Spatio-temporal Analysis of Land Cover Changes of Izmir Province of Turkey Using Landsat TM and OLI Imagery

Abstract

Izmir province has experienced a tremendous increase in population such as other metropolitan cities of Turkey due to migration issues in last 30 years. Land use and land cover (LULC) of Izmir was affected from this unplanned population increase in province scale, natural ecosystems were destroyed by settlement, and agricultural activities of that doubled population. The aim of this study was determining LULC changes in Izmir and 30 m spatial resolution Landsat images were used as the main data along with geographic information systems (GIS) and remote sensing (RS) procedures. Satellite images from 1986 to 2015 were used to investigate LULC change patterns and a transition matrix was created. Object-based image classification technique was used for creating segments before classification. Segments, created by multi-resolution segmentation approach, were classified by visual analysis by the help of old maps, high-resolution imagery, Google Earth, and Landsat images. GIS-RS based hybrid method provided a very high accuracy (about 93%) in satellite image classification and maintained a transition matrix of changing dynamics. The results showed that increased impervious surfaces and decreased natural areas such as forests and meadows were the main features of LULC change in study area.