Glacier Surface Dynamics from GNSS Observations of the Grenzgletscher

Université Lyon 1 & Ecole Centrale Internship Report LMU : Ludwig Maximilian University of Munich Department of Earth and Environmental Sciences Geophysics 5th May – 31st July 2025 Glacier Surface Dynamics from GNSS Observations of the Grenzgletscher Student : Enzo Desseaux Supervisor : Pr. Mantelli Elisa Home supervisor : Dr.Couston Louis Abstract I completed my internship at LMU the Ludwig Maximilian University of Munich, from May 5th to July 31st, 2025, within the Glaciology Group at the Department of Earth and Environmental Sciences and Geophysics. This internship was part of the European Ressearch Concil (ERC) project PHAST A physics based study of ice stream dynamics. The main objective was to study the surface velocity of the Grenzgletscher glacier using GNSS (Global Navigation Satellite System) stations installed on the surface. The work involved two complementary approaches : — A technical approach, focusing on the processing of GNSS station measurements to derive high-precision horizontal and vertical positions through time. — A scientific approach, centered on the interpretation of these positional data to characterize glacier dynamics, including displacement patterns and velocity. This internship was my first immersion in a field research environment. I was particularly impressed by the challenges and logistics of conducting measurements in harsh alpine conditions. It allowed me to develop skills in glaciology, fluids dynamics, data analysis, and remote sensing techniques. Acknowledgments I would like to express my sincere gratitude to Professor E. Mantelli, who supervised me throughout this internship. Her guidance, availability and valuable advice were instrumental in ensuring the high quality of this experience. I would also like to thank Dr Couston, my university supervisor, for his regular updates, constructive feedback and constant support over the past three months. Finally, I would like to extend my special thanks to the entire Glaciology group, especially Dr Cristina Gerli and Dr Joaquim Wassernan, for their help, kindness and collaborative spirit. Their support enabled me to settle in quickly and progress in such a stimulating and complex environment. Glacier Surface Dynamics from GNSS Observations of the Grenzgletscher 1 Table of Contents Abstract 1 Acknowledgments 1 1 Introduction 3 2 Scientific Context 4 2.1 Ice Stream Dynamics and Sliding Onset . . . . . . . . . . . . . . . . . . . 4 2.2 Polythermal Glacier Behavior . . . . . . . . . . . . . . . . . . . . . . . . . 4 3 Study Site : Grenzgletscher 5 3.1 Geographic and Glaciological Setting . . . . . . . . . . . . . . . . . . . . . 5 3.2 Relevance to PHAST Objectives . . . . . . . . . . . . . . . . . . . . . . . . 6 4 GNSS Techniques and Data Acquisition 7 4.1 GNSS Positioning Principles . . . . . . . . . . . . . . . . . . . . . . . . . . 7 4.2 Instrumentation and Deployment . . . . . . . . . . . . . . . . . . . . . . . 7 4.3 Processing the GNSS data stations . . . . . . . . . . . . . . . . . . . . . . 9 5 Results 10 5.1 SurfaceVelocityFields............................. 10 5.2 Temporal Evolution of Displacement . . . . . . . . . . . . . . . . . . . . . 12 5.3 Comparison of Base Station . . . . . . . . . . . . . . . . . . . . . . . . . . 13 6 Discussion 15 7 Conclusion 16 Bibliography 17 Annexes 18 .1 GNSS stations movement . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 .2 Basestationcomparison ............................ 23 .3 Velocity of the upper plateau and the lower plateau depending on time . . 24 Glacier Surface Dynamics from GNSS Observations of the Grenzgletscher 2 1 Introduction 1 Introduction Glacier dynamics play a key role in the evolution of alpine glaciers and in regulating meltwater input to downstream environments. A central challenge in glaciology is to understand the processes that control how glaciers deform and slide over their bed, as these mechanisms govern their flow and long-term stability. Despite significant progress, many aspects of glacier motion and its variability over time remain poorly understood. The PHAST project (A Physics-based study of ice stream dynamics), funded by the European Research Council (ERC) and led by Pr. Elisa Mantelli at LMU Munich, aims to address this knowledge gap by investigating the physical processes underlying the onset of basal sliding. Recent theoretical advances by Mantelli et al.