A Decisive Air Force Map

The nautical chart from the  Mitsuo Fuchida papers  entitled 日本海及黄海 or Nippon Kai and Kô Kai (the Sea of Japan and Yellow Sea) was compiled from surveys of the Imperial Japanese Navy and British, Chinese, and Russian charts up to 1934 and features soundings (ocean depths) in meters. However, the annotations on this map are what makes it so unique. Appearing to be from the later stage of World War II, the handwritten notes detail the military assets of the Imperial Japanese forces.

The Ketsugo (Decisive) Operation was an all-out joint operation by the Imperial Japanese Army and Navy, including their air units, to counter the possible US Army landing on the Japanese Home Islands. The directives were adopted at an Imperial Conference (Gozen Kaigi) on June 8, 1945. By this stage of the Asia Pacific War, the Imperial Japanese Navy’s striking power rested overwhelmingly in its air components. The surface ship fleet was catastrophically depleted. A shortage of oil meant the few major units remaining (aircraft carriers, battleships, and cruisers) were nearly useless, save for use of their antiaircraft armament for local defense. The submarine fleet was very diminished. To an important degree, the independent existence of the Imperial Navy hinged on its air fleet. This is why the navy resisted army plans to bring air units of both services under one command, for fear this would effectively extinguish the navy’s independent existence.

The figures presented on this chart appear to be the total available aircraft strength of the Imperial Navy in the Home Islands (the term used by the United States to describe the Japanese archipelago that forms the country’s homeland, apart from its other territories and colonies). The combined figures for the Tenth, Third, and Fifth Air Fleets appear to be 3,680. This total is significant, as it appears to be well below the calculation of the postwar United States Strategic Bombing Survey (USSBS) that attempted to figure the number of Japanese aircraft available to fight against the Allied invasion which was halted when Japan surrendered. The USSBS figures for the Imperial Navy were 2,700 training aircraft converted for “special attack” or kamikaze missions and 3,200 combat aircraft for conventional missions. The total estimate was 5,900. Postwar, the commander of the Imperial Navy air units stated that ultimately every aircraft would have been expended in special attack missions.[1]

While it is not explicitly stated on the map, the figures for types of aircraft are consistently in round numbers which strongly implies these are estimated totals they had on hand and not precise operational figures. Japanese planning aimed to make available the maximum strength to use against invasion. Except in limited circumstances, in the months of June 1945 through the end of the war, in August 1945, Japanese aircraft were generally held in reserve and not counterattacking US ships offshore or attempting to intercept American aircraft flying over Japan. Part of this preservation effort was the wide dispersal of aircraft geographically to keep them from being destroyed on the ground. This dispersal program, however, must have had a detrimental effect on maintaining aircraft serviceability and would have made keeping updated figures on availability hard to ascertain.

• Fighter aircraft: 190
• Reconnaissance aircraft: 20
• Bombers: 80
• Torpedo bombers: 60
• Attack bombers: 45
• Fighter bombers: 30
• Yokosuka K5Ys: 550
• Total: 975

Japanese planning for Operation Ketsugo allowed for a number of scenarios for the expected American landing in the Home Islands. It made no allowance for a Soviet landing from the Sea of Japan. The planners very accurately expected the most likely scenarios: first, a landing on Kyushu about November 1945, which would be followed by a landing in the Tokyo region. 

In view of this planning, the Tenth Air Fleet are generally depicted on this chart not in the northernmost regions of Honshu but further south, where they could be employed against a landing near Tokyo. No bases are visible on Hokkaido for Imperial Navy aircraft. Hokkaido does appear to house army transport aircraft (the Twelfth Air Fleet), and the significance of this and its listing on this chart are addressed below. 

One major factor in aircraft deployment was the severe shortage of aviation fuel.  Available evidence shows that fuel was gathered at active bases and not stored at bases not actively occupied. Even if this was done in some cases, there clearly was not enough fuel to create a large alternate network of bases. The significance of this factor is important in contemplating the planned Soviet landing on Hokkaido.

