Abstract Using land surface temperatures (LST) derived from remotely sensed data, numerous studies have shown that not only urban areas are warmer than surrounding rural areas, but within urban areas, there are large spatial variations in temperature. Few studies, however, have investigated the spatial variation of air temperature within a city. Using air temperature data collected from 117 weather stations in Shenzhen for the year of 2017, we quantified the within-city spatial pattern of air temperature and its nocturnal and seasonal variability. We further investigated the quantitative relationship between air temperature and LST and compared their spatiotemporal patterns. We found that: 1) Air temperature varied greatly in space, showing very strong spatial heterogeneity and having large nocturnal and seasonal variability; 2) The spatial and temporal variation is more prominent during extreme heat days, during which the difference of hourly maximum temperature among stations could reach 8.7 °C around 16:00 and remain as high as 8.6 °C until 23:00, indicating that locations in cities with higher temperatures may suffer more intensive and longer excessive urban heat, and thereby global warming may exacerbate the difference in air temperature at the local scale; 3) LST had significantly positive correlations with air temperature, especially in the summer night, but their cool and hot spots did not always overlap in space. Results can enhance understanding of the spatial heterogeneity of the urban thermal environment and have practical implications on excessive urban heat mitigation and adaptation in an increasingly warming global climate.

中文翻译：

---

城内气温时空异质性及其与地表温度的关系

摘要 利用遥感数据得出的地表温度（LST），大量研究表明，不仅城市地区比周边农村地区温暖，而且在城市地区内，温度空间变化很大。然而，很少有研究调查城市内气温的空间变化。利用深圳市2017年117个气象站的气温数据，量化了市内气温的空间格局及其夜间和季节变化。我们进一步研究了气温和 LST 之间的定量关系，并比较了它们的时空模式。我们发现：1）气温在空间上变化很大，空间异质性非常强，夜间和季节变化较大；2) 极端高温天时空变化更为突出，16:00 左右站间每小时最高气温差异可达 8.7°C，23:00 前仍高达 8.6°C，表明各站位在气温较高的城市，城市高温可能会更加密集、持续时间更长，从而全球变暖可能会加剧局部尺度的气温差异；3）LST与气温呈显着正相关，尤其是在夏夜，但冷热点在空间上并不总是重叠。结果可以增强对城市热环境空间异质性的理解，并对在日益变暖的全球气候中过度城市热缓解和适应具有实际意义。