基于局地气候区的城市热岛空间格局研究

Spatial Pattern Study of Urban Heat Island Based on Local Climate Zones

摘要: 本研究针对快速城市化进程中的热环境问题，以郑州市为例，基于局地气候分区(LCZ)框架，集成Sentinel-2与MODIS遥感数据，分析了郑州市2023年四季昼夜LST变化，并揭示了地表城市热岛强度(SUHII)的时空格局。研究首先构建了郑州市LCZ分类图谱，总体精度85.39%，清晰呈现出中心高密度–外围开敞–边缘自然的圈层结构；进一步基于LCZ量化了郑州市各季节地表温度均值，年均温度极差达2.94℃。建筑密集型区域明显高于自然覆被型。夏季白昼温差最大，LCZ2的温度高达38.76℃，比LCZA高出6.72℃。建筑区的夏季昼夜温差较大，LCZ2达到14.66℃，而植被覆盖型LCZG的夏季昼夜温差仅为12.95℃。冬季，部分建筑区域出现“冷岛”现象，LCZ1冬季夜温降至−0.50℃，低于多数自然覆被类型。建筑区在夏冬季节的温度波动大于植被或水体覆盖区。进而定量刻画了不同LCZ类型四季昼夜的热岛强度(SUHII)。研究发现，SUHII时空分异显著，春季热岛覆盖范围最广，占比约37%。夏季热岛强度最高，秋季冷岛效应凸显，强冷岛面积达全年峰值。冬季整体热岛效应最弱。昼夜对比表明，高密度建成区LCZ 1、LCZ 2始终是强热岛的稳定核心，其夏季昼间SUHII显著高于其他类型；而水体与林地在日间均表现出显著的降温效应，可使局部地表温度降低3℃~6℃。本研究从LCZ视角为理解城市热岛提供了定量化的空间单元与证据，成果可为郑州市及同类大城市的空间结构优化、气候适应性规划与热岛缓解策略制定提供科学依据。

Abstract: This study focuses on the thermal environment issues in the context of rapid urbanization, using Zhengzhou, a national central city, as a case study. Based on the Local Climate Zone (LCZ) framework and integrating Sentinel-2 and MODIS remote sensing data, it analyzes the seasonal diurnal and nocturnal LST variations in Zhengzhou in 2023, revealing the spatiotemporal patterns of surface urban heat island intensity (SUHII). The study first constructed an LCZ classification map for Zhengzhou, with an overall accuracy of 85.39%, clearly presenting a concentric structure of central high-density, peripheral open, and edge natural areas. Further, the study quantified the seasonal average surface temperatures for Zhengzhou based on LCZs, finding a maximum annual temperature range of 2.94˚C. The temperature in densely built-up areas is significantly higher than in natural cover areas. The largest daytime temperature difference occurs in summer, with LCZ2 reaching 38.76˚C, which is 6.72˚C higher than LCZA. The summer diurnal-nocturnal temperature difference is also more pronounced in built-up areas, with LCZ2 having a difference of 14.66˚C, while the temperature difference in vegetation-covered LCZG is only 12.95˚C. In winter, some built-up areas experience a “cold island” phenomenon, with LCZ1’s winter nighttime temperature dropping to −0.50˚C, lower than most natural cover types. Built-up areas generally have larger temperature fluctuations in both summer and winter compared to vegetation or water-covered areas. The study further quantifies the seasonal diurnal and nocturnal SUHII for different LCZ types. It finds significant spatiotemporal differentiation of SUHII, with the widest coverage in spring, accounting for approximately 37%. Summer exhibits the highest SUHII intensity, while autumn shows a pronounced cold island effect, with the strongest cold island area reaching its annual peak. The overall heat island effect is weakest in winter. Comparisons between day and night show that high-density built-up areas, LCZ 1 and LCZ 2, are the stable core of strong heat islands, with summer daytime SUHII significantly higher than other types. In contrast, water bodies and forests exhibit a significant cooling effect during the day, lowering local surface temperatures by 3˚C~6˚C. This study provides quantitative spatial units and evidence from the LCZ perspective for understanding urban heat islands, and the findings can serve as a scientific basis for spatial structure optimization, climate adaptation planning, and heat island mitigation strategies in Zhengzhou and similar large cities.