A new band ratio technique for mapping debris-covered glaciers using Landsat imagery and a digital elevation model

Abstract

Mapping of debris-covered glaciers using optical remote-sensing data is a difficult task due to the fact that the debris cover on the glacier surface has a similar reflectance to surrounding rocky or sandy areas in the visible to near-infrared wavelength region. Therefore, a new method was developed for mapping debris-covered glaciers that considers thermal and slope information in addition to visible and near-infrared information. The method was tested for delineating the Koxkar glacier and Yengisogat glacier using Landsat Enhanced Thematic Mapper Plus and Landsat Thematic Mapper imagery and Shuttle Radar Topography Mapping Mission (SRTM) data. Specifically, the method combines a new band ratio image with slope information derived from the SRTM digital elevation model to better distinguish periglacial debris and supraglacial debris. The new band ratio image considers the near infrared (NIR), shortwave infrared (SWIR), and thermal infrared (TIR) bands, and is given by TIR/(NIR/SWIR). The debris-covered glacier boundaries are generated by thresholding the new band ratio image and overlaying it onto a binary slope map. Accuracy assessment was carried out through comparisons of the classified maps with reference maps from the Randolph Glacier Inventory, Glacier Inventory GIS data of China, and a manual delineation done using Google Earth™ images. The accuracy assessment shows that the result from the proposed method has a good agreement with the different reference data sets.