Modeling of land use and land cover change with remote sensing data and artificial neural networks (İzmit sample)

This study; learning and identifying land use and land cover change; forestry, agriculture, industry, city planning, rural and urban management, natural resource management etc. It is an important decision-making tool to be created for management studies. The Land Use and Land Cover (LULC) change of Izmit province and its close circle, which has been the subject of a rapid industrialization process, between the years 2000-2010 and 2020 has been revealed by Remote Sensing (RS) and Artificial Neural Networks (ANN) approach and mapped by making an estimation of the year 2030. The purpose of the estimate is stated. Scope of work; land use change by four land use classes (forest, water, agriculture and construction area) using satellite images; It is simulated using elevation and slope variables. Landsat 5 TM, Landsat 7 ETM and Landsat 8 OLI images were used for simulation. The images obtained in this study were classified and Kappa values were calculated as 91% for the year 2000, 87% for the year 2010 and 94% for the year 2020. The validation value of the 2030 simulation was determined as 89.2%. According to the results of this study, it is estimated that from 2020 to 2030, forest areas will decrease by 0.41%, agricultural areas by 4.38% and water areas by 0.04%. It is estimated that forest and water areas have decreased at a low rate in spatial and temporal planes, but agricultural areas tend to decrease more rapidly and this situation will continue.