Will climate change in the Arctic increase the landslide-tsunami risk to the UK?

Lead Research Organisation: University of Cambridge
Department Name: Scott Polar Research Institute


Submarine landslides can be far larger than terrestrial landslides, and many generate destructive tsunamis. The Storegga Slide offshore Norway covers an area larger than Scotland and contains enough sediment to cover all of Scotland to a depth of 80 m. This huge slide occurred 8,200 years ago and extends for 800 km down slope. It produced a tsunami with a run up >20 m around the Norwegian Sea and 3-8 m on the Scottish mainland. The UK faces few other natural hazards that could cause damage on the scale of a repeat of the Storegga Slide tsunami. The Storegga Slide is not the only huge submarine slide in the Norwegian Sea. Published data suggest that there have been at least six such slides in the last 20,000 years. For instance, the Traenadjupet Slide occurred 4,000 years ago and involved ~900 km3 of sediment. Based on a recurrence interval of 4,000 years (2 events in the last 8,000 years, or 6 events in 20,000 years), there is a 5% probability of a major submarine slide, and possible tsunami, occurring in the next 200 years. Sedimentary deposits in Shetland dated at 1500 and 5500 years, in addition to the 8200 year Storegga deposit, are thought to indicate tsunami impacts and provide evidence that the Arctic tsunami hazard is still poorly understood.

Given the potential impact of tsunamis generated by Arctic landslides, we need a rigorous assessment of the hazard they pose to the UK over the next 100-200 years, their potential cost to society, degree to which existing sea defences protect the UK, and how tsunami hazards could be incorporated into multi-hazard flood risk management. This project is timely because rapid climatic change in the Arctic could increase the risk posed by landslide-tsunamis. Crustal rebound associated with future ice melting may produce larger and more frequent earthquakes, such as probably triggered the Storegga Slide 8200 years ago. The Arctic is also predicted to undergo particularly rapid warming in the next few decades that could lead to dissociation of gas hydrates (ice-like compounds of methane and water) in marine sediments, weakening the sediment and potentially increasing the landsliding risk.

Our objectives will be achieved through an integrated series of work blocks that examine the frequency of landslides in the Norwegian Sea preserved in the recent geological record, associated tsunami deposits in Shetland, future trends in frequency and size of earthquakes due to ice melting, slope stability and tsunami generation by landslides, tsunami inundation of the UK and potential societal costs. This forms a work flow that starts with observations of past landslides and evolves through modelling of their consequences to predicting and costing the consequences of potential future landslides and associated tsunamis. Particular attention will be paid to societal impacts and mitigation strategies, including examination of the effectiveness of current sea defences. This will be achieved through engagement of stakeholders from the start of the project, including government agencies that manage UK flood risk, international bodies responsible for tsunami warning systems, and the re-insurance sector.

The main deliverables will be:
(i) better understanding of frequency of past Arctic landslides and resulting tsunami impact on the UK
(ii) improved models for submarine landslides and associated tsunamis that help to understand why certain landslides cause tsunamis, and others don't.
(iii) a single modelling strategy that starts with a coupled landslide-tsunami source, tracks propagation of the tsunami across the Norwegian Sea, and ends with inundation of the UK coast. Tsunami sources of various sizes and origins will be tested
(iv) a detailed evaluation of the consequences and societal cost to the UK of tsunami flooding , including the effectiveness of existing flood defences
(v) an assessment of how climate change may alter landslide frequency and thus tsunami risk to the UK.

Planned Impact

Our project will provide the scientific basis for decisions by three major types of stakeholder.

UK Flood-Risk Management (Environment Agency; Department for Environment Food and Rural Affairs; Scottish Government): Project results will be disseminated to DEFRA and the Environment Agency (EA) who manage the risk of flooding in England and Wales, and Scottish Government that has responsibility for policy on flood management in Scotland. Our results will be used within a multi-risk framework for UK flooding that includes storm surge and rainfall sources. The EA will be involved in Work Block 6, which will utilise their 'National Flood and Coastal Defence Database' of both government and third party assets. Our analysis of potential magnitude of landslide-tsunami generated flood inundation, frequency and societal cost would be incorporated with their previous initiatives such as 'Risk Assessment for Flood and Coastal Defence Strategic Planning (RASP)'. RASP provides a flexible hierarchical method for assessing flood risk from multiple sources, and strategic prioritisation of flood defences and targeting of flood warning and emergency preparedness. We will inform the Department of Business, Innovations and Skills (as they have interest in foresight and analysis of future risks to the UK), Department of Energy and Climate Change (for offshore energy structures), and Department of Transport of project results.
The Intergovernmental Oceanographic Commission of UNESCO (IOC) has a mandate from the international community to co-ordinate tsunami early warning and mitigation in the North Atlantic. They have established a tsunami information centre and have a series working groups, including those for 'hazard assessment, risk and modelling', and 'regional tsunami warning system architecture'. They provide the correct route for project results to inform future warning and mitigation strategies for landslide-tsunami. They also provide a forum for exchange of technical information with European partners (such as those involved with the EU Transfers FP6 project in 2006-2009). We will report to the two IOC working groups on our tsunami modelling and estimates of UK risk and vulnerability. Key project results and final report will be disseminated through the IOC.

