This study investigates the relationship between local climate zones (LCZs) and land surface temperature (LST) in Mashhad, a cold semiarid metropolis in Iran, to address a key knowledge gap in understanding urban heat dynamics in dry climates. Rapid urbanization and the expansion of impervious surfaces have intensified the urban heat island (UHI) effect, contributing to higher energy demand, degraded air quality, and increased health risks. Using cloud-free Landsat-8 imagery and the single-channel (SC) algorithm, we generated an LST map for 2023. LCZs were classified via the random forest method in System for Automated Geoscientific Analyses (SAGA) geographic information system (GIS), and their relationship with LST was analyzed using spatial statistical tools. Moran’s I and Anselin Local Moran’s I were applied to detect spatial autocorrelation, clusters, and outliers, while hotspot analysis and zonal statistics in ArcMap quantified temperature variation across LCZs. Results showed that water bodies had the lowest average LST (27.88C), while bare soil/sand exhibited the highest (398C). Peripheral industrial areas showed consistently high LSTs, whereas cooler zones were concentrated in the city center. Compact midrise zones, common in central Mashhad, averaged 30.38C, in contrast to lightweight low-rise areas}mainly informal settlements on the urban fringe}which averaged 37.48C. These findings emphasize the role of urban morphology in modulating surface temperatures. The study advocates for compact, midrise (between 3 and 9 floors) development models as a more climate-responsive approach to urban planning in cold semiarid cities, offering potential for mitigating UHI impacts and enhancing climate resilience

کلمات کلیدی