Of particular note in the Tenth Air Fleet inventory are the Yokosuka K5Ys, given the code name “Willow” by the Allies. This is the standard biplane intermediate trainer aircraft of the Imperial Navy. It was first produced in 1933, and despite later development of monoplane trainers, it was produced until the end of the war. Given its trainer role, the original design provided for two seats. Total production was 5,770. It was used in both land plane and sea plane configurations, the latter in very limited production. It's construction is described as “mixed”: given the time frame of development, clearly large parts of the aircraft were wood. 

The significance of this listing in the available aircraft for the Tenth Air Fleet is that the Imperial Navy, like the Imperial Army, converted most operational training aircraft for the “special attack” or kamikaze role in 1945, primarily for Operation Ketsugo. The original design had one fixed machine gun, fitted to fire forward, and one flexible mount, presumably fitted to the rear cockpit area. As originally designed, the aircraft could carry smaller explosives, up to two 30 kg (66 pounds) or ten 10 kg (22 pounds) bombs. Sources are limited and do not indicate the type or types of bombs fixed to aircraft of this type in the special attack role. Presumably it could carry at least 100 kg (220 pounds) of bombs if modified, perhaps more based on the weight saved when flown by only one man. Nor is it clear whether the planes’ structure would require this weight to be divided between the wings, or whether one heavier bomb could be fitted below the fuselage.

Operationally, in the special attack role, this aircraft possessed the great disadvantage of slow speed. This meant much longer exposure to antiaircraft fire if the aircraft was spotted. As a trainer, it was rated for 115 knots at sea level, or 132 mph. A heavy bomb load would seriously reduce this speed—a good guess would be to around 100 knots or less. Roughly, this would put the aircraft within range of antiaircraft fire for about three times as long as first-line aircraft.

The K5Ys did have two advantages, as reflected in some attacks during the Battle of Okinawa by wooden float planes sent on special attack missions. When combined, their ability to fly very low and slow presented a difficult target to fighter aircraft. The wooden construction also meant that US search radars would not pick them up easily, or at all, so they might be very close before being detected. Further, the deadly US proximity (variable time, or VT) fuzes that made larger-caliber US antiaircraft shells so effective would not obtain a strong enough “return” to detonate the shell from the radio transmitter that triggered the fuze. These types of aircraft sank one American destroyer, with very heavy loss of life. American reports on this incident picked up on the fact that the wooden construction had largely nullified radar detection and effectiveness of the VT fuzes. Another factor giving the K5Y an advantage would be if the aircraft staged attacks under low-visibility conditions, at dawn or dusk. Small aircraft would be particularly difficult to see under those circumstances.

Japanese planners looked to launch special attack planes in massive waves and saturate Allied defenses. These aircraft of the Tenth Air Fleet, stationed so far north, however, would have had to redeploy southward to engage in the projected landing in Kyushu. This gets into the fuel issue discussed above. Also, these aircraft were committed toward defense of the Tokyo region, so they may not have been committed against a landing on Kyushu.

One other special aspect about these aircraft may have been of considerable significance, although we can only conduct some informed projection here. The map shows 150 K5Ys at Kōriyama in Honshu, farther to the north. The Soviets seriously contemplated and planned for a landing on the west coast of Hokkaido in August. As events turned out, before their projected date of action—which appears to have been around August 22 or 23—the emperor announced his decision to end the war and the Japanese government surrendered. The Imperial Army, before surrender, had deployed all regular units on the east side of Hokkaido to deal with a US landing from the Pacific Ocean side. There were no regular units anywhere near the projected Soviet landing site. It would have been a long march to get Japanese troops in position to confront the Soviet landing on the Sea of Japan coast of Hokkaido.[2] 

In reaction to such a landing, it is likely a Japanese counter would have been to try to redeploy special attack planes, probably among them Imperial Navy K5Ys, against the Soviet landing. It is uncertain whether the northernmost K5Y units could have attacked from their bases—which would have been a problem given the shortage of aviation fuel—or whether they would have had to redeploy closer to the planned Soviet landing site. Deployment of these K5Ys might have stood in the way of a successful Soviet landing and the likely division of Japan into US and Soviet occupation zones had Japanese surrender not taken place first.

