The Mulberry harbours were floating artificial harbours designed and built to support the invasion of Europe by allied forces in 1944 after D Day.

Share on twitter
Share on facebook
Share on linkedin
Share on reddit
Share on print

Although there are many good examples of amphibious operations before WWII and in other WWII theatre’s one cannot look at the subject of over the shore logistics without considering the D Day Normandy landings of 1944 and beyond.

After Dunkirk, and with the Battle of Britain won, thinking turned to the return fixture. It was realised very soon that a logistics element of unparalleled proportions would be required. It was also a widely held assumption that the single most significant logistic challenge would be that of ports.

During the First World War, the then Major Bruce White (Royal Engineers) was responsible for port operations and the assembly of a secret military port alongside Richborough Castle in Kent. It was at Richborough that the first electric cranes were used and the development of RORO train ferries evolved.

Richborough Secret Port

The Royal Engineers had always had a responsibility for ports and inland waterways

From Winston Churchill;

Seamen go to sea in ships and it is their absence that landsman build harbours and refuges to which are brought back the fruits of their service

At the start of WWII, Bruce White (by then, Bruce White MBE) returned to service as a Brigadier, still in the Royal Engineers. He was appointed to the post of Director of Ports and Inland Water Transport. Well before D Day planning had commenced, port repair, creation and expansion was already problem being grappled with.

Because the Channel ports were under threat, damaged by German bombers or being heavily used by the Royal Navy, the country needed ports on the East of the country that could handle maritime traffic from the USA and Canada. One of Bruce White’s early projects in his new post was to oversee the design and construction of two dedicated military ports in Scotland. Military Port Number 1 was at Faslane and Military Port Number 2, Cairn Ryan.

Both were impressive constructions; Number 1 having six deepwater berths and Number 2, four. They had the full range of road and rail connectors, lighterage, material handling, storage and repair facilities. The port at Faslane is now of course HMNB Clyde, home to the Royal Navy’s nuclear submarines. 

He also ordered 360 harbour cranes from Messrs Stothert and Pitt of Bath(now Clarke Chapman), each prewired for use with a generator if mains power was not available or intermittent. These would go on to provide invaluable service in ports across the world, many still being in service today, such as those at Marchwood military port.

Port Repair Vessels were designed and built for the invasion of Europe in the expectation that repairing ports would be quicker than making them. 

The design of these vessels was subsequently used in many theatres, US forces copying the design. In order to provide an area for training and berthing the repair vessels two Port Repair Depots were established, one at the ‘secret port’ at Richborough and the other at Marchwood, near Southampton.

All this experience would prove to be invaluable and it is the bedrock on which the story of the Mulberry Harbour rests. There are many worthy claims for whose idea the Mulberry Harbour was, Churchill, proposed a seaborne invasion of the German islands of Borkum and Sylt using flat bottomed barges that would form wave barriers, and Operation Hush proposed flat bottomed invasion barges for an operation against Belgium.

In 1940 Lieutenant Colonel Wilson in the War Office remembered the Churchill idea and asked the engineer Guy Maunsell (of towers fame) whether it was feasible. In response, he produced some outline sketches but nothing further was forthcoming.

The Welsh civil engineer Hugh Iorys Hughes proposed a solution for piers in 1941 that would allow ships to discharge vehicles onto a shoreline. In 1941 the British Army set up a small team called ‘Transportation 5’ or Tn5 for short to look at the problem of port repair, Brigadier Bruce White MBE, Royal Engineers, in charge.

Immediately upon appointment, he set about ensuring that this domestic remit was expanded to include possible future invasions of mainland Europe because he correctly surmised that utilising existing ports would be vital.

The Royal Navy officer John Hughes -Hallet was also credited with suggesting the idea after the Dieppe raid but by then, the other various ideas were already in their early stages.

Brigadier White also realised that in order to repair damaged ports an accurate survey would be required and due to time constraints, as much information prior to the on-site survey would be needed. Making use of a loaned office in the Institute of Civil Engineers he assembled a team of translators, engineers and industry experts that would create a volume on every single port likely to be used in the forthcoming invasion.

What is certain is that over a period of time, these different strands and collections of engineers would eventually coalesce in one form or another to produce Mulberry although it is arguable that Guy Maunsell’s contribution was not fully recognised.

The US declared war on Germany on December 11 1941 and Russia was impatient for a second front against Germany but with painful memories of wasteful attrition warfare and failed amphibious operations from WWI, the UK was reluctant to launch ill-conceived raids without thorough preparation, adequate logistics and overwhelming firepower.

