The mapping of debris in glacierized terrain is required for managing the water resources, glacier mass-balance studies and the monitoring of glacier health. Two types of debris i.e. Supraglacial debris (SGD) and periglacial debris (PGD) are derived from the same source i.e., surrounding valley rock and have similar reflectance which makes it difficult to differentiate between them. Hence, in this study a novel integrated approach is proposed where spectral information and thermal data from Landsat 8 Satellite image in conjunction with geomorphometric and topographic parameters extracted from SRTM DEM are utilized to classify SGD and PGD along with other classes in Chandra River Basin (CRB) covering the area of 2422.1 km2 in western Himalayas. Nearly one fourth of the study area is glacierized region while SGD and PGD cover nearly 7% of the study area. Accuracy of the classified data is assessed through comparison with manually digitized data set and minimal difference in area is observed. Results are validated with high resolution (10 m) Sentinel 2a image and data collected from field observations. The SGD is precisely demarcated with 93% accuracy with an overall 83.50% accuracy of classification. Thus, this work presents an efficient, better and prompt method for classifying glacierized areas more effectively than manual delineation at basin/sub-basin level.
Keywords: Basin level; Glacier mapping; Integrated image classification; Periglacial debris (PGD); Supraglacial debris (SGD); Western Himalayas.
Copyright © 2021 Elsevier B.V. All rights reserved.