abdulrauffuseini2023-dot/prestea-land-degradation

Google Earth Engine pipeline analyzing Sentinel-2 imagery (2021-2026) to quantify 7,590+ hectares of mining degradation in Prestea, Ghana. # Satellite-Based Mining Land Degradation Pipeline (Prestea, Ghana) ## 📌 Project Overview This project presents an automated cloud-computing framework built in **Google Earth Engine (GEE)** to detect, map, and quantify land degradation caused by illegal/artisanal surface mining in Prestea, Western Region, Ghana. By conducting a multi-temporal change detection analysis between **2021** and **2026** using Sentinel-2 Surface Reflectance (10m resolution) data, the pipeline isolates spectral signatures corresponding to canopy loss, soil exposure, and open-pit water accumulation. --- ## 🔑 Key Quantitative Findings * **Total Degraded Surface Area:** **7,590.74 Hectares** * **Primary Drivers:** Open-cast excavation, forest clearing, sediment-laden runoff, and artificial pit formation. * **Spatial Resolution:** 10-meter pixel precision. --- ## 🛠️ Methodology & Technical Workflow 1. **AOI Definition:** Loaded boundary geometry for the Prestea mining catchment area. 2. **Data Acquisition & Filtering:** Composited cloud-free (<15% cloud cover threshold) Sentinel-2 imagery for baseline (2021) and monitor periods (2026). 3. **Spectral Indices Computed:** * **Normalized Difference Vegetation Index (NDVI):** $\text{NDVI} = \frac{\text{B8 (NIR)} - \text{B4 (Red)}}{\text{B8 (NIR)} + \text{B4 (Red)}}$ * **Normalized Difference Water Index (NDWI):** $\text{NDWI} = \frac{\text{B3 (Green)} - \text{B8 (NIR)}}{\text{B3 (Green)} + \text{B8 (NIR)}}$ * **Difference Vegetation Index ($\Delta\text{NDVI}$):** Isolated severe loss ($\Delta\text{NDVI} < -0.25$). 4. **Area Calculation:** Multiplied degraded binary mask by pixel area (`ee.Image.pixelArea()`) and reduced regions to extract exact hectarage. --- ## 📁 Repository Structure * `prestea_degradation_pipeline.js`: Complete Earth Engine JavaScript source code ready for execution in the GEE Code Editor.