Combined Operations HQ had since the mid-thirties been researching amphibious raiding and under Admiral Sir Roger Keyes launched a number of small scale raids to ‘take the fight to the enemy’. The experiential base was being built in anticipation of future larger-scale operations. At the end of 1941 Commodore Lord Louis Mountbatten was placed in charge of Combined Operations.

[the_ad id=”53392″]

Piers for Use on Beaches

Initial experiments focussed on means of connecting ships to the shore such that they could unload vehicles, wave attenuation would come later.

Mountbatten was an advocate of the artificial harbour concept that had been bouncing around Combined Operations HQ that would allow the allies to attack where the Germans were weak but it was not a universally accepted position.

Work had continued in the UK through 1942 with various concepts and a progress update for Churchill prompted his famously impatient ‘Piers for use on Beaches’ memo.

Piers for use on Beaches

It is often thought that this memo from Churchill was the start of the process but it was not, conceptual work was already well underway but somewhat rudderless.

Churchill provided the vital impetus.

It was after this that the War Office Tn5 group was given a broader scope that included the development of piers for use on beaches, rather than simply means of repairing and expanding additional ports although as described above. Although it must be noted, Brigadier White was already doing that in any case. Tn5 cast the net wide and set out to investigate three concepts, testing each before making decisions on moving forward.

Each of the concepts for connecting ships to shore for the purpose of rapid unloading of vehicles and stores; one each from the Admiralty, Tn5 and Hugh Iorys Hughes.

Before meaningful testing could be carried out, a suitable location was needed. Garlieston in Scotland was chosen after an exhaustive search and survey process because it matched the likely Normandy beach gradients (1 in 200), tidal range (30ft) and soil conditions, was sparsely populated and located away from major cities. Two beaches were used for the tests, Cairnhead and Rigg Bay.

There are a couple of interesting stories about Garlieston and the Mulberry harbour.

First was the requirement for military personnel to act as ‘test subjects, essentially, spare bodies to play the role of invading soldiers and help with installation. Units were requested to send their best men but many apparently took this to mean ‘troublemakers you want to get rid of’. 

Stationed at Glasserton House the locals nicknamed them ‘goons’ because of their generally rowdy behaviour, behaviour that included petty theft and nearly burning the house down. After the great storm though, the ‘goons’ would be instrumental in returning Mulberry B to operation.

The other story involves the efforts of an MI5 agent posing as a tramp who tried repeatedly to gain information about what was going on from the local residents and military personnel, none was forthcoming, typical of the time.

The disastrous August 1942 raid on Dieppe was a test of tactics and equipment at scale. The failure of this raid should be viewed in the context of it providing answers to a number of key questions those planning the main invasion had. It also emboldened Hitler’s position on the effectiveness of static defences as defined by the ‘Atlantic Wall’ against the opinions of his Generals who thought investing resources in a mobile reserve and counter-attacking force would be the better option.

After building prototypes, testing started in Garlieston.

First, was the Admiralty design called Swiss Roll that used a flexible roadway constructed of timber and canvas, secured using steel cables.

Swiss Roll

The tests showed that although the concept was sound, it could not support anything heavier than a lightly loaded small truck. It was discounted from further development although actually used very briefly for personnel offload during D Day at Mulberry B (file that one under ‘not a lot of people know that)

The ‘Hughes Pier’ used a series of concrete caissons called Hippos that would sit on the sea bed with rigid steel roadways attached, called Crocodiles.

Hippo and Crocodile - Mulberry Harbour

Hipo and Crocodile Drawing 07

Hipo and Crocodile Drawing 05

Hipo and Crocodile Drawing 04

Hipo and Crocodile Drawing 03

Hipo and Crocodile Drawing 02

Hipo and Crocodile Drawing 01

Tests revealed that the caissons would not form a sufficiently stable platform for the roadways, which suffered severe bending and distortion.

It too was eventually discounted.

The Design Matures

At a dinner at Chequers, Bruce White explained to Churchill how he could help with his ‘piers for use on beaches’ problem, recalling the particulars of a storm at Valparaiso harbour in 1924 during which a ship belonging to the same firm he worked for had survived when no other ship had simply because it was equipped with spud legs.

The ship’s master had seen the damage that the storm waters were doing to other ships so simply raised his above the waves on its spud legs.

The ship was the rock cutting dredger, the Derocheuse, built in 1888 by Lobnitz and Co in Renfrewshire.

Rock cutting dredge, DEROCHEUSE built by Lobnitz of Renfrew as used in the Suez Canal


The idea of using spud legs and a floating roadway to shore evolved over several months with the Lobnitz company of Renfrew playing a key role in the design of the pier head, with Major Allan Beckitt RE and WT Everall RE working on the roadway.

