Coastal Flooding in the United Kingdom, 1953 and Now
Feb 21, 2013
Editor's Note: This month marks the 60th anniversary of the Great Flood of 1953, one of Europe's most devastating natural disasters. In this article, Dr. Peter Sousounis (Principal Scientist, Atmospheric Science), Raulina Brito (Engineer, Flood Modeling), and Nan Ma (Senior Editor, Communications) take a look back at the event and evaluate the UK's coastal flood risk today. This summer, the AIR Coastal Flood Model for Great Britain will be available in AIR's detailed modeling software CLASIC/2™ and Touchstone™. The model was previously available only in CATRADER®.
Sixty years ago, a powerful storm in the North Sea sent waves crashing over coastal defenses in the UK and the Netherlands. Floods inundated thousands of square kilometers of land, damaging some 70,000 properties. More than 2,500 people lost their lives in the Great Flood of 1953 and it remains one of Europe's worst natural disasters in several centuries. Physical damage in the UK at the time was estimated at approximately GBP 50 million.
A confluence of many conditions made the storm and resulting surge especially powerful. But just how rare was it, and could a similar disaster occur today? While flood defenses and warning systems have improved dramatically since 1953, the population and exposure at risk have multiplied many times over. Furthermore, the level of protection is not uniform along the vulnerable coast and there may always be waves capable of overtopping even the most solid defenses.
This article will take a look back at the 1953 floods, explain the factors that influence storm occurrence and surge potential, and examine coastal development and flood defenses before and after the event. Finally, we'll discuss what might happen the next time a large storm surge reaches the east coast of the UK.
The 1953 Event
The Great Flood of 1953 was the result of several meteorological, oceanographic, and even astronomical factors. From a meteorological perspective, the North Atlantic Oscillation (NAO) Index, which is well known to influence the tracks and intensities of storms that affect Europe, was highly positive. This allowed the extratropical cyclone to form sufficiently north and track over the British Isles before diving southeastward, and also created atmospheric conditions that supplied more energy to the storm.
The storm's formation over the central North Atlantic was unremarkable, but a set of conditions came together—including the phasing and super-positioning of an upper-level wave of low pressure and strong upper-level jets—that influenced its track and intensification. In fact, the cyclone underwent bombogenesis, a term reserved only for storms that intensify at the impressive rate of 24 mb or more in 24 hours. The cyclone deepened from 987 mb to 963 mb between 1200 UTC January 30 and 1200 UTC January 31. As a result, wind gusts in some locations exceeded 100 mph. In Costa Hill, Orkney, for example, a peak gust of 126 mph was recorded.
The narrowness of the upper-level wave drove the cyclone southeastward into the North Sea instead of eastward across it and also helped elongate it in the north-south direction. These two characteristics resulted in strong northwesterly winds down a long fetch of the North Sea for a prolonged period of time.
From an oceanographic perspective, the storm surge and flooding in the Netherlands was the direct result of these northwesterly winds driving (dragging) water onshore near the time of high tide. Near where the center of the storm came ashore, the low pressure of the cyclone itself contributed almost a third of a meter to the storm surge. Maximum storm surge values exceeded 5 m in the Netherlands.
Significant storm surge, exceeding 4.5 m in places, also occurred along the southeastern coast of the UK, even though winds were parallel to the coast. The nearly closed shape of the North Sea basin was one reason. Water that is dragged southward here will try to exit through the easiest opening it can—in this case the English Channel. As some of the water was redirected westward and pushed through this narrow channel, some of it piled up against the east coast of the UK as storm surge.
Furthermore, the Coriolis force, which results from the rotation of the earth, drives water at a 90° angle to the right of the direction of the wind (e.g., toward the west, for a south moving storm) in a process known as Ekman transport. The Coriolis force may also have enhanced a feature known as a coastal Kelvin wave. This type of wave is long in wavelength and results in water piling up along right-hand side coasts in the northern hemisphere. (Both Ekman transport and coastal Kelvin waves contributed to the storm surge early on with Hurricane Ike in 2008; these effects are even more significant at higher latitudes.)
Finally, from an astronomical perspective, the storm surge in many of the affected regions occurred at the time of high tide. This alone accounted for water heights of 2-3 m above mean sea level in many places. Furthermore, it was not an ordinary high tide but rather a spring tide, occurring just after a full moon. This accounted for an additional meter of seawater rise.
