On the night of 31 January/1 February 1953, severe flooding from the North Sea claimed over 300 lives along the east coast of England and over 2000 lives along the Dutch and Belgian coasts. This article describes the background to the event and some of its effects, and looks at the lessons that have been learned from it.
We often hear today, sometimes even from apparently authoritative sources, that extreme weather is occurring more frequently than used to be the case; such comments tend to be the staple diet of the media whenever a storm breaks or high/low temperatures are reported. It is very difficult to prove or disprove such statements (for the UK, let alone globally which is beyond the scope of this article), but mostly they appear to be based on ignorance. A riposte for anyone seriously concerned that there may be an element of truth in them is to consider the weather between 1947 and 1953: extreme followed extreme during this period in the United Kingdom. This article is concerned in detail with just one of those events, which in its effects was the most disastrous of them all, but first it will be apposite to briefly refer to the other extremes during this period. The data come from a variety of sources studied by the author over the years – especially the erstwhile Daily and Monthly Weather Reports of the British Meteorological Office and British Rainfall. Central England Temperature data can be accessed at http://www.metoffice.gov.uk/hadobs/hadcet/cetml1659on.dat and England and Wales rainfall at http://www.metoffice.gov.uk/climate/uk/datasets/Rainfall/date/England_and_Wales.txt
Summary of extreme weather 1947–1953
The late winter of 1947 was probably the most extreme of the twentieth century (though it has a rival in that of 1962/1963). It was remarkable partly because there had been little out of the ordinary about it until 22 January. February, though, was the coldest of that century: the central England mean temperature of −1.9 °C was nearly a degree lower than that of any other February throughout the twentieth century, and there was no sunshine in London from the 2nd to the 22nd inclusive. March was the coldest of the century in Scotland, and the wettest in England and Wales, and there was widespread severe flooding as a rapid thaw of the snow was followed by frequent heavy rain.
Then came a brilliant summer, peaking in August: for the combination of lack of rain, abundant sunshine and high temperatures, the month was unprecedented for over 75 years over practically the whole of the British Isles (but was eventually surpassed by August 1995). Later in the year, the wedding day of Princess Elizabeth on 20 November was marked by a most unseasonable temperature of 18 °C in London – a few degrees higher than it was on Coronation Day in June 1953.
There then followed the wettest January of the twentieth century (by some margin) in England and Wales in 1948, a temperature of 24 °C in London on 9 March, a severe westerly gale in southern Britain on 7 August and unprecedented flooding and immense damage across southeast Scotland on 12 August after up to 160 mm of rain; rail links between Berwick-upon-Tweed and Edinburgh were suspended for several weeks owing to the destruction of bridges.
The Saturday of the Easter weekend in 1949, 16 April, brought a temperature of 29 °C in London, and the September that year was the warmest of the century. On 21 May 1950, a major tornado (by British standards) tracked from Wendover to Linslade (Buckinghamshire). September 1950 was the wettest of the century (by far) over Scotland, whilst the very cold, snowy December in 1950 was only ‘beaten’ by that of 1981.
After an unremarkable 1951, mid-August 1952 brought catastrophic flooding and much loss of life in the Lynmouth area of north Devon, following up to 225 mm of rain (Marshall, 1952). There was a notable smog in London from 5 to 8 December 1952: this reduced visibility to a few yards for long periods (Bonacina, 1953; Meetham, 1953). Overall, the weather in the weeks prior to the North Sea flood disaster was mostly quiet and cold: there were only short interludes of mild, ‘westerly weather’ in December 1952. However, there was what might be seen as a precursor to this event when a very deep depression moved down the North Sea on 17 December giving a severe northwesterly gale, including a gust of 96kn at Cranwell (Lincolnshire). January 1953 was quiet and anticyclonic until the closing days.
