According to photographic imagery and marine forensic and photogrammetric analyses, the photo below indicates the presence of at least one midget submarine in Pearl Harbor on the morning of 7 December 1941. The same analysts who support that theory now are concentrating their search for evidence on a timeline derived from translated Japanese documents that has allowed them to determine which midget submarines were capable of entering the harbor that morning.
The following represents the latest research expanding on a project undertaken in 1993 by a team of imagery intelligence analysts and photogrammetric scientists from Autometric, Inc. (now Boeing-Autometric). It was conducted as a public service on behalf of the historian of the USS Arizona Memorial, National Park Service, Daniel Martinez, who requested identification of an object in a well-known Imperial Japanese Navy aerial photograph. He wanted verification that it was a Japanese Navy Type-A (Ko-hyoteki)-class "midget" submarine. Martinez cited a local historian and newspaper correspondent, Burl Burlingame, who had postulated that the object was indeed one of the midget submarines. Burlingame had presented his opinion at a symposium sponsored by the National Park Service in 1990, and he had incorporated his thoughts on the subject into his monograph, Advance Force Pearl Harbor: The Imperial Japanese Navy's Assault on America (Pacific Monograph, 1992, see page 199).
The Boeing-Autometric team analyzed the photograph and presented its findings to Martinez at a National Park Service-sponsored press conference on 6 December 1994. The report concluded that a Type-A submarine was present in the photograph and that it did participate in the attack on Battleship Row. The report also stated that this submarine had fired both of her torpedoes successfully. One torpedo could be seen striking the USS West Virginia (BB-48), and the second is seen moving in the direction of the USS Oklahoma (BB-37).
Subsequently, the team incorporated its findings into two articles for Naval History magazine, published in its December 1999 and June 2000 issues, respectively. Further analysis in mid-2000 led to this article, and we hope to present our body of work at the upcoming events marking the 60th anniversary of the Pearl Harbor attack in December 2001.
The Approach to the Harbor
Based on translations of the official Japanese history of midget submarine operations against Pearl Harbor (no longer in print and without an official English translation), we have examined the movement of the submarines, from the time of their release to their approach to the entrance of the harbor. Yoshinori Nakai of Japan's Mainichi newspapers supplied a copy of the history, and his English translation of it validated another conducted by Tadashi Hoida. Table 1 shows the release times for each midget submarine and the distance and the bearing to Pearl Harbor's main channel buoy for each one. We plan to have an official translation for the next update of our research, particularly with regard to radio communications from the midget, the 1-16 tou, to the mother submarine.
As this table shows, the fleet submarine 1-16 released her midget at 42 minutes past midnight, local time, on 7 December, followed by four other releases. The proximity of the submarines to the harbor and patrolling U.S. Navy ships, together with the time of sunrise, shows that only the first two submarines had the best chance to penetrate the waters surrounding Ford Island. The 1-16 tou could be the submarine identified in the photograph.
Figure 1 illustrates each midget submarine and release time as shown in Table 1. Assuming that each mother submarine calculated her bearing and range to the channel-marker buoy correctly, and allowing for a circular positioning error of one nautical mile, the first two boats released that morning had the best chance of reaching the channel-marker buoy without being detected. The release times of the last three probably precluded their entrance into the harbor.
Excepting the 1-24 tou, each submarine is projected at two speeds: the one in red is at 5 knots; the second, in black, is at 4 knots. The red asterisk marks the estimated time for each midget to reach the channel-marker buoy. The black dash lines and arrows represent the estimated additional time and the arrival time for each submarine at 4 knots. These times take into consideration the time each submarine needed to correct her trim, determine her heading to the channel entrance, and take additional sightings to correct for any error in dead reckoning. This assumes that all boats were submerged and used their periscopes to update their dead reckoning. It does not take into account the local currents, but the speeds selected are speeds made good through the water, regardless of shore currents. In addition, it does not factor in possible mechanical problems that could have occurred during the transit.
