Examining the nexus between land surface temperature and urban growth in Chattogram Metropolitan Area of Bangladesh using long term Landsat series data

Abstract

In this study, we assessed the impact of Land-Use/ Land-Cover (LULC) changes on Land Surface Temperature (LST) in Chattogram Metropolitan Area (CMA) of southeastern Bangladesh. Landsat-5 TM and Landsat-8 OLI images were used to extract six dominant LULC categories and LST over a 28-years period. The Random Forest classifier was applied to derive LULC categories, whereas the Single-Channel and the Split-Window algorithms were applied to extract LST for the respective years. Three bio-physical parameters such as Normalized Difference Vegetation Index (NDVI), Modified Normalized Difference Water Index (MNDWI) and Normalized Difference Built-up Index (NDBI) representing three dominant LULC categories were extracted from the images to assess their interaction with the LST. Further, the Urban Heat Island (UHI) effect was calculated to assess the impact of LULC changes on the UHI distribution. Additionally, the class level matrices were extracted to assess their relationship with LST and UHI and these were further used in the development of a mathematical model to provide a probable solution for mitigating UHI effect. The result depicted that the urban area (built-up area and impervious surface) in CMA is increasing with a considerable rate (3.55 km2/yr.). The mean LST of the urban area continued to increase from 1990 (20.17 °C) to 2018 (25.83 °C) and the period 2000–2010 experienced the highest rise of LST (normalized LST difference + 0.94 °C) in the study area. The relationship between the studied biophysical indices and LST showed that NDBI positively affects the LST. Further, the analysis of UHI effect showed the number of growing administrative units in terms of intensifying LST which could be reduced by increasing urban patch numbers and edge perimeter with decreasing urban patch size.