The pierhead was based on a Lobnitz dipper dredger called the AB95 Lucayan. The Lucayan used the same type of electrically driven spud legs as the Derocheuse, built in 1923.

It was not scrapped until 1980.

AB Lucayan

The pierhead was to be a relatively simple design, steel construction, 200 feet long by 60 foot wide with each corner having a 90-foot spud leg that could be raised and lowered by electric motors, thus raising or lowering the platform.

A pair of diesel generators provided power for the motors.

MOD 951

Because it rested on its spud legs, it would provide a stable platform for ships to unload onto to, in effect, a quayside.

Although it could be raised clear of the waterline and waves the normal mode of operation would see it partially submerged, the water providing support, reducing the load to be transmitted to the seabed by the legs.

Once a ship had unloaded its vehicles or stores at the pier head they would need to be transferred to shore by some form of flexible roadway or bridge span.

Mulberry Harbour Whale 02

The Tn5 concept, neatly summed up in the diagram above, used floating roadways attached to pierhead platforms that would rise with the tide and be supported either by buoyant pontoons when floating or the beach when not. Because the pier head and beach would be fixed in position the total length of the causeway would change between the high and low watermark.

The solution proposed was a telescoping span at the pier head end.

Bruce White continued to build his organisation, recruiting many specialist consulting engineers and technicians; the technical and industrial foundations of Mulberry were being built and built well.

One of the innovations was the use of two committees, a technical committee and a manufacturing committee, the former discussed and agreed on the designs and proposed a range of solutions to the manufacturing committee, who examined them in terms of materials and labour requirements.

The best technical solution would be no good if it could not be built in war-ravaged Great Britain.

If the pier head was relatively simple, the roadway was far from it.

They would be required to accommodate lateral as well as vertical movement in each span and as a whole. Torsional flexibility would also need to be considered and this would require the use of complex spherical bearings developed by William Everall and Allan Beckett in conjunction with the Military Experimental Engineering Establishment (MEXE).

Tn5 Design Bridge Span on Test 15

Even the tread pattern on the bridge decks was subject to much discussion and testing.

Tn5 Bridge Section

Supporting bridge spans would be a series of floating pontoons, each required to support a total weight of 56 tons each, including a 25-ton single load, usually a tank.

Mulberry Harbour Whale 01

The original concept used a design derived from a Thames barge and testing these initial concepts at Garlieston provided invaluable insight not only into their suitability and use but also their towing characteristics and general seaworthiness.

Tn5 Design 06

Tn5 Design 02

Tn5 Design 04

Tn5 Design Testing



After continuous testing the designs evolved.

To allow multiple pier heads to be connected together so that more than one ship could be accommodated and a ‘one-way’ configuration for vehicles an intermediate pontoon was designed that connected using one of the telescoping bridge spans.

A further addition allowed the unloading and loading of ramp equipped LST and LCT’s (x2) using a sloping deck pontoon called a buffer pontoon that was connected using a flexible hinge.

Spud Pierhead 02

Shock absorbing fenders and a ramp assembly would allow simultaneous unloading from the LST’s front and side ramps.

Spud Pierhead 03

A collapsible mooring dolphin was designed and built but not used.

Intermediate Pontoon

Intermediate Pontoon

LST Buffer Pontoon

Spud Pierhead 07

Connected Pierheads

Tn5 Design 14

Because of a shortage of steel, many roadway pontoons were made using Ferroconcrete, a technique pioneered by Messrs Wates at the Vickers Barrow in Furness shipyard.

Steel pontoons were still used and modified with simple spud legs for areas where rocks may be prominent.

Mulberry Harbour Beatle 02

Mulberry Harbour Beatle 01

The telescopic span was also modified to include an ‘erection tank’ that allowed faster coupling to the pier head.

Mulberry Harbour Whale 03

Tn5 Design 07

During trials it was found that correct mooring of the, by now called, Whale and Beetle components were vital and a specially developed anchor designed by Allan Beckett was used, the now-familiar kite anchor.

These anchors were placed and maintained by small low profile boats called shuttles or Surf Landing Under Girder (SLUG) boats.

The importance of SLUG boats was stressed many times, although there are scant references to them in much of the published history.

Wave attenuation, or the production of sheltered water, was a particularly difficult problem for the design team.

The committee worked on a number of options for the production of sheltered water, blockships, concrete caissons, air bubbles (called the Brasher System after the American engineer, Philip Brasher) and a partially submerged balloon called a ‘Lilo’, the latter concept being developed by the Admiralty.

