“VC Attack”, by Graham Wragg, illustrates Flight Lieutenant David Hornell’s valiant attack on a U-Boat during the Battle of the Atlantic.
THE LAST NORTH ATLANTIC CONVOY BATTLES
In the North Atlantic the last great convoy battle was fought by three groups of U-boats, appropriately named Stürmer, Dränger, and Raubgraf. They formed a combined force of thirty-eight U-boats to attack two convoys, SC122 and HX229, from 16 March. At that time British cipher officers had been unable to decode the enemy’s signals, but in sharp contrast Germany was using British signals from either the Admiralty or the C-in-C Western Approaches that gave the enemy the vital details of not only the position of convoys, but also their courses across the Atlantic. Thus the enemy was able to deploy its U-boats accordingly.
It was not until air cover from Coastal Command appeared on 17 March that the U-boats were restricted, ‘which allowed only five boats to make submerged attacks’, and ‘ever-increasing air defence caused us to lose contact with both convoys’. For Admiral Dönitz there had been no air cover on the night of 16/17 March, ‘and the U-boats took full advantage of the fact’. Of the signals sent by the British Admiralty and the C-in-C Western Approaches, Dönitz comments: ‘These signals were most useful to U-boat Command.’ Of air cover then available, from the limited successes achieved by Coastal Command, Sunderlands and Fortresses of No. 18 Group were deployed, damaging three U-boats, which were not identified.
Although at the Washington Conference in March it had been decided to reorganise cover for convoys from Newfoundland, this had not taken effect at the time of the sailings of HX229 and SC122. Thus the RCAF patrols at that time by No. 5 Squadron from Gander were short by 60 miles from the U-boats’ first attack.
It was apparent that from both Coastal’s bases in the east and from Gander in the west long-range Liberator aircraft were still lacking. As the official Canadian Air Force historian has stated: ‘Even one VLR Liberator, able to extend that patrol by another 200 miles, might have made a crucial difference.’
Three causes may be attributed to the serious losses in those two Atlantic convoys, therefore: the enemy’s use of the Admiralty or C-in-C North Western Approaches signals giving positions and courses of the convoys, the lack of deployment of VLR aircraft on both sides of the Atlantic for convoy protection and the inability at that time of the Allies to decode the enemy’s signals. Former Coastal Command aircrew, who were forbidden to transmit by W/T when near convoys, might well ask, ‘Why was the Admiralty so blasé in transmitting courses and positions of our convoys?’
Various accounts differ in respect of the number of ships sunk in the U-boat attacks on Convoys SC122 and HX229, but the naval historian gives the figure of twenty-one merchant vessels for one U-boat sunk. This battle demonstrated the need for cover by aircraft rather than escort vessels, as only one ship was sunk when once air cover was maintained. These serious losses did not gain the publicity given to the ‘Channel Dash’, although there was infinitely more need. The movement of the three warships was not life threatening to Britain, but the rate of loss of merchant vessels in March was so. Sir John Slessor gives the losses in that month alone as 620,000 GRT. As he rather understates, ‘It clearly could not go on.’
What followed in the following two months in the Battle of the Atlantic resulted in the defeat of the German U-boats, which was to be acknowledged by Admiral Dönitz on 24 May 1943. During the period 26 April to 23 May fourteen groups of U-boats were involved in attacks at various times against four Atlantic convoys.
The most significant fact about those convoy battles was that they were in the area south of Cape Farewell, and thus within what had been the ‘Mid-Atlantic Gap’, which had been a ‘happy hunting ground’ for U-boats. Now it was to become a grave-yard for U-boats rather than merchant ships.
Between 3 and 6 May six U-boats were sunk directly south of Cape Farewell in operations against ONS5; two in attacks against HX237; two more were sunk during attacks on SC129; four were sunk in operations between 15 and 20 May against SC130; and two between 21 and 23 May against HX239. The reasons for the Allies’ success were greatly improved provision of surface escorts, including aircraft-carriers, and improved air escorts.
The American surface escorts had HF/DF (high-frequency direction finding), with their aerials located higher up than was the case with the Royal Navy’s escorts, and were able to detect U-boats on the surface and before the U-boats had seen such vessels.
The Royal Navy, with aerials set lower on their vessels, lost some of their advantage. Allied Intelligence had enabled the convoys to be aware of the U-boats, and air escorts were thus laid on accordingly. It was Air Marshal Slessor’s policy to be fully prepared to deploy maximum effort in support of threatened convoys, and such was to prevail at this time.
