Chapter 50 - Assessing agriculture land-use change using remote sensing data in the Gilan Province, Iran

Abstract

Land use is a dynamic phenomenon, and the knowledge of how to utilize land-use changes and study their causes and factors in several periods can be extremely important to planners and city managers. This study aimed to determine the probability of land-cover change in gardens and agricultural land uses, residential areas, barren lands, and pastures between 1990 and 2015. In this study, a land-use change map of the Gilan Province in northern Iran for the western and eastern parts from 1990 to 2015 was prepared. ETM+images from 1990 and Landsat OLI images from 2015 were used. After preprocessing through ultrawide classification, the land-use map was extracted. Using satellite imagery, the area was divided into seven user units (forest cover, agricultural lands, grasslands, marine areas, lands without vegetation, residential areas, and canebrake) and a user map was prepared separately for each date. In this study, land-use classification maps were evaluated with an overall accuracy of 80% and a kappa coefficient higher than 0.8 from the error matrix calculation. A significant part of the conversion of agricultural and forest lands is related to the replacement of residential land use with these lands, which is mostly in the eastern part of the study area. Factors affecting land-use change in the study area can be divided into economic, governance, demographic, and natural factors. The results confirm that satellite information for extracting land-use maps and detecting land-use changes is an effective method for planning natural resource management.