Comparisons with recent years and perspectives
A look at the records for the period 2006–2012 reveals rather more instances of record-breaking months than there were between 1947 and 1953 (for temperature and rainfall), but fewer extreme individual events. It should also be noted that changing times bring changing impacts of particular types of weather. The 1947 winter was extreme by any standards, but it hit a country on its knees after the war, with fuel and food supplies already severely restricted. Some of the loss of life in the Lynmouth disaster could be attributed to cottages having been built far too close to the River Lyn: lessons were learned from this – but we still build on flood plains. London's air was cleaned up after the smog (which owed much to chimney and industrial smoke), and there has not been anything like it since – indeed, thick fog in central London is virtually unknown now as a result of the warmth given off by buildings on cold nights preventing the temperature falling to its fog point. Smog is now rather different: visibility may not be greatly reduced, but vehicle exhaust can create toxic conditions during a period of still weather. Athens and Beijing are perhaps the cities best known for this problem.
As will quickly become apparent when we look in detail at the North Sea Surge of 1953, there was a large element of ‘circumstances conspiring’. It could not happen today to anything like the same disastrous effect (at least in terms of lives lost) – though that does not mean we should be complacent. The natural world can catch us out in so many ways, and whilst it is commonplace now to blame weather disasters on human interference that is by no means always the cause.
Most of the rest of the material in this article is based on ‘North Sea Surge, the story of the East Coast Floods’ (Pollard, 1978), an excellent book which sets the scene in its first sentence: The night of Saturday 31 January 1953 brought to Britain one of the greatest peacetime disasters in her history. (Throughout the following sections, direct quotes from the book are in italics).
Tragedy at sea
Figures 1-3 show how the weather pattern developed to bring the disaster. Initially, the depression off southern Iceland looks innocuous, but as it deepened it began to turn towards the southeast and by early morning on 31 January a very tight pressure gradient was apparent on its western flank. Those living on the east coast of England, and on the Dutch coast, remained completely unaware of impending doom – which would not be the case if a similar event were to occur nowadays.
But before any of this happened, there was a major tragedy in the northern Irish Sea. In the early afternoon of 31 January, the British Railways car ferry Princess Victoria, on its regular crossing from Stranraer to Larne, keeled over and sank after a six-hour struggle against the severe gale. 132 of the 174 people on board lost their lives including all the women and children, who were thrown into the sea when their lifeboat overturned as it was being launched. Several smaller ships also sank off our northwestern coasts. Over land, large tracts of Scottish forest were laid waste: in his book, Michael Pollard notes that the number of trees felled by the wind during Friday night and Saturday equalled the normal cull of a whole year.
Saturday night in the early 50s
In 1953, television was in its infancy – only to be found in a few houses – and there were just three British radio networks, all run by the BBC, with infrequent news and weather bulletins. The news was full of the sinking of the Princess Victoria, a horrific enough story in its own right, which shocked many. It is doubtful if warnings of ‘northerly gales of exceptional severity’ along the coast really registered – and what to do if they did? It was a rather different story on the other side of the narrow North Sea, where warnings of ‘dangerously high tides’ had been issued in the late afternoon. There was at that time no formal arrangement for tidal warnings in Britain.
A full moon in late January meant a spring tide on this night – although this was not predicted to be as high as either of the new moon (neap) tides a fortnight either side of it. An official government report, the Waverley Report, later noted that coincidence with those neap tides would have meant that, given the same distribution and sequence of winds and atmospheric pressure, the disaster would have been worse. It was, though, not just the high tide that was responsible, but the surge that preceded it by about 90 minutes. A tidal surge is produced when high winds build up a wall of water, and is exacerbated by the effects of the pressure gradient itself. All things being equal, sea levels are higher under low pressure than under high pressure, so where pressure changes very rapidly at right-angles to the flow significant unstable variations in the depth of water can develop: the earth's rotation also plays a part (much more detail on the complex causes of surges may be found in Harris (1963)). On this night, it was estimated that the pre high-tide surge may have exceeded ten feet in places. The effect of surge followed by high tide was to create a prolonged exceptionally high tide which would, even without the storm-driven waves, have severely tested the sea-defences.