The goal of ‘effective joint working’ was not yet resolved.

A minute of a committee meeting on the 18th August 1943 records;

It would appear that Lt. Commdr Steele’s main object in attended the Committee had been to acquire information for the Admiralty regarding the activities of this committee rather than for the purpose of sharing information with them

Work continued, the optimal design for the air bubble system used short lengths of pipe suspended from floats rather than anchored or fixed to the seabed. It was also found that having the perforations facing down produced the best effect. A great deal of experience and input came from the compressor company, Ingersoll Rand.

A letter to a member of the committee on 9th September 1943 records their confidence in one of their engineers that would be travelling to Garlieston;

As well as assisting with the layout of the individual compressors and main supply lines he will take charge of the complete compressor plant, supervise operation and maintenance during the period of your experiments. We feel sure Mr Ashby can relieve you of all all anxiety regarding the supply of air

Blockships offered the advantage of being self-propelled but they were no more effective than concrete caissons and in any event, would need tugs for placement. The concrete caissons would provide excellent wave attenuation but likely be difficult to tow. 

The air bubble system showed a great deal of promise but the amount of power needed to generate sufficient air volume was thought too considerable so it was dropped at the end of 1943.

The floating breakwater was original of an inflatable design from Robert Lochner, hundreds of experiments confirmed it was capable of reducing waves if used with a suitable concrete anchor.

Despite its promise, vulnerability to enemy gunfire or accidental damage resulting in further experiments with rigid construction.

Mulberry Harbour - Flexible Breakwater

Lilo became Bombardon, the cruciform steel shape.

Bombardon 01

Bombardon 02

Mulberry Harbour - Bombardon under test

The Admiralty thought the combination of blockships and Bombardon would be sufficient but the civil engineers on Tn5 thought otherwise, in fact, they were less convinced about Bombardon because of the significant quantity of steel required and blockships because of their unpredictability when sinking, and when sunk.

The Admiralty, on the other hand, thought the civil engineers on Tn5 simply could not appreciate the difficulty in towing large concrete caissons and that none would survive the journey.

The concrete caissons went through a number of design revisions but fundamentally, they were large concrete hollow forms that would be towed into position and flooded.

Phoenix Caisson 04

Resting on the seabed, their sheer mass would provide sheltered water for the pier heads.

MOD 86

The final design was a joint UK/USA one that would displace 6,000 tons and 200f feet long by 50 feet wide.

MOD 581

A shortage of materials meant compromises were made, reinforcement and wall thickness were often the subject of much discussion but the driving factor was their short intended lifespan and the simple fact that they had to be available to the required quantity in time for D Day, perfect was most definitely the enemy of good enough.

Phoenix Caisson 01

There were many arguments and disagreements between Tn5 and The Admiralty on the subject of providing sheltered water that would result in counterproductive delays and poor use of finite resources.

Ironically, it was Air Chief Marshal Sir Charles Porter that provided the view of risk v resource that helped settle the matter.

In the end, all three systems were used; concrete caissons (Phoenix), floating cruciform (Bombardon) and sunken blockships (Corncob).

Each system was used to their advantage, blockships for example, worked better in shallow water and were used for providing sheltered water for landing craft, the ‘Gooseberries’

93 Bombardons were made, consuming a total of 20,000 tons of steel. 147 Phoenix were required in 6 depth variations for D Day but as the Mulberry Harbours operation time was extended more were needed to double bank them, these latter types also had top covers to alleviate the problem of overtopping causing internal pressure to burst them.

Some of the PHOENIX caissons had anti-aircraft guns and barrage balloons, with the necessary crew quarters built into the structure.

At the end of August 1943, the Quebec conference was convened between the allies. During the voyage, Lord Mountbatten staged a makeshift demonstration of the effectiveness of breakwaters. Professor J.D. Bernal was one of Mountbatten’s scientific advisers and responsible for the lecture theatrics in one of the bathrooms. A fleet of paper ships were floated in of the large baths, providing the wave motion, using a broom, was Lt Cdr Grant.

Upon the command of ‘more waves,’ the broom was vigorously pumped up and down in order to simulate the English Channel swell.

The paper ships were all lost.

The exercise was then repeated but this time, the replacement ships were protected with an inflated Mae West life preserver, simulating the breakwaters. Of course, despite the efforts of Lt Cdr Grant, no ships were lost.

Everyone was thus convinced.

The complete concept was duly presented at the Quebec Conference in 1943.

Winston Churchill requested that Brigadier White and a small team join the conference and after flying over in a bomber, the discussions had a greater level of detail than originally planned.

