Читать книгу The Combined Bomber Offensive 1943 - 1944: The Air Attack on Nazi Germany - L. Douglas Keeney - Страница 11

Presently the effectiveness of first-phase bombardment is to be examined from another point of view, but first one of the most important problems of first-phase operations deserves some attention. As a statement from an Eighth Air Force intelligence report indicated, the U. S. heavy day bomber posed a formidable problem to the German Air Force:

The first operations of Flying Fortress formations in high altitude daylight precision bombing attacks over Europe opened a new chapter in the oldest conflict in warfare. That conflict is the perpetual struggle between offensive and defensive technique. Our Fortresses, as used here, were a new offensive weapon.

Against them the Germans confidently pitted the three-fold mechanism of defense which aerial bombing has itself generated: Radar detection and counter attack by both flak and highly maneuverable heavily armed fighter planes. This defensive technique as developed by both Germans and British was believed to have precluded large scale daylight bombing in this theater forever.

But the Fortress brought to its work two unique features. The first was its ability to operate effectively above the maximum effectiveness of light flak. The second was its formidable armament of .30 caliber machine guns, which, compounded by the mutual support of close formation flying, provided withering air power against conventional fighter plane attack.

It is true that the countermeasures taken against the Fortresses and the Liberators threatened to restore the balance between the offense and the defense which the first operations had upset. The complete redress of the balance would have nullified the bomber offensive. Our air leaders were forced to great efforts in the defense of the day bombers to keep the offensive going. It is necessary to note some of these efforts as they developed during the first phase.

The simplest statement of the problem is presented in the figures for the heavy bombers. The operations of the last five months of 1942 cost 31 heavies missing in action, 26 attributed to enemy aircraft and 5 to flak or a combinations of flak and enemy aircraft; and the first quarter of 1943 saw 59 of our bombers lost on operations. During the first month of the period under review, April 1943, 20heavies were lost; in May there were 69, and in June 85, giving a total of 161 for this first phase of the CBO. Of these losses, 9 were due to the action of enemy aircraft, 98 to flak or a combination of flak and enemy aircraft, and the remainder to accidents and unknown causes.

The rate of loss is more revealing than the absolute loss. Taking these planes that were lost due to combat (lost in action or to operational salvage) as a percentage of planes attacking (completing effective sorties) we find the following:

Another side of the bomber loss problem is seen in the attrition rate. This considers all kinds of lost aircraft (operational and nonoperational) as a percentage of unit equipment. During 1942 the attrition rate for the B-17’s was 3.3% and for B-24’s was 2.1%. For the first quarter of 1943 for both types of heavy bombers it was above 7.2% That the attrition rate turned sharply upward for B-17’s is evident from the figures for April, May, and June. It is to be remembered that the Fortresses constituted the bulk of the bomber force during this period. Attrition as a percentage of unit equipment was:

Another measure of the severity of the opposition encountered by the day bombers is found in the amount of battle damage sustained by aircraft which were able to return from missions. During the 1942 operations, 27% of heavy bombers completing “credit sorties” (that is, sorties entering areas where enemy action was normally anticipated) were damaged in varying degrees of severity; in the first quarter of 1943 the rate was 31% and for April, May, and June it was approximately 30.7. That meant that three out of every 10 heavy bombers entering enemy territory were damaged.

Loss and damage figures can be interpreted in various ways. “The Statistical Summary of Eighth Air Force Operations in the European Theater, 17 August 1942-9 May 1945” and various studies by the Operational Research Section are excellent sources for these figures. The interest here is in that these date tell about the intensity of enemy opposition. The evidence indicates that the effectiveness of that opposition was increasing during the first phase of the bomber offensive.

It is evident also that the defensive problems presented by this severe opposition were recognized by AAF leaders and that energetic steps were taken to overcome the formidable efforts of the Germans to halt the day bomber offensive.

It had been noted that the three principal German defense were the radar installations for detecting the approach of hostile forces, the German fighter force with its elaborate control system, and antiaircraft artillery (flak). Since the principal defensive efforts of the Eighth during the first these were concerned with fighters and antiaircraft fire, those measures will be considered here. Radar countermeasures are treated in connection with the third phase when they were first used.