Figure 1 also shows a U.S. Navy event and contact timeline applied along with that of the midgets. The USS Ward (DD-139) was at sea throughout the period. She was patrolling within a 2-by-2-nautical-mile exclusion zone just south of the harbor's channel entrance. After 0200, two minesweepers, the USS Condor (AMC-14) and the USS Crossbill (AMC-9), joined the Ward. Both minesweepers swept for mines until they reentered the channel at around 0525. The Boeing-Autometric team places the USS Antares (AKS-3) within the general vicinity of the exclusion zone at 0500.5 Also indicated are the times the submarine/antitorpedo net gates, astride the channel and just inside the harbor entrance, were opened.
The well-documented activity of the 1-24 tou discounts her as the mystery submarine seen in the attack photograph. This boat suffered from a critical gyrocompass failure that delayed her release and slowed her approach. She also suffered a nearly catastrophic failure in her trim after her release. The result of all this was that she struggled in her approach throughout the period, beginning with her release and through to the end of the timeline. Finally, this midget was the one that eventually beached herself on the north shore near Bellows Army Airfield. Historians have identified her as Midget C, and she can be seen on display at the Admiral Nimitz Museum in Fredericksburg, Texas.
In 1960, U.S. Navy divers discovered another midget submarine, approximately 1 nautical mile offshore and submerged in 76 feet of water. This boat still had her torpedoes, and after they were removed by separating the bow from the rest of the submarine, a new bow was reconstructed, and the submarine was presented to the Japanese Maritime Self-Defense Force. It is on display at the Japanese Naval Academy at Eta Jima. This boat, Midget D, could be either the 1-18 tou or the 1-20 tou. Most likely it is the latter.
Substantial evidence supports this claim. A comparison of the salvaged submarine's location (as stated in a December 1961 article in the U.S. Naval Institute Proceedings) with the approach headings (reciprocal bearings to those in the table) and time of release indicates that the submarine in question must have been either the 1-18 or the 1-20. Both were launched near the same bearing, and their subsequent courses were comparable. The only difference was the release time and the distance each boat had to travel before she reached the channel-marker buoy.
In his book, Pearl Harbor Ghosts (New York: William Morrow, 1991) Thurston Clarke details an account he found in the December 1967 issue of Our Navy magazine. In the article, a former U.S. Navy officer, Ellsworth Boyd, interviewed a man who claimed to be the crewman Sasaki. Sasaki, together with Lieutenant Sakamoto, were the crew of the 1-20 tou. In the article, this man stated that the midget had malfunctioned and sunk. His description of the interior of the submarine, including the failed scuttling charge fuse, matches the Proceedings article.
The Our Navy article further quotes the alleged Sasaki as saying that both crew members heard depth-charge attacks while they sat on the bottom. The first occurred at 0645 local time, when the Ward commenced her attack on a submarine reported to be off the starboard quarter of the Antares, which was approaching the entrance to Pearl Harbor from the east. This would have placed Sasaki's submarine between the shore and the Antares. The team compared the position given in the Proceedings article with a nautical chart of the area, then compared it with the release point calculated from Table 1. It appears that this submarine ran for a short period before she sank. If this account can be accepted fully, then another midget submarine was in position to trail the Antares. This could have been the 1-18 tou, which was released 12.6 nautical miles from the channel-marker buoy and 42 minutes prior to the release of the 1-20 tou. Her position in relation to the Antares makes her a candidate.
The next submarine is what historians identify as Midget B, and this could be one of the first midgets released. This is the boat that, according to most accounts, entered Pearl Harbor, fired her torpedoes unsuccessfully, and was subsequently sunk by the USS Monaghan (DD-354).
Figure 1 illustrates exactly how close these submarines were to Pearl Harbor's entrance. When one factors in the release times and distances to Pearl Harbor's channel buoy, however, the chances of success for the last three boats are greatly reduced. Adding sunrise to the calculation, together with the need for all three boats to run a 3-nautical-mile narrow channel, with three defensive barriers across it (this is even before they entered the waters surrounding Ford Island), the last three boats would have had to increase their approach speed to make it to the channel and then to their attack positions before sunrise.9
This would have increased their risk of detection dramatically. Considering that need and the increase in U.S. naval activity after 0200, it is no wonder the detections and engagements occurred between 0340 and 0645.
Each midget submarine skipper knew his orders were to proceed with dispatch, but they were not to be reckless in their approach. Detection would warn the Americans and jeopardize the success of the aerial attack. Each skipper had to have his midget in position before the aerial attack, scheduled just before 0800 local time.
