Groundwater Potential Zone Mapping in Mueda District, Northern Mozambique: Integrating Remote Sensing, GIS-Based AHP and Water Point Validation
Description
This dataset supports the manuscript entitled “Groundwater Potential Zone Mapping in Mueda District, Northern Mozambique: Integrating Remote Sensing, GIS-Based AHP and Water Point Validation”. The dataset includes spatial and tabular materials used to map groundwater potential zones in Mueda District, Cabo Delgado Province, Mozambique, through the integration of remote sensing, GIS-based multi-criteria analysis, the Analytical Hierarchy Process (AHP), weighted overlay modelling, and validation using water point data. The data package contains the main thematic layers and derived outputs used in the analysis, including rainfall, geology, Topographic Wetness Index (TWI), drainage density, lineament density, land use and land cover, soil type, groundwater potential classes, and water point validation data. These layers were harmonized, reclassified, weighted using AHP, and integrated in a GIS environment to produce the final groundwater potential zone map. The final classification identifies areas of very low, low, moderate, and high groundwater potential in Mueda District. The dataset also includes information used to calculate the spatial distribution of each groundwater potential class and to assess the spatial relationship between the final map and 17 observed water points. The data are intended to support transparency, reproducibility, and further research on groundwater assessment, hydrogeological planning, and spatial decision-making in data-scarce environments.
Files
Steps to reproduce
1. Define the study area by obtaining the administrative boundary of Mueda District, Cabo Delgado Province, Mozambique, and use it to clip all spatial datasets to the same extent. 2. Acquire and prepare the input datasets, including rainfall data, a Digital Elevation Model, Landsat 9 imagery, geology data, soil data, and water point data from SINAS/DNAAS. 3. Harmonize all spatial layers by projecting them to the same coordinate reference system, aligning their spatial extent, and converting them to a common raster format and resolution. 4. Derive the thematic layers used in the groundwater potential model: rainfall, geology, Topographic Wetness Index, drainage density, lineament density, land use and land cover, and soil type. 5. Reclassify each thematic layer into a common suitability scale from 1 to 5, where 1 represents very low groundwater potential and 5 represents very high groundwater potential. 6. Apply the Analytical Hierarchy Process to assign weights to the seven conditioning factors. The final weights used were: rainfall 34.3%, geology 21.7%, Topographic Wetness Index 15.0%, drainage density 8.9%, lineament density 8.1%, land use and land cover 6.8%, and soil type 5.2%. The consistency ratio was 0.100. 7. Generate the groundwater potential index using GIS weighted overlay analysis according to the following equation: GWPZ = 0.343P + 0.217G + 0.150TWI + 0.089DD + 0.081LD + 0.068LULC + 0.052S, where P is rainfall, G is geology, TWI is Topographic Wetness Index, DD is drainage density, LD is lineament density, LULC is land use and land cover, and S is soil type. 8. Classify the final groundwater potential raster into groundwater potential classes. In Mueda District, the observed classes were very low, low, moderate, and high potential. 9. Calculate the area and percentage of each groundwater potential class using the raster attribute table or zonal statistics. 10. Validate the final groundwater potential map by overlaying the 17 observed water points with the classified groundwater potential raster and extracting the corresponding potential class for each point. 11. Summarize the spatial relationship between water points and groundwater potential classes, including the concentration of points in low and moderate potential zones and their distribution across localities such as Nanhala and Chapa. 12. Use the resulting maps, tables, and validation outputs to reproduce the figures, class distribution statistics, and interpretation presented in the associated manuscript.