Investigation and prediction on Forests Covers Changes Using Fuzzy Object-Based Satellite Image Classification and CA-Markov (case study: City of Romeshkan)
Subject Areas :رحمان زندی 1 , Hajar Shehabi 2 , Ebrahim Akbari 3
1 - عضو هیات علمی دانشگاه حکیم سبزواری
2 -
3 - Ferdosi University
Keywords: Forest cover, Romeshkan, Landsat, Object-Based image classification CA-Markov,
Abstract :
Forest is a valuable heritage and one of important factors in the ecosystem of an area. In addition to using and exploiting them, measures must be taken to preserve them. Zagros’ forests, especially Lorestan province’s forest due to negligence have been destroyed throughout past decades. The aim of this research is to examine, detect and predict Romeshkan’s forests cover changes. changes that were taken place between 1987-2017 were extracted by satellite Images (Landsat) and using Object-Based image analysis (OBIA) method, then, were classified in 5 classes (Farmlands, Forest, Range, water and Residential). Finally, classification results show that there is a sharp decrease of forested areas (81.17 km2) and an increase of Range and Farmlands over past 30 years in the study area, also most of the forested areas were occupied by farmlands during 30 years. In the first period (1987-2002) forest cover of the study area had not faced a major change, but most of the rangelands turned to farmlands. While in the second period (2002-2017) forest cover dramatically dipped and its area decreased from 122.58 in 2002 km2 to 43.42 km2 in 2017, which the rate of forest covers decrement was around 79.16 km2. Moreover, in order to predict forest cover changes in the future (2030) CA-Markov model was applied to data that indicates 10.70% of current forest covers will be reduced in 2030, and the main Landuse/Landover change will be occurred between forest to farmlands and forest to range lands by 6.901 and 9.172 km2, respectively.