This project analysed land use and land cover change in a study area in western Kenya using multi-temporal Landsat 8 OLI imagery spanning 2013 to 2023. Cloud-free composites were generated in Google Earth Engine using median pixel compositing over dry season windows.
Supervised classification was performed using the Random Forest algorithm with 500 trees, trained on 300 ground truth points collected via GPS and verified with Google Earth historical imagery. Six LULC classes were defined: cropland, forest, grassland, wetland, built-up, and bare soil.
Post-classification change detection matrices were computed to quantify transitions between classes, and a Kappa coefficient was calculated to validate classification accuracy.
Persistent cloud cover over the study area required careful selection of dry-season imagery and use of cloud-masking algorithms in GEE to ensure sufficient clear-sky coverage.
Analysis submitted as remote sensing coursework and achieved top marks. Methods adopted as reference workflow by two other students in the cohort.