LST-Based Heritage-Sensitive Greening for Surface Heat Mitigation in Zabid, Yemen: A Spatial Decision-Support Framework for Earthen Historic Cities

Historic earthen cities face a dual challenge: surface thermal stress reduces livability, yet conventional greening can intensify the moisture processes that threaten earthen fabric. This study examines that tension in the Historic Town of Zabid, Yemen, a UNESCO World Heritage earthen city. Rather than a canopy-layer heat-island or comfort assessment, it develops an LST-based decision-support framework for heritage-sensitive greening, using two hot-season Landsat scenes (23 May and 8 June 2024). The method integrates land surface temperature (LST), the normalized difference vegetation index (NDVI), hotspot extraction, relative wetness-risk screening with an independent moisture cross-check, and earthen-fabric sensitivity in a GIS rule logic. Mapped green space occupies only 0.85% of the walled historic core. The core showed a recurrent daytime cool-core, 1.69 °C and 1.78 °C below the surrounding buffer on both dates, and LST declining 2.76 °C across the NDVI gradient. Overheating concentrated on vegetation-deficient surfaces: 98.40% of hotspot area fell within the two lowest NDVI classes, and 93% of hotspot area persisted across both dates. Wetness-risk screening flagged 14.10% of the buffer as the highest caution class. These diagnostics translate into five intervention classes—safe hotspot greening, courtyard micro-greening, contained planting, shade-first response, and protection/exclusion—a planning-level method whose indicative zones require field verification before implementation.