What caused the flood?
Posible causes include high tides, crumbling sea walls, storm surge and a tsunami. The tide was certainly at its highest, but this alone would not be
enough. The sea defences may have been in a poor state of repair, but are unlikely to have given way on such a scale right across the estuary. A storm
surge is quite likely, especially combined with high tides, but there is also evidence that there may have been a tsunami.
Sea walls
Tides
Weather - storm surge
Tsunami
Sea Walls
Since most of the Gwent and Somerset Levels are below the height of the the highest tides, any breach of the sea defences would be disastrous. It is often
stated (though without reference to the source of the information) that the sea wall collapsed at Burnham and thirty villages were flooded.(E.g Somerset
by Ralph Whitlock, p122 ISBN 0 71342905 4; and www.burnham-on-sea.com). Historically the
sea defences had been maintained by the churches, such as Wells and Glastonbury. and with the dissolution of the monasteries under Henry VIII, they may
have fallen into disrepair.
However, there is clear evidence from John Stradling's epigrams that one sea wall had only recently been constructed at
Aberthaw in Glamorganshire, yet this was destroyed.
A document dated 1626, entitled "A survey of the Sea Walls" among the Afan Wallia MSS, held at Glamorgan Records Office, details the poor state of many
sea walls in what is now the Port Talbot area. Although the handwriting is hard to read, there is evidence of rebuilding in 1607, but it is not clear if
this is before or after the great flood.
A paper by J.R.L. Allen on Reclamation and Sea Defence in Rumney Parish mentions that the Records of the Court of Augmentations for 25 February 1590/1
say "Note that the Manor of Rompney lies adjoining the sea. There is a wall between the sea and the lordship for the defence of the same, which wall
being about 2 years past in great decay, was by commission new made and placed more into the land than before it was."
Thus there seems to have been quite a bit going on in the way of repairing sea defences, although the fact that certain stretches were repaired doesn't mean
that neighbouring stretches might not have been in just as bad a state, but not repaired.
The preface to the 1829 reprint of the Lamentable newes out of Monmouthshire pamphlet has two pages about Sea Walls.
It states they are "admirably built, to the height of twelve to eighteen feet" (but presumably this applies to 1829 rather than 1607). Its final paragraph
casts light on the maintenance of the defences:
The Commissioners of Sewers, originating with Henry IV, and Henry VII. were, according to Mr. Williams, continued and regulated
to preserve the moors, frequently inundated in times of confusion and civil war; and though the regulations were on summary principles,
they have not been since altered; because proceedings to check inundations, particularly those of the sea, must be secured by prompt obedience of
the tenants and landholders, in repairing the first breaches or injuries of dykes and sea-walls, which cannot be insured without powers of great extent
in the commissioners.
By the Laws of the Court of Sewers, the Commissioners are invested with powers equal, if not superior, to the Sovereign. For neglect of duty, or payment
of rates, they can not only seize upon the premises, but make a claim to, and alienate the property.
Tides
Knowing the precise date (see calendars page), the state of the tides can be calculated. In fact there was an exceptionally high spring tide on the morning of the flood. Even though this may have been the highest tide of the year, it would not by itself be a likely cause, since it would be only inches higher than other high tides in days and months before, and equally high tides would have occurred regularly once or twice a year without such disastrous consequences.
High tides occur twice a day, and on that day high tide was about 9am, just before the flooding occurred.
Swansea tides on 16th January 1607 calculated by S-Tide. The Gwent and Somerset areas would have been about an hour behind.
Twice a month, the high tides are at their highest (spring tides), with smaller neap tides on the weeks in between.
The highest of the spring tides that month occured on the day of the flood.
Tides for January 2006 are quite similar to those of 1607.
The blue line records the rise and fall of the tide twice a day. The height of the highest tides varies - with a smaller peak in the early part of the month and a bigger peak later on.
The moon's orbit is not quite circular - its closest point to the earth is called perigee, and its furthest is apogee.
When perigee occurs at full moon, then apogee is roughly at new moon and vice versa. In such a case one spring tide of the
month is particularly high (perigeean spring tide), while the other is less so.
(When perige and apogee occur at half moon, then both spring tides are an average height.)
The 27th January 1607 was both the second closest perigee of the year, and a full moon. The highest tides follow two or three days later, and in this case the morning of the 30th
would have had the highest tides of the season.
Although one of the highest tides of the year, this was still only
inches higher than the highest tides of previous days and months, and presumably not significantly higher than the perigeean
spring tides of previous years. The Gwent and Somerset Levels are about 6 metres above mean sea level. A high spring tide would be higher than this, but not enough to flow
over the top of the sea walls. Tides alone could not be sufficient to cause this amount of destruction.
Of course, a tsunami is no less likely to happen at this point than any other, but it does seem a coincidence that one
should strike at precisely this time.