The disaster unfolds
Figure 4, taken from a short article by Maurice Crewe (2003), shows the area affected by the disaster. There was localised flooding along the coast of northeast England during the Saturday afternoon, but initially the raging sea did not unduly overflow the sea-walls of Lincolnshire, except at Cleethorpes and Grimsby: it was not yet an exceptional event. It was the north Norfolk coast that caught the full fury of the water first. Residents, well-used to batterings from the elements, had spent the afternoon preparing for a ‘normal’ storm as best they could, securing their boats, shoring up shrubs and trees, and generally battening everything down in readiness for a wild night. Darkness fell and they retreated to the fireside. Soon after 5 pm, Wells-next-the-Sea suffered the first major invasion of the sea, forcing people to move upstairs. Reports indicate that it was shortly after this that the Lincolnshire coastal villages surrounding Mablethorpe and Sutton-on-Sea were inundated, suggesting that as the surge hit the north Norfolk coast it was impeded sufficiently to ‘back-up’ further north. 40 people lost their lives along this stretch of the Lincolnshire coast, and there were over 60 deaths in Kings Lynn and along the west-facing stretch of coast to Hunstanton, especially in a settlement of bungalows and chalets by Snettisham Beach. A train on the single-track line from Hunstanton to Kings Lynn collided with a floating bungalow which rendered the engine immovable. The lights failed, the engine fire went out, chunks of debris battered the stricken train, and passengers had to stand on the seats to keep clear of the water. It took six hours for the train to reverse to Hunstanton.
The surge rounded the coast of northeast Norfolk in mid-evening, hitting the village of Sea Palling badly and claiming seven lives. Great Yarmouth and Lowestoft were the next towns of any size to be hit – twice in the case of Great Yarmouth, as the banks of the five-mile long Breydon Water to its west gave way, so that water flooded into the town from its rear shortly after the sea had battered it from the front. 9 people died and around 10 000 had to be evacuated. Lowestoft fared better: a new sea-wall, built at a cost that was apparently criticised by many ratepayers, blunted the full force of the sea. The town was flooded but not overwhelmed as were so many communities that dreadful night. If the wall had not held, the town, with Oulton Broad behind it, might well have been trapped between two waters as happened at Great Yarmouth.
The effects of the surge became yet more serious as it reached the villages along the coast of Suffolk and Essex. Many of these were populated by elderly people in bungalows (originally never intended for winter occupation), and it was now the middle of the night so that most had retired to bed and thus had little chance of surviving the ordeal that was about to befall them. Southwold, with only one route into it, was cut off for 48 hours, whilst Felixstowe, like Great Yarmouth, was attacked by the water on two fronts, leading to 39 people losing their lives, including several whole families. The tide was 8 feet above its expected height when it swept into Harwich, taking 8 lives and forcing the evacuation of over 1000 residents. Figures 5-8 illustrate some of the flooding along the Essex coast, scenes that were multiplied along over 100 miles of the coastline of eastern England.
One of the worst-hit communities was the bungalow beach settlement of Jaywick, south of Clacton (Figure 8). Again, these bungalows were intended only for summer use, but about 250 of the 1800 chalets were occupied that night. 37 people, with an average age of 66, died there. Rescuers had little clue as to which chalets were occupied and which were empty. But the last place to be hit was to suffer the worst fate of all that night. 58 people died on Canvey Island, another unplanned settlement to the west of Southend with just one road linking it to Benfleet. It had a population then of about 10 000, many once again in housing intended for summer occupation only, and here too there was no record of which houses were permanently occupied. Canvey Island was, in a word, overwhelmed quite suddenly, some time after midnight. It was Jaywick all over again, but on a much larger scale. For the six hours or so between the sea's invasion and dawn the situation on the island was too confused and too appalling for anyone to handle. During the next 48 hours, the entire population of Canvey Island was evacuated.
There was some flooding around the Kent coast, but London escaped – just – probably because there was sufficient west in the wind to offset its thrust up the Thames Estuary, and maybe because so much water had already settled on to land areas in its passage around East Anglia. That same westerly component in the wind meant there was no escape for Holland, much of the coastal strip of which is below sea-level. An official account there stated that nearly 1800 people are known to have been drowned, over 50 000 evacuated, and about 50 000 cattle and 100 000 hens were lost.
