The US Navy is building a fleet of small unmanned vessels to act as pickets and outer escorts for carrier and surface groups despite having no evidence, whatsoever, that the unmanned vessels can effectively carry out their task. Since the Navy refuses to conduct any actual experimentation to validate the concept prior to committing to production, is there any other source of information that might allow us to assess the concept? Of course there is and it’s our most reliable source … history! Specifically, it’s the battle for Okinawa and the role of the Kamikaze and the US Navy picket ships.
Okinawa was a major battle for both the US and Japan. US losses during the 3 month battle were steep with 768 aircraft lost in addition to staggering naval losses.(5) From Wikipedia:
At sea, 368 Allied ships—including 120 amphibious craft—were damaged while another 36—including 15 amphibious ships and 12 destroyers—were sunk during the Okinawa campaign. The US Navy's dead exceeded its wounded, with 4,907 killed and 4,874 wounded, primarily from kamikaze attacks. (5)
Just as astounding to us, today, was the magnitude of the forces assembled for the battle. Consider this partial Allied naval order of battle at Okinawa (1)
US Navy combat ships:
11 fleet carriers
6 light carriers
22 escort carriers
8 fast battleships
10 old battleships
2 large cruisers,
12 heavy cruisers
13 light cruisers
4 anti-aircraft light cruisers
45 destroyer escorts
Amphibious assault vessels:
84 attack transports
29 attack cargo ships
LCIs, LSMs, LSTs, LSVs, etc.
52 submarine chasers
23 fast minesweepers
Royal Navy combat ships:
5 fleet carriers
7 light cruisers
The 265 US combat ships, alone, nearly equals the size of our entire present day fleet and that was just the force for a single operation. We’ve truly forgotten the size of the force necessary to wage total war.
Japan, too, fielded a large force, mainly aerial, along with the battleship Yamato, a cruiser, and several destroyers. During the course of the battle, Japan launched 10 large scale kamikaze attacks against the US Navy Fifth Fleet guarding the Okinawa amphibious invasion fleet. Each attack consisted of hundreds of aircraft. For example, the first attack consisted of 355 kamikaze aircraft and 344 escort fighters and lasted for five hours. The US Combat Air Patrol (CAP) did an amazing job but could not stop every attacker.
Twenty-two kamikazes penetrated the combat air patrol shield on April 6, sinking six ships and damaging 18 others. Three hundred fifty U.S. crewmen died. (2)
Only 22 attacking aircraft managed to penetrate the aerial defenses but the damage they did was enormous.
What was the overall result of the kamikaze attacks?
The Japanese fell short of their goal of “one plane one ship,” but sank 36 American warships, and damaged 368 other vessels at Okinawa. The Navy’s losses were the highest of the Pacific war: 4,907 sailors and officers killed, and 4,824 wounded. Japan lost an estimated 1,600 suicide and conventional planes at Okinawa. (2)
The sheer number of attacking aircraft represented what we would, today, call a saturation attack intended to overwhelm the defensive capacity of the US fleet.
The Navy’s answer to the kamikaze saturation attacks was to establish a ring of radar picket ships around the island and the invasion fleet extending out as far as 80 miles. The pickets provided early warning, fighter direction, and direct engagement. Each picket ship was tied to a circular station of 5000 yds radius.
Wikipedia describes the Okinawa radar picket system.
A ring of 15 radar picket stations was established around Okinawa to cover all possible approaches to the island and the attacking fleet. Initially, a typical picket station had one or two destroyers supported by two landing ships, usually landing craft support (large) (LCS(L)) or landing ship medium (rocket) (LSM(R)), for additional AA firepower. Eventually, the number of destroyers and supporting ships were doubled at the most threatened stations, and combat air patrols were provided as well. In early 1945, 26 new construction Gearing-class destroyers were ordered as radar pickets without torpedo tubes, to allow for extra radar and AA equipment, but only some of these were ready in time to serve off Okinawa. Seven destroyer escorts were also completed as radar pickets. The radar picket mission was vital, but it was also costly to the ships performing it. Out of 101 destroyers assigned to radar picket stations, 10 were sunk and 32 were damaged by kamikaze attacks. The 88 LCS(L)s assigned to picket stations had two sunk and 11 damaged by kamikazes, while the 11 LSM(R)s had three sunk and two damaged. (3)
Okinawa Picket Stations
Note: Some picket diagrams show a 16th station located near station 12.
Understanding the basics of the situation at Okinawa, what can we learn that is applicable to today’s Navy? The foundation of any analysis is the recognition that the Kamikaze was the functional equivalent of a guided anti-ship missile. The guidance, obviously, was in the form of a human pilot and the ‘missile’ was very powerful, rivaling a modern guided missile in terms of destructive impact. This functional equivalency allows us to assess the attacks and defense in modern terms. Further, the dynamic of the Kamikaze and the picket ships gives us insight into the Navy’s plans regarding its unmanned picket/escort vessels.
As you recall from previous posts, the Navy intends to procure two types of unmanned vessels. One will be a small version which is intended to act as a picket for a larger group by providing surveillance and reconnaissance – much the same as the picket ships did at Okinawa. The second will be a somewhat larger vessel which is intended to stay with the main group and act as a missile barge.
So, what does the Okinawa Kamikaze and picket ship scenario tell us about the Navy’s plans for its unmanned vessels today? There are several lessons, factors, and considerations for us.
Lethality – We need to recognize that today’s anti-ship missile will be every bit as lethal, if not more so, than the Kamikazes. In fact, the situation is far worse today due to the complete absence of armor on modern ships. The Okinawa picket ships routinely absorbed multiple hits, kept fighting, and often survived. Does anyone seriously believe that a Burke, FFG(X), or LCS can take multiple hits and not sink? Astonishingly, one of those ships is actually designed to be abandoned at the first hit! A missile attack against our ships will be devastating and we need to factor that into our ship designs and cost and we need to accept that naval battles will involve a significant degree of attrition. The lethality will absolutely stun us.
Saturation – The Kamikaze was used as a saturation attack with each of ten major attacks consisting of several hundred aircraft. This is a lesson we have completely forgotten. Peer warfare requires huge numbers of munitions – dwarfing any estimates we may have