Half a dozen meetings later, the decision was made to produce two complete Mulberry harbours, A (American) and B (British).

Mountbatten soon realised the enormity of the task would be beyond Combined Operations and so he reorganised the project with more involvement from the War Office.

After returning from Quebec, Brigadier White found a memo on his desk titled ‘Artificial Harbours’ describing the decision from the conference, he was horrified at the complete lack of operational security and immediately went to the War Office, requesting a meeting with the head of security to report the breach.

The ‘big book of codenames’ that was used for assigning operation names was consulted.

Next available from the list was Mulberry, and so, the Mulberry system of artificial harbours and landing piers was officially born.

Between the end of 1942 and the end of 1943, most of the effort was design and testing, perfecting the final system to be produced.

Upon the conclusion of this phase, the Allies had agreed on the design of the four key elements of Mulberry;

  • The means of producing sheltered water (blockships, floating breakwaters and concrete caissons)
  • A pierhead that allowed a number of different ship types to quickly offload
  • A pier that could accommodate the movement of wind, wave and tide
  • A connector from the pier to the beach (this was vital, but like the SLUG boats, not often heard about)

The objective offload for Mulberry A at Omaha Beach was 5,000 tons per day and Mulberry B at Arromanches, 7,000 tons per day.

Production of the Mulberry Harbour

Production started in December 1943, 6 months to D Day.

23 pier heads were ordered including 14 buffer and 24 intermediate pontoons. 10 miles of floating roadway (Whales) would need 120 eighty-foot spans and 670 floating pontoons (Beetles).

Hundreds of manufacturers and tens of thousands of personnel were involved in the immense production engineering challenge and it was far from plain sailing.

Despite continued protests from Bruce White, the Admiralty continued to insist on Bombardon. Building Bombardon caused a number of unwanted secondary effects.

First, the use of dry docks for Bombardon meant they were unable to be utilised for Phoenix construction, the King George V dry dock in Southampton could be used to build 8 Phoenix in one go for example.

Bombardons under construction King George V Southampton

This meant Phoenix were often constructed in unsuitable locations which caused a number of losses.




Phoenix caisson under construction 07


Second, the amount of steel needed meant an increasing number of Beetle pontoons had to be constructed on concrete, an unsatisfactory compromise.

Beetle assembly at Marchwood 01

Beetle construction at Wates Engineering

Whale Build

Mulberry Whale spans at Marchwood

Whale spans at Marchwood under construction 01

Whale spans at Marchwood under construction 02

D Day Plus

Once construction had concluded the myriad components would need to be assembled and sailed across the English Channel to Normandy, both significant feats of organisation, skill and seamanship in their own right.

Getting Ready to Cross the Channel

The diagram below shows the assembly sites and routes.

Mulberry Harbour Assembly

The assembly task required many Phoenix caissons to be re-floated because they were sunk in various locations for concealment purposes.

Mulberry Harbour Towing 03

Mulberry Harbour Towing 04

A 24084

The Whale roadways were assembled, complete with their Beetle floatation pontoons, and towed across the channel as complete strings.


Mulberry Harbour Assembly 04

H 39297

A 24116

Bombardons were usually towed in pairs.


A force of 132 tugs was used to tow the various parts of the two Mulberry harbours across the English Channel.


H 39295

Some of these components were lost to weather and enemy action but for the most part, arrived as planned.

Operational security was paramount, the captains of the blockships were told they were going to the Bay of Biscay and a model of the Mulberry harbour and invasion beaches in the headquarters of the Automobile Association at Fanum House was made by toymakers who were confined to the building until well after the invasion.

The construction plan was equally complex, accounting for combat and weather losses, and coordinating with available labour and overall invasion sequence.

The Mulberry construction teams in the UK and US were given the go-ahead to sail on the afternoon of the 6th of June, D Day.

First to go were the Mulberry B team, the British No1 Port Construction and Repair Group. Following the relatively light resistance at Arromanches, they started work quickly.

Mulberry A Plan

The US forces at Omaha were much less fortunate and faced both an array of difficult obstacles and a reinforced defending force but nevertheless, both build teams got cracking, making use of mobile command posts such as the paddle steamer HMS Aristocrat.

Mulberry B Layout Diagram 02

Assembly off the Normandy Coast

The first task for the constructions teams was to confirm the initial findings of the covert surveys conducted prior to D Day and then to carry out more extensive surveys for the caissons and pier heads.

Positions were then marked using buoys in readiness for the construction to begin proper.