Of the four U-boats sunk in attacks against SC130, three were claimed by Coastal Command’s aircraft–two by Liberators, and one by a Hudson–that operated from Iceland as far west as 25º48’W. Convoy SC130 was considered by the naval staff to be the last convoy to be seriously menaced, and Roskill stated: ‘This fine achievement was largely due to the almost continuous presence of air cover during the time when the convoy was being threatened.’
Thus was Air Marshal Slessor’s policy justified. The German Commander Hessler’s view was: ‘The most surprising feature of the enemy’s success was that according to our radio intelligence there were never more than one or two aircraft in the air at the same time.’
Admiral Dönitz’s opinion was that the success of the Allies against the convoys was due to the escorts working in ‘exemplary harmony’ with support groups that were specially trained, but supported by ‘continuous air cover’ with aircraft equipped largely with the latest radar, although the Hudson, which sank U-273, had sighted it through a gap in the clouds from 8 miles. It had used baiting tactics, which appear as a feature of No. 269 Squadron’s anti-U-boat successes. Aircraft equipped with radar had for Dönitz ‘robbed the U-boats of their power to fight on the surface’.
After the attacks against those four convoys, during which sixteen U-boats had been sunk and others seriously damaged, Dönitz considered that Germany had lost the U-boat battle and ordered those in the North Atlantic on 24 May 1943 to sail with caution to the Azores area.
1943 – BATTLE OF THE ATLANTIC WON!
The Battle of the Atlantic against U-boats was won in 1943; the Allies had gained the initiative. This was due to a number of factors: politically there was agreement with both Canada and USA for a North Atlantic convoy system, with cooperation between St John’s, Newfoundland, and Liverpool, but with the concurrence of ‘Cominch’ in the USA, VLR aircraft were to operate from Newfoundland in support of North Atlantic convoys. This resulted in the closing of the Mid-Atlantic Gap south of Cape Farewell. Later in the year, the Azores Gap was closed by Portugal acceding to the Allies having bases in the Azores.
Long-range aircraft, notably the Liberator, became available to the Command, which had begun the year with only half a squadron of those aircraft, but by the end of the year had seven squadrons so equipped. There was increasing availability of aircraft equipped with Leigh Lights, and the Mark III radar (9.1 cm wavelength), which initially could not be detected by U-boats, was coming into service.
While Coastal Command was gaining more and more aircrew with experience in antisubmarine warfare, and with increasingly high morale, Germany was losing its more experienced U-boat commanders, and its crews were being demoralised.
In 1943 Coastal Command had effective weapons, which were lethal to U-boats–the 250 lb Torpex depth charge, fitted with a shallow-depth pistol, the 35 lb rocket projectiles, and the acoustic homing torpedo.
Strategy and tactics had been developed, with the C-in-C Coastal Command opting for the offensive in two areas–the Bay of Biscay against U-boats in transit (Slessor’s ‘trunk of the tree’), and near convoys that were threatened with attacks. Tactics employed by Coastal Command in the Bay of Biscay were on one or two occasions to ‘swamp’ a given area with radar transmissions. Tactics that were modified were to opt for a 100 ft spacing of sticks of depth charges, to be released along the track of the U-boat, thereby increasing the prospect of a kill by 35 per cent. Aircrew, on encountering severe anti-aircraft fire, would circle and await assistance from at least one other aircraft, or possibly from surface craft.
The armistice with Italy in 1943 did not reduce the requirements of the Command, rather did it mean that some squadrons, such as Nos 500 and 608, remained in the Middle East area in support of invasion operations. Still lacking in 1943 for Coastal Command were a low-level bombsight and radio altimeters. There was still a need for closer cooperation between the Admiralty and the Command in respect of keeping Coastal Command informed about the movement of convoys and the Navy’s vessels, and also to provide hunter/killer groups of such vessels as sloops to operate with aircraft.
A weakness shown in both the German U-boat Command and the Admiralty was the lack of security in transmitting by W/T signals, in the Admiralty’s case by giving courses and positions of convoys, albeit in code.