(2019) [1] have identified a novel class of feedbacks that may govern the initiation and temporal behavior of ice streams — a potential breakthrough in ice dynamics theory. PHAST combines field observations, first-principle modeling, and large-scale simulations to : (i) develop a coherent theory of ice stream dynamics, (ii) assess the role of basal sliding onset, and (iii) implement a new generation of ice sheet models that can capture ice stream behavior. In this context, a dedicated field campaign was conducted from March to June 2025 on Grenzgletscher, near Zermatt in the Swiss Alps. Sixteen GNSS (Global Navigation Satellite System) stations were deployed in the accumulation zone to obtain high-precision surface velocity measurements, complemented by radar and seismic data. These observations are crucial for identifying signatures of subtemperate sliding (a form of glacier movement occurring even at low basal temperatures) and for testing theoretical predictions from the PHAST framework. This field campaign is an important step in preparing for a larger project planned for summer 2026. This project will involve deep ice drilling to measure vertical velocity profiles and study how sliding changes with depth. The long-term aim is to establish a link between surface motion and basal processes in order to improve our understanding of the factors that control glacier dynamics at different scales. This report presents a detailed, step-by-step account of GNSS data acquisition and processing carried out during the 2025 campaign, with a focus on deriving reliable surface velocity fields essential for the PHAST project’s broader scientific goals. Glacier Surface Dynamics from GNSS Observations of the Grenzgletscher 3 2 Scientific Context 2 Scientific Context 2.1 Ice Stream Dynamics and Sliding Onset Ice streams are narrow, fast-flowing corridors of ice within otherwise slowly moving ice sheets or glaciers. They serve as the primary conduits for ice discharge from the interior to the margins and significantly influence the mass balance and sea-level evolution (Benn and Evans, 2008) [3]. Basal sliding occurs when the glacier base moves relative to the underlying bedrock or sediment. It is controlled by basal temperature, subglacial water pressure, bed geometry, and sediment properties. In particular, meltwater lubrication and subglacial till deformation can significantly accelerate glacier motion (Czurda and Flowers, 2020) [4]. Recent theoretical advances by Mantelli et al. (2019a) [1] have introduced the concept of thermally activated sliding, where basal sliding may occur under subtemperate conditions that is, when the ice is below but close to the pressure melting point. This regime allows partial sliding while preventing full refreezing of the base. In a companion paper, Mantelli and Schoof (2019b) [2] demonstrated that subtemperate zones may exhibit thermalmechanical instabilities that contribute to the spontaneous emergence of fast-flowing ice stream patterns. Building on this framework, Schoof and Mantelli (2021) [5] modeled the development of ice stream spacing and shear margins under near-cold basal conditions. They showed that even minor variations in basal temperature can trigger large-scale patterning due to strong feedbacks between sliding, thermal gradients, and basal stresses. Hydrological feedbacks also influence basal sliding. For instance, frictional heating increases basal melt (positive feedback), while cavity expansion due to sliding can reduce water pressure (negative feedback). Hoffman et al. (2014) [6] highlight that such interactions can regulate short-term speedups and are essential for understanding the dynamic response of ice streams. These physical insights are central to the PHAST project, which seeks to quantify the onset of sliding under subtemperate conditions such as those observed at Grenzgletscher. 2.2 Polythermal Glacier Behavior Polythermal glaciers contain both cold ice, below the pressure melting point, and temperate ice, at or near the melting temperature. This thermal structure strongly affects glacier dynamics, basal sliding, and internal deformation. Acold glacier is characterized by ice that remains below the pressure melting point throughout its entire thickness, meaning that basal ice is frozen to the bedrock, which limits basal sliding. In contrast, a temperate glacier has ice at the pressure melting point throughout, often resulting in abundant basal water and enhanced basal sliding. In polythermal glaciers, the coexistence of both cold and temperate ice layers or zones enables localized basal sliding and can enhance glacier flow. The interaction between cold and temperate ice results in complex thermal regimes, including areas of basal freeze-on or melting, which influence the distribution of water and sliding behavior. Grenzgletscher is one of the few glaciers in the Alps exhibiting this polythermal structure, making it an ideal natural laboratory to study the onset of sliding under subtemperate conditions. Observations of surface velocity combined with basal temperature and hydrological measurements can provide insights into the physical processes driving glacier Glacier Surface Dynamics from GNSS Observations of the Grenzgletscher 4 3 Study Site : Grenzgletscher motion and ice stream formation. Understanding polythermal glacier dynamics contributes to the broader comprehension of ice sheet behavior, especially in regions where similar thermal conditions prevail, such as parts of Greenland and Antarctica. 