With surveys complete, the first component to arrive were the blockships, scuttled in overlapping patterns to avoid wave penetration and excessive scour. By the 13th Gooseberry 1 at Utah was complete and as a shelter for small to medium-sized ships allowed dry-shod unloading of a large number of personnel and material using pontoon causeways before the pier heads were established.

However, against British advice, the US construction team allowed gaps between the blockships to facilitate small craft movement, this would later prove to be a serious mistake. The second Gooseberry was completed during the same week, also at Omaha.

Mulberry harbour Corncob Blockship 05

Mulberry harbour Corncob Blockship 07

Gooseberry 3 at Arromanches, 4 at Courselle (Juno) and 5 at Ouistreham (Sword) were then completed in short order.

Mulberry harbour Corncob Blockship 06

By the 17th of June, both Bombardon strings were in place.

Mulberry Harbour - Bombardon at St Laurent


Mulberry Harbour Installation 15 (Bombardon)

Phoenix towing started on the 10th of June, by the 18th, 75 were in position, planted exactly according to plan except one that swung out of position as a result of a collision between it and a tug, in the dark.

Mulberry Harbour Installation 02 (Phoenix Caisson )

Mulberry Harbour Installation 01 (Phoenix Caisson)

Mulberry Harbour Phoenix Caisson 02

Mulberry Harbour Phoenix Caisson 03

A 24083

B 5726

Pier heads and Whale roadways started to arrive on the 9th of June and by the 14th, vehicles were rolling off ships, onto the pier head and down the roadway towards the beach.

Mulberry Harbour Installation 06 (Spud Pierhead)

B 5699

Mulberry Harbour Whale Roadway 02

Mulberry Harbour Whale Roadway 04

B 7240

Mulberry Harbour Whale Roadway 30 (3)

Mulberry Harbour Installation 10 (Spud Pierhead)






Mulberry Harbour Spud Pierhead 08 Crusader Tank

Mulberry Harbour Spud Pierhead 11

The build operations were not without their fair share of problems; failures and late arrivals competed with unexpected seabed conditions and poor weather to slow things down.




Mulberry Harbour Spud Pierhead 01-6

The Great Storm and its Aftermath

Between the 19th and 22nd of June, a storm of unprecedented magnitude hit the invasion beaches.

The forecasters did provide some warning that allowed tugs to be stocked with rations, pier heads raised, moorings checked and doubled and all other ships ordered back to England.

The tugs were also armed with PIAT’s and ordered to sink any Bombardon or ship that became a hazard to the Mulberry harbours as it was thought this might be likely.

The storm was actually the worst for 80 years, comparable to that which scattered the Spanish Armada several hundred years earlier.

Damage was considerable.

Mulberry Harbour A Storm 02

Mulberry Harbour A Storm 04

Mulberry Harbour A Storm 10

Mulberry Harbour A Storm 05


Many of the Phoenix caissons were overtopped; not a problem in itself but when the tide subsided the internal water pressure caused five of them to burst and one was damaged by scouring.

The Bombardons fared much worse as the weather conditions exceeded their design specification, every single one broke loose. Many hundreds of small craft were tossed around like matchwood and some of the piers were damaged.

Mulberry A was more exposed than B and therefore suffered a great deal more damage.

Pier heads and roadways were smashed. 21 of the 30 Phoenix caissons were destroyed by a combination of severe scouring, internal water pressure and being battered by free-floating Bombardon.

The lack of attention to the correct mooring procedures was also thought to be a contributing factor to the damage on the roadways but much of this was also caused by free-floating landing craft, 5 of which were actually British. Out of 650 LCT’s in the area, only 330 survived although the DUKW’s were able to ride out the storm by the simple expedient of driving onto the beach.

Damaged ships were cut open to get at their cargo and ships were deliberately beached regardless of their ability to be re-floated, desperate times needed prompt action.

Mulberry Harbour A Storm 07

Mulberry Harbour after the Storm 09

B 6060

Mulberry Harbour after the Storm 17 (A)

Mulberry B Damage

A 24374

Mulberry Harbour after the Storm 04

Mulberry Harbour after the Storm 05

The construction force and a number of US officers with relevant experience of salvage and Mulberry construction thought it could be repaired but more senior officers differed.

The decision was made to salvage what could be used on Mulberry B to aid in its reinstatement.

Mulberry Harbour B (Arromanches) 03

The Gooseberries at Mulberry A were to be reinforced and the Phoenix caissons at Mulberry B doubled up so they could continue to operate until the Winter, much longer than anticipated. The caissons were also filled with dredged sand, covered with steel sheeting, gaps filled with rock-filled steel mesh gabions.