The map indicates a very large number of army transport aircraft annotated mainly on northern Honshu, with apparently some at Chitose on Hokkaido. Totaling 260 throughout the map, these aircraft were gathered for a massive special forces operation to target the Allies’ B-29 bases in the Marianas. In Japan, this was called Katana Kengo and translates as Operation Sword. In US messages and intelligence summaries, it was called Operation Damocles.

The Japanese plan was that a raiding force of three thousand men, both Imperial Army and Imperial Navy, would be flown down to the Marianas in these transport planes and stage a massive commando raid to destroy as many B-29s as possible, along with the infrastructure of their bases. Preparations for the operation were tracked by US code breaking, and when it seemed the preparations were complete and the operation set to start soon, Admiral William F. Halsey, Jr.’s Task Force 38 hit the bases of these aircraft. Over two days, starting on August 9, 1945, they destroyed the great majority of the planes and forced the cancelation of the operation.[3]

It can be surmised that the participation of Imperial Navy sailors in the plan raid was why Fuchida particularly noted these Imperial Army aircraft on this map, which he did not normally do.

The Kyushu L11W was a single-engine bomber training aircraft that was designed for a crew of five. From November 1942 to August 1945 a total of 798 were built. The aircraft was originally all metal, but due to shortages an all-wooden version followed that was built in small numbers. Given the general program of converting training aircraft into special attack aircraft, presumably a number of these aircraft were also dedicated to this mission. They could be fitted with one 250 kg (551 pound) bomb for kamikaze missions, which confirms that point. Further, the numbers listed on this chart would carry the inference that these were aircraft for special attack missions, since it appears the chart only lists aircraft used in combat roles. The general comments about the Yokosuka K5Y broadly apply to the K11W. The major differences are that almost all K11Ws were metal, so they lacked the advantages of the wooden aircraft. They were also notably larger and easier to detect and probably would have been more easily shot down by fighters or antiaircraft fire. 

• Fighter aircraft: 270
• Reconnaissance aircraft: 15
• Bombers: 90
• Fighter bombers: 65
• Torpedo bombers: 40
• Seaplanes: 30
• Flying boats: 3
• Yokosuka K5Ys: 140
• Kyushu K11Ws: 110
• Total: 873

Located in central Honshu, the Third Air Fleet covered three major regions identified on the map in large red letters and delineated by red lines: San'in Skies (山陰空) along the western coastal area, Kinki Skies (近畿空) in the greater Osaka-Kyoto area, and Tokai Skies (東海空) in the area around Nagoya and east to the Izu peninsula.

The Third Air Fleet deployment appears to have been intended to permit attacks against an invasion of either Kyushu or the Tokyo region. It is not clear whether all the aircraft could mount attacks in either direction. There would need to be better evidence on available fuel and the effective ranges of all the aircraft, especially the K5Y and K11W, when outfitted for a kamikaze mission.

• Fighter aircraft: 320
• Reconnaissance aircraft: 20
• Bombers: 80
• Fighter bombers: 45
• Torpedo bombers: 40
• Attack bombers: 20
• Fighter bombers: 20
• Seaplanes: 90
• Flying boats: 7
• Yokosuka K5Ys: 1,020
• Kyushu K11Ws: 170
• Total: 1,832

In the southernmost area of aircraft staging on this chart, the Fifth Air Fleet included three regions delineated by red lines and named in large red letters. They include: Naikai Skies (内海空) in southern Honshu and all of Shikoku; Kyushu Skies (九州空) in Kyushu and the Tsushima Straights, and Chōsen Skies (朝鮮空) in southern and eastern Korea.