The plan for the bomber offensive which was drawn up in the European theater in April called attention to the growth of the German fighter force and the changes that had taken place in the composition of that force. The GAF was increasing its fighters at the expense of other types, and it was concentrating these on the Western Front at the expense of other theaters. Both these trends were to continue for many months.

The German fighter force was not only increasing its numbers and concentrating then against the bomber offensive; it was improving its standard pursuit tactics and even making some rather startling tactical innovations. Two of these innovations began to appear during the first half of 1943. The first was air-to-air bombing, both dive and level, by individual planes and in formation. Apparently the first use of this tactic against the Eighth’s heavy bombers was on 16 February 1943, in a mission against St. Nazaire. Reports of its use were made in March, and the four April heavy missions encountered it. Air-to-air bombing was met on at least nine missions in May and several times in June. This German measure was never successful because of the great difficulty of insuring the detonation of a projectile at just the right instant sufficiently near a rapidly moving target. Undoubtedly the reason for its use was that the size of the lethal burst of a bomb was very great – perhaps six times as large as that of the best flak.

The other new tactic of the German fighters which was initiated during the first phase was the employment of 21-cm. rockets by both FW-190’s and Me-109’s. Their use was first reported positively on 21 May 1943 during the raid on Wilhelmshaven, but there is some evidence that they were used earlier. The danger to the heavy bombers from the rocket projectiles lay in the fact that the rocket gun could be carried to 35,000 feet by either of the enemy’s two best single-engine fighters, and once at that altitude could fire an 80-pound shell very accurately at ranges well beyond those of the .50-caliber machine guns carried by U S. bombers. Another great danger from the rocket projectile lay in the fact that it had a lethal burst of 150 feet and its detonation could be well controlled.

The greatest danger from enemy fighters during the first half of 1943 was not due to these innovations but to improved pursuit tactics and that these fighters were armed with larger caliber guns. During 1943 there were 40 cannon hits for every 100 machine-gun hits; during the first half of 1943 there were 77 for every 100 machine gun hits. The hits by heavier guns helps to explain the increased loss rate of bombers to fighter action. The damage rate due to enemy fighters became particularly formidable during the first two months of expanded VIII Bomber Command operations (May-June 1943) than more than 15 of the bombers crossing into enemy territory were damaged by fighters. This was approximately one and one-quarter times the rate for all operations prior to May.

An indirect product of damage from enemy aircraft was self-inflicted damage to the heavy bombers. This was caused both by the guns on the damaged planes and by guns on friendly planes in the same or near-by formations. This category of battle damage became prominent first in May 1943. On missions 55 through 60, carried out between 13 and 21 May, 24 of 125 aircraft damaged by machine-gun fire were reported damaged by their own guns. Most of the self-inflicted battle damage came from empty shell cases falling from higher planes. Most of the self-inflicted machine-gun fire damage occurred to B-17’s and was caused by waist gunners firing into the horizontal stabilizers and elevators on their own planes.

The other German weapon for combating the day bomber was flak. Although the damage rate from flak was higher than from enemy fighters, the damage from the forger was generally less severe. It was often true, however, that flak damage was a contributing factor in many cases of aircraft reported lost to enemy aircraft, for an initial hit by antiaircraft frequently caused a bomber to straggle behind its formation and make it easy prey for fighters. All important German targets were heavily defended by flak; these included the submarine yards and airdromes that were the particular targets during the first phase of the offensive. In France, the most heavily defended areas were around Paris and St. Nazaire, both of which were attacked by the Eighth during the second quarter of 1943.

Defense of the day bomber involved both materiel and tactical developments. Two of the more outstanding materiel problems that demanded attention during the first phase were the modification of the heavy bomber to increase fire power and the development of escort for the bomber formations.

During the early months of the bomber offensive certain complaints were voiced against the armament of the heavy bombers. The most serious was that the forward fire power was entirely inadequate. B-17d’s and F’s came to the European theater with a single hand-held .39-caliber center nose gun or with no center nose gun at all. B-17E’s had no side nose guns and some F’s had two hand-held side nose guns. B-24D’s generally had one .50-caliber gun in the nose. Such armament was not a sufficient deterrent to the German fighter armed with 30-mm. cannon.

When Marshall Goering was interviewed as a prisoner of war, he said that tail attacks on the day bombers were preferred to head-on attacks because the closing-in speed involved in the latter allowed too short a firing time. Tail attacks were the rule during the period from August 1942 to November 1942, but head-on attacks predominated in December 1942, and during the whole first half of 1943. The evidence is quite clear, and it was recognized in the theater at the time.

