Effect of the reference rural landscape on annual variations in surface urban heat island intensity

Abstract

The urban heat island effect poses a significant climate risk associated with urbanization. Existing research primarily focuses on the spatiotemporal variations in land surface temperature (LST) within urban areas and their effects on surface urban heat island intensity (SUHII). However, investigations into how annual dynamic changes in rural landscapes impact reference LST and SUHII remain limited. This study examines 28 representative cities worldwide that experience seasonal snow cover and are predominantly surrounded by croplands or forests. The objective is to explore the patterns of continuous annual variations in SUHII and to elucidate the underlying physical mechanisms through which changes in rural landscapes affect these variations. The results reveal that SUHII consistently exhibits a double-peak pattern, with peaks occurring during the transition from winter to spring and in mid-summer. The first, moderate-intensity peak is attributed to differential snowmelt between urban and rural areas, while the more pronounced second peak is associated with the enhanced cooling effect of vegetation evapotranspiration (ET) during the growing season. Furthermore, a strong correlation between leaf area index, ET, and SUHII underscores the significance of rural vegetation in shaping SUHII variations. These findings highlight the necessity for detailed descriptions of rural reference landscapes to improve the interpretability and comparability of SUHII results. Additionally, the study suggests the need for differentiated SUHII evaluation thresholds tailored to cities with diverse rural landscapes.