Ice speed & Artificial Intelligence (AI): Using satellite data and machine learning methods to detect change on the Greenland Ice Sheet

Global sea level rise and the associated flood and coastal realignment that accompanies it, is recognized as posing the greatest climate change risk to the UK and coastlines around the world. Satellite observations have a critical role to play as they offer the only possible method for continuous monitoring of sea level rise, and to mitigate our response to this major challenge. Over the past century, global sea levels have risen by 1.7 ± 0.3 mm per year on average, increasing to 3.2 ± 0.4 mm per year during the last 30-years. Climate models predict that this acceleration is only set to continue in the future. While thermal expansion of the ocean is currently the largest component of the sea level budget, our knowledge of the size and timing of the future contribution from the Antarctic and Greenland Ice Sheets is much less certain.

Satellite Earth observation has revolutionized our understanding of the remote and inaccessible Polar Regions. Without this critical resource we would have a far less complete understanding of which regions are changing & how fast, and what the physical mechanisms are responsible for driving the associated change. In Greenland, satellite data has been vital for revealing the ice sheet-wide spatial pattern of ice flow. Since 1992 ice velocity in Greenland has sped up markedly, with individual ice streams such as Jakobshavn Isbrae tripling in speed over the 32-year period. Despite a clear long-term trend for increasing ice velocity, speed up has not been constant through time, with neighbouring glaciers speeding up and slowing down in response to different forcing mechanisms. In Greenland satellite data has been instrumental in discovering and documenting seasonal speed variability where glaciers can speed up by up to 50% in summer months in comparison to winter. This demonstrates the clear need for continuous monitoring of all glaciers, and that ice speed acts as the early warning indicator of ice sheet imbalance.

This project offers an exciting opportunity to work at the interface of climate and Space science, making an important contribution to international efforts to study the effects and impact of climate change. In this PhD, you will work closely with world-leading experts in satellite observations, glaciology ice flow modelling, and machine learning, to better understand the Greenland Ice Sheet. Through supervision by Professor Hogg, you will use satellite observations to measure ice speed and then the mass balance of the Greenland Ice sheet, quantifying the ice sheet sea level contribution over the last 30-years. Synthetic Aperture Radar (SAR) data, from Earth observation satellites including ERS-1/2, TerraSAR-X and Sentinel-1, will be used to track changes in ice speed in Greenland, using intensity feature tracking and interferometry. Ice velocity measurements will be combined with surface and bed topography measurements to determine ice flux, and then this will be converted to mass balance using the Input-output-method (IOM). Through co-supervision by Dr Surawy-Stepney at the University of Leeds, you will pioneer the use of machine learning to make measurements of crevassing, and combine your satellite observations with the BISICLES ice flow model to investigate the drivers of ice dynamic change. Through co-supervision by Professor Ahlstrøm, you will have access to the PROMICE in-situ data network, which will provide invaluable resource for calibrating and validating your satellite observations, and investigating change on short time scales.

During your PhD you will lead at least three journal papers on these important science topics, and you may have the option to undertake a Polar field campaign. The PhD will be based in the School of Earth and Environment at the University of Leeds, and you will therefore have valuable opportunities to work closely with both the European Space Agency (ESA), and European collaborators through the projects affiliation with the ESA and NERC research projects that are already funded. The successful applicant will have access to a broad spectrum of specialist training in Earth Observation and glaciology, in addition to the extensive University of Leeds workshops on a range of topics, including scientific programming through to managing your degree. Applicants will hold good first degree (first or high 2.1) or Masters degree in physics, maths, Earth science, climate science, computer science, Earth observation or a related discipline. We welcome applications from a wide range of backgrounds, including those with non-traditional qualifications or from industry – please contact us to have a chat about your suitability for the programme.

https://yes-dtn.ac.uk/research/ice-speed-artificial-intelligence-ai-using-satellite-data-and-machine-learning-methods-to-detect-change-on-the-greenland-ice-sheet/