CARDOSO, Renata    AMORIM, Margarete. Estimating spatial patterns of air temperature using the local climate zone (LCZ) scheme and statistical regression models. []. , 12, pp.75-99. ISSN 2182-1267.  https://doi.org/10.17127/got/2017.12.005.

The transformation of natural surfaces due to built urban forms affects several climatological variables, especially air temperature. Temperature differences are traditionally measured in urban and rural areas and are commonly related to the urban heat island (UHI) intensity; this leads to simplistic interpretation of the landscape universe and therefore the UHI. We applied the local climate zone (LCZ) scheme in Presidente Prudente, São Paulo, Brazil, to analyze the city’s thermal climate. Our first step was to map potential LCZs. We then performed mobile traverses during summer and winter evenings to gather temperature data, and used statistical regression models to calculate the correlation between LCZs and air temperature, and to predict intra-urban thermal patterns. The results show a positive correlation (0.76) between urban morphology, land cover, and air temperature, compact built zones with higher temperatures, and maximum average inter-zone thermal difference of 4.4°C (â^†T _{LCZ 34-A}).

: Urban climate; potential LCZs; mobile measurements; Presidente Prudente (SP).

        · |     · |     · ( pdf )