Re-Insurance Sector: Willis and their Research Network will be strongly involved as formal project partners, especially in building of geospatial data bases and estimates of societal cost. Willis is the world's 3rd largest insurance and re-insurance broker and the Willis Research Network (WRN) that they created is the world's largest collaboration between academic partners worldwide and the insurance industry. They facilitate access to other WRN members working on related risks, and access to the global insurance sector through WRN meetings. Willis will provide their staff time for project meetings, workshops, and ad-hoc discussions with project members. The consortium project will have access to the global insurance sector through additional meetings hosted and organised by Willis. Willis will host two of the PDRA positions in their offices in London, and during these visits the PDRAs will learn how geospatial tools are used by industry.

Wider Users: The subject of infrequent but potentially high impact landslide needs to be conveyed carefully to a wider (non-scientific) audience, as shown by previous press coverage of landslide-tsunamis in the Canary Islands. Project results will be disseminated by press releases (from both NOC and NERC Arctic Research Programme Office) and by a dedicated website. The cruise will form part of the NOC Classroom-at-Sea project to involve school children. We seek to involve the NERC programme's Knowledge Exchange co-ordinator with this wider dissemination of the project and its results.
Description The project has made considerable scientific advances in understanding the recurrence times of past large submarine landslides, which is a key part of assessing associated tsunami hazards. A paper in Geology (Clare et al., 2014) showed that large submarine landslides may have a common frequency distribution of recurrence times that is temporally random, suggesting that they are not strongly correlated with non-random processes (such as climate or sea-level change). The time to the next slide is independent of the time since the last slide. Analysis of an updated global database of submarine landslides ages concluded that there is either no statistically significant correlation between landslide frequency and sea-level, or the error bars on available ages are too great to tell (Urlaub et al., QSR). A paper currently in review (Pope et al., QSR) suggests that it is the latter, and that combination of locations or
sequential triggering processes can easily generation temporally random landslide recurrence intervals. Initial field work analysing tsunami deposits in the Shetland Islands suggests that a tsunami may have occurred ~5,000 years ago. This suggests that since the large Storegga Slide tsunami 8,150 years ago, two additional and more localised tsunamis at 5,000 and 1,500 years BP may have affected northern UK coastlines. Rapid progress has been made in tsunami modelling, with the adaptive-mesh fluidity model results now benchmarked against other tsunami models and field data on Storegga Slide tsunami deposits.

An innovative process has been trialled to identify the geographical location of tsunami sources that will cause greatest impact for a given site on the UK coastline. Initial work is analysing how ice unloading affects stress patterns on faults, and its implications for future seismicity. A PDRA is now working on how gas hydrates may affect the mechanical behaviour of generalised types of sediment at high latitudes. Recent publications have explored whether the sudden loss of structure within marine sediment can lead to large-scale slope failure (Urlaub et al., 204, in review). The project has just submitted an
initial case for the inclusion of tsunami hazards on the UK National Risk Register, through the Natural Hazards Partnership. Stakeholders have been engaged at annual and project meetings, with a particularly productive workshopheld in Edinburgh with the Scottish Environment Protection Agency (SEPA). The project is also now also well connected to European tsunami warning stakeholders through the F7 ASTARTE project, which started in 2013, through attendance at ASTARTE meetings.
Exploitation Route March 2014: Submission to the Cabinet Office (via Natural Hazards Partnership) on inclusion of Landslide-Tsunamis in the UK National Risk Register.

November 2013: Workshop with Scottish Environment Protection Agency (hosted in Edinburgh) on 'tsunami flooding'.

September 2013: Workshop with participation from the Environment Agency and
international experts.
Sectors Environment

Description The findings of this work have informed government agencies about tsunami risk
First Year Of Impact 2014
Sector Education,Environment
Impact Types Societal,Economic

Description Expert panels and talk at Nobel Dialogues Day, preceding the Nobel Prizes Ceremony in Stockholm, December 2018 
Form Of Engagement Activity A formal working group, expert panel or dialogue
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Public/other audiences
Results and Impact About 100 people listened to the Nobel Dialogues presentations which were also streamed internationally. Dowdeswell talked about the role of the Arctic and Antarctic in global environmental change.
Year(s) Of Engagement Activity 2018
URL https://www.nobelprize.org/events/nobel-prize-dialogue/
Description Seminar and Q & A session as part of Cambridge Advanced Leadership Programme 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach International
Primary Audience Professional Practitioners
Results and Impact Wide-ranging discussion of how environmental change affects the cryosphere and how that, in turn, affects humankind through, for example, sea-level rise

Followup from several industrial participants
Year(s) Of Engagement Activity 2011,2012,2013,2014
Description Talks to groups visiting the Polar Museum of the Scott Polar Research Institute 
Form Of Engagement Activity A talk or presentation
Part Of Official Scheme? No
Geographic Reach Regional
Primary Audience Schools
Results and Impact Talks to at least three schools and general public visits to the Institute's museum each year
Year(s) Of Engagement Activity 2014,2015,2016