Land is a vital natural resource for human survival and the basis of all ecosystem services. However, land degradation in the form of land use and land cover changes has been a severe problem in the world. Rural-to-urban migration in developing countries, as well as the conversion of many nearby villages into urban areas, has largely caused a rapid increase in the size of urban settlements. The evidence shows that the North of Iran is experiencing rapid changes in land use, which has endangered the stability of the region during the last three decades. This article aimed to identify and compare the changes in the growth of the city and the expansion of the city for the study case of Amol located in Mazandaran province in three decades from 1986 to 2016. Therefore, in order to produce land cover changes images Using Landsat time-series image data (TM, ETM+ and OLI/TIRS sensors), the Linear spectral unmixing model has used with the extraction of endmembers and two classes of urban and non-urban lands were identified for each time section. Spectral unmixing method as a new method in the analysis of land cover changes has been less discussed in research related to urban issues. Accuracy assessment of land cover classification, using the confusion error matrix with the calculation of overall accuracy above 90% and kappa coefficient above 80% for the obtained images, shows high and very good accuracy. The findings showed that the doubling of the city's population in these four decades has not only been accompanied by a horizontal expansion of 60% in the surrounding green and open lands, but has also resulted in an increase in building density and the number of floors. In these four decades, the expansion of Amol city has generally been from the center to the periphery, and the least changes have occurred in the central parts of the city and the most changes have occurred in the northern and eastern parts of the city.