Application of SAR Interferometry to Himalayan Glaciers
Gopalan Venkataraman(1) , Rao Y.S(1)
, and Rao K.S(2)
Indian Institute of Technology,
CSRE, IIT Bombay, Powai,
(2) Kuwait University, P.O. Box No. 5969,Safat 13060, Kuwait, 13060,Safat , Kuwait
Most of the work related to SAR Interferometry for Ice and Glacier are generally confined to Polar Regions. Mountainous region like Himalayan region with high relief causes layover and shadow in SAR images. Application of SAR Interferometry is very critical for these regions. Hence an attempt is made in this study to apply SAR Interferometry for Gangotri and Siachen, the largest glaciers of the Himalayan region. Gangotri glacier is about 24 km long with a maximum width of 2.5 km. This glacier is found to be retreating at a rate of 20 m/year in the recent past. This is a cause of great concern, as this glacier is a major source of water for many important rivers of north India. Siachen glacier is the longest glacier which is located at the highest altitude with lot of strategic importance to India. This glacier is 72 km long with a maximum width of 8 km.
We studied these two glaciers with InSAR data available from ERS-1 & 2 tandem experiment and also with ENVISAT ASAR data. DEMs were generated with ascending pass of ERS-I & II tandem data (25 & 26 March, 1996) for Gangotri glacier. We observed good coherence over the glacier while the surrounding areas show de-correlation due to melting. Comparison of estimated and toposheet height values, indicates a difference of 50 m for Gangotri glacier. Similarly DEMs were generated for Siachen glacier using both descending (1&2 April 1996) and Ascending (2&3 May,1996) pass of ERS-1&2 tandem data . We observed some phase jumps due to problems in phase unwrapping. As we did not have GCPs only qualitative comparison was done using contour maps.
Differential fringes representing glacier movement were also generated with ERS data using two pass techniques for Gangotri and Siachen glaciers. We have used the simulated SRTM DEM for this purpose. Ascending pass exhibits more differential fringes than descending pass in case of Siachen glacier. The velocity profile from highest elevation to lowest along with distance was also plotted. Maximum velocity profile observed for Gangotri is 10cm/day and for Siachen 20cm/day. Positive and negative velocities based on the direction of the movement have also been separately expressed in the map.
Based on ENVISAT ASAR data spanning from 2003 to 2004 with 35 days to 350 days temporal variation, about 5 InSAR pairs were processed for Gangotri glacier. The pair with 35 days gap had shown good coherence at some slopes away from the radar look direction where differential InSAR studies can be employed to observe glacier motion. The other pairs were highly de-correlated due to long temporal gaps and many surface changes associated with that. DEMs from SRTM, ASTER and toposheets were also used for observing height variation along the Gangotri glacier. Further studies with more suitable InSAR pairs of ENVISAT ASAR data are in progress and the final results will be presented in the workshop.