3 Study Site : Grenzgletscher 3.1 Geographic and Glaciological Setting Grenzgletscher is one of the largest valley glaciers in the Swiss Alps, located near Zermatt in the canton of Valais. It originates on the eastern flank of the Monte Rosa massif — the second highest peak in the Alps — and flows northward before merging with the Gornergletscher at approximately 2,500 m elevation. The glacier spans an elevation range from about 4,450 m at its source near the Dufourspitze to its terminus at roughly 2,400 m, with a total length of over 13 km and a surface area of approximately 32 km2[7]. The accumulation zone is heavily fed by avalanches and wind redistribution from the surrounding steep ridges, leading to a thick and active snowpack. The flow regime is characterized by converging tributaries, steep surface gradients in the upper parts, and a transition to a more complex flow structure in the middle glacier where lateral compression and shear become prominent. Ice thicknesses are estimated to exceed 300 m in the upper basin, based on radar sounding campaigns [8]. In recent decades, Grenzgletscher has shown evidence of strong retreat and surface lowering, typical of many Alpine glaciers. This makes it a valuable site for process-based studies on glacier dynamics. Figure 1 – Satellite map of the location of the Grenzgletscher Glacier Glacier Surface Dynamics from GNSS Observations of the Grenzgletscher 5 3 Study Site : Grenzgletscher Figure 2 – Grenzgletscher Glacier valley. on the background the peak is the Dufourspitze 3.2 Relevance to PHAST Objectives Grenzgletscher offers a unique combination of accessibility, substantial ice thickness, active flow, and thermal complexity. These features make it particularly well suited for testing hypotheses related to basal sliding and subtemperate ice stream initiation. The 2025 field campaign, conducted as part of the PHAST project, focused on the upper accumulation zone, between 3,400 m and 3,900 m elevation. Data collected during this campaign — including high-resolution GNSS time series, seismic ambient noise, and ground-penetrating radar profiles — will support the interpretation of basal conditions and the identification of zones potentially undergoing sliding. These results are essential in preparation for the summer 2026 drilling operations, which aim to recover temperature profiles and deploy englacial instrumentation to directly assess sliding rates at depth. Ultimately, Grenzgletscher serves as a controlled alpine analogue for studying sliding onset in conditions that approximate those found beneath larger polar ice masses, such as Greenland and Antarctica. Glacier Surface Dynamics from GNSS Observations of the Grenzgletscher 6 4 GNSS Techniques and Data Acquisition 4 GNSS Techniques and Data Acquisition 4.1 GNSS Positioning Principles Global Navigation Satellite Systems (GNSS) allow for precise positioning on Earth’s surface by receiving radio signals from a constellation of satellites. The most widely used system is the American GPS, but it can be combined with other satellite networks such as the European Galileo, the Russian GLONASS, and the Chinese BeiDou. This multi-constellation approach improves positioning accuracy and reliability, particularly in mountainous environments like the Alps where satellite visibility may be limited due to topography. Unlike basic GPS devices, which typically have an accuracy of 2 to 3 meters, GNSS techniques used in glaciology can achieve centimeter-level precision. This improvement is made possible through the use of differential GNSS. In this method, a reference receiver (called the base station) is installed on solid ground and remains fixed, while another receiver (the rover) is placed on the glacier surface and follows its motion. Since the base station’s position is known and does not change, its data can be used to correct the positional data from the moving rover. This correction compensates for common sources of error such as atmospheric delays, satellite orbit uncertainty, and clock drift. The GNSS receivers determine their positions by measuring the travel time of signals from multiple satellites. The more satellites available and the better their spatial geometry, the more accurate the positioning. For high-precision applications such as monitoring glacier motion, these methods enable position estimates with horizontal accuracy better than one centimeter and vertical accuracy within a few centimeters, especially when data is averaged over daily periods. Figure 3 – Scheme showing how GNSS techniques work.. Here the car represents the stations which are moving along the glaciers 4.2 Instrumentation and Deployment During the PHAST 2025 field campaign conducted on Grenzgletscher, a network of 16 GNSS stations was deployed across the upper accumulation zone of the glacier, spanning elevations between 3600m and 4000m above sea level. The primary objective was to obtain Glacier Surface Dynamics from GNSS Observations of the Grenzgletscher 7 4 GNSS Techniques and Data Acquisition high-resolution surface displacement data over a two-month period, from March to May 2025. Each station was equipped with an ArduSimple simpleRTK3B Pro receiver, integrating the Septentrio Mosaic-X5 multi-band GNSS chip, capable of tracking all major satellite constellations (GPS, Galileo, GLONASS, BeiDou, suitable for high-precision geodetic measurements under harsh alpine conditions. The GNSS units were mounted on aluminum poles drilled into the glacier surface to ensure stable and secure installation. Stations were strategically distributed to capture both longitudinal and transverse displacement patterns, with particular attention to