The first rescue efforts
Before a major relief programme could be organised there were many tales of heroic rescue attempts, which met with varying degrees of success. In Lowestoft, a party of some 40 children, with adult helpers, was stranded in a church hall where they had been enjoying a social event. Three police officers, alerted by their cries, found them marooned above the altar steps in the adjacent church: the aisle was waist-deep (to adults) in water and the pews were adrift in it. There was little they could do, other than seek to keep everyone calm, but after several more minutes a local boatman turned up. It took several boat trips, each one involving carrying the children and adults one by one from the altar to the boat, to effect the rescue. One police officer recalled that one of the adult helpers seemed to have no concern for her companions, least of all the children, but when it was her turn to be carried to the boat ‘gave me explicit instructions not to get her fur coat wet. Had I not been in a House of God I would probably have dropped her into the water and made her walk to the boat’.
The aforementioned beach settlement of Snettisham, on the northwest Norfolk coast, was the scene of prolonged rescue operations throughout the night, with the Station Inn becoming a sort of advance medical station. Here too, one woman was concerned for her fur coat: she insisted on going back for it, and it cost her her life. Her body, still without the coat, was found next morning in a hedge in the field next to her bungalow. One rescuer commented that they had to contend not just with the wind and water as they drove from beach to inn, but also with an airborne barrage of straw, hay, branches and other debris. Salt-spray covered the rescuers and their cars, and ‘to make things more unpleasant it started to snow and sleet for a time’.
At Dunwich (Suffolk), a 91 year-old invalid refused to be moved from a downstairs room of a school house – until the sea reached the level of her bed, at which point two Special Constables forced their way into her room through the window (having been unable to open the door because of the force of the water), and carried her to the relative safety of an upstairs room. At Felixstowe, a pregnant woman, her husband and two young children had to fight their way onto their roof and wait for about five hours, cold and wet, before they were rescued by boat. There was a ‘magnificent sight’ in Harwich, where a local man was spotted rowing a dinghy which was loaded with cats at the stern, dogs in the middle, and a most magnificent parrot perched on the stern and shouting its head off: ‘Get out of here you buggers’.
But, of course, the main story of the night was one of death and devastation on a horrendous scale.
Coping with the aftermath
The first two paragraphs of chapter four of Michael Pollard's book are worth quoting in full. When the dawn came on Sunday, the east coast of England was an appalling sight: just how appalling could be appreciated only by the RAF crews who were sent out at first light to take aerial photographs of the devastation in ‘Operation Floodlight’, which was to continue for several days. The monstrous surge in the night had made nonsense of the maps. Along three-quarters of the coastline, the defences had been damaged, and huge lengths – like the 34 miles from Cleethorpes to Skegness – almost completely destroyed. South of Mablethorpe and Sutton-on-Sea, some 22 000 acres of prime agricultural land lay under the sea water that poured in through a gap in the sea wall one-third of a mile long, the biggest of the many breaches. The bank between Kings Lynn and Hunstanton, with its formerly neat line of beach huts and bungalows, had been reduced to a shambles of driftwood, with caravans and shanties lying on their sides in the water like toys. The north Norfolk villages of Salthouse and Cley, normally well inland, had become waterside settlements which looked as if they had been blitzed. The cliffs of northeast Norfolk showed several new shallow bays that the sea had carved out during the night. Great Yarmouth and its hinterland were unrecognisable from the air. Further south, Foulness and Canvey had become half-submerged offshore islands, with many of the residents still a day or two from rescue. The surge had driven up the Thames Estuary and taken over large stretches of industrial land close to the river. Pressing on, it had made incursions round the Kent coast as far as Deal.
The tallies of damage were added up later: over 300 lives had been lost, over 24 000 houses damaged, 200 industrial premises flooded and put out of action, along with 12 gasworks and 2 power stations, 11 trunk roads made impassable, 200 miles of railway track put out of use, 160 000 acres of agricultural land made sterile for a season at least and in many cases for years, 9000 sheep lost as well as 1100 head of cattle, 2600 pigs, 34 000 head of poultry, 70 horses…Perhaps it was fortunate for morale that the statistics were not available that Sunday morning.
