Determination of Land Surface Temperature with Thermal Remote Sensing Images: A Case Study Kocaeli Province

In parallel with urbanization, the temperature in dense residential areas is higher than in rural areas around the city. This heat event, which causes the weather to be warmer than the natural areas around urban areas, is defined as the urban heat island. With the use of thermal bands in remote sensing platforms, the determination of land surface temperatures (LST) and spatial distributions of related parameters have started to be represented better than the old methods. The information obtained from the remote sensing platforms, because of the low cost and wide area coverage has been used as data source in various applications. In this study, in Kocaeli province maps of LST were created by using thermal bands of 5 Landsat 8 satellite images taken between July 2015 - 2019. According to the created maps, the LST values of the western – high industrial zone – and southern – geothermal activity – regions of the province were found to be high during the 5-year time period. In addition, various industrial facilities in İzmit and Kartepe districts were determined to have high LST values. Wind maps were created with wind speed and direction data from air quality stations, and it was found that they did not have any reducing effect on the LST. Finally, the Soil Adjusted Vegetation Index (SAVI) and Normalized Difference Built-up Index (NDBI) images were created from the satellite images and the statistical relationship between LST was investigated. As a result of this investigation, it was determined that there is an inverse with SAVI and a linear relationship with NDBI.