Resolving seasonal ice velocity of Himalayan glaciers using SAR remote sensing

Abstract

High Mountain Asia’s cryosphere including many mountain glaciers is the origin to more than ten major rivers (Bhattacharya et al. 2016) and are a source of water for more than 1.3 billion people living downstream (Bolch et al. 2012). These glaciers have experienced heterogeneous rates of mass loss since the 1970s. Yet, the associated changes in ice flow during differenent seasons that lead to mass redistribution and the importance of the glacier's sensitivity to climate are not well understood. Monitoring glacier movement is essential for the understanding of mass transport from up glacier to downstream areas, for ice thickness modeling and for the assessments of glacier health and glacial hazards. Field-based ice velocity measurements are expensive and a logistical challenge, so that measurements with a high temporal resolution are not feasible. On the other hand, optical remote sensing techniques are limited by the cloud coverage that is quite common at high altitudes. Radar remote sensing offers uninterrupted imaging as it is not effected due to cloud and bad weather. In this thesis, we used space-borne Sentinel-1 C-band radar images over 6-24 days days throughout the year 2021 to estimate and resolve seasonal ice velocity changes of Drang Drung glacier, western Himalayas, India. During autumn-winter periods, we found high InSAR coherence and resolved 6-day ice velocities in the ablation areas only. Summer months are known for surface melting, which resulted in the coherence loss. In May- October period, we applied SAR offset tracking over 1-month SAR image pairs. Combining these ice velocities for different seasons, we find that the ice velocities vary with elevation showing higher velocities upstream as compared to the areas close to glacier terminus. Glacier sped up with the onset of surface melting and attained velocity peak during midsummer followed by late summer velocity minimum. We also find second autumn acceleration. We hypothesize that these changes are related to subglacial hydrology, the increase in velocity in summer is likely due to an increase in air temperature and a subsequent availability of meltwater in drainage system, which further pressurized the system leading to basal sliding and faster velocities. When the drainage channel becomes more efficient by late summer, velocities decreased due to decrease in pressure. We interepret that the cycle repeated in autmn-winter when the meltwater from summer stored in supraglacial ponds, firn body is available as runoff and pressurized the ineffiencent drainage system leading to increase in velocities. Glacier ice velocity varies from 9.41 m/year to 25.4 m/year in different seasons with the overall mean velocity of the glacier estimated as 6.96 cm/day for the entire glacier and 3.07 cm/day for the ablation region in summer, 2.58 cm/day, 2.74 cm/day for the ablation region during January to May and October to December respectively. This thesis presents the feasibility of InSAR phase methods based on short-term InSAR pairs in combination of SAR offset tracking for summer scenes in order to resolve ice velocity of a Himalayan glacier. Our proposed data combination shall be applied to the future Indo-US NISAR mission with 12-days global repeat pass with higher wavelength such as S-band (India) and L-band (global).