Evaluating the urban climate of a typically tropical city of northeastern Brazil

Abstract

This study attempted to assess a bioclimate index and the occurrence of an urban heat island in the city of Campina Grande, northeastern Brazil, using data taken from mobile measurements and Automatic Weather Stations (AWS). The climate data were obtained during two representative months, one for the dry season (November 2005) and one for the rainy season (June 2006) at seven points in an urban area. Ten-minute air temperatures recorded by an AWS installed in urban areas were compared to those from a similar station located in a suburban area to assess the urban heat island (UHI). The data were collected using a 23X data logger (Campbell Scientific, Inc.) programmed for collecting data every second. The thermal discomfort level was analyzed by Thom's discomfort index (DI), and an analysis of variance was applied for assessing if there was any statistically significant difference at the 1% and 5% significance level of thermal comfort among points. Mann-Kendall statistical test was used for identifying possibly significant trends in a time series for air temperature, relative humidity, and class A pan evaporation for the city of Campina Grande. The present study found UHI intensities of 1.48°C and -∈0.7°C for the months taken as representative of the dry and rainy seasons, respectively. Summer in the city has partially comfortable conditions while the winter is fully comfortable. There are significant changes in DI hourly values between seasons. Only during the rainy season did all points of the city have a comfortable condition until 8:19 h, at which time they become partially comfortable for the rest of the day. Results indicated that there was a 1.5°C increase in air temperature and a 7.2% reduction in relative humidity throughout the analyzed time series. The DI also showed a statistically significant increasing trend (Mann-Kendall test, p∈