Land-use and land-cover transitions and associated vegetation-greenness dynamics in Kenya's North Rift, 1995–2023

This dataset supports the manuscript "Land-use and land-cover transitions and associatedvegetation-greenness dynamics in Kenya's North Rift, 1995–2023." It comprises a 4,831-pointextraction spanning the North Rift region of Kenya (approximately 34.01–38.06°E, 0.20°S–5.40°N),with 102 variables per point. Contents:- Land-use/land-cover (LULC) classification at seven nominal benchmark years (1995, 2000, 2005, 2011, 2016, 2020, 2023), coded 0–4 (0 = Waterbodies, 1 = Cropland, 2 = Natural Forest, 3 = Bareland, 4 = Grasslands). - Normalized Difference Vegetation Index (NDVI) at the same seven benchmark years.- Annual precipitation (1991–2023).- Annual maximum temperature (1990–2023).- Minimum temperature, bare-soil index, and soil-moisture index at selected years (see accompanying data dictionary for exact coverage per variable).- Point coordinates (longitude/latitude, geographic WGS84). Known data-quality notes- No independent classification-accuracy assessment exists for the source LULC classification.- NDVI values span three Landsat sensor generations (Landsat 5, 7, 8); cross-sensor radiometric harmonization was tested but not fully resolved. This is a point-level tabular extraction, not the original wall-to-wall classified rasters. Included files:- cleaned_dataset_[author].xlsx / .csv — the point-level dataset- data_dictionary.csv — variable definitions, units, and year coverage- analysis_scripts/ — Python scripts for transition-matrix construction, Welch's t-tests with Benjamini–Hochberg correction, Hedges' g effect sizes, cluster-robust and wild-cluster-bootstrap inference, and Getis–Ord Gi* spatial hotspot analysis- README.txt — description of file structure and how to reproduce the manuscript's tables/figures