A Supervised Classification of Geomorphometric Parameters to Extract the Geomorphologic Landforms in Dry Regions. (Case study: Maranjab region)

Study of Landforms and their spatial distribution in Geomorphology and other environmental science is important. In this study presented a quantitative method using morphometric parameters that extracted from digital elevation model for the classification of arid region landforms. Geomorphometric parameters used include slope, profile curvature, longitudinal curvature, and minimum and maximum curvatures that are appropriate for Identification of landforms morphometric Properties and are largely associated with geomorphological processes. Map of geomorphometric parameters in the raster format extracted using SRTM digital elevation model in the region Maranjab. Classification was done by supervised classification Maximum likelihood classification (MLC) using landform training samples from area of study. The results of the classification showed in seven class include: alluvial fans, cross and mixed sandy Dune, barkhan, stone knives, stone hills, and playa so result show playa have the greatest expansion in the region. Results of statistical evaluation of landform classes with three parameters minimum, maximum and average are presented. The assessment accuracy of classification was performed using the base Geomorphology map of Maranjab region. The results show that the classification overall accuracy is 81% and the kappa index is 0.73.