The storm surge breached more than 1,000 flood defenses in England, and damage occurred along a 1,600 km stretch of the coast from Yorkshire to Kent. An estimated 24,000 homes were damaged.
A Brief History of Coastal Development
The east coast of England is no stranger to flooding from the sea. In medieval times, Anglo-Saxon settlers in this region built dirt embankments before reclaiming salt marshes and inland wetlands for farming purposes. Surviving evidence of these defenses can be found along the Lincolnshire coast, upon the Fens, and around the Thames Estuary in Essex and Kent.
It was only after the medieval period that it became standard practice to build flood defenses to protect existing settlements, and the development of land in coastal floodplains accelerated in the 19th and 20th centuries with the expansion of coastal towns and seaside resorts. Settlements that originated on relatively higher ground eventually spread into lower elevations to accommodate a growing population. This led to the building and strengthening of coastal defenses, which encouraged further expansion in low-lying areas.
One example of this tragic development cycle is Canvey Island in the Thames Estuary, situated below the mean level of high tides in the spring. The population of the island surged from about 300 at the turn of the 20th century to more than 11,000 in just fifty years. Canvey Island was one of the worst affected areas in the UK during the 1953 floods, when waters inundated the entire island.
London, too, has expanded throughout history with the reclamation of marshlands and the construction of increasingly sophisticated tidal defenses on the Thames Estuary, usually in response to flood events. The earliest recorded account of flooding along the Thames Estuary was in 1099, and later events were noted in London in 1236 and 1663. More recently, a storm surge in 1928 coinciding with high tide produced a peak height of 5.55 m on the Thames in London. A large portion of the city was flooded and repairs took years to complete.
Now, in the 21st century, London continues to expand. As part of the Thames Gateway urban regeneration project, commercial and residential development is heading eastward into marshlands, some of the best natural protection against flooding.
Flood Defense and Warning Systems
Since the flood in 1953 (and very much because of it), the UK government has devoted significant resources to improving coastal defenses. In the last decade, the government has spent more than GBP 250 million per year in the counties of Lincolnshire, Norfolk, Suffolk, and Essex. Along most portions of the coast, flood defenses implemented by the Environment Agency are designed to withstand a 1-in-200 year surge event.
The center of London was spared from major flooding in 1953, but the embankments along the Thames came dangerously close to being overtopped, which served as an urgent wake-up call that stronger defenses were needed to protect the capital. Legislation approving the construction of the Thames Barrier was enacted in 1972, and it became fully operational in 1982. The Barrier is designed to protect against a 1-in-1,000 year flood event until the year 2030, allowing for an annual sea level rise of 8 mm per year. Its gates have been closed a total of 76 times (51 since 2000 alone) to protect against tidal flooding.
In addition to coastal defenses, the UK has also made significant improvements in forecasting and alert capabilities, the absence of which was a major reason why there was such a high loss of life in 1953. The Environment Agency uses forecasts from the Met Office's Storm Tide Forecasting Service to provide flood alerts to local authorities and directly to residents and business owners. Approximately 1.2 million people are signed up to receive automated warnings.
Will There Be a Next Time?
The 1953 event required many conditions to come together that—taken individually—were not all that unique. The wintertime NAO exhibits considerable variability on multiple time scales. For example, the NAO Index was positive a few days before the 1953 cyclone formed, but dropped to near zero just two days later. And, from the 1970s through the 1990s, the NAO was positive in a vast majority of years during the winter. Since 2000, the split between NAO+ and NAO- years has been near even.
Storm tracks that carry cyclones down into the North Sea rather than across it are also not that unusual, occurring every ten years or so (Rijkswaterstaat and KNMI 1961). Oceanographic conditions, including the bathymetry and funnel shape of the North Sea, represent a fixed hazard. High tides occur twice a day and spring tides occur twice a month, at the time of new moon and full moon.
