Machine learning–driven groundwater contamination risk mapping for the Niger Delta using GIS and synthetic hydro-environmental data, producing spatial risk zones and GeoTIFF outputs to support groundwater protection, environmental monitoring, and evidence-based decision-making in data-scarce coastal regions.
Groundwater Contamination Risk Mapping in the Niger Delta Using GIS and Machine Learning
📌 Project Overview
This project develops a spatially explicit groundwater contamination risk map for the Niger Delta using GIS and machine learning. Synthetic hydro-environmental datasets are integrated to model contamination susceptibility and generate raster and vector risk outputs.
🎯 Objectives
Model groundwater contamination risk using ML classifiers
Integrate hydrogeological and anthropogenic factors in GIS
Produce GeoTIFF and shapefile risk maps for decision support
🗂️ Project Structure
├── data/
│ └── groundwater_contamination_dataset.xlsx
├── scripts/
│ └── groundwater_contamination_risk_ml.py
├── outputs/
│ ├── groundwater_contamination_risk_niger_delta.tif
│ └── groundwater_contamination_risk_zones.shp
├── README.md
🧪 Dataset Description
Synthetic but realistic variables include:
Depth to groundwater
Nitrate concentration
Electrical conductivity
Land use intensity
Distance to pollution sources
Soil permeability
Target variable: Groundwater contamination risk (Low, Moderate, High)
🧠 Methodology Summary
Data preprocessing and normalization
Supervised ML classification (Random Forest)
Rasterization and spatial prediction
Risk zoning and GIS visualization
🗺️ GIS Outputs
GeoTIFF: Continuous groundwater contamination risk surface
Shapefile: Classified contamination risk zones
🛠️ Tools & Libraries
Python, NumPy, Pandas
Scikit-learn
Rasterio, GeoPandas, Shapely
QGIS / ArcGIS for visualization
📍 Study Area
Niger Delta region, Nigeria (WGS84 – EPSG:4326)
👤 Author
AGBOZU EBINGIYE NELVIN
LinkedIn: *
linkedin.com