Phoenix upgrade

At Omaha, the Gooseberries were reinforced and this allowed US forces to actually exceed the intended offload rate compared to the original plan.

By D + 30 it was operating at an average of 9,000 tons per day throughput.

Because the piers at Omaha were destroyed LST’s had to be beached but favourable geology and construction work allowed them easier access than previously thought possible. The Rhino ferries, LST’s, landing craft and DUKW’s continued their relentless operations and at times exceeded the offload rate of Mulberry B.

Mulberry B was in almost constant use for 5 months and in excess of 2 million men, half a million vehicles and 4 million tons of supplies were landed.

Mulberry Harbour B (Arromanches) 01

As Winter approached the larger Liberty Ships were increasingly used instead of the smaller coasters and LST’.

These 7,000-ton ships were offloaded at Mulberry B until the 19th November 1944, when the artificial harbour at Arromanches was closed, its war over.

Post War

A number of Mulberry Phoenix caissons still exist today, the one off Shoeburyness an example of a Phoenix that broke away from its anchor during repositioning after construction. Several Beetles remain on the shore near Garlieston and Marchwood

Mulberry Beatles used as coastal storm defence at Hythe and Marchwood 01

Mulberry Beatles used as coastal storm defence at Hythe and Marchwood 02

wigtown bay garlietson

and another Phoenix pair, in Portland harbour.

Mulberry Phoenix Caisson Portland Today 03

Two Phoenix were unable to be re-floated and are a dive attraction off Selsey Bill. The Crocodile at Garlieston could be clearly seen until 2006 when a storm destroyed it. A number of Whale roadways were used as bridges in Northern France, some still in use.

Mulberry Harbour Whale used as a bridge 01

Mulberry Harbour Whale used as a bridge 05

As part of the D-Day 70 commemoration activities, the UK Hydrographic Office conducted a very detailed survey of the remains of the Phoenix caissons and blockships at Arromanches, Mulberry B.

Mulberry Phoenix Caisson Arromanche Today 03

Mulberry Phoenix Caisson Arromanche Today 04

Mulberry Phoenix Caisson Arromanche Today 05

Mulberry Phoenix Caisson Arromanche Today 06

Mulberry Phoenix Caisson Arromanche Today 02

Mulberry Phoenix Caisson Arromanche Today 01

In 1953, the UK, Belgium and the Netherlands suffered at the hands of a severe flood (Watersnoodramp) caused by a heavy storm and high tides. Wikipedia has good background but to summarise, the devastation was massive. The UK death toll exceeded 300 but in the Netherlands, it was over 1,800.

The war had seen many of the dikes used for military fortifications and maintenance activity had slowed down or ceased completely. During the post-war rebuilding phase, many of the repairs were of the expedient type and it was later noted that some of these areas were the first to give way to the combined effect of storm and tide.

After the immediate rescue and recovery activities had been completed there were many gaps in the sea defences to close. Most of these were completed in a relatively short period but some of the larger and more complex gaps would need a great deal of heavy-duty construction. Compounding the reconstruction was the twice-daily tide and amount of damage.

Mulberry Harbour Phoenix Caisson used for Dutch Flood Defences 07

Mulberry Harbour Phoenix Caisson used for Dutch Flood Defences 06

Zuid Beveland. Sluiting Veerhaven Kruiningen met Phoenix caisson, type AX

Mulberry Harbour Phoenix Caisson used for Dutch Flood Defences 04

Mulberry Harbour Phoenix Caisson used for Dutch Flood Defences 01

The Allies had previously used surplus D Day Phoenix Mulberry Harbour caissons for a similar task in 1945 and 1946 on the island of Walcheren so the same technique was proposed. After extensive scale modelling, eight Phoenix caissons were floated over from the UK although some were lost in heavy seas during the journey. Over a period of several months, they were used to close the gaps in destroyed sea defences in a number of locations.

One of the more fascinating aspects of this little known story is that they are still there. The actual sea defences no longer rely on the caissons but the Dutch decided to turn them into a museum, a museum that commemorates the floods and those involved. On 6 November 2003, 50 years after the closing of the last breach at Ouwerkerk, the four caissons and the surrounding area were awarded National Monument status by the Minister of the Interior, Johan Remkes; and from that day were known as the National Monument Watersnood 1953.

NL-Ouwerkerk Schouwen-Duiveland luchtfoto caissons foto: Joop van Houdt



Since 2001 a museum has been sited in one of the caissons, click here to view the museum’s website.