Weather
Rainfall
Unlike many recent floods, this one was not caused by high rainfall. The source of the water was clearly the sea, not the rivers. Much has been made of
the statement that the sun was "fayrly and bryghtly spred" to show that the weather was not too bad on the day. Certainly there was no evidence that it
was raining at the time. However, there are other reports that it was a wet month and the Severn was in flood.
Evidence of the weather comes from one of the first expeditions from Britain to colonise America.
According to http://home.att.net/~englishamerica/places/va607002.htm (with my italics)
"December 20, 1606 The first three ships ("104 men and the crews") departed London for Virginia." by "January 5, 1606/7, the ships of the Virginia
expedition anchored at the Downs (England). They were stormbound for about a month. Due to taking a longer route around the Canary Islands to the
West Indies, and exceptionally stormy weather, they didn't arrive at the "capes of Virginia" until April 26, 1607"
According to the Monmouthshire pamphlet, 400 ewes perished at Llandaff, 4 miles inland, by the river Taff. This is quite a long way from the sea, and at
about 14 metres above sea level is higher than anywhere else that was affected, which could indicate that the river was also in flood.
The Somerset pamphlet states "In January last [towards the end of the moneth,] the sea at a flowing water meeting with Land-floudes, strove
so violently together, that bearing downe all thinges yt were builded to withstand and hinder the force of them" This suggests that the
surrounding land was already somewhat flooded.
Storm Surge
However, it does not require rainfall for a storm surge to occur, which, combined with a very high tide, could quite possibly overflow the sea defences.
A storm surge is the result of the combined effect of low pressure and winds driving the sea water towards the land.
The following is taken from a book whose title I have lost:
Strong winds from the SW drive water into the Irish Sea and raise sea level. The effect is more significant in the shallow extremities such as the
Liverpool/Morecambe Bay areas and Severn Estuary. Surges of 1m or more can be expected perhaps ten times per year. In January 1991 a surge peak of
2.3m was recorded at Heysham.
Of course a 1 metre increase in sea level would go unnoticed at low or medium tide, and even at most high tides would be within the scope of the sea walls. But if such a surge happened on 30th January at 10 in the morning, it would have combined with the high tide, and could easily have overflown the sea defences over a wide area.
A similar flood occurred in January 1953 in eastern England as a result of a storm surge.
Evidence
The records published at the time fairly consistenly blame a combination of stormy winds and a spring tide:
The Arlingham parish register mentions "an exceeding great fludd, and the greater by reason of the south west winde".
The Monmouthshire pamphlet states "In the month of Januarie last past upon a Tuesday, the Sea being very tempestuously moved by the windes, overflowed his ordinary Bankes."
The Barnstaple parish register says "there was such a mightie storm and tempeste from the river of Barnstaple with the comminge of the tyde"
The document by John Paul, the Vicar of Almondsbury states "the ryver of Severn rose upon a sodeyn Tuesday mornyng the 20 of January beyng the full pryme day and hyghest tyde after the change of the moone by reason of a myghty strong western wynd."
Walter Yonge states in his diary "The 20th of Jan 1606-7, by reason of a great tempest, the sea brake in at divers places on the north side of this country."
Camden in his major historical work, Britannia, published in 1607, writes: "the Severn-Sea after a Spring-tide, being driven back by a Southwest-wind (which continued for 3 days without intermission) and then again repuls'd by a very forcible Sea-wind, it raged with such a tide, as to overflow all this lower tract,"
Burton's Admirable Curiosities says "A mighty west wind continuing 16 hours brought the Sea into the Severn (after a great rain and a spring tide)"
Tsunami
Recent work by Simon Haslett and Ted Bryant (see
burnham-on-sea.com and the
BBC), suggesting the flood was the result of a tsunami, has received much publicity. There is much
evidence on the ground that a tsunami, with a much greater power than just a storm surge, has occurred in the channel. However, fixing a date to this
event is not so easy. There is clear evidence in the pamphlets written at the time (sources page), of the violence of the
flood. Whether this is partly exaggeration, or might be true also for a storm surge that overflowed the sea walls is a matter of debate.
The idea of a tsunami is not new. In Archaeologia Cambrensis vol 13 for 1913, there is an account of a field trip by the Cambrian Archaeological
Association to St Brides Wentloog. After mentioning the inscription in the church giving the date of the flood, it states that "This is a record of the
huge tidal wave, caused by some seismic upheaval which swept the whole Wentloog Level." Unfortunately they do not give any justification for why they
believe it was a tidal wave.
There was an earthquake in May of 1607. This is mentioned in Walter Yonge's Diary as happening around the 12th of May. The Lost Chronicle of Barnstaple describes a small earthquake on 19th May.
This was too late to have caused the January tsunami, but might be an aftershock of an earlier quake. However, it is odd that this minor quake should get into the written records, yet one large enough to cause
a tidal wave is not mentioned.