Seasonal flow variability of Greenlandic glaciers : satellite observations and numerical modeling to study driving processes

• Main Author: Derkacheva, Anna
• Other Authors: Institut des Géosciences de l’Environnement (IGE), Institut de Recherche pour le Développement (IRD)-Institut national des sciences de l'Univers (INSU - CNRS)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA), Université Grenoble Alpes 2020-., Jérémie Mouginot
• Format: Doctoral or Postdoctoral Thesis
• Language: English
• Published: HAL CCSD 2021
• Subjects: Remote sensing; Seasonal variations; Physical processes; Glacier; Greenland; Télédétection; Variations saisonnières; Processus physiques; Groenland; [SDE.MCG]Environmental Sciences/Global Changes; Petermann Gletscher; Upernavik Isstrøm; glacier; greenlandic; groenlandais; Ice Sheet; Petermann gletscher; Upernavik
• Online Access: https://theses.hal.science/tel-03508093; https://theses.hal.science/tel-03508093/document; https://theses.hal.science/tel-03508093/file/DERKACHEVA_2021_archivage.pdf

The recent changes of outlet glaciers flow speed have vast control of the undergoing mass loss of the Greenland ice sheet. The processes driving the flow variability on different time scales, as well as the associated consequences and feedbacks, are not yet entirely understood. This is partly because the lack of frequent, precise, and large-scale observations limits the development of the numerical models. It is particularly difficult to resolve seasonal speed fluctuations, yet it is crucial to better constrain the physical processes controlling the ice flow.This thesis aims to address (i) the difficulties that exist in establishing robust seasonal time-series of Greenland glacier surface velocities from satellite observations, and (ii) the use of these time-series in numerical models for better understanding of the flow drivers.Satellites are able to cover large areas in a relatively short time and uniform way. Continuous time-series with seasonal temporal resolution have only started to be used recently, due to the limited number of image acquisitions made previously. Nevertheless, the time-series of ice surface velocity derived from individual sensors remain temporally incomplete and relatively noisy. Taking together three suitable satellites (Landsat-8, Sentinel-2, and Sentinel-1) across three case study sites in Greenland (Russell sector, Upernavik Isstrøm and Petermann Gletscher), we demonstrate that it is possible to obtain continuous year-around time-series only by combining results from multiple satellites. It is also shown here that by applying post-processing based on the data redundancy to such multi-sensor datasets, we are able to achieve persistent tracking of ice surface motion with a temporal resolution of about 2 weeks and mean accuracy of about 10 m/yr. With such parameters, we can resolve the seasonal variability of greenlandic glaciers where previous studies had limited success.Elaboration of reliable numerical models which would correctly represent the ice flow processes requires suitable ...