variations in altitude and the proximity to crevasses or flow anomalies. To ensure reliable operation in cold and remote conditions, each station was powered by a battery pack housed in an insulated container. Data were recorded locally at a 1Hz sampling rate, allowing for detailed post-processing and high temporal resolution analysis of glacier surface motion. This deployment strategy was designed to balance measurement precision and system durability with spatial coverage, enabling a comprehensive investigation of both shortterm and long-term surface displacement within the accumulation zone of Grenzgletscher. It was in place from 20 March to 14 May. Two main areas were monitored (see Figure 4) : —Lower plateau : GNSS3, GNSS4, GNSS5, GNSS6, GNSS7, and GNSS8. All of these stations were additionally equipped with seismic sensors for complementary studies. —Upper plateau : GNSS11, GNSS12, GNSS14, GNSS15, and GNSS16. Figure 4 – Satellite image showing the positions of the GNSS rover stations on Grenzgletscher. Glacier Surface Dynamics from GNSS Observations of the Grenzgletscher 8 6 Discussion 6 Discussion This section outlines potential directions for further investigation had the internship continued, highlighting aspects that warrant deeper exploration. First, a more detailed analysis of the velocity patterns would be valuable. The initial assessment was relatively brief, and a more thorough examination could clarify whether there is a consistent acceleration during nighttime and whether the time of day significantly influences glacier surface velocity. The filtering process could also be improved. In this study, solutions located too far from others were excluded using a broad distance criterion ; however, this threshold could be more precisely defined. A more rigorous approach would enhance the reliability of the dataset and reduce potential biases. Here, I simply used a filter to extract the solutions, but we need to read more papers to get a better idea. The Zermatt station appears particularly promising for future field campaigns, as it enables highly precise position estimates using the developed processing code. Daily position estimates exhibit remarkable accuracy, especially when a single daily solution is used for a rover station. The positional error remains within the millimeter range compared to the newly calibrated GNSS1 station (see Figure 23). However, caution is warranted when processing data referenced to the Zermatt station. A greater number of observations are filtered out due to the long baseline and the significant altitude difference between base and rover stations. Additionally, as illustrated in Figure 23, displacement estimates based on Zermatt can be two to three times larger than those obtained with the traditional base station. For rigorous analyses, it is advisable to use both reference stations depending on the parameter of interest and to compare results to evaluate their consistency. Further investigation of the results from GNSS8 (see Figure 17) would also be beneficial to determine whether the observed movement is primarily driven by snowmelt processes or glacier ice dynamics. One possible approach would involve measuring snow depth at the time of receiver installation and retrieval to better interpret the displacement signals. The observed velocity patterns and displacement characteristics suggest potential evidence of subtemperate sliding beneath the glacier. Variations in speed, especially temporal accelerations during specific periods, may indicate basal motion influenced by transient meltwater input or changes in basal water pressure. Correlating these velocity changes with environmental variables such as temperature, precipitation, or seismic activity could help confirm the presence and dynamics of subtemperate sliding processes in this glacier. Uncertainties and Limitations Despite the high precision of GNSS measurements, several uncertainties and limitations remain. The spatial distribution of stations was limited, restricting the ability to interpolate a continuous velocity field and fully capture spatial heterogeneities in glacier motion. Position estimates are also sensitive to the choice of base station, as demonstrated by the differences observed between using the historical GNSS1 base and the Zermatt reference station. Furthermore, data filtering criteria, such as excluding stations based on proximity thresholds, introduce potential biases and may exclude relevant information. Environmental factors such as snow accumulation, melting, and sensor stability can affect measurement Glacier Surface Dynamics from GNSS Observations of the Grenzgletscher15 7 Conclusion quality and require careful consideration. Finally, the relatively short observation period limits the ability to assess seasonal and longer-term glacier dynamics. Extending measurement duration and increasing station density would reduce these uncertainties and provide a more comprehensive understanding of glacier behavior. 