4 Jul 2011

UK tsunami risk: an overview of the evidence

Last week an unusual tidal occurrence along the coastline of Cornwall, Devon, Dorset and Hampshire prompted interest and discussion in the media.  Tidal gauges from Cornwall to Hampshire recorded a 2ft-high column of water moving from west to east, and some witnesses reported a strong static charge in the air.  In the Yealm estuary near Plymouth water surged at four times the normal speed, as shown in the video below.


Credit:  Youtube/SDFitch

Dr Mark Davidson, an expert in coastal processes at the University of Plymouth, was quoted in support of the most likely theory:  that the phenomenon was a mild tsunami of some kind caused by an undersea sand or mud slide on the continental shelf 200 miles west of the Cornish coast.  It is unlikely that an earthquake was involved because British Geological Survey seismographs did not detect any activity during the relevant period.

An oceanographer who was at sea when the tsunami struck, Dr Simon Boxall, has put forward an alternative theory that the phenomenon was actually a 'seiche', i.e. a standing wave caused by an area of very low or high pressure moving across a body of water.  A low-pressure system would explain the reports of static electricity.  However other experts speculate that a piezoelectric effect caused by rock vibrations during an earthquake or landslide could also explain the static charge.

Whatever the precise cause, this phenomenon is trivial in its effects when compared to the Indian Ocean earthquake and tsunami of 2004 or the earthquake and tsunami in Japan earlier this year.  According to the British Geological Survey there is unequivocally no potential for an event of that magnitude striking the UK.  'Mega-thrust' earthquakes of that kind can happen only at plate boundary subduction zones where the Earth's tectonic plates jam up against one another.  Japan, Sumatra and South America are known to be at risk, but Britain sits squarely in the middle of a tectonic plate and the nearest subduction zones are some considerable distance away.

There is however a body of research into the risk of tsunamis of lesser magnitude affecting the United Kingdom, much of it developed as a result of concerns raised following the devastating Indian Ocean event as well as some evidence of tsunami-like events in the UK's historical record.

In 2004 two academics, Dr Simon Haslett of Bath Spa University College and Dr Ted Bryant of the University of Wollongong in Australia, published a study that examined historical records of the 1607 coastal flood event in the Bristol Channel and Severn Estuary.  Due to the widespread loss of life this event is arguably the most serious sudden natural catastrophe to affect the UK in the past 500 years, and is therefore of substantial interest to risk modellers.

The Severn Estuary has the second highest tidal range in the world, and most experts take the view that the 1607 flood event was caused by a massive storm surge.  However Haslett and Bryant found inconsistencies in the historical evidence and concluded that they could not rule out tsunami as a possible cause because of "the ambiguity of the regional meteorological conditions, the lack of documentary evidence for hurricane winds, and the nature of the damage inflicted."

The initial work from Haslett and Bryant was based on a reading of contemporary accounts of the flooding, such as a religious leaflet illustrated with the woodcut shown below.  However the two academics published an additional paper in 2007 in which they presented additional geomorphic evidence from the British Channel that they argued was consistent with a tsunami there during the timespan of the 1607 event.


In early 2005, prompted mainly by the Indian Ocean event, DeFRA commissioned a consortium of UK institutions led by the British Geological Survey to assess the potential for tsunami-type events originating in the North Sea and North Atlantic and to consider their potential threat to the UK.  Two reports were published:  'The threat posed by tsunami to the UK' in 2005 and a supplementary study 'Tsunamis - Assessing the Hazard for the UK and Irish Coasts' in 2006.  Both are available on the DeFRA website.

The DeFRA study considered four potential sources of tsunami that could affect the UK:  in the North Sea, the Celtic Sea, offshore of Lisbon and at La Palma in the Canary Islands.  The summary conclusions of the 2005 report were as follows:
  • A strong, potentially damaging tsunami reaching the coasts of the UK resulting from a passive margin earthquake in the Sole Bank area (western Celtic Sea) or associated with the Rockall Trough, or in the North Sea Fan area, is credible.
  • A tsunami reached the UK following the Storegga slide, but the geological model suggests that another glaciation (time scale ~100,000 years) is needed to re-establish the conditions required for a similar failure at that location. However, there are other sections of the neighbouring continental slope that have the potential for a landslide, possibly triggered by a passive margin earthquake.
  • A tsunami reaching the UK from the Azores-Gibraltar region (responsible for the earthquake and tsunami causing the destruction of Lisbon in 1755) is a possible event. However, the likelihood of a future tsunami from this source being worse than that of 1755 is negligible.
  • Landslides around the Canary Islands are probably less frequent than those on the continental slope of northwest Europe. Should a Canary Islands landslide occur, its effect at distance would be heavily dependent on the nature of the collapse. The evidence points to a gradual collapse being most likely, and this will have little effect at the distance of the UK.
Broadly the study concluded that the modelled heights of tsunami waves arriving close to the UK shore would be comparable to those in typical storm surges scenarios, and that major centres of population should therefore be able to cope using their existing flood defence infrastructure.  The report did however note that the impact of a tsunami might differ from that of a storm surge due to the absence of a link to meteorological conditions.  One of the signal characteristics of tsunamis, as with earthquakes, is of course the lack of warning time.

Some lingering doubts remain.  In 2008 Haslett and Bryant published a review, 'Historic tsunami in Britain since AD 1000' in which they argued that the DeFRA-backed reports had ignored some written historical records and that British tsunami risk required "a more careful evaluation".

The potential for a catastrophic collapse of the Cumbre Vieja Volcano at La Palma in the Canary Islands remains a subject of debate.  Early work suggested collapse of the volcano could cause a trans-Atlantic tsunami that would effectively wipe out New York, Washington DC, Boston and Miami in addition to wide-scale impacts in Southern England and Europe.  However subsequent modelling seems to indicate any collapse of the volcano would be gradual and that wave effects would therefore be more localised.