Monitoring glacier surface velocity in West Kunlun Mountain using offset tracking methods based on ALOS/PALSAR images

The motion of mountain glaciers on the Tibetan Plateau is one of the key parameters in the research of environmental changes, especially the surface velocity variation over a period of years. Due to the abominable geographical conditions of steep terrain and freezing temperature, it is hard to acquire enough information using traditional ground survey. Optical remote sensing images could be useless during some special period with too much cloud cover. The main objective of this paper is to evaluate the performance of the alternative method-tracking offsets from SAR images, and is trying to add to the lacking information of glacier monitoring in west Kunlun Mountain. Three complete and detailed maps of glacier flow velocity fields during 2007-2010 are obtained by procedures of offset detection using cross correlation in Fourier domain. We found that winter glacier motions on the north slope are averagely 1 cm/day faster than south slope, well corresponding to the local topography. Accuracy of the results could be evaluated based on the offsets extracted from the regions obviously stationary, like the flat ground or the bed rock. The potential deviation is on the order of 0.12-0.25 pixels, which is acceptable in such large scale.