In this paper, we evaluate the possibility of using optical remote sensing techniques to survey remote glacier areas, where conventional surveying techniques are difficult to carry out. The extent of different spectral classes on the Tyn-dall glacier area in the Southern Patagonian icefield, Chile, is evaluated through classification with respect to five different snow and ice types on the surface of the glacier. Different classification approaches are tested for their eligibility to identify these snow and ice types. As no labeled data is available for the investigated remote area, a novel method is tested to obtain labeled Sentinel-2 compliant data from theoretical spectral reflectance curves. The achieved classification results show that all examined classification approaches are suitable for detecting different spectral snow and ice classes on the glacier surface.
