Resolving subglacial hydrology network dynamics through seismic observations on an Alpine glacier.

This is my PhD manuscript on " Resolving subglacial hydrology dynamics through seismic observations on an Alpine glacier ". This file can be also uploaded from: https://hal.univ-grenoble-alpes.fr/tel-03191311v1.

You can also find my PhD presentation PDF and the associated video can be found here or copy/paste this link: https://www.youtube.com/watch?v=IuEv7JKiYrg&t=475s or search for Ugo Nanni on youtube.

Summary:

The way in which water flows in the subglacial environment exerts a major control on ice-bed mechanical coupling, which strongly defines glacier sliding speeds. Today our understanding on the physics of the subglacial hydrology network is limited because of the scarcity of field measurements that yield a partial representation of the heterogeneous subglacial environment. The aim of my PhD work is to use passive seismology to help overcome common observational difficulties and quantify the evolution of the subglacial hydrology network pressure conditions and its configuration. Recent works show that subglacial turbulent water flow generates seismic noise that can be related to the associated hydrodynamics properties. These analyses were conducted over a limited period of time making it unclear whether such approach is appropriate to investigate seasonal and diurnal timescales, I.e. when subglacial water flow influences the most glacier dynamics. In addition, previous studies did not consider spatial changes in the heterogeneous drainage system, and until now, almost no study has located seismic noise sources spatially scattered and temporally varying. In this PhD work I address those seismological-challenges in order to resolve the subglacial hydrology dynamics in time and space.We acquired a 2-year long continuous dataset of subglacial-water-flow-induced seismic power as well as in-situ measured glacier basal sliding speed and subglacial water discharge from the Glacier d'Argentière (French Alps). I show that a careful investigation of the seismic power within [3-7] Hz can characterize the subglacial water flow hydrodynamics from seasonal to hourly timescales and across a wide range of water discharge (from 0.25 to 10 m3/sec). Combining such observations with adequate physical frameworks, I then inverted the associated hydraulic pressure gradient and hydraulic radii. I observed that the seasonal dynamics of subglacial channels is characterized by two distinct regimes. At low discharge, channels behave at equilibrium and accommodate variations in discharge mainly through changes in hydraulic radius. At a high discharge rate and with pronounced diurnal water-supply variability, channels behave out of equilibrium and undergo strong changes in the hydraulic pressure gradient, which may help sustain high water pressure in cavities and favor high glacier sliding speed over the summer.We then conducted a one-month long dense seismic-array experiment supplemented by glacier ice-thickness and surface velocity measurements. Using this unique dataset, I developed a novel methodology to overcome the challenge of locating seismic noise sources spatially scattered and temporally varying. Doing so, I successfully retrieve the first two-dimensional map of the subglacial drainage system as well as its day-to-day evolution. Using this map, I characterize when and where the subglacial drainage system is distributed through connected cavities, which favour rapid glacier flow versus localized through a channelized system that prevents rapid glacier flow. In addition, I also use high frequency seismic ground motion amplitude to study glacier features such as crevasses, thickness or ice anisotropy in a complementary way to what is traditionally done with seismic phase analysis.The first outcome of this cross-boundary PhD work is that one can analyse passive seismic measurements to retrieve the temporal evolution of subglacial channels pressure and geometry conditions over a complete melt-season. The second is that dense seismic array measurements can be used to resolve the subglacial drainage system spatial configuration and observe the switch from distributed to localized subglacial water flow. Such advances open the way for studying similar subglacial process on different sites and in particular in Greenland and Antarctica. This also concerns numerous sub-surface environment that host similar process such as volcanoes, karst, and landslides.