principal Component Analysis (PCA) and Hierarchical Cluster Analysis (HCA). 4. **Machine Learning** Random Forest and Support Vector Machine (SVM) classifiers with median imputation, an 80:20 stratified split, 5-fold GridSearchCV, and SHAP explainability. ________________________________________ **Results** **Water Quality** • Mean pH: **6.81** • pH range: **4.28–8.91** …