NDVI-Detected Vegetation Dynamics, Spatial Morphology and Fixed-Population Co-Location Proxies in Riyadh Governorate, Saudi Arabia, 2018–2026: A Multimetric Framework for Urban and Cultural-Landscape Monitoring

Abstract

This study examines January–March Normalized Difference Vegetation Index (NDVI)-detected vegetation, its spatial morphology and its co-location with a fixed population surface in Riyadh Governorate, Saudi Arabia, for 2018–2026. Seasonal Sentinel-2 Level-2A median composites were thresholded at NDVI 0.20, 0.30 and 0.40 and analysed with box-counting dimension, lacunarity, landscape metrics and a fixed Global Human Settlement Layer population (GHS-POP) 2025 grid. Detected vegetation (NDVI ≥ 0.20) ranged from 126.70 to 205.75 km2, peaked in 2023 and was closely associated with Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS) January–March precipitation (Pearson r = 0.844, exact permutation p = 0.0013, false-discovery-rate q = 0.0035; r = 0.798 after detrending). After adjustment for precipitation, the total detected area showed no independent temporal trend. In contrast, vegetation within ESA WorldCover built-up pixels increased by 31% (NDVI ≥ 0.20) to 82% (NDVI ≥ 0.40), peaked in 2026 and retained a positive precipitation-adjusted trend, whereas the 2023 maximum was dominated by ephemeral vegetation on bare/sparse land. The temporal pattern was robust to alternative vegetation indices and ±1 km boundary perturbations. The workflow provides a reproducible, explicitly descriptive baseline for monitoring urban and cultural-landscape settings in arid cities.