Skip to main content
European Commission logo print header

Climate Sensitivity of Glacial Landscape Dynamics

Periodic Reporting for period 3 - COLD (Climate Sensitivity of Glacial Landscape Dynamics)

Reporting period: 2021-01-01 to 2022-06-30

How do erosion rates in glacial landscapes vary with climate change and how do such changes affect the dynamics of mountain glaciers? Providing quantitative constraints towards this question is the main objective of COLD. These constraints are so important because mountain glaciers are sensitive to climate change and their deposits provide a unique history of Earths terrestrial climate that allows reconstructing leads and lags in the climate system.
The climate sensitivity of mountain glaciers is influenced by debris on their surface that impedes ice melting. Theoretical models of frost-related bedrock fracturing predict that rates of debris production are temperature-sensitive and that its supply to mountain glaciers increases during warming periods. Thus a previously unrecognized negative feedback emerges that lowers ice melt rates and potentially buffers part of the ice retreat due to warming. However, the temperature-sensitivity of debris production in glacial landscapes is poorly understood. Specifically, we lack robust erosion rate estimates for these landscapes, which are key for testing models of frost-related bedrock fracturing.
Here, I propose an innovative combination of new tools that capitalize on recent developments in cosmogenic nuclide geochemistry, landscape evolution modelling, and planetary-scale remote sensing analysis. I will use these tools to quantify headwall erosion rates in mountainous glacial landscapes and to gauge the sensitivity of mountain glaciers to variations in debris supply. Expected results will provide a basis for assessing the impacts of global warming, for improved predictions of valley glacier evolution, and for palaeoclimate interpretations of glacial landforms. COLD will focus on glacial landscapes, but the inverse modelling approach I will develop is applicable to any landscape on Earth and has the potential to fundamentally transform how we use cosmogenic nuclides to constrain Earth surface dynamics.
In the period covered by this report, the first 2.5 years of the project, the research team was assembled, equipment acquired, field work conducted, and samples processed in the lab. The team members are ~2 years into their employment and currently working on their first manuscripts related to the COLD project.

We conducted extensive field work in the European Alps with the goal to determine long-term headwall erosion rates with the help of cosmogenic nuclides. We collected a total of 90 supraglacial debris samples from 8 Swiss valley glaciers during two field campaigns to the European Alps in the summers of 2018 and 2019. To obtain erosion rates, we quantify the concentration of in-situ produced cosmogenic 10Be in quartz minerals of the debris samples. By now, 54 samples have been processed at the GFZ Potsdam and measured at the accelerator mass spectrometer (AMS) at the University of Cologne. The results are currently under evaluation. We furthermore collected a total of 32 bedrock samples for cosmogenic nuclide analysis at varying altitudes/lithological environments/temperature regimes in the Western European Alps in January and September 2019. We developed a multiple cosmogenic nuclide approach, in which we measure cosmogenic 3He, 10Be, and 14C on the same quartz samples, to simulatenously determine rock temperature, erosion rate and erosion style. Concurrently to sample collection and processing, we developed a 1-D numerical model for simulating the thermal evolution and cosmogenic nuclide production, decay and diffusion during the erosional history of a bedrock cliff. This model allows us to explore the sensitivity of the three-nuclide system to changes in erosion rate and style and will be an important tool for evaluating the field-based data. To compare our cosmogenic nuclide derived erosion rates with existing erosion rate estimates, we compiled published data on bedrock erosion rates in alpine regions globally.

Apart from field and lab work, we conducted experiments with a fully coupled numerical ice and landscape evolution model to evaluate the field-based empirical data. For that purpose, we developed a module that includes the production and decay of cosmogenic nuclides like 10Be and which allows tracing individual rock particles through the glacier. An application of the model to the Chhota Shigri Glacier shows promising results and we plan to use the model also on the glaciers sampled in the European Alps.

Finally, we used remote sensing tools to quantify the global distribution of debris-covered glaciers and developed methods to investigate debris cover with thermal band imagery. We developed code and applied for automatized mapping of the global distribution of supraglacial debris-cover using Google Earth Engine. We tested both Landsat 8 and Sentinel-2 imagery, as well as different band combinations, including band ratios, the normalized difference snow index, and linear spectral unmixing. We also tested the possibility to extend the debris cover mapping through time by exploiting older Landsat imagery (i.e. Landsat 5 and 7), but our analyses have shown that the results are highly sensitive to both the sensors, as well as the observation time and geographic region. This means that robust results would require manual adjusting these thresholds, which is at odds with our aim of planetary scale automated processing. We are currently developing codes for the Earth Engine that exploit thermal band data from Landsat imagery, starting with Landsat 5 in 1984. Concurrently, we acquired a thermal sensor that we mounted to an unmanned aerial vehicle (UAV), and conducted field experiments over several debris-covered glaciers. We furthermore developed a surface energy balance model to be used with the UAV data to invert the acquired thermal images for debris cover thickness, and we collected debris thickness measurements from the surveyed glaciers for ground-truthing.
• Within the COLD project, we developed codes for planetary-scale mapping of supraglacial debris cover using Google’s Earth Engine and used this method to generate the first global assessment of supraglacial debris cover extents. We currently expand these techniques to thermal band imagery.

• Using cosmogenic nuclide geochemistry in order to quantify headwall erosion rates in glacial landscapes is not entirely new, but it has not been frequently applied, and the sampling strategy implemented in this project has not been performed in the same detail by any other study. Similarly, measuring cosmogenic 3He and 14C measured in quartz grains is not novel, but combining these three nuclides the way we do it is new. Our work will demonstrate additional applications for these techniques, contributing to a more thorough understanding of both the methodology as well as its applications and limitations.

• In the field of coupled ice and landscape evolution modeling, we have developed a Lagrangian particle tracing approach for simulating the production and transport of glacial debris and tracking the production and decay of cosmogenic nuclides. This is the first time that rock particles are treated as distinct elements in a numerical ice-flow model, with properties that evolve through time. The model provides a robust basis for interpreting cosmogenic nuclide concentrations in samples collected from glacier surfaces.
Oberaletsch Glacier