The 1-16 tou, under command of Lieutenant Yokoyama and his crewman, Warrant Officer Eyed, separated from her mother submarine at 0042. As recorded in the official history, the distance and bearing of the separation point from the buoy marking Pearl Harbor's channel entrance was 7 miles at 212 deg. The Boeing-Autometric team estimates that once Yokoyama found his bearings, he headed toward the entrance at nothing less than a 5-knot speed (if he approached at a higher rate of speed then it was much the better for him). The team discounts the notion that he would have pressed on faster than 10 knots, however, because all of the midget skippers needed to calculate the endurance capability of their craft. Foremost in their minds must have been how to get into position before the aerial attack commenced without being detected. Once they fired their torpedoes, they were to proceed to a rendezvous point 7 miles west of the island of Lanai. Their boats had a range of 100 nautical miles at 4 knots. A speed beyond 10 knots would drain their batteries too quickly.
As seen in Figure 1, the last three submarines would have faced a very difficult chance of success. In the case of the 1-24 tou, besides having a gyrocompass failure and trim problems she was more than two hours late to the harbor entrance. After being attacked repeatedly, the 1-24 lost her way and moved around the east side of Oahu. She was found later, beached on the north side of the island, near Bellows Airfield.
All of this evidence supports the proposition that the 1-16 tou and the I-22 tou were in the best position to make it into the waters surrounding Ford Island. Both boats had enough time and darkness to shield their approach, enter the channel, and penetrate the defensive nets. Aiding them was the fact that only the destroyer Ward was operating in the vicinity of the harbor entrance prior to 0200. At approximately that time, the minesweepers Condor and Crossbill began their operations inside the exclusion zone established seaward of the harbor entrance. The 1-16 tou could have been at the entrance of the channel at the time the minesweepers sortied. Just another one or two knots of speed easily could have placed both boats well inside the two-mile exclusion zone seaward of the harbor entrance if not in the channel. Both craft would have been in a position between the channel entrance and the operating U.S. Navy units.
At 5 knots, the 1-16 tou could have reached the buoy and the harbor-channel entrance just around 0200. This coincides not only with the minesweeper sortie but also with the opening of the antisubmarine and torpedo nets. At this time, however, Yokoyama and the 1-16 tou had approximately another 1.5 nautical miles to reach the first two defensive barriers.
The nets were closed after the minesweepers sortied, but at this writing, the exact time of that closing is unknown. The record shows that the nets were opened on or about 0458 for the returning minesweepers, and they remained open until 0840, more than 40 minutes into the aerial attack. At 5 knots, Yokoyama would have made it to the first two nets sometime around 0220. At this point it was irrelevant whether the nets were open, because each midget crew planned on their being closed. They anticipated that each boat would need to dive below the nets (intelligence had told them how far down the nets went and how deep the channel was at this point). If the 1-16 tou found the nets open, her final run through the channel would have been simpler. Based on the calculations of the Boeing-Autometric team, the "Electric Light Barrier" was located approximately one mile farther in the channel, and just a half-mile away from the opening to the harbor proper. Traveling at 5 knots, the 1-16 tou could have reached this barrier well before 0300 local time. Apparently, the Japanese did not know that this "high tech" barrier existed. According to Park Service Historian Martinez, this would have been a surprise to the Japanese midgets. In fact, he has interviewed an American who was one of the operators of the barrier during the early-morning hours of 7 December.
At this point, the value of information surrounding this Electric Light Barrier becomes more clear. One submarine definitely passed through it, and it was the one sunk by the Monaghan. Boeing-Autometric team members believe that historical data describing the barrier must exist. In the Martinez interview, the gentleman indicates that the barrier was activated by something passing through it during the early-morning hours of 7 December. But the warning was discounted, because the barrier was prone to false alarms. If there are times associated with the gentleman's account, then the Boeing-Autometric team will be in a better position to back its timeline and refine its estimates. The journey continues!
Boeing-Autometric’s Rodgaard, Lucas, and Biache—who have analyzed evidence of midget submarines at Pearl Harbor since 1993—added Techmatics/Anteon marine forensic shock analyst Hsu to the team in 1998.