For his period of command, Sir Philip Joubert gives, in respect of the anti-U-boat campaign, 825 sightings, which resulted in 607 attacks but only twenty-seven U-boats sunk, plus three shared. Additionally, 120 were so damaged that they had to return to base. Against those figures Coastal Command lost 233/4 aircraft, of which 116 were due to hazards such as the weather. Out of the 233 losses, 179 were from No. 19 Group and could be attributed to operating over the Bay of Biscay and encountering many enemy fighters, in addition to heavily armed U-boats.
The comparatively few sinkings in relation to sightings could be attributed to lack of suitable weapons, but also lack of suitable aircraft. To a lesser extent, inexperienced crews were operating in the earlier period of Joubert’s tenure. The part played in the defeat of the U-boats in the Battle of the Atlantic by aircraft under Coastal Command’s control amounted to ninety U-boats sunk, and fifty-one damaged.
Of the ninety that were sunk, fifty-three had been in the area of convoys; only one ship was sunk in a convoy when an aircraft was in close support. Overall in 1943, Germany lost 258 U-boats due to all causes. As the American Air Force decided after a survey in 1943, ‘Attack from the air against the U-boat at sea had been the most effective single factor in reducing the German submarine fleet.’
In 1944 there were three major operations in Western Europe. They were the Allied landings at Anzio in January, Operation Overlord (the Normandy landings in June), and the last German counter-attack during December in the Ardennes. A comparatively minor operation by the Allies was ‘Anvil/Dragoon’–landings in southern France in August.
Operation Overlord was to be uppermost in the minds of all Allied services from January onwards. In that month General Eisenhower was appointed Supreme Commander of the Allied Expeditionary Forces, with headquarters in England, whence the Allied invasion of Europe would be launched. This was to have an immediate bearing on Coastal Command, as, although victory in the Battle of the Atlantic had been achieved in 1943, U-boats still posed a threat, particularly against a vast armada of shipping required in taking troops and equipment to the Continent.
Air Chief Marshal Sir Sholto Douglas succeeded Air Marshal Slessor as AOC-in-C Coastal Command in January, and summed up his anti-submarine forces as 430 aircraft, including ten squadrons of Liberators (including three from the USN), and five squadrons of Leigh-Light Wellingtons.
There were also two squadrons each of Halifaxes, Hudsons, Fortresses and Catalinas, and seven squadrons of Sunderlands. Most notable was the improvement in the availability of Liberators, as in 1943 Air Marshal Slessor had begun his tenure with only half a Liberator squadron. There was also an improvement by now, not only in having the ‘right type’ of aircraft, but also having effective weapons generally available against U-boats, including the 250 lb Torpex depth charges with shallow-depth pistols, and rocket projectiles. The improved radar (Mark III) was also becoming more generally available.
There were still sixty U-boats in the Atlantic well able to pose a serious threat to Allied shipping. U-boats in 1944 were better armed against air attacks, with 37 mm and 20 mm cannon. When on the surface they could use improved detection devices against Coastal Command’s radar such as the Mark IIIs. There were an increasing number of U-boats equipped with Schnorchels, which comprised two tubes extending above the surface to allow exhaust gases from the diesels to escape, with the second tube taking in fresh air. Dönitz had looked upon the Schnorchel-equipped type as the ‘one hundred per cent underwater boat’, but it was to be proved otherwise.
Although, as Dönitz had hoped, it enabled U-boats to operate in the English Channel, a great strain was imposed on the U-boat crews due to their remaining submerged for so long in hostile waters. In rough sea conditions the Schnorchel was automatically closed, causing at least severe discomfort to the crews. Furthermore, Coastal Command aircrew were able to detect the Schnorchel boats with their Mark III radar, and also visually, not infrequently reporting seeing ‘white smoke’ on the surface of the sea. One such sighting, made from 1,000 ft altitude, included not only mention of the white smoke, but that it was emitted from a grey object 1 ft in diameter, projecting about 2 ft and moving at 12–15 knots. In that specific example the U-boat was sunk with cooperation from a naval escort group. There were, however, two other natural forms of disturbance with which Schnorchels might be confused–waterspouts and spouting whales.
The Command’s response to Schnorchel U-boats was to operate what was coded ‘High Tea’. This was to release a pattern of sonobuoys in the area of a suspected submerged U-boat. Sonobuoys could be thought of as the air equivalent of the Navy’s Asdic: that was, they could detect sound waves, with signals sent to the aircraft by electromagnetic means. Against such devices, Schnorchel U-boats were at a disadvantage, as the noise of their diesels would mask the sonobuoys’ signals, but the additional noise from the U-boat would aid detection by the aircraft.
