Home About Membership Contact Despatches Supplies Forum Gallery News

Mulberries and Gooseberries

 temporary harbours at Overlord

by Chris Bridges




Following the evacuation of the British Expeditionary Force (BEF) from Dunkirk in 1940, the British began planning their return to continental Europe, seeing it as the only way to conclude the war with Germany.


The disastrous raid on Dieppe in 1942, proved to the Allies that a frontal assault on a heavily defended port/harbour was out of the question. The invasion, therefore, would have to come on the beaches. As we know, the invasion planners chose the Normandy coastline on which to make the landings.


For any invasion of Europe, however, a great deal of supplies would be required by the fighting divisions. The nearest port was Cherbourg which, if captured, was not expected to discharge more than 1300 tons per day, as it was predominantly a passenger port and not suitable for cargo. Even for the initial plan of a three-divisional assault a supply rate of 10,000 tons per day was required by D-Day plus 5. When the number of assault divisions were later increased to five, this supply problem was heightened.


The only way to supply the Allied armies until enough ports had been captured was, therefore, from the beach. This was risky, however, due to the unpredictable nature of the English Channel. Some form of protection was required, therefore, to save the smaller craft from sinking in rough weather, and to allow larger ships to carry on discharging.


Thus, the Mulberry Harbour was born. The actual proposer of the idea is disputed, but it is believed to be Commodore John Hughes-Hallett, naval commander of the Dieppe raid and Naval Chief of Staff to the Overlord planners. At a meeting following the Dieppe raid he declared that if a port could not be captured, then one should be taken across the Channel. Although this was met with derision at the time, the concept of the Mulberry Harbours began to take shape when Huges-Hallett moved on to the Overlord planning team.

The Development of the Harbour Components


The Piers


Due to nature of the shallow beaches within the invasion area, at low tide large stores ships could not get within a mile of the beach. To unload these ships, therefore, a pier was required. But unlike traditional piers it would have to float up and down with the tide as this could rise and fall up to 20 feet during spring tides (see Figure 1).  


Figure 1 - Floating Roadway at High and Low water


The specification for the pierheads were that they had to be capable of berthing three 2,000-ton coasters, whilst the piers themselves had to allow a continuos flow of traffic and would not be less than a mile long.


Whales: The piers (or whales) were actually bridge sections designed for trains (see Figure 2). These were 80 feet long and incredibly flexible allowing torsional displacement of 40° along the span and an angular movement between spans of 24°. Some spans were even telescopic to take into account of the: variation in the height of the tides; variation in the angle of the roadway at high and low tides (the bridge would be slightly longer at low-tide); movement of the bridge sections during heavy storms; and the variable distance between pierheads which were to be linked together.


Figure 2 - detail of Floating Roadway


Beetles: The floats, or pier pontoons, supporting the piers were called Beetles (see Figure 3). They were required to support 56 tons, which included a 25-ton tank. Due to a war time shortage of  steel (for the 460 floats needed some 5,000 tons of steel was required) reinforced concrete was used instead. Wooden fenders were added to the sides and bottom to protect the floats from damage.


Figure 3 - Concrete Float (beetle)

Figure 4 - Arrangement of Spud Pontoon


Spud Pontoons: The pierheads were pontoons designed with legs that would fix to the seabed and allow the pontoon to move up and down (see Figure 4). The pierhead could, therefore, be used at low and high tides with no problems. The legs were called spuds, hence the name spud pontoon. The pontoon itself was 200 feet long, 60 feet wide and 10 feet deep, and weighed over 1,100 tons. It could accommodate 21 men and had a kitchen, sanitation facilities and heating appliances. Types of pierheads can be seen on Figure 5.


Figure 5 - Arrangement of Stores Pierhead


The Breakwaters


To protect the ships unloading at the piers from the changeable channel weather a system of breakwaters were developed.


Bombardons: A number of ideas were considered for breakwaters. The first of these involved the use of compressed air. Perforated pipes were to be laid on the sea bottom around the area to be protected and compressed pumped through it. The result was a calm area within which ships were able to discharge at the piers. Although the principle was quite sound, the trials proved that the pipes would be moved in any heavy swell and that an enormous amount of power was required (at least 1.5 million horse-power for an area half that required).


Robert Lochner, who was in charge of this project, thought about what the were trying to achieve and realised that waves were in fact only surface deep and, therefore, only a shallow barrier was required. From experiments in his garden pond with a lilo, Lochner and his team went on to produce a floating, steel hollow cruciform “tube” which would form a suitable breakwater (code name Bombardon - to suggest a link with anti-aircraft defences) (see Figure 6).


Figure 6 - Design of Bombardoons


Pheonix: The primary breakwater was to be provided by concrete caissons. These were to be from 25 to 60 feet high and 174 to 204 feet long. The displacement was up to 6,000 tons.


Due to lack of space for mooring, the caissons had to be sunk until required. This was not taken into account during the design stage and a hurried conversion programme was implemented. Unfortunately, after sinking a number of caissons it was realised that the onboard pumps provided by the Royal Engineers and required to raise the Phoenixes were inadequate. The Royal Navy was, therefore, appointed to raise the Phoenixes as they had the right equipment and the salvage knowledge.


Gooseberries: With the increase in the number of assault divisions and beaches following the appointment of Eisenhower and Montgomery, the use of three beaches, as well as the Mulberries, were required to discharge supplies for the Allied armies.

As the Mulberries were located in the centre of the invasion area, small boats unloading at the beaches at the flanks would be unable to get to the Mulberries in the event of bad weather. Also, as over 4,000 small ships were expected to be used in this operation, the Mulberries just did not have the space to protect them all.


