
This dataset provides projections of urban land expansion in Sub-Saharan Africa from 2022 to 2050 at 30 m resolution. Pixel values indicate the year in which each non-urban pixel is projected to become urban.
The projections were generated using the STUrban spatial-temporal fusion model, with the 2022 Annual Global Land Cover dataset as the baseline and future urban land increments estimated from Africapolis-derived country-level growth rates. Model performance was validated using RF AUC and Figure of Merit metrics. The AUC values achieved 0.955–0.976 in training and 0.943–0.965 in simulation. FoM values exceeded 0.2 in most training regions and 0.35 in some regions, indicating satisfactory spatial agreement between simulated and observed urban expansion.