
This repository contains the data processing and analysis scripts used in the study of land use/land cover (LULC) dynamics and environmental variable relationships in Abijata–Shalla Lakes National Park, Ethiopia.
The code supports two main components of the research:
A supervised classification workflow based on Landsat 8 imagery (2024) using the CART (Classification and Regression Tree) algorithm. The script includes cloud masking, image compositing, training data integration, land cover classification, and accuracy assessment, including Overall Accuracy, Kappa coefficient, Producer’s Accuracy, and User’s Accuracy. The final classified map is exported as GeoTIFF, and accuracy metrics are exported as CSV files.
An R-based workflow for computing and visualizing correlation matrices of bioclimatic and topographic variables. The script generates a publication-quality heatmap using the ggcorrplot package and exports a high-resolution figure suitable for journal publication.
LULC_Classification_GEE.js)Correlation_Matrix_R.R)requirements.txt)