From the book “Hell To Pay, Operation DOWNFALL and the Invasion of Japan, 1945-1947” by D. M. Giangreco regarding the impact of typhoons on the planned invasion of Japan. p. 175-176
“If there had been no atom bombs and Tokyo had attempted to hold out for an extended time (a possibility that even bombing and blockade advocates in Washington granted), the Japanese would have immediately appreciated the impact of the storm [Typhoon Louise of October 9, 1945] in the waters around Okinawa. Moreover, they would know exactly what it meant for the follow-up invasion of Honshu, which they had predicted as accurately as the invasion of Kyushu. But even with the storm delay plus friction of combat on Kyushu, the Coronet schedule would have propelled U.S. engineers to perform virtual miracles to make up for lost time and implement Y-day as early in April as possible. Unfortunately the divine winds packed a one-two punch.
From May 27 to April 7, 1946, yet another typhoon raged in the Pacific. On April 3 Barbara struck Luzon, where it inflicted only moderate damage—ripping roofs off of Base M warehouses at Lingayen Gulf, grounding an Army tugboat, and sinking a ship in Manila Bay, where waves briefly reached an unusual thirty-five feet in the harbor—before pounding toward Taiwan. Coming more than six months after the war, it was of no particular concern. The Los Angeles Times gave it several short paragraphs on the bottom of page 2 and didn’t even mention the storm’s name. But if Japan had held out, this typhoon would have and profound effects on the world we live today.
[Typhoon] Barbara would have been the closest-watched weather cell in history. If the delayed invasion of Honshu was not already in the process of being launched, the typhoon’s long, lumbering approach to the Philippines would allow First and Eighth Army soldiers (many of whom would have lived in tents instead of barracks because it was expected that they would have moved north a month earlier) to make the best preparations they could under the circumstances. Ships and craft that could not be sent south would be secured and likely ride out the storm with minimal losses. However, if Coronet was in the middle of its execution from the twenty-five-day window Y – 15 to Y + 10, chaos would ensue because the storm’s track and intensity could only be guessed at within the parameters of the limited data available.
Would slow, shallow-draft landing craft be caught at sea or in the Philippines, where loading operations would be put on hold? If they were already on their way to Japan, how many would be able to reach the Koshiki Retto anchorage and Kyushu’s sheltered bays or get back to Luzon? And what about the breakwater caissons for Ironhorse, the massive artificial harbor to be assembled east of Tokyo? The 1945 construction of the harbor’s prefabricated components carried a priority second only to the atom bomb, and the first packages of this precious towed cargo would have begun arriving in the western Pacific at this time. They could not be allowed to fall victim to this and other seasonal storms and be scattered across the Philippine Sea.
Whatever stage of deployment U.S. forces were in during those first days of April, a delay of some sort—certainly no less than a week and perhaps much, much more—was going to occur. A delay that the two U.S. field armies invading Honshu could ill afford and that the Japanese militarists would see as yet another sign that they were right after all. And while much of the land around Tokyo today contains built-up areas not there during the war and deceptively smooth terrain, thanks to the delays over which the United States had absolutely no control, any soldier or Marine treading this same flat, dry “tank country” in 1946 would, in reality, have been up to their calves in muck and rice shoots by the time the invasion actually took place.”