Determination of the effect of land cover change on land surface temperature through remote sensing and geographical information systems integration: Case of Aksaray province

Worldwide use of land cover and land use (LC/LU) is changing for many different reasons, while green areas are declining, urban growth, construction, and industrialization are also increasing. This change in LC/LU affects land surface temperature (LST) and increases temperature values.With the developing technology, remote sensing methods have become frequently used in the identification and evaluation of LC/LU and LST changes. Aksaray is one of the rapidly growing cities in Turkey. This situation also causes the increase in construction rapidly in the city. This study aims to examine temporally the effect of the change in LC/LU of Aksaray city on land surface temperature between 1990-2018. In this direction, first of all LC/LU maps were created using CORINE data for 1990, 2000, 2006 and 2018. LST values were calculated and mapped using Landsat 5 TM and Landsat 8 OLI/TIRS satellite data for the same years. In addition, Normalized Difference Vegetation Index (NDVI) for green areas, Normalized Settlement Area Index (NDBI) for residential areas and Modified Normalized Difference Water Index (MNDWI) for wetlands were also extracted from these images, and all results were correlated with LST values. In order to confirm the accuracy of the LST calculated from the Landsat satellite images, the LST maps were created using MODIS MOD11A2 and the correlation between the two data sets was examined. In addition, the measurements taken on the same day as the satellite images used by a meteorology station in the city center, which measures soil temperature, were examined and compared. As a result of the study, Aksaray province in LC/LU; artificial regions, agricultural areas and salt marshes have been found to increase, while forest areas, herbaceous areas, bare areas, terrestrial marshes and water structures have decreased. When the LST data are correlated, with LC/LU, the temperature values in wetlands and forest areas are significantly lower than the in artificial areas, bare areas, moors and heathland and natural grassland areas and agricultural areas.