ABSTRACT
Urbanization leads to massive land use/land cover (LULC) change and contributes significantly to increases in urban land surface temperature (LST). The rapid, unplanned transformation of urban LULC significantly affects the functions of the associated ecosystem and biodiversity. This study aims to identify the variation in LULC, and its influences on the urban thermal environment, such as LST, urban heat island (UHI), and urban field variance index (UTFVI), for Chattogram city, Bangladesh. Multi-temporal Landsat Thematic Mapper/OLI: Operational Land Imager (TM/OLI) satellite images were used for the years 1999, 2009, and 2019, to classify LULC classes, identify the transition between different LULCs and to quantify variation in LST, UHI and UFVI distribution in the study area. Using the support vector machine algorithm, the LULC was classified, and the accuracy of the categorized maps was more than 85%. The spectral radiance model was used to extract urban thermal information from satellite images. The analysis of LULC estimation suggests a significant net increase in urban built-up areas (+3.51%) and a reduction in vegetation cover (−6.81%). The mean LST distribution shows that the high-temperature zones were associated with built-up areas, with the lowest temperatures being associated with vegetated cover and water bodies. The UHI and UTFVI estimation also indicate a trend toward increased surface heat stress in the study area. The study demonstrates that a rise in the non-evaporative surfaces (built-up area) and a reduction in the green cover has significantly increased the LST, UHI and UTFVI effects in the study area. Remote-sensing techniques were found to be very useful, particularly in minimizing the time required to evaluate urban expansion and to assess the influence of urbanization on LST.
