Monitoring deforestation in Iran, Jangal-Abr Forest using multi-temporal satellite images and spectral mixture analysis method

Abstract

Satellite remote sensing assessment provides a systematic and object means to quantify the rate and form of forest changes. The main objective of this study was to quantify the forest change and forest degradation in one of the oldest forests of Iran named Jangal-Abr Forest at the north of Shahrood city. Based on Landsat-TM and OLI images acquired from 1987 to 2013, four vegetation and soil fraction images were produced and compared to map the land-cover changes occurred over 26 years. Spectral mixture analysis (SMA) method was applied to estimate fraction images of the selected components over the study area. This technique examines the direction of physical changes between multi-temporal images. Then, change vector analysis (CVA) approach was used for detecting forest changes. Results revealed that during the last 26 years, 147.5 km2 of forests were degraded mostly with medium severity. Part of the deforestation related to the northern and eastern regions of the study area in the vicinity of residential areas in contrast, forest regrowth area during the study period was estimated to be 62 km2 typically classified with low and medium severity in the eastern parts of the study area. CVA demonstrates that deforestation in the Jangal-Abr area has significantly accelerated from 4.6 km2/year in period 1987–2000 to about 9.6 km2/year in period 2010–2013. The rate of regrowth has been decreased gradually from 3.3 km2/year in period 1 (1987–2000) to about 0.76 km2/year in period 3 (1987–2000). By comparison the results of SMA method with Normalized Difference Vegetation Index (NDVI) changes method, it was found that except for some negligible differences, the results of both methods was very closed in terms of both spatially and the area of degradation and regrowth regions.