Overview

Project Highlights:

  • Assessment of changes in Antarctic Peninsula ice shelves since 2000.
  • State-of-the-art statistical model of ice shelf-collapse risk.
  • Novel glaciology, big data analysis elements, and Earth System modelling.

Ice shelves form the floating extensions of the Antarctic Peninsula ice sheet and play a crucial role in regulating ice sheet flow and global sea- level rise. The presence of an ice shelf provides resistive (back or buttressing) forces, which partly compensate the driving forces of inland ice flowing to the sea [1]. Since the mid-20th century and during the satellite observational era, several ice shelves in the Antarctic Peninsula have
substantially retreated or even catastrophically collapsed [2]. This has resulted in acceleration of inland ice flow [3] by a factor of up to eight [4], with some basins still adjusting to pre-collapse velocities some twenty years after disintegration [5]. Acceleration of inland ice following shelf collapse has resulted in significant contributions to sea- level rise from this region [5,6], with future contributions expected to be heavily dependent on the state and fate of the remaining shelves in the peninsula and elsewhere in Antarctica [7]. Forecasts of future sea-level rise require ice sheet models incorporating realistic predictions of the timing of future ice shelf collapses. Risk estimation of Antarctic ice shelf collapse thus remains a important goal of the cryospheric sciences.

This project will utilise satellite and climate model to construct a statistical model of ice shelf collapse risk. Despite many satellite observations and proxy reconstructions of previous collapse episodes, the complexity of governing processes occuring within ice shelves so far precludes the use of physically-based forecast models. However, the emerging and substantial observational record of ice shelf properties, and surveys of more than half a century of ice shelf collapse episodes [2], lend themselves well to combination within a statistical model framework. Bayesian nonparametrics provide a class of data- led statistical models that adapt their complexity to the data itself. This approach incorporates existing observations to model a phenomenon, yet is flexible enough to allow future inclusion of new datasets. This quality is essential to modelling ice shelf collapse risk, where new observation and information are often made available. The project will make use of ice shelf physical properties, environmental conditions and collapse timing histories to estimate the risk of future collapse events. In particular, we will seek to assign probabilities to major collapse events at individual ice shelves over the course of the next 100 years. These probabilities can then be used in physically-based ice sheet models to improve forecasts of the Antarctic ice sheet contribution to sea-level rise.

Methodology

The project will undertake a variety of state-of- the-art quantitative analyses to assess recent collapse episodes, elicit expert judgement [e.g.8] on future ice shelf collapse risk, produce a   data-led framework for selection, analysis, archiving and retrieval of environmental and glaciological datasets, and derive probabalistic estimates of ice shelf collapse risk.

The main aim is to combine data observed in the past three decades with expert opinions, coherently through a Bayesian analysis. In order to achieve this goal we will apply a spatio-temporal covariance regression model coupled with Gaussian process priors. This Bayesian nonparametric procedure effectively tunes the complexity of the model to the information present in the data. In addition, it propagates the uncertainty in physical observations and inputs from climate models coherently, and yields risk estimates for future collapse events under various projections, averaging latent observables according to their plausibility.

Training and Skills

At Birmingham, the student will receive training in glaciology and remote sensing, specifically in monitoring of ice shelf processes (PI Barrand). At Cambridge, the student will work in collaboration with Co-I Bacallado to receive training in  Bayesian numerical methods. Immersion in the research groups of both Barrand and Bacallado and attendance at national and international conferences and meetings will encourage the student to build a supportive research network.

Additional residential and field skills training in glaciological fundamentals and techniques  may be possible through attendance at international spring and summer schools operated by UNIS (Svalbard), IMAU (Utrecht) and UAF (Alaska). Additionally, each student will attend and benefit from a variety of CENTA specific training courses.

Timeline

Year 1: Collate datasets from new  satellite sensors (ESAs Cryosat-2 and Sentinel-1, NASA Landsat 8) to assess post-2010 ice shelf retreat, collapse and thinning. Conduct expert elicitation exercise on ice shelf collapse risk  with participants of speciality research themes (e.g. Forum on Ice Shelf Processes [FRISP], Antarctic Peninsula Climate Variability [APCV], relevant EGU and AGU session attendees).

Year 2: Publication on ice shelf retreat and collapse history. Ice shelf collapse risk expert elicitation publication. Construct and tune Bayesian model.

Year 3: Publication on Bayesian model approach. Tune model and calculate ice shelf collapse risk estimates under future climate model (emissions) scenarios.

Year 4: Publication on ice shelf collapse risk. Thesis writing and submission.

Partners and collaboration (including CASE)

This project will benefit from the research collaboration network of PI Barrand, in particular, liaison and collaboration with leading research groups at Utrecht, NASA JPL / UCI Irvine, BAS, and Scripps Institute of Oceanography. Project non-academic partner Engdahl will provide remote sensing data to the project via the European Space Agency. By nature, this project is inter-disciplinary, requiring collaboration with glaciologists, statisticians, engineers, atmospheric scientists and oceanographers.

Additionally, there are numerous possibilities for public engagement work resulting from this science.