Modifications were performed on the heavies throughout the first half of 1943 in order to increase their forward fire power. On the B-17’s either one or two .30-caliber guns with ring-and-post sights were installed in the center nose section. On B-24’s an additional .50-caliber nose gun was mounted.

These hand-held guns were the best answer the Eighth found to the problem during the first phase. They were recognized as a temporary measure, for the manufacturers of the B-17 were working on a power-driven chin turret for the aircraft. Turrets possessed several advantages over hand-held guns in field of fire, accuracy, and fire power per gunner. Since the chin turret was not expected until late in 1943, another temporary experiment in the form of a reworked Bendix lower turret was tried out during the first phase. An airplane with one of these experimental turrets arrived in the theater on 27 June and was given thorough tests. The preliminary results indicated some superiority over the hand-held guns, but combat tests proved the later better. The reworked turret would not react quickly enough on simultaneous attacks from several directions. Orders were given from Headquarters, AAF to stop installation of the makeshift lower turret and to continue with the development of the regular chin turret. Meanwhile it should be remembered that the chief dependence for forward fire cover by heavy bombers during the first phase was upon the hand-held .50caliber gun provides with direct sighting. Other armament modifications were performed on the heavier but those designed to increase forward fire power seen most significant.

Coupled closely with the effort to increase the firepower of the bomb-carrying heavies was the attempt to convert a member to an escort cruiser by substituting fire power and armor for bomb-carrying capacity. In the spring of 1942, Maj. Gen. Carl A. Spaatz had speculated on the possibilities of such an aircraft: “Auxiliary (expendable) tanks offer the only immediate solution for extending the range of fighters, unless it can be developed that the bomber, with its fire-mover, can substitute ammunition for bomb-load and act as an accompanying fighter. The escort cruiser which was developed – a heavily armed and armored B-17 – was designated the XB-40. This craft had the chin turret, upper turret, ball turret, direct-sighting power-boosted twin .50’s on each side of the waist, twin .50’s in a Martin electric upper turret in the radio compartment, and twin .50’s in a power-boosted tail gun mount with reflector sights. It carried 40,000 rounds of ammunition. General Eaker was quite interested in the development and in mid-March 1943 expressed a desire to try it out against the Germans, but at the same time voiced doubt about the economy of taking bombers that did not carry bombs on combat missions. In the operating groups there was more enthusiasm for the escort airplane, particularly for the chin turret which was one of its most promising features.

In mid-April, when the German fighters were concentrating on the lead formations of his bomber forces, the Eighth’s commander became very anxious to get some of the YB-40’s as quickly as possible. He was told that 13 YB-40’s were to be sent to the United Kingdom just as soon as the crews had completed their gunnery training.

These heavy cruisers arrived in the theater in the first part of May and most of them were assigned to the 92d Bomb Group (II). After additional training and modification they embarked on their first mission on 29 May, when seven were dispatched to St. Nazaire. The initial dictated the necessity for modification of waist and experience tail gun feeds and ammunition supplies, and indicated the basic defect of the craft, which was its inability to keep up with normal B-17’s – especially after the latter had dropped their bombs. Modifications were complete by 15 June, and YB-40’s took part in five missions during the last half of the month. On the 22 June raid on Huls, 11 were dispatched and one was lost. On the 25^th, only four of the seven dispatched were able to accompany the formations to Northwest Germany. Five took off on the 26^th, but not one was able to complete the attack. Their record was better against St. Nazaire on the 23th, when all six dispatched completed the mission. The mission to Le Mans on 28 June found the two that were dispatched abortive.

Although the YB-40’s were to see further action in the theater, the June experience was sufficient for the Eighth’s commander to form an adverse judgment on this version of the escort cruiser. His report condemned it because it had different flight characteristics (it was tail-heavy) and was so much heavier than the normal B-17 that it could not fly formation with them. And 40’s had been tried in their own formations and on either side of the lead ship in a combat wing, but in neither case were they successful. The crews of the YB-40’s did not like the planes because they had to occupy the hot spots in the formation and because they carried no bombs. General Eaker was strongly sold on one feature of the YB-40 – the chin turret – which was presently to be placed on all operating heavies. He did not entirely condemn the whole idea of such an escort, but the June experience convinced him that such a craft required two features that the 40’s did not have: the ability to carry bombs and the same flight characteristics as other bombers.

At the same time that the development of the heavy escort was taking place, attention was given to the build-up of a fighter force and the extension of the range of our fighter craft for escort purposes.

The fighter situation in the Eighth Air Force during the first quarter of 1943 was pitiful. It has been noticed that three P-38-equipped groups had been sent from the Eighth to the Twelfth Air Force in October 1942. Four additional fighter groups were trained by the Eighth for the North African theater during the last part of 1942. There was but one operational group left to the Eighth and that was the 4^th (“Eagle”) Fighter Group, equipped with Spitfires. This situation prevailed until 3 April 1943, when three groups became operational. These three (4^th, 76^th, and 56^th) were all using the P-47 Thunderbolt at the time. The 4^th first had had some P-47’s assigned to it in January, and was converted to this craft in March. The 78^th camp to the theater equipped with P-38’s but was presently reequipped with the P-47. The 56^th came to the United Kingdom on 13 January 1943, already out-fitted with the Thunderbolt. The tardiness in getting the P-47 into operation was due partly to radio and mechanical difficulties in the aircraft and to the necessity for “selling” it to the pilots. Moreover, the Fighter Command had to proceed cautiously, for it did not want to run the risk of a serious setback in making a new airplane operational.

Three operational groups of fighters were all the Eighth had during the first phase of the bomber offensive. Although the 353d Group arrived in the theater in June, it did not become operational until August.

Inquiry as to why VIII Fighter Command was no better off in the early part of 1943 is not very profitable. Inter-theater competition drained the groups away and there seems to have been some weight of opinion that the B-17 could carry out daylight operations without escort. The CBO Plan had provided no build-up rate for fighters as it had for bombers, although it had called attention to the need for the latter craft. General Bradley, chief author of the build-up and troop basis for the Eighth, had made the assumption that strategic bombing would generally be unsupported by fighters because of the fighters’ deficiency of range. Whatever the reason, VIII Fighter Command was sadly lacking in planes all during the first phase.

The realization that they were needed, however, became strongly impressed upon the minds of the air force leaders on both sides of the Atlantic. The attrition and damage figures were easy to read. In April the Commanding General of VIII Fighter Command expressed the need for 30 groups and the opinion that opposition to the bomber offensive would grow heavier unless the German fighters were neutralized by an American fighter force. The German fighter tactics of concentrating against the bombers during the bombing run made it clear that protection would have to go all the way to the target. This was set forth by Assistant Secretary of war for Air Robert A. Lovett after a trip to the United Kingdom in June. He declared, “The greatest single factor differentiating the 8^th AF operations from those in other theaters is the extremely high proportion of battle damage resulting from combat with the best German fighters … . There is an immediate need for long range fighters. This may be met by tanks for the P-47s for now but ultimately P-38s and P-51s will be needed.”

The story of increasing fighter range by use of drop tanks is well covered in the histories of VIII Fighter Command and VIII Air Force Service Command. Here all that is necessary is to note the progress that was made in this development during the first phase of the bomber offensive. Expendable tanks were not a development of the European theater, but had been used for some time for ferrying purposes. One of the first tanks used by the P-47 was a 200-gallon-capacity paper ferry tank. The tests which were made on this tank in March 1943 were only partially satisfactory because fuel could not be extracted above 20,000 feet. Moreover, its structure was much that it was not capable of pressurization, and it was not good aerodynamically. However, the range of the Thunderbolt could be extended by using fuel from this tank while climbing to altitude. Such tanks were employed on the first escort mission in which VIII Fighter Command used droppable tanks. This event did not take place until 28 July.

Besides the test on the 200-gallon ferry tanks, there were two other developments in the expendable tank problem during the first phase. One was the testing and provision for procurement of a smaller paper tank of British design. VIII Air Force Service Command’s maintenance division designed a steel tank which was approved by Fighter Command on 29 May. British agencies were asked to manufacture these, but because of production difficulties, the British proposed the substitution of 103-gallon paper tanks which could be pressurized and which had been successfully used on the Mustang. Tests were run on the tank and it was officially approved by the Fighter Command on 26 June. The British began delivery of this tank on 13 July. The third development in the range-extension problem was experimentation with a variety of metal tanks raging in capacity from 75 to 150 gallons. All these were capable of leak-proofing and pressurization but required the installation of certain fixtures on the P-47. Although the solution to the fighter range problem was not reached during the first phase, the need for extending the range was thoroughly realized and progress in design and production was made on both sides of the Atlantic.

The fighter protection given to the heavy bomber operations during the first phase was provided by both RAF Fighter Command and VIII Fighter Command. As a matter of fact, the latter was under the control of RAF Sector Controller until 30 June 1943. All four of the April missions of VIII Bomber Command were given fighter support by Spitfires of the RAF.

The first escort of heavy bombers by P-47 Thunderbolts was furnished on 4 May during the mission against the Ford and General Motors plants at Antwerp. On this mission six squadrons of P-47’s and six of RAF Spitfires were used. The Bomber Command described the support as excellent and attributed the mediocre enemy fighter opposition to the presence of the P-47. During May and June the Thunderbolts escorted the day bombers on at least five other occasions. The other operations of VIII Fighter Command consisted of sweeps, or forays into enemy territory, and diversions designed to give indirect support to the heavies. The P-47’s flew 4727 sorties in 31 operational days during the fires phase and about 30 of them were in support of heavy bombers.

Defense against enemy fighters involved tactical deployment of bombers as well as material developments. Principal tactics were the use of the formation, the diversion, and the feint.

The formation as a defensive measures witnessed much experimentation and development in the ETO in 1943. In the early operations bombing was done by elements of three aircraft in formation and later by squadrons of six. The intensity of enemy fighter attacks forced the bringing together of more aircraft to secure mutual fire support. From January to April 1943 the formations flown were boxes of 12 to 14 aircraft rather widely spaced. The formations were generally loosely flown, for individual evasive action was encouraged. After May the basic group formation was a combat box of 18 to 21 aircraft flown much more tightly than in the earlier period. Such a box ideally flown would occupy 1,200 feet in altitude, 900 feet in breadth, and 600 feet in trail. Actually it was always much larger in breadth and trail. Two or three combat boxes stacked together constituted the combat wing. Experience indicated that three groups was the maximum number that could be practically flown together in a defensive formation. The combat wing was not maintained during the bombing run, since bombing was done by groups. After bomb release the combat boxes would reassemble for support on the route to base.

Formation flying required great skill on the part of the pilots, and the Eighth constantly emphasized the necessity for better training in this specialty at operating altitude. The use of group bombing also influenced another aircrew position. One or two bombardiers in a group became very important because generally all planes released bombs on one sighting operation, that of the bombardier in the lead aircraft. Others simply toggled their bombs on his signal.

The diversion as a defensive tactic was employed by both fighters and bombers. The chief defensive value of the diversion lay in its proper timing to attract enemy fighter action to itself from the main bombing effort of the day. The diversionary force was at times dispatched ahead of the main effort in order to arouse the enemy ahead of time, have him use up his fuel, and so find his fighters grounded when the main attack was delivered. It was also dispatched after the main effort had gone out in order to lessen enemy pressure on a withdrawing force. The diversion was employed at least five times by VIII Bomber Command and several times by VIII Fighter Command during April, May, and June. During the raid on the Erla aircraft plant at Antwerp on 5 April, a force of 25 B-24’s conducted a diversionary sweep over the North Sea and toward the French and Belgian coasts. This measure brought up no fighter opposition, but a similar venture on 4 May was highly successful. On that occasion the bombing objectives were the Ford and General Motors plants at Antwerp. Seventy-nine B-17’s were dispatched on the main effort and 20 B-1’s and 13 B-24’s were dispatched on a diversionary feint toward the French coast. The diversion as well as the main effort was provided with fighter cover. The feint attracted more than 100 fighters (about half the Germans had in the region) and kept many of them airborne so long that they could not attack the main effort. The principal force encountered out 30 enemy aircraft.

VIII Fighter Command engaged in diversionary indirect support of heavies even before it undertook direct close support. On the 17 April raid against Bremen, 76 P-47’s carried out a sweep of the Holland coast while the bombardment mission was in progress.

The diversion not only had defensive value; it gave excellent training to participating air crews. The bomber diversion was but a variation of what has been called the “split-target” technique. This involved the dispatch of strong forces to several different objectives at the same time, thus causing the enemy to disperse his forces. Although practiced during the first phase, the best use of the tactic had to await the build-up of the Eighth to greater strength.

Another defensive variation of the diversion was the feint or fake by the main attack force toward one target with a sharp turn toward the intended objectives at the last moment. Route to and from the target was not left to the groups but was carefully prescribed by higher headquarters.

Except for radar countermeasures, the defensive tactics against flak were fairly well developed by the time of the first phase. These involved selection of operating altitude and route, reduction of bombing run, control of the number of groups bombing together, and evasive action. Further protection from flak was provided by armor plate and the development of special flak curtains and flak suits for crews.

Bombing altitude of heavy bombers during the first phase was generally above 20,000 feet. The lowest reported bombing altitude was 19,000 feet on the Le Mans mission of 20 June, and the greatest, 27,590 feet during a June raid against northwest Germany. Although it is not possible to state an average, it is safe to say that most of the day bombing was from about 23,000 feet. This was high enough to take the planes out of the range of the enemy’s light flak. Bombing altitude, as is the case with many other operational policies, was a compromise. Much better bombing accuracy could have been achieved and many operational problems could have been eased at lower levels. On the other hand, the damage from heavy flak could have been greatly decreased by operating at greater altitudes. The Commanding General of the Eighth stated that the reasons for not going higher were that above 25,000 feet increasing difficulties were experienced with oxygen equipment and with propeller and supercharger action. Operations above 20,000 feet brought to the Eighth a host of problems connected with oxygen, electrically heated flying suits, machine guns, and the functioning of internal combustion engines. Many of the complaints in the United Kingdom about the lack of training of new crews and replacements were due to the altitudes to which the enemy forced operations.

Intelligence and antiaircraft officers played a prominent part in the planning of routes designed to avoid as much flak as possible. With the enemy’s chief emplacements known, not only the best route from base to target could be selected but the best angle of final approach could be charted. Angles of approach and departure were determined by calculating the effectiveness of all possible AA batteries along each 30-degree direction about the target. Charting of such courses became a very important and exacting portion of mission planning. Likelihood of flak damage was also diminished by making the turn along the predetermined axis of attack at the last moment and the production of the bombing run to the shortest possible time commensurate with accurate bombing.

The common method of AA fire control used by the Germans were the predicted barrage and the continuous-following method. In the predicted barrage each gun fired at a particular spot in the sky through which the formation is due to pass; hence the Allies sought protection from barrage fire by increasing the spread of the entire formation in altitude and breadth. Protection from continuous-following fire control was secured by closing up the units in trail and thus saturating the enemy’s flak defense at a particular time. This tactic reduced the risk to any single unit.

Various types of evasive action were taken by the bombers to reduce the likelihood of hits by flak. One was a change of altitude by at least 1000 feet on the approach to the bombing run or after bombs were released. Another was the making of irregular changes in course of at least 20 degrees every 20 to 40 seconds except on the bomb run. Such tactics gave protection because of the tracking time necessary in continuous-following fire control and because of the time required for a projectile to reach operating altitude.

In addition to protection offered by armor plate, further protection was given to air crewmen by the development of flak suits and flak curtains which were capable of stopping low-velocity projectiles. In the autumn of 1943, an analysis of the wounds to crew members on combat missions had showed that 70 were due to missiles of relatively low velocity. The Surgeon of the Eighth Air Forces, Col. Malcolm Crow, was of the opinion that many of these wounds could be prevented by body armor. As a result of researches by the medical department and by the British, had the Wilkinson Sword Company of London make up simple suits of body armor. These were made of laminated (20 gauge) manganese steel plates 1/16 inch in thickness, approximately two inches square, which overlapped 3/8” on all sides. This armored suit covered the vital parts of the trunk, and was designed with a quick release mechanism to facilitate removal in case the wearer was required to abandon the aircraft. The armor as used formed a loose fitting outer garment which weighed about 20 pounds. By the end of May enough of these were being manufactured in the theater to outfit some 30 Heavy crews, and orders were to be placed in the United States for additional quantities.

Defensive measures and devices constantly called for the best efforts of air leaders. That these efforts brought forth both makeshift and very clever contrivances is already evident. Subsequent defensive developments will be taken up as they appeared in later phase of the offensive.