It was the strategy of Dönitz always to deploy his U-boats where they were likely to sink ships with the minimum danger to themselves, and in January thirty-three U-boats were sent to operate against Russian convoys, where in those Arctic waters they could gain the support of the Luftwaffe operating from Norway.
Germany had been losing 20 per cent of its U-boats every month, and of 70 per cent that returned from operations, some were seriously damaged. While in transit through the Bay of Biscay, U-boats were remaining submerged for as long as twenty to twenty-two hours per day and were taking twelve days in transit. Dönitz had exhorted his crews to accept the harsh conditions as their only option against the Allies’ superior forces in both aircraft and escort groups, and as an ex-U-boat captain, Commander Hessler, stated, it was a matter of ‘containing the enemy’, and ‘a more cautious strategy was required’.
In the early months of 1944, although U-boats had ceased making pack attacks on shipping, there was some concentration of U-boats west of Scotland and Ireland. Furthermore, in the Bay of Biscay, fewer sightings were being made due to the U-boats remaining submerged, and spending a minimum time surfaced to recharge their batteries. Sir Sholto Douglas therefore reduced the density of his patrols in the Bay of Biscay to concentrate his forces off the west coasts of Scotland and Ireland. For this purpose, the operations of his Groups Nos 15, 18, 19 and Iceland were coordinated under No. 15 Group at ACHQ Liverpool. It was a logical decision, as Allied convoys of necessity converged in that area. There was the advantage, also, that his medium-range aircraft such as Hudsons and Wellingtons could effectively operate to a considerable distance into the Atlantic from both Iceland and the north-western bases that by then were available.
The success of this strategy could not be assessed solely on the number of sightings and ‘kills’ made, however, as just to keep the U-boats submerged could be rated a success. In fact, it resulted in U-boats being dispersed away from the shores of the United Kingdom, ‘beyond 15ºW where conditions were somewhat better’.
The Command’s aircrew came to accept that flying for hours over the Atlantic, now sighting convoys rather than empty lifeboats, was more than justified. The last attack on a North Atlantic convoy was in February. It is notable not just for being the last attack, but for the contrast in points of view by German and British historians. Hessler dismisses it in one short paragraph as the loss of two U-boats in the period 16–19 January. Captain Roskill gives a chart for 29 January to 24 February and states that eleven U-boats were sunk, six by escort groups, against the losses of one straggler and one sloop. Within that period, two U-boats were sunk by Coastal’s Wellingtons.
Features of note concerning one of those sinkings were that earlier, on 7 February, one of the Wellingtons had obtained a radar contact at 9 miles that was not lost until the ¼-mile range of the scanner had been reached. This showed what a great improvement in radar had occurred. With the earlier Mark I, a U-boat might well have been missed, or lost in ‘sea returns’ well before a ?-mile range. The attack on the U-boat that followed did not result in a kill, but later the same Wellington captain was successful. Many operational records indicate that when once a captain has made an actual attack (rather than a practice run), he was likely to be successful in subsequent attacks.
The Wellington successes were notable also in that they used six depth charges spaced at 60 ft rather than what had earlier been advocated–100 ft spacing. When 100 ft was specified, the Command, perhaps, was thinking in terms of four rather than six depth charges. This success in the final convoy battle demonstrated what had been advocated at the beginning of the war: that was that to combat the U-boat menace a combined effort of both surface vessels and aircraft was required. It had been shown earlier during July 1943 in the Bay of Biscay; but due then to the shortage of escort vessels, the system of combined operations had been discontinued.
At this stage, the German point of view became: ‘There was no point in disposing the U-boats primarily for attacks on convoys, and the sole remaining purpose was to discomfort the enemy, while at the same time providing … sufficient freedom to escape the concentrations of A/S forces.’ And the ‘Costly convoy attacks were therefore abandoned.’ Despite the freedom given to U-boat commanders, their losses ‘remained high’ and it was ‘believed’ that most were being lost in an area 700–800 miles south-west of Ireland. This indicated that U-boats could not have made W/T contact with the U-boat Command before being lost, and possibly that they were sunk in a surprise attack by Leigh-Light aircraft, which gave no opportunity for the U-boats to signal their headquarters. Because of such losses, U-boats were ordered on 27 March to avoid that area of the Atlantic.