Perhaps the most interesting post-war story is that of the pier head design. It was based on a dredging platform design and if it ain’t broke, don’t fix it. The Berghavn dredger was built in 1980 by the very same Lobnitz company that designed the Mulberry Pier heads and after a refurbishment in 1980 is still in service with the Norwegian company, Secora

Does it look familiar!



Perhaps the least discussed aspects of Mulberry was the counterproductive inter-service and international rivalry, hubris, resistance to change and a general ‘not invented here attitude that beset the project from start to finish.

The Admiralty insisted it knew about port construction (it didn’t), the War Office (Army) insisted it knew about maritime towing (it didn’t) and the US Navy insisted the whole scheme was a waste of time (it wasn’t.

This conspired to rob Mulberry of much of its potential.

It was only the inspirational leadership and dogged persistence of a collection of civilians, British Army, Royal Navy, US Army, US Coastguard and US Navy officers that overcame these difficulties to produce the end result.

At a technical and operational level collaboration and joint working were excellent but this was not always apparent at higher levels and the problems were compounded by labyrinthine command and control structures used by both forces. For a number of reasons, not least a general antipathy to the concept from the US Navy, the US construction teams did not have enough time for training and thus, did not follow the correct mooring procedures for the Whale/Beetles and placing of blockships.

They were advised by British personnel but the warnings were not heeded, and Mulberry A suffered because of it.

It has become somewhat fashionable to compare the post-storm offload using DUKW, Rhino ferries and LST at Omaha to Mulberry B and declare Mulberry a failure, but that misses several critical elements.

The Mulberry Harbour was a system, the Phoenix caissons and blockships provided the calm water that enabled the Rhino, DUKW and LST equipment to be used. Without these, only a tiny fraction of the offload rate could have been achieved, certainly not enough to sustain the invasion force which would have been severely hampered.

Phoenix and blockships were essential to both methods of offload.

This leaves the other components, namely Bombardon and the Pierhead/Whale offloading platform and roadway.

The Bombardons were the greatest failure, although, in their defence, they were operated in conditions far exceeding their known parameters. In hindsight, they were a complete waste of steel, manpower and building facilities that could have been used for other purposes such as shipping or in lieu of the concrete Beetles that proved so troublesome.

After the storm, they were not reinstated.

Which leaves the pier head and roadway open to question.

None of the Allies knew beforehand about the favourable soil conditions at Omaha that allowed heavy construction plant to cut beaching channels for the LST’s that allowed turnaround times to be reduced, a tactic that was simply impossible to predict possible until after the storm, so there may well be an element of being wise after the fact in praising the LST beaching method.

Tonnage offload rates are also rather a simplistic measure that does not consider low volume stores or the importance of one particular cargo or the other. Mulberry B would eventually be used for the vast majority of vehicular and bridging offload, for example.

However, a valid criticism of the pier head was that offload rates were not improved by employing experienced stevedores or employing modern electric cranes, despite being suggested by none other than Churchill himself.

The whole Mulberry system was designed around the British LST Mark 2 and 3.

The LST buffer pontoon was almost an afterthought but proved vital. They allowed the LST to approach, discharge, return to England and be halfway back to Normandy in the same time it would take one to the beach, discharge and then wait for the tide to re-float it. Although the US beaching of LST’s at Omaha would enable over the beach throughput rates to be maintained at comparable rates to the pier heads, it created problems elsewhere. 

The LST’s being used at Omaha were unable to be redeployed to other locations further up the coast and in the Pacific theatre as per the original plan.

Using LST’s longer than anticipated at Omaha, therefore, simply shifted problems elsewhere.

Cherbourg validated the Allies assumptions that port denial is both very easy to do and very hard to correct, there is no doubt an alternative was needed i.e. Mulberry

In general, Mulberry was a success that proved its worth, staying open longer and enabling greater offload rates than originally planned.

Mulberry is often seen as a wholly British affair but this is incorrect, US expertise played a key role at various stages, a joint success in the truest sense of the word, despite the numerous problems described above.

Perhaps the final word is best left to someone from the other team;

To construct our defences we had in two years used some 13 million cubic meters of concrete and 1.5 million tons of steel. A fortnight after the landings by the enemy, this costly effort was brought to nothing because of an idea of simple genius. As we know now, the invasion forces brought their own harbours, and built, at Arromanches and Omaha, on unprotected coast, the necessary landing ramps.
Albert Speer

Table of Contents

[adinserter block=”2″]

[adinserter block=”1″]

Change Status

Change Date Change Record
 23/07/2021 initial issue

Further Reading

Research on the storm and waves

Discover more from Think Defence

Subscribe to get the latest posts to your email.

Leave a Reply