Answers are grounded in this paper and cite their sources.
Rainfall-season maps are widely used in agricultural planning, although rainfall occurrence does not establish the persistence of soil-water availability under evaporative demand. We developed a crop-neutral daily rainfall–reference evapotranspiration (ET₀)–storage framework to characterise the occurrence and recurrence of qualifying moisture seasons across Kenya during 1991–2020. CHIRPS rainfall and FAO-56 Penman–Monteith ET₀ were combined with a uniform 100 mm storage scenario and a spatially variable, SoilGrids-derived plant-available water capacity (AWC) for 0–100 cm. The final analysis covered 18,296 grid cells (562,587.64 km²). The class representing no reliably recurring qualifying moisture season occupied 75.02% of modelled area under fixed storage and 74.45% under AWC. Class assignments changed across 44,511.29 km² (7.91%) between scenarios. Across 24 alternative classification-threshold configurations, 71.79% (fixed) and 71.27% (AWC) of national modelled area retained the baseline class in every alternative; 14 hydrological perturbations were evaluated separately in a stratified 360-cell sample. In a fractional county overlay, the no-reliable-season class occupied 82.18% of AWC-modelled area allocated to 23 designated arid and semiarid counties, compared with 6.73% in nonmember counties. An ESA CCI surface soil-moisture comparison exhibited scale- and detrending-dependent correlations and did not provide independent quantitative validation of the simulated storage. The framework separates rainfall onset from sustained modelled moisture and provides transferable, explicitly assumption-dependent descriptors of seasonal recurrence; the mapped regimes are not crop-suitability or yield forecasts.
No takes yet. Share an insight, caveat, or question.
Ngetich et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: