Decoding glacial landscapes using automated geomorphological mapping and machine learning

Melting of ice from polar ice sheets and mountain glaciers will be the largest contributor to 21st Century sea level rise, but uncertainties remain in projections of future rates and patterns of ice mass loss. The response of the cryosphere to past episodes of climatic change in Earth history provides an important analogue that can be used to help develop more robust predictions of future behaviour.

Landscapes in the Arctic, Antarctica, and mountainous regions provide a valuable record of historical glacial and fluvial erosive activity over a range of spatial and temporal scales (e.g., Rose et al., 2013; Paxman et al., 2021). This, in turn, can shed important insights into past ice extent and dynamics. However, owing to their inaccessibility, the landscape evolution and glacial history of many of these regions is poorly understood. With the recent acquisition of large subglacial topography datasets (e.g., MacGregor et al., 2021) and the development of high-resolution digital elevation models of exposed terrain (e.g., the ‘ArcticDEM’), there are now significant opportunities for systematic analysis of regional- and continental-scale topography.

The aim of this project is to use automated techniques to map the morphology of subglacial and/or subaerial landscapes and in turn reconstruct patterns of erosion and past ice extent and dynamics. The student will build on recently developed methods such as continuous valley width measurement (Clubb et al., 2022) and the use of automated classification schemes to characterise subglacial environments (Jamieson et al., 2014). Geomorphological interpretations will be integrated with numerical ice sheet modelling and (where available) chronology from offshore sediment records to constrain past glacial and climatic conditions.

The project is multi-disciplinary, with opportunities for the student to develop expertise in landscape morphometric analysis, use of geostatistical techniques and machine learning, and numerical modelling. There is also an option to undertake fieldwork to ground-truth the automated methods.

The student will be closely embedded in the ‘Sea Level, Ice and Climate’ and ‘Catchments and Rivers’ research clusters in the Department of Geography in Durham, which is world-leading in the study of glacial and fluvial landscapes and ice sheet history. The student will also spend time working on ice-sounding radar data and geomorphological interpretation with supervisor Ross at Newcastle University. More broadly, this project will address key goals of international research programmes such as INSTANT (INStabilities and Thresholds in ANTarctica; within the Scientific Committee on Antarctic Research), and involve collaboration with international researchers in geomorphology, glaciology, and palaeoclimate.

Click on an image to expand

Image Captions

(a) Terrain beneath the Greenland Ice Sheet. (b) Dry Valleys, Antarctica (credit: USGS). (c) Radar image of subglacial valleys in eastern Greenland.


The project will make use of freely available high-resolution digital elevation models (for exposed topography) and ice-sounding radar data (for subglacial topography) for automated mapping of geomorphometrics (quantitative measures of landform/landscape morphology). Landscape visualisation and analysis will be carried out in Geographic Information Systems (GIS) software. The student will develop existing computer code (from supervisors Paxman and Clubb) to extract metrics such as valley depth, width, and shape. They will also gain experience in writing and implementing their own Matlab and/or Python code to explore further metrics indicative of glacial and fluvial signatures in the landscape, and in applying machine learning techniques to link morphometries to surface processes.

There is also an option (depending on the interests of the student) to ground-truth the automated mapping by undertaking fieldwork in a representative mountain landscape. This would involve mapping of valley shape and smaller-scale landform morphology along a number of valley transects, with the aim of identifying the transition between glacial and fluvial erosive regimes. The results from the fieldwork would provide independent verification of interpretations made using the automated methods.

Numerical ice sheet modelling will then be performed to compare the inferred regional-scale erosion patterns to simulations of past ice behaviour under possible palaeoclimates. To do so, the student will use the state-of-the-art Parallel Ice Sheet Model (PISM) on the Hamilton8 supercomputer at Durham University. This modelling will help us to understand the relationships between erosion signatures in the landscape and past ice sheet behaviour, and as a result will lead to improved constraints of the state of the Earth’s cryosphere during past warmer intervals.

The methods described here can be applied to multiple target regions on Earth and other terrestrial bodies in the solar system. Options for study areas include the Himalayas (or other mountainous regions), Greenland, Antarctica, and Mars.

Project Timeline

Year 1

Assessment and selection of potential study areas using satellite imagery, ice-sounding radar echograms, and digital elevation models. Attend workshop in best practises for data science and machine learning. Develop skills in valley metric extraction and analysis using Matlab and/or Python. Option: conduct fieldwork in a representative mountain landscape to quantify valley morphology along key transects and ground-truth the automated methods.

Year 2

Refine and implement a methodology to map and classify valley morphology using supervised machine learning. Work on processing and analysis of ice-sounding radar data at Newcastle University. Attend summer school on climate and ice sheet dynamics. Begin to perform ice sheet modelling. Present work at a UK glaciology and/or geomorphology conference.

Year 3

Complete morphometric analysis. Integrate results with numerical ice sheet models, field data, and published chronological data to constrain past glacial extent and palaeoclimate. Present outcomes at international conference. Begin to draft papers and thesis.

Year 3.5

Write up and submit thesis. Finalise manuscripts for publication. Attend international conference.

& Skills

Techniques in geomorphic mapping, machine learning, Geographic Information Systems (GIS) analysis, and numerical modelling will form the core of this project. Development of the necessary skills will be facilitated through in-house expertise in Durham and Newcastle, via NERC researcher training events, and via internationally recognised summer schools.

Further training in transferable skills, including project management, scientific writing, oral and written communication, and media and public engagement, is available via the award-winning Durham Centre for Academic, Researcher & Organisation Development (CAROD). The student will also benefit from cross-disciplinary training provided as part of the IAPETUS2 DTP.

The student will be encouraged to present their work at conferences and seminars at Durham and attend relevant national and international conferences throughout the course of their PhD research. The student is also encouraged to apply for small grants from the Royal Geographic Society, British Society for Geomorphology, and Scientific Committee on Antarctic Research, which will allow them to gain experience with writing funding proposals.

References & further reading

Clubb, F.J., Weir, E.F., Mudd, S.M., 2022. Continuous measurements of valley floor width in mountainous landscapes. Earth Surface Dynamics 10(3), 437-456.

Jamieson, S.S.R., Stokes, C.R., Ross, N., Rippin, D.M., Bingham, R.G., Wilson, D.S., Margold, M., Bentley, M.J., 2014. The glacial geomorphology of the Antarctic ice sheet bed. Antarctic Science 26(6), 724-741.

MacGregor, J.A., Boisvert, L.N., Medley, B., Petty, A.A., Harbeck, J.P., et al., 2021. The Scientific Legacy of NASA’s Operation IceBridge. Reviews of Geophysics 59(2), 1-65.

Paxman, G.J.G., Tinto, K.J., Austermann, J., 2021. Neogene-Quaternary uplift and landscape evolution in northern Greenland recorded by subglacial valley morphology. Journal of Geophysical Research: Earth Surface 126(12), 1-24.

Rose, K.C., Ferraccioli, F., Jamieson, S.S.R., Bell, R.E., Corr, H., Creyts, T.T., Braaten, D., Jordan, T.A., Fretwell, P.T., Damaske, D., 2013. Early East Antarctic Ice Sheet growth recorded in the landscape of the Gamburtsev Subglacial Mountains. Earth and Planetary Science Letters 375, 1-12.

Apply Now