The effect of urban morphology on the intensity of urban heat islands;Case study: Tabriz

UHI is a phenomenon whereby urban regions experience warmer temperatures than their rural surroundings. UHI influences well- being and welfare Average energy consumption and consequently, pollution and social equity of cities. Many factors contribute to urban heat island formation, as time (day and season), synoptic weather (wind, cloud), city form (materials, geometry, greenspace), city function (energy use, water use, pollution), city size (linked to form and function), geographic location (climate, topography, rural surrounds). Due to UHI adverse impacts on urban metabolism, ecological environment, the favourable living condition of cities and overall livability of cities, it has been an important research topic across various field of study and scholars gave more and more attention to it.  UHI has been studied for a long time, it was first described by Luke Howard in the 1810s. During the last decade Significant research efforts have been performed to evaluate the urban heat island phenomenon's impact on the urban environment. According the literature review the main goal of this study is; exploring the effect of Urban Morphology on UHI, in the Tabriz city.

This study is a correlation one. Be. In this research, ArcMap, ENVI and SPSS software have been used to generate data and apply relationships. To conduct this research, Landsat 8 images of OLI sensors at different dates for summer and winter have been used. In this study, to evaluate the UHI and influenced area of the city, the satellite images of land sat 8 OLI/ TIRS (thermal band 10) were used. The land sat 8 OLI/TIRS images that covered Tabriz summer and winter in the year of 2014 to 2019 were provided by USGS.
To perform radiometric correction of images from ENVI 5.1 software using FLAASH method. Flash is the first atmospheric correction tool that corrects visible wavelengths and infrared and infrared wavelengths of up to 3 micrometers. In the flash method, the Meta Data file is used to correct the desired bands, which include multispectral bands and thermal bands. For multispectral bands, radiance and reflection operations were performed, but for thermal bands, only radiance operations were performed. In this context, the Lowest and Highest  Position, Spatial Autocorrelation, Hot and Cold spots and finally multivariate Regression analysis were used.

The results of this study showed that the high temperature is most widespread in suburban areas especially in north west and south east rather than central parts of the city. According to the research findings, the average temperature of Tabriz in summer for the studied periods is equal to 37.7 ºC. also the average temperature varies in different years and does not show a specific trend. The average temperature of the city during the study period in winter is equal to 11.8 ºC. But according to the finding, the average temperature of the city in summer and winter is low compared to the surrounding areas. The average temperature difference between the city and surrounding areas is 33.7 ºC and 22.5 ºC in winter. Findings related to the autocorrelation pattern of Moran spatial analysis also show that the distribution of UHI in the city of Tabriz is clustered. The present study also showed that urban morphology can affect the intensity of Heat Islands. 
Based on the findings of regression analysis and calculated F (17.65) and the coefficient of significance obtaind at the level of 0.00, the predictor varizbles can well satisfy the behavior of the research dependent variable in the summer. For winter, the whole model can be generalized according to the calculated F (9.36) and significance coefficient (0.00).  on the other hand, the present study showed that the distance from the green space has an effect on the intensity of UHI, so that based on the findings of the study and calculated F(7.596) and significant level(0.00) this can be confirmed.

The present study sought to investigate the effect of urban morphology on the intensity of UHI. For this purpose, we used Landsat 8 satellite images and the technique of separate window algorithm to estimate the surface temperature. Spatial statistical analyzes such as Moran and Hot & Cold spots and multivariate linear regression were also used for analysis. In line with previous studies conducted in Iran, this study also showed that the temperature inside the city is cooler than the surrounding temperature and in a way in a city like Tabriz, we are facing cold islands instead of heat islands. The reason can be related to the compactness and high density of buildings in the cities, which requires further research. This study also showed that the surface temperature is affected by urban morphology and distance from green space. The research opens new field for future researches.