How does submarine melting affect marine-terminating glaciers?


Glaciers and ice sheets will be the largest contributors to 21st Century sea level rise, but estimates of ice mass loss remain highly uncertain. One of the largest challenges lies in predicting the retreat of tidewater glaciers (i.e. those glaciers that drain directly into the sea). These glaciers can retreat very rapidly in response to climate warming, discharging large quantities of ice into the ocean. Improving our ability to predict tidewater glacier retreat is therefore vital if more accurate estimates of future ice loss, and hence sea level rise, are to be developed.

A likely cause of the widespread retreat of tidewater glaciers is an increase in submarine melting (i.e. melting of the submerged part of the glacier terminus) in response to warming ocean waters. Despite widespread interest in this process however (e.g. Truffer and Motyka, 2016), there remains considerable uncertainty as to how important submarine melting is for tidewater glaciers, and what determines the relative sensitivity (or insensitivity) of individual glaciers to changes in submarine melt rate (e.g. Benn et al, 2017).

It is well recognised that tidewater glaciers can show marked differences in behaviour. For example, Greenland’s largest tidewater glaciers may extend far into very deep fjords, whilst smaller glaciers draining into Svalbard’s warmer fjords may be unable to advance beyond water a few tens of metres in depth. The causes of this contrasting behaviour lies in the balance between the rate at which ice is transported down glacier (the ice flux) and the rate at which it is lost through iceberg calving and submarine melting at the terminus. It is however difficult to disentangle these controls in the messy complexity of the real world, or to assess the implications for how these glaciers will respond to ongoing ocean warming.

Recent advances in numerical modelling techniques mean that we are now able to use models to simplify and explore the processes occurring at tidewater glaciers with unprecedented realism. In particular, the ability to model tidewater glaciers in three dimensions, including iceberg calving and erosion by submarine melting, is providing new insight into fundamental behaviour of tidewater glaciers (e.g. Cowton et al, 2019, Todd et al, 2019). This PhD project will harness these developments to provide critical insight into the role of submarine melting in controlling the rate of tidewater advance and retreat, addressing research questions including:

– What determines the sensitivity of glaciers to submarine melting?
– Can glaciers be placed on a continuum between those that are highly sensitive to submarine melt rate, and those that are unaffected by submarine melting?
– What determines the rate of advance or retreat of a glacier subject to submarine melting? To what extent can this be estimated based on the balance of ice velocity and submarine melting?
– Can these criteria be applied to real tidewater glaciers to identify their likely sensitivity to submarine melting and response to ocean warming?


The project will use the glaciological model Elmer/Ice, a state-of-the-art tool for glaciology research which provides a highly-advanced suite of functionality for simulating tidewater glacier processes (Todd et al, 2018). The model will be run using the world-class Archer2 supercomputer, to which access will be provided through the Natural Environment Research Council (NERC).

The model will be used to explore the influence of submarine melting (and feedbacks affecting the rate of iceberg calving) on tidewater glaciers during phases of terminus advance, stability and retreat. Experiments will be run using idealised domains, in which the influence of key controls can be isolated. Glacier dynamics are known to be sensitive to the form of the underlying bedrock, so experiments will consider a range of idealised bed topographies, from simple slopes to more complex undulations.

Glacier advance:
In the first instance, the modelled glaciers will be allowed to advance down fjord whilst subject to a range of submarine melt rates. The rate of advance, the location of eventual stability (at which the terminus advances no further), and the water depth at the stable terminus will be examined in relation to the applied submarine melt rates. The experiment will be repeated for glaciers with higher or lower ice flux, to assess differences in behaviour between large and small tidewater glaciers.

Glacier retreat:
The stable glacier configurations generated in the advance scenarios will be subject to increasing submarine melt rates until retreat is triggered. Key properties such as the melt rate required to trigger retreat, the subsequent retreat rate, and the new stable terminus position will be analysed in conjunction with fundamental glacier properties such as ice flux and velocity.

Application to real glaciers:
The findings from the modelling experiments will be used to assess the potential impact of submarine melting across a range of real-world glaciers. Key glacier characteristics, such as ice flux, velocity and thickness, along with information on bed topography and ocean properties will be extracted from existing databases and supplemented with remote sensing observations as required. This will allow a novel classification of glaciers based on their fundamental sensitivity to submarine melting, and assessment of the likely implications of an increase in melt rates. Records of terminus change at these glaciers will be used to identify different glacier behaviours and test the findings of the model.

Project Timeline

Year 1

Literature review and research design. Develop skills with Elmer/Ice and begin modelling experiments.

Year 2

Modelling experiments and analysis of results.

Year 3

Continued modelling experiments and comparison with observations. Write up of thesis.

Year 3.5

Write up of thesis.

& Skills

The student will develop the skills necessary to set up, run and analyse the results from simulations using Elmer/Ice. This will be delivered through a combination of in-house expertise and external training on a dedicated Elmer/Ice course. No prior use of computer models or high performance computing systems is required. Additional support can be provided for the use of software for general data analysis and presentation, such as Matlab or Python.

Further training in transferable skills, including project management, oral and written presentation and media and outreach engagement is available through the Centre for Educational Enhancement and Development (CEED) at the University of St Andrews. The student will be required to present their work at conferences and seminars within St Andrews, and will be expected to attend appropriate national and international conferences throughout their PhD research.

References & further reading

Benn, D.I., Cowton, T., Todd, J. and Luckman, A., 2017. Glacier calving in Greenland. Current Climate Change Reports, 3(4), pp.282-290.

Cowton, T.R., Todd, J.A. and Benn, D.I., 2019. Sensitivity of tidewater glaciers to submarine melting governed by plume locations. Geophysical Research Letters, 46(20), pp.11219-11227.

Todd, J., Christoffersen, P., Zwinger, T., Raback, P., Chauche, N., Benn, D., Luckman, A., Ryan, J., Toberg, N., Slater, D. and Hubbard, A., 2018. A full-Stokes 3-D calving model applied to a large Greenlandic glacier. Journal of Geophysical Research: Earth Surface, 123(3), pp.410-432.

Todd, J., Christoffersen, P., Zwinger, T., Raback, P. and Benn, D.I., 2019. Sensitivity of a calving glacier to ice-ocean interactions under climate change: new insights from a 3-D full-Stokes model. The Cryosphere.

Truffer, M. and Motyka, R.J., 2016. Where glaciers meet water: subaqueous melt and its relevance to glaciers in various settings. Reviews of Geophysics, 54(1), pp.220-239.

Further Information

For enquiries, please contact Tom Cowton at

Apply Now