Impact of the urban design on Urban Heat Islands: The Use and Possibilities of Thermal Satellite Imagery

Abstract

This thesis is about the use of thermal imagery sensors mounted to satellites to retrieve Land Surface Temperature data in the research to Urban Heat Islands. The purpose of the thesis is to see whether the UHI data derived from the LST will be similar to current and accepted models of UHI’s based on the air temperature. Furthermore different urban design features such as, the percentage of green, the height of the green, the amount of water surface, the surface area of low albedo surfaces, the building age, the Sky View Factor, entropy and slope will be researched if they have a significant impact on the Urban Heat Island effect based on Land Surface Temperature. For this research several methods are used. To see whether the UHI data derived from the LST corresponds with the UHI data from the models a spatial autocorrelation is done to see whether the LST data is clustered together. Also a geographically weighted regression is done. The results of the geographically weighted regression show if the UHI effect based on the LST explain and correspond with the differences in the UHI effect based on the air temperature. In addition a multiple linear regression is caried out to get an insight on which of the urban design features have an effect on the UHI effect based on the LST. The most important results are that the UHI effect based on LST on both the neighbourhood and the sub neighbourhood level show a clustered pattern which is expected from temperature data. Also the UHI effect based on LST has a very strong correlation with the UHI effect based on air temperature. The urban features that significantly correlated with the UHI effect based on LST are the percentage of green, the amount of water surface, the surface area of low albedo surfaces and the building age. From these the low albedo surface area has a positive correlation and the other features have a negative correlation. Recommended for future studies is to expand the research area and also include dates which are less warm and sunny to compare the UHI effect based on LST with the UHI effect based on air temperature.