7 Conclusion This internship offered a valuable opportunity to contribute to the PHAST project’s objectives by investigating surface velocities of the Grenzgletscher glacier using GNSS data. Through both technical and scientific approaches, we developed a reliable workflow for GNSS data processing and conducted a preliminary analysis of glacier surface motion. The analysis revealed daily and sub-daily variations in displacement, as well as spatial differences in surface velocity between plateau zones. Although limited by the relatively short observation period and sparse station distribution, the results provide important insights into glacier behavior during the spring-summer transition. The integration of reference data from the Zermatt station notably improved the precision of positional estimates, allowing for refined comparisons and highlighting the sensitivity of displacement results to base station selection. Several limitations were identified, including the need for improved filtering strategies, more rigorous data validation, and better spatial coverage. Despite these challenges, the findings lay a solid foundation for future field campaigns. Specifically, increasing the density of GNSS stations, extending the measurement period, and integrating snow and seismic observations will enable more comprehensive assessments of glacier dynamics and basal processes. Overall, this work demonstrates the critical role of high-precision GNSS observations in advancing our understanding of glacier flow, particularly within the broader framework of the PHAST project. It represents an important preparatory step toward future fieldwork and modeling efforts aimed at unraveling the physics of sliding onset and ice stream behavior. Glacier Surface Dynamics from GNSS Observations of the Grenzgletscher16 Références Bibliography Références [1] Mantelli E., Haseloff M., and Schoof C. (2019). Ice sheet flow with thermally activated sliding. Part 1 : The role of advection. Proceedings of the Royal Society A : Mathematical, Physical and Engineering Sciences, 475(2230), 20190410. https: //doi.org/10.1098/rspa.2019.0410 [2] Mantelli E., and Schoof C. (2019). Ice sheet flow with thermally activated sliding. Part 2 : The stability of subtemperate regions. Proceedings of the Royal Society A : Mathematical, Physical and Engineering Sciences, 475(2231), 20190411. https://doi.org/10.1098/rspa.2019.0411 [3] Benn D. I., and Evans D. J. A. (2008). Glaciers and Glaciation. London : Hodder Education. [4] Czurda C., and Flowers G. E. (2020). Subglacial hydrology and basal sliding : A critical review of feedbacks and controls. Journal of Glaciology, 66(256), 947–965. https://doi.org/10.1017/jog.2020.57 [5] Schoof C., and Mantelli E. (2021). The role of sliding in ice stream formation. Proceedings of the Royal Society A : Mathematical, Physical and Engineering Sciences, 477(2250), 20200870. https://doi.org/10.1098/rspa.2020.0870 [6] Hoffman M. J., Andrews L. C., Price S. F., Catania G. A., Neumann T. A., Lüthi M. P., and Gulley J. D. (2014). Links between acceleration, melting, and supraglacial lake drainage of the western Greenland Ice Sheet. Journal of Geophysical Research : Earth Surface, 119(3), 604–626. https://doi.org/10.1002/2013JF002943 [7] GLAMOS (2023). Swiss Glacier Monitoring Network. https://www.glamos.ch [8] Bauder A., Funk M., and Huss M. (2003). Ice thickness measurements and bedrock topography of the Grenzgletscher (Valais, Switzerland). ETH Zürich, Versuchsanstalt für Wasserbau, Hydrologie und Glaziologie (VAW) Glacier Surface Dynamics from GNSS Observations of the Grenzgletscher17 Références Appendix .1 GNSS stations movement Figure 12 – Enter Caption Figure 13 – Trajectory and daily horizontal displacement of GNSS4. Glacier Surface Dynamics from GNSS Observations of the Grenzgletscher18 Références Figure 14 – Trajectory and daily horizontal displacement of GNSS5. Figure 15 – Trajectory and daily horizontal displacement of GNSS6. Glacier Surface Dynamics from GNSS Observations of the Grenzgletscher19 Références Figure 16 – Trajectory and daily horizontal displacement of GNSS7. Figure 17 – Trajectory and daily horizontal displacement of GNSS8. Glacier Surface Dynamics from GNSS Observations of the Grenzgletscher20 Références Figure 18 – Trajectory and daily horizontal displacement of GNSS11. Figure 19 – Trajectory and daily horizontal displacement of GNSS12. Glacier Surface Dynamics from GNSS Observations of the Grenzgletscher21 Références Figure 20 – Trajectory and daily horizontal displacement of GNSS14. Figure 21 – Trajectory and daily horizontal displacement of GNSS15. Glacier Surface Dynamics from GNSS Observations of the Grenzgletscher22 Références Figure 22 – Trajectory and daily horizontal displacement of GNSS16. .2 Base station comparison Figure 23 – GNSS1 position differences. The location of the base station GNSS is shown in green and the new correction using the Zermatt base station is shown in purple. There is around 3m differences. Glacier Surface Dynamics from GNSS Observations of the Grenzgletscher23 Références .3 Velocity of the upper plateau and the lower plateau depending on time Figure 24 – Velocity speed for the lower plateau station. Figure 25 – Velocity of the upper plateau station. We can see that the upper plateau stations are much moving together compared to the lower plateau on Figure 24 Glacier Surface Dynamics from GNSS Observations of the Grenzgletscher24