The location and numbers of aircraft in the Fifth Air Fleet is striking evidence of the astute Japanese calculation that the initial American landing on the Japanese Home Islands would be on Kyushu. The most direct evidence is the deployment of 60 percent of all of the K5Y trainer/suicide planes to the Fifth Air Fleet. 

A few comments are warranted about the US perspective of events corresponding with the positioning of Japanese military assets at this time: The United States created the Downfall Plan for an initial invasion of the Japanese Home Islands, which in its original draft called for a two-phase invasion. The first phase, Operation Olympic, was scheduled to land at three or four locations (one was deemed optional) on Southern Kyushu around November 1, 1945. The purpose of this effort was to secure the southern third of Kyushu. This area would then afford air bases and sea bases to support the second phase, Operation Coronet, about March 1, 1946. This would target the Tokyo-Yokohama region. Its goal was to end Japanese resistance.

When the plan was drafted, the United States projected opposition to Olympic at its outset as only three Japanese field divisions in Southern Kyushu. The assigned American landing force totaled fourteen divisions, two regimental combat teams, and an array of artillery, tank, and combat engineer units. The scenario allowed that the Japanese might build up to eight divisions with 350,000 men, but that the reinforcements would arrive piecemeal, so the ground combat component of the invasion force would continue to enjoy great numerical superiority during the campaign. Its firepower superiority would be even greater, implying that US casualties would not be severe. In connection with these estimates, total Japanese air strength against Olympic was expected to be no more than 2,500 aircraft. Although the original planned strength of the Olympic invasion force was set at 766,700 men, only about 218,000 of this landing force were ground combat soldiers and Marines who would capture the southern third of Kyushu. The remainder were a massive array of support troops, especially construction units, to repair existing air bases or create new ones for 2,794 aircraft and likewise set up port facilities and other logistical facilities.[4]

President Truman, greatly disturbed by the losses on Okinawa, called for a major meeting with his military and naval leaders on June 18, 1945. The memorandum calling for this meeting stated that the principal criterion for the president’s decision on the issue of invasion would be casualties. We now know that the briefing prepared by the army omitted a series of relevant prior estimates pointing to high or massive US casualties.[5] Instead, in a statement delivered by General George C. Marshall, the army claimed it would be wrong to give an explicit number because casualty experience in the Pacific had been “so diverse.” In place of an explicit number, a series of analogies were advanced. Marshall emphasized the example of Luzon, with the most favorable ratio of five Japanese to one American casualty. 

The senior naval officer, Admiral Ernest J. King, presented himself as supporting Marshall. But we also know now that King was absolutely determined to stop any actual invasion of the Japanese Home Islands. This was in accord with decades-old navy views that an actual invasion would likely lead to unacceptable casualties or even outright failure. King only agreed that preparations for invasion as an option should be permitted to proceed. He explicitly told the other members of the Joint Chiefs of Staff in April 1945 that they would return to the question of whether to invade the Home Islands in August or September. (King was not thinking about the atomic bombs, but he calculated that by those dates the fleet would be free of obligations at Okinawa and able to clamp down on a very rigorous blockade of Japan. He thought the evidence at that point would arm him to better refute the army's position on invasion presented before President Truman.) King carefully hedged his ostensible support for Marshall before Truman on June 18 by adding that of course even if Truman authorized Olympic at that time, the operation could be canceled later.[6] Truman was effectively asked at the meeting to approve the entire two-phase, Downfall plan, but he only sanctioned Olympic.[7]

In the weeks around the June 18 White House meeting, radio intelligence (supplemented by aerial photographic reconnaissance) detected the stupendous build-up of Japanese forces on Kyushu. By August these numbered thirteen divisions, ten assorted brigades, and various other units. Total Kyushu garrison was between 700,000 and 900,000 troops, virtually all of whom would be committed to confronting the expected invasion. Most of these were located proximate to the selected Olympic landing beaches. The accurate location of the defenders stemmed not from a security breach but from the Japanese deduction that any American landing on southern Kyushu would have as its primary aim the seizing of air bases or ground suitable for air bases. Any ordinary topographical map could identify the level, low ground suitable for an airfield on southern Kyushu. These locations were few and obvious, and that is where the Japanese massed their defenders.

