Analysis of the urban heat island under different synoptic patterns using local climate zones

Characterising the spatio-temporal variability of the Urban Heat Island intensity (UHII) is of utmost importance to understand processes related to air pollution dispersion and thermal comfort. However, this feature is still under-researched in most global cities due to instrument limitations or lack of expertise. This study aims at investigating the influences of standard and subclasses of Local Climate Zones (LCZ) on the behaviour of the UHI in a Brazilian city, using LCZs classification/sub-classification and in situ near-surface air temperature observations. Cluster analysis applied to daily meteorological data split the study period into five groups with similar weather characteristics. Each cluster was dominated by a combination of weather systems acting over the region which, in turn, determined the strength of the UHII. The overall diurnal UHII was strong during the nights and early mornings, with mean maximum values of 6.0 degrees C, and around 0 degrees C or negative during the rest of the day. The largest UHII (11.0 degrees C) was registered under anticyclone conditions, whilst the smallest ones (around 0 degrees C) were always observed under overcast and rainy conditions regulated by the passage of troughs, cold fronts and mesoscale complex systems. The spatial distribution of the nocturnal UHII based on the sampled and predicted LCZ temperature analysis showed that the central areas were substantially warmer than the suburban and rural areas. The introduction of six LCZ sub-classes revealed to be a suitable mapping approach to describe the land use and cover and evaluate the air temperature field in areas with mixed urban fabric . The LCZ map helped identify key climatic areas which could be used in planning strategies to target the UHI effects and create both aesthetically pleasant and healthy cities.