Detection of Population Density, LULC Variation and Cross-Regional Similarities Using K-Means Clustering Algorithm in Istanbul Example

Abstract

In this study, the effect of urban sprawl on land change in Istanbul was examined using Geographic Information System (GIS) technologies and the CORINE Land Cover (CLC) data set produced for the years 1990-2018 and population information. According to this; It has been determined that urban sprawl in the study area has increased due to population growth, especially industrial units, city structures, mines and construction sites have increased by approximately 9%, while maquis areas, arable, mixed agricultural areas and forest areas have decreased by 9%. According to the K-means application, similarities in the districts were revealed between 1990 and 2018. According to the results obtained, it was determined that the districts that were in clusters with similar characteristics in the 1990s changed over time and were located in different clusters. As a result, it is predicted in the study that urban sprawl will increase further due to population growth in Istanbul.