Microscale Urban Climate of a Hot-Arid City: LCZ-Specific Air Temperature and Heat Stress Patterns from PALM Large-Eddy Simulations

Building-resolving urban climate simulations require detailed three-dimensional geospatial data that are often unavailable for hot-arid cities. This study examines whether a Local Climate Zone (LCZ)-based workflow can provide a physically interpretable urban representation under such conditions. The LCZ-wizard was used to generate a heterogeneous 2×2 km domain containing ten built and four natural LCZ classes in an Aswan-motivated hot-arid setting. A clear-sky summer simulation was conducted with the PALM model system at 2 m horizontal resolution. LCZ-conditioned distributions of near-surface air temperature and the Universal Thermal Climate Index (UTCI) were analysed. Mean daytime air temperatures differed by only 0.57 K among built classes, whereas the nighttime difference between compact high-rise LCZ 1 and lightweight low-rise LCZ 7 reached 5.42 K. Daytime UTCI showed a clearer morphological signal: LCZ 1 had the lowest built-class mean (36.33 °C), while LCZ 7 had the highest (38.72 °C), a pattern consistent with the combined radiative, geometric, aerodynamic, and thermal properties of the compact LCZ representation. This ordering reversed at night, with compact classes exhibiting greater nocturnal heat retention. The daytime UTCI 95th percentile exceeded the 46 °C extreme-heat-stress threshold in LCZs 1–8, showing that exposed locations remain hazardous even where class-mean conditions are more favourable. The mean daytime UTCI of the dense-tree class was 4.74 K lower than that of bare soil or sand under the prescribed soil and vegetation conditions. The results reveal a diurnal trade-off for the prescribed case: compact classes are associated with lower daytime radiant exposure but greater nocturnal heat retention. The numerical values represent one idealised clear-sky, low-wind summer situation and should not be interpreted as a validated climatology of Aswan or as directly transferable to other wind and stability regimes. The study provides a proof of concept for using LCZ-derived PALM domains in controlled process studies of data-scarce hot-arid cities.