C041-04
Process identification preceding and during a glacier surge in the Karakorum Mountains enabled by high-temporal resolution image processing

Friday, 11 December 2020: 16:12
Virtual
Ian Arburua Delaney1, Saif Aati2, Flavien Beaud3, Surendra Adhikari4 and Jean-Philippe Avouac2, (1)University of Lausanne, Lausanne, Switzerland, (2)California Institute of Technology, Division of Geological and Planetary Sciences, Pasadena, CA, United States, (3)University of British Columbia, Vancouver, BC, Canada, (4)California Institute of Technology, Jet Propulsion Laboratory, Pasadena, CA, United States
Abstract:
Glacier-surging occurs due to mechanical instabilities within certain glaciers that cause them to periodically accelerate. The precise processes responsible for these instabilities are poorly understood. Yet, examining glaciers surges can help to understand a wide range of glaciological processes, such as glacier hydrology and sliding. Here, we present velocity maps from the Shisper glacier in the Karakorum Mountains of Pakistan as it surged. We leverage 100 open-access images from Landsat 8 and Sentinel-2 from 2013-2018 (pre-surge) and 2018-2019 (syn-surge). By implementing a nearly automated workflow in the COSI-Corr feature-tracking software package, we create velocity maps from 2013 to 2019 with time-intervals as short as 5 days. Furthermore, a unique algorithm in this version of COSI-Corr enables us to better filter artifacts from the data. The resultant dataset shows several distinct features surrounding the glacier surge. For instance, (1) spring speed-up velocities progressively increase in the two years leading up to the surge, (2) velocities in the winter remain 2-5 times higher than quiescence velocities , and (3) the middle of the glacier sped-up first, and the upper glacier subsequently accelerated from debuttressing a month later. Such an observation allows us to identify the location of the instability which initiated the glacier surge. Additionally, we document that hydrological events, such as high melt and a lake drainage, coincide with the glacier’s accelerations and slow-downs, respectively. This dataset, along with glacier thickness estimates, can help us identify different regimes of glacier dynamics, an important step to better understanding glacier sliding (see F. Beaud C024). Increased frequency of satellite observations and development of image processing algorithms enables us to identify these distinctive processes and holds promise in other glaciological applications.