Groundwater Potentiality Prediction Using AHP-MCDA and GBT Model in a Typical Basement Complex, Nigeria: Insights From Remote Sensing and Geophysical Datasets

Delineation of geologic features that are capable of hosting water in economic quantity in the Basement Complex has been a major concern because they are usually localized due to restricted fractured and weathered rock. To effectively evaluate the groundwater potentiality prediction index (GPPI) accuracy of an area, solely depends on the groundwater potentiality predictors (GPPs) considered and the statistical model used in analyzing the data. Therefore, the acquired remotely sensed and geophysical depth sounding database processed using autopartial curve matching software and computer aided iteration to determine was analyzed using the conventional Analytical Hierarchy Process (AHP) model and the machine learning Gradient Boosting Tree (GBT) data driven model. Such a data driven model (GBT) is efficient in solving complex and cognitive problems in high uncertainty and complex environments. Twelve (12) groundwater potentiality predictors (GPPs) namely: Digital Elevation Model (DEM), Slope (S), Drainage Density (Dd), Land Use (Lu), Aquifer Resistivity (ρa), Aquifer Thickness (h), Overburden Thickness (b), Aquifer Hydraulic Conductivity (k), Aquifer Transmissivity (Tr), Aquifer Storativity (St), Aquifer Diffusivity (D), Aquifer Reflection Coefficient (Rc). The efficacy of GBT model was applied using the Salford Predictive Modeler 8.0 software. The data were partitioned into training and test dataset in ratio 90:10 using k-10 cross validation techniques. Their prediction importance was determined and the groundwater potentiality prediction index calculated and processed in the ArcGIS environment to produce the groundwater potential prediction index (GPPI) map of the investigated area. The investigated area was classed into three (3) zonations of low, moderate and high groundwater potential with about 56% classed within the low groundwater potential zone. Fifteen (15) water column measurement from wells was used to validate the developed model by calculating the predictive correlation accuracy (PCA) using the spearman's correlation analysis. The AHP-GPPI and GBT-GPPI model gave a correlation of (rs = 0.66; p = .007) and (rs = 0.74; p = .002) respectively. In conclusion, the model has proven that the drop in aquifer resistivity doesn't necessitate the presence of groundwater but rather several parameter should be integrated together to better understand the true nature of the aquifer.

Mapping of groundwater potentiality index parameters using remote sensing and GIS techniques in the southern mountain, Yogyakarta Special Region

In the hydrogeological map sheet of the Special Region of Yogyakarta, the Mountain Zone is categorized as an area of scarce groundwater. This research is intended to determine the parameters of groundwater potential in the area of scarce groundwater according to the Groundwater Potentiality Index (GPI) methods, including; fractures, lithology, slope, topography, and rainfall. Fracture parameters, distribution, and topography were collected from the Indonesia Geospatial Portal and the Digital Elevation Model (DEM). The lithological parameters were obtained from data from the Geological Agency due to the Interpretation of Remote Sensing Images. Rainfall data for the last ten years was obtained from reports. Results show that most of the research area is a fairly massive rock area, and there are some local faults. The lithological parameters indicate that the research area is composed of breccias, sandstones, and tuffs. Distribution parameters obtained information that most distribution is notated river orders 1, 2, and 3 with several river orders notation 4, 5, and 6. The slope varies from <3% to> 65%, and the intensity of rainfall almost evenly ranges from 1600-2100 mm/year.

Spatial Prediction of Groundwater Potentiality in Large Semi-Arid and Karstic Mountainous Region Using Machine Learning Models

The drinking and irrigation water scarcity is a major global issue, particularly in arid and semi-arid zones. In rural areas, groundwater could be used as an alternative and additional water supply source in order to reduce human suffering in terms of water scarcity. In this context, the purpose of the present study is to facilitate groundwater potentiality mapping via spatial-modelling techniques, individual and ensemble machine-learning models. Random forest (RF), logistic regression (LR), decision tree (DT) and artificial neural networks (ANNs) are the main algorithms used in this study. The preparation of groundwater potentiality maps was assembled into 11 ensembles of models. Overall, about 374 groundwater springs was identified and inventoried in the mountain area. The spring inventory data was randomly divided into training (75%) and testing (25%) datasets. Twenty-four groundwater influencing factors (GIFs) were selected based on a multicollinearity test and the information gain calculation. The results of the groundwater potentiality mapping were validated using statistical measures and the receiver operating characteristic curve (ROC) method. Finally, a ranking of the 15 models was achieved with the prioritization rank method using the compound factor (CF) method. The ensembles of models are the most stable and suitable for groundwater potentiality mapping in mountainous aquifers compared to individual models based on success and prediction rate. The most efficient model using the area under the curve validation method is the RF-LR-DT-ANN ensemble of models. Moreover, the results of the prioritization rank indicate that the best models are the RF-DT and RF-LR-DT ensembles of models.

Mapping Groundwater Potential for Irrigation, by Geographical Information System and Remote Sensing Techniques: A Case Study of District Lower Dir, Pakistan

The changing climate and global warming have rendered existing surface water insufficient, which is projected to adversely influence the irrigated farming systems globally. Consequently, groundwater demand has increased significantly owing to increasing population and demand for plant-based foods especially in South Asia and Pakistan. This study aimed to determine the potential areas for groundwater use for agriculture sector development in the study area Lower Dir District. ArcGIS 10.4 was utilized for geospatial analysis, which is referred to as Multi Influencing Factor (MIF) methodology. Seven parameters including land cover, geology, soil, rainfall, underground faults (liniment) density, drainage density, and slope, were utilized for delineation purpose. Considering relative significance and influence of each parameter in the groundwater recharge rating and weightage was given and potential groundwater areas were classified into very high, high, good, and poor. The result of classification disclosed that the areas of 113.10, 659.38, 674.68, and 124.17 km2 had very high, high, good, and poor potential for groundwater agricultural uses, respectively. Field surveys for water table indicated groundwater potentiality, which was high for Kotkay and Lalqila union councils having shallow water table. However, groundwater potentiality was poor in Zimdara, Khal, and Talash, characterized with a very deep water table. Moreover, the study effectively revealed that remote sensing and GIS could be developed as potent tools for mapping potential sites for groundwater utilization. Furthermore, MIF technique could be a suitable approach for delineation of groundwater potential zone, which can be applied for further research in different areas.