Abstract
Abstract. In recent years, forests in the north of the country have been attacked due to human interference. Increasing population and development of residential and agricultural areas have led to deforestation. Change detection is one of the most common methods for evaluating natural resources. The aim of this study is to monitor changes in forests of Golestan province in two period times from 1990 to 2019, using Landsat images. Accordingly, by incorporating those data sets land use maps are produced. Also, the SVM algorithm is used with six different classes including forest (F), urban area (U), agriculture (A), uncultivated land (UL), water (w) and bare soil (BS). The achieved overall accuracies are 85.48% and 89.86%. Then the map and matrix changes were obtained by post-classification comparison method. The results showed that the Golestan province's forests were reduced and converted to agricultural and urban land uses.
Highlights
Ground cover has not been constant over time and is subject to change by human activities (Quintero, et al 2016)
The method used in this paper consists of three basic steps: (1) Pre-processing of image data sets, (2) Producing of land use map of each data sets, and (3) Doing change detection
Based on the methodology described above, the results are as follows: Figures 5 illustrate the land use map prepared for the years 1990 and 2018
Summary
Ground cover has not been constant over time and is subject to change by human activities (Quintero, et al 2016). Various research has been done on deforestation using satellite imagery: (Khoi, Murayama 2011) modeled the changes in the forests of northern Vietnam using neural network and Markov random field, and concluded that forest degradation rates were much higher than in farmland and residential areas. (Sudhakar, et al 2016) used satellite imagery and the Land Change Modeler method to investigate the extent of deforestation in India between 1930 and 2013, and concluded that about 40% of India's forests during these years they are removed. Determining the location and rate of changes occurring in each area can help to manage and monitor this valuable forest. By this investigation, we can control the condition of forests in the past and plan for their restoration. The present study aimed to determine forest land-use change in Golestan province using satellite imagery and post-classification method (Hasanlou and Seydi, 2018)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