The solution was to provide five shelters (Gooseberries), one at each beach, formed by blockships (see Figure 7). These were later developed to provide a sort of landing ship service station, with facilities for repairs, refuelling, and first aid as well as accommodation.


Figure 7 - Plan of Typical Gooseberry


The Construction of the Harbours


On the afternoon of 6 June the Mulberry forces were given the order to sail (see Figure 8).


Arromanches, which was the location of Mulberry B, was not one of the assault beaches. The engineer vanguard was, therefore, landed to the east and had to make their way there by foot. Luckily no opposition was encountered.


Their US counterparts had a much harder time as the location of Mulberry A was Omaha beach where, apart from the fighting, there were a number of beach obstacles to clear.


The first task was to carry out a hydrographic survey of the beaches and mark the locations of the various elements. Next, on 7 June, the blockships for the Gooseberries were despatched. Apart from the problem of skippers of tugs releasing their tows too early, at Utah three blockships were sunk prematurely by the fire from ‘88s’. Two were luckily sunk in the right position.


At Omaha, the Americans decided to leave two gaps in the line of blockships attached to Mulberry A, to allow easier access to the beach. This was against the advice of the Royal Navy who warned that the sea would pour through in the event of a storm - as it did on 18 June.


All the Bombardons were in place by 17 June, but they did not start placing the Phoenixes until 10 June. The piers began being placed on 9 June.


It should be emphasised that all the components for the harbours were towed across the English Channel. This was not easy as one can imagine with around 40% of all the tows lost. The sea and weather being the main problem, but there was always the possibility of E-boat attack.


By 16 June the Germans had discovered the purpose of the Mulberries and began to launch air attacks. The biggest threat was, however, mines. These were laid by ship and often aircraft in such numbers that, due to the intensity of traffic, were bound to be set off sooner or later. Mine-spotting parties were, therefore, established.


The Storm


A storm of Force 6-7 raged through the English Channel on 19-22 June. On receiving the weather forecast the British took extreme precautions, even threatening to sink ships that were not unloading if they refused to leave the harbour. Ships or Bombardons that broke loose were to be sunk with PIATs if they were to threaten the integrity of the harbour. The effect on the British was five destroyed Phoenixes, damaged piers, the disintegration of the Bombardons, and about 250 small craft sunk or battered onto the shore.


The effect on the Americans was much worse due to: (i) the more exposed position of Mulberry A; (ii) their failure to construct their Gooseberries to design; (iii) their failure to construct the Mulberry to specification; and (iv) their failure to take adequate precautions on hearing the forecast of the storm. They lost their piers as well as twenty Phoenixes, seven blockships and over 100 small craft. Eisenhower ordered that Mulberry A should be abandoned and what was salvageable was used to reinforce Mulberry B.


The Gooseberries stood up rather better and survived the storm.  


Figure 8 - Assembly of Mulberry Components


Success or Failure?


There have been many arguments as to whether the Mulberry Harbours were worth the effort. Hastings(1984) is one of the more recent historians who criticise the decision to build the Mulberries, and their failure to provide the goods.


Apart from the fact that the harbours took up a considerable amount of the nations resources for their construction, there are considerable questions over its effectiveness, especially when compared to the rate of unloading on the beaches. The following figures are usually quoted to show that the Mulberries were not, in fact necessary:




(long tons)



British Beaches

(excl. Mulberry B)













The Discharge of Supplies, Vehicles and Troops, 6-19 June 1944 (Hartcup, 1977)


During this period, however, the Mulberries were still under construction with Mulberry B unable to take vehicles due to the LST pier being incomplete. During the first two days of the storm over 2,000 tons of ammunition was unloaded at Mulberry B whereas nothing could be unloaded on the beaches (Walter, 1986). This was at a crucial time in the battle for the Allies.


Over the long term Mulberry showed its true value. Up to the end of August, Mulberry B unloaded 40% of British stores. By the end of October, 25% of all the Allies supplies, 20% of personnel, and 15% of vehicles had past through the Mulberry B. The remainder of supplies etc. came through Port en Bessin, Ouistreham, later through the ports of the Pas de Calais, and of course, across the beaches.


Mulberry A and B were designed to facilitate the unloading of 12,000 tons of stores per day (Mulberry B to take 7,000 tons) (Mitchell, 1994). In the event, Mulberry B averaged 6,000 tons per day occasionally reaching 10,000 tons with 12,000 tons in reach if bottlenecks hadn’t occurred ashore (Harris, 1994).


With the Allies failure to capture intact ports within the first few months of the fighting, the Mulberry harbour must be regarded as a necessity. It undoubtedly helped the Allies survive over this period and must be regarded as an unqualified success.





Below is a list of various references that I have consulted.  A great deal of detail was obtained from Hartcup, but for a general overview of the Mulberry Harbours I recommend White, Mitchell or Harris.


Harris, A. The Mulberry Harbours. Royal Engineers Journal, Vol. 108, April 1994.


Hartcup, G. Code Name Mulberry: The Planning, Building and Operation of the Normandy Harbours. (David & Charles, Newton Abbot, 1977).


Hastings, M. Overlord: D-Day & the Battle for Normandy. (Simon & Schuster, New York, 1984).


Mitchell, H. The Mulberry Harbours - civil engineering’s greatest project. Civil Engineering Surveyor, June 1994.


Walter, A.E.W. A Harbour goes to France. Royal Engineers Journal, Vol. 100, March 1986.


White, B. The artificial invasion harbours called Mulberry. (Viking Group, England, 1980).


Winney, M. (Ed.) New Civil Engineer: The Mulberry Harbour 1944-1994. (Thomas Telford Ltd., London, 1994).


back to the contents page    back to world war II