The most recent threat to the east coast of the UK came in November 2007 (during a NAO+ phase). On the heels of the extratropical remnants of Hurricane Noel, a low pressure system (also known as Cyclone Tilo) arrived in northwestern Europe. Atmospheric conditions at the time, including a ridge of high pressure to the west of the British Isles and the position of the jet stream, created a steep pressure gradient similar to that in 1953. This resulted in a northwesterly flow that diverted the system into the North Sea. Arriving around high tide, the system was forecast to overtop many sea defenses, and flood warnings went into effect along the east coast of the UK and in the Netherlands. Fortunately, the storm surge was less severe than predicted, and damage was relatively minor.
A 2008 paper by Zong and Tooley studied the frequency of coastal flooding in the UK since the 1780s based on archival newspaper records and synoptic weather reports published by the Met Office. They found that the east coast, defined as the segment between Aberdeen, Scotland, and Ramsgate, Kent, experiences the highest frequency of coastal flooding in the UK. In the AIR Coastal Flood Model for Great Britain, damaging storm surges affect the east coast an average of once every ten years. In terms of losses trended to 2007 (to account for inflation and exposure growth, but not reflecting current flood defenses), the 1953 event causes total economic losses of approximately GBP 3.3 billion, corresponding to an annual exceedance probability of 0.7% (or a return period of around 140 years). While it is the most damaging storm surge for the UK in living memory, it should not be considered an exceedingly rare level of loss. In fact, more damaging scenarios are possible.
Present Day Scenarios
Extratropical systems with lower central pressures than the 1953 event and arriving with higher tide conditions are entirely feasible. In fact, tide heights for the 1953 event were estimated to be 1 to 3 feet lower than they could be at other times of the year, depending on location.
The scenario shown in Figure 3 represents a 1,000-year loss event for the Greater London area. The Thames Barrier itself holds up in this scenario, but many smaller downstream defense structures fail. Damage is mostly confined to outside the administrative boundary of Greater London (outlined by the dark black line), and insured losses in the region total approximately GBP 0.5 billion.
In an even more extreme scenario, representing a 2,000-year return period for the Greater London area, the Thames Barrier fails. As shown in Figure 4, flooding on both sides of the river extends throughout London, all the way upstream to the Teddington Lock. In this scenario, modeled insured losses in the region total more than GBP 4.5 billion.
According to the Environment Agency's director of flood and coastal risk management, 1.3 million people in England and Wales are at risk of coastal flooding. Expected annual economic losses from coastal floods has been estimated at GBP 0.5 billion, but under current flood management practices, this figure is projected to increase dramatically over the next century because of climate change (Hall et al. 2006).
The UK has been identified as one of the most vulnerable countries to sea level rise in Europe. A study by Lowe and Gregory (2005) used physical models to simulate the effects of greenhouse gas emissions and sea level rise on storm surge heights in the UK. They project that in the 2080s, the number of winter storms that cross the UK could increase from five to eight. This increased storminess is accompanied by higher surge levels, which are most pronounced off the southeast coast of England, where the 50-year return period storm surge height is projected to increase by approximately 0.7 m.
There is significant uncertainty, however, in all climate change projections and the historical record is insufficient to conclude whether there is a long-term trend toward more extreme surge heights or storminess.
As it stands, much of UK's coast relies heavily on manmade flood defenses, and an estimated 10% to 15% of the coastline is less than 5 m above sea level. Coastal flooding will remain a prominent issue, for the population at risk, the government, and the insurance industry.
English Heritage (2011). Roman and Medieval Sea and River Flood Defences, Introductions to Heritage Assets
Hall, J.W., Sayers, P.B., Walkden, M. Panzeri, M. (2006). Impacts of climate change on coastal flood risk in England and Wales: 2030-2100, Phil. Trans. R. Soc. A, 364(1841), 1027-1049
Lavery, S., and Donovan, B. (2005). Flood risk management in the Thames Estuary: looking ahead 100 years. Phil. Trans. R. Soc. A, 363(1831), 1455-1474
Rijkswaterstaat and KNMI (1961). Verslag over de stormvloed van 1953 (Report on the 1953 Flood),'s-Gravenhage: Staatsdrukkerij en Uitgeverijbedrijf
Steers, J.A. (1953). The East Coast Floods, The Geographical Journal, 119(3), 280-295
Zong, Y. and Tooley, M. J. (2003). A historical record of coastal floods in Britain: frequencies and associated storm tracks, Natural Hazards., 29(1), 13-36