By the last two weeks of July and into the first week of August, the intelligence—primarily code breaking—about the Japanese defenses of southern Kyushu had identified most the units in the build-up. The numbers were already estimated as up to 680,000 and new identifications of Japanese units were being made almost every day. Likewise, code breaking disclosed a huge buildup of Japanese aircraft, half or more detailed for kamikaze attacks. This intelligence demonstrated that the basis for assuming a huge numerical superiority over the Japanese defenders was grossly wrong.[8] The intelligence was so alarming that General Marshall inquired by coded message on August 7 (the day after the bombing of Hiroshima) whether General Douglas MacArthur still supported Olympic. MacArthur replied on August 9 that he did not believe the intelligence picture was accurate and urged Olympic to proceed as authorized. 

At this point Admiral King pounced. He took the exchange of messages between Marshall and MacArthur and sent them to the senior naval officer in the Pacific, Admiral Chester W. Nimitz. King formally asked Nimitz for his views on the viability of Olympic. But King knew from a private communication from Nimitz in May (not shared with the army) that Nimitz, based on the Okinawa experience, no longer supported any invasion of the Home Islands. 

King’s message to Nimitz was on August 9, the day of the bombing of Nagasaki. The next day, August 10, the first Japanese message indicating that the Japanese might surrender was received. By August 14 the war with Japan was over. There is no record of any reply from Nimitz back to King. The obvious conclusion is that when Nimitz realized there was a real possibility of Japan surrendering in the near term, he withheld any response.[9]

To a degree this map confirms the large and unanticipated massing of Japanese air power to meet a US invasion such as Olympic, it also reinforces the American perception in July and August 1945 that Olympic was untenable unless there were some radical change in the balance of combat power between the invaders and the defenders. There is another aspect we know now that was also unknown for many years after the war. On August 13, Marshall initiated a dramatic effort to alter the balance of combat power for Olympic. Given his message to MacArthur on August 9, Marshall clearly recognized that the intelligence picture raised severe doubts about the viability of Olympic. Practically speaking, there simply were not enough additional ground units that would be available by November 1 to increase the number of combat troops in the invasion force. But Marshall thought of another option to restore the viability of Olympic.

We have the transcript of a conversation between a staff officer detailed by Marshall to discuss the situation with an official at the Manhattan Project, which built the atomic bombs. That transcript shows the staff officer relayed Marshall’s belief that either the Japanese government would surrender after two cities had been hit with atomic bombs, or that no additional atomic bombing of Japanese cities would secure Japan’s surrender. Therefore, the staff officer explained that Marshall wanted to know the number of atomic bombs that would be available by November 1. The staff officer went on to explain that Marshall now contemplated using the additional atomic bomb production in what would now be called tactical use, as a direct means of supporting Olympic. The number given in the transcript was seven bombs. The transcript also shows that the issue of safety in the use of atomic bombs in this fashion came up. But the only safety concern discussed was what to do if one of the bombs failed to explode. There was no discussion at all of any radiation hazard.[10]

Unless otherwise stated, the data on all Japanese aircraft is drawn from Rene J. Francillon, Japanese Aircraft of the Pacific War (Annapolis, MD: Naval Institute Press, 2000).

1. Richard B. Frank, Downfall: The End of the Imperial Japanese Empire (New York: Random House, 1999), 182–85.
2. For background on this, see Frank, Downfall, 323–34.
3. Frank, 158, 395.
4. Frank, 117–22, 145.
5. Frank, 131–39.
6. Frank, 36–37.
7. Frank, 139–148.
8. See Frank, chapter 13.
9. Frank, 273–77.
10. Frank, 311–13.

Additional Resource:  Senshi sōsho , the National Institution for Defense Studies