It was to be a slow, painful journey out of the disaster. At first, the news spread slowly as befitted a winter Sunday morning in the early 1950s: the loss of the Princess Victoria was still the main story. Even on Monday morning, The Times reportedly dismissed the events of the weekend in 600 words. Furthermore, since the county administrative centres of Lincoln, Norwich, Ipswich and Chelmsford were not affected by the floods, it took time for them to react. There was much, understandable but perhaps unfair, criticism about the slow reaction of officialdom: there was at the time no recognised call-out system to deal with emergencies. There was just so much to tackle: where do you start? (Nothing to do with meteorology, but the author has retained an image in his mind of another tragedy around this time – a multiple train-crash at Harrow-on-the-Hill (northwest London) in August 1952. The scene that befell rescue workers defies comprehension. It is a credit to all involved, and to human resilience, that we do carry on after the most horrendous disasters – and it is so easy to criticise initial apparent tardiness). Local government workers joined the armed forces (British and American) in the reconstruction work, and a steady stream of contractors in all the relevant fields was soon involved. Of course, the other side of human nature also reared its head. There were sporadic outbreaks of looting in areas from which shopkeepers and residents had been evacuated, despite police reinforcements sent to these areas not only to deter looters but also to keep residents away until it was safe for them to return. There was also a certain amount of trouble with parties of sightseers, though not as much, no doubt, as there would be in the car-owning society of today, or if disaster had struck at any other time but midwinter.
One of the most urgent tasks was to repair sea-wall breaches to prevent the twice-daily tides reinforcing the flooding – and certainly before the major neap tide was due in mid-February. Evacuation of the survivors in the most devastated locations was also paramount: most of the population of Canvey Island was taken to safety by midnight on Sunday, by means of coach, bus, taxi, private car and boat. A few miles to the northeast, the only contact the survivors on Foulness island had had with the outside world for some 30 hours was a BBC message broadcast to them at Sunday midnight promising them rescue in the morning.
Life slowly – very slowly – returned to normal in these east coast communities, amongst many recriminations about how relief funds were spent. It was obvious that ways had to be devised to prevent anything like it ever happening again. Clearly, weather and tide was perfectly capable of staging a repeat. Less widespread, but still locally serious, flooding had affected the east coast several times prior to 1953 during the twentieth century – see Jensen (1953a; 1953b) and Lawrence (1953), for accounts of these and more ancient North Sea floods. What was the best way of tackling it next time? Sturdy, well-designed sea defences are one, very costly, way; it was decided some years ago that this was impracticable all along the coast, and that stretches that can ‘safely’ be left to flood should be allowed to do so – safely that is for humans, but maybe at serious deletrious effects to agriculture and related industries. For a detailed account of the latest policy see the Environment agency's Coastal Handbook (http://publications.environment-agency.gov.uk/PDF/GEHO0610BSUE-E-E.pdf).
It was also recognised that London had had a lucky escape in 1953. Some Londoners had not been so lucky in a flood in 1928, and a major tidal incursion into the centre of London could drown thousands as well as having severe effects on all manner of infrastructure. The Thames Barrier at Greenwich is the outcome of such concerns (McRobie et al., 2005), and is effective at not only keeping excessively high tides at bay on one side, but also occasionally at limiting the outflow of a rain-swollen non-tidal Thames from adding to high tides. But any excess of water, tidal or not, still has to be allowed to dissipate somewhere, and the effects of a slowly-sinking southeastern coastline and rising sea-levels consequent on a melting polar icecap still pose major concerns.
Forecasting a major North Sea surge is a key factor in saving lives (http://www.pol.ac.uk/ntslf/model.html). In the worst-case scenario you may not be able to do much about the water, and serious flooding may still result with devastating effects to infrastructure in some localities. But no lives should be lost – provided the warnings are heeded. A comprehensive, frequently-reassessed warning system, involving the Met Office, Environment Agency, local government, river authorities and emergency services is now in place. Practice exercises are held at intervals, and everybody living in the vulnerable communities should be well aware of how they should react if a real warning is issued. The practical problems of any threatened repeat may be that not all residents will react as they have been advised. Some may refuse to move, others may panic: a concern on Canvey Island has been that an inappropriate rush to the mainland could create havoc with possibly disastrous consequences. Nevertheless, a repeat of the 1953 disaster is surely unimaginable.
I am grateful to Steve Jebson at the National Met Library, Exeter, for allowing me the extended loan of Michael Pollard's book, and thank Councillor John White of Tendring District Council for facilitating the use of Figures 5-8.