Monitoring urban heat island intensity of Dhaka mega city using multi-temporal and multi spectral satellite imagery

Parvin, Nigar Sultana (2024). Monitoring urban heat island intensity of Dhaka mega city using multi-temporal and multi spectral satellite imagery. University of Birmingham. M.Phil.

Parvin2024MPhil.pdf
Text - Accepted Version
Available under License All rights reserved.
Download (5MB)

Abstract

Land surface temperature (LST) with high spatial and temporal resolution is a significant parameter for Urban Heat Island (UHI) determination. Moreover, such data are not currently available due to sensor design or compromises between temporal and spatial resolution. Therefore, the development and corroboration of certain data are highly valuable. Landsat and Moderate Resolution Imaging Spectroradiometer (MODIS) data are frequently fused to produce Landsat-like imagery using the spatiotemporal fusion technique, which may be used to increase spatiotemporal resolution. However, integrated MODIS and Landsat satellite datasets for LST estimation is still a unique approach for Dhaka, Bangladesh. This study evaluated the effectiveness of using multi-temporal and multi-spatial satellite imagery for assessing the intensity of Dhaka city’s UHI from 2011 to 2022. To obtain the seasonal changes and ensure frequent monitoring, the UHI was estimated for the winter-to-summer transitional period (January to May) of the year 2011 and 2022, respectively. The study demonstrates the usefulness of combining MODIS and Landsat (5TM and 9 TIRS /OLI) to obtain high spatial and temporal resolution imagery required for accurate UHI estimation. The Spatial and Temporal Adaptive Reflectance Fusion Model (STARFM) was used to fuse the satellite imagery, which generated synthetic Landsat imagery by applying the STARFM algorithm to MODIS daily datasets. These synthetic images, along with the original Landsat images, were used to calculate Dhaka's LST for the respective seasons and years. Human activities continuously influence land use. It is important to observe and detect the changes of land use to protect and sustain the environment. Dhaka megacity is considered one of the world's most populous cities, which is Bangladesh's capital. The study also focused on the dynamic changes in land use which substantially transformed urban areas between the years 2011 and 2022. Population and Housing Census for the respective years has been used to explore the transformation trends of land use within the city. This study also attempts to investigate the interaction between population growth and land use change patterns, especially by examining the density of the population and their distribution within the study period. This investigation categorises land use into four distinct feature classes: built areas, greenspaces, water bodies and bare soil. Within the limit of this study, an accuracy assessment has been carried out to ensure the accuracy and clarity of the classified maps illustrating land use and land cover. Between the years 2011 to 2022, the built-up area within Dhaka has experienced a notable increase of 24.9%, contrarily, there has been a significant decline of -7.36% in greenspaces, -2.48% in water bodies and -15.06 % in bare land. Dense vegetation cover has declined and very few have been replaced with moderate vegetation rather than built areas. The study’s findings provide a concise and accurate depiction of the urban growth of Dhaka over the past twelve years.

Type of Work:

Thesis (Masters by Research > M.Phil.)

Award Type:

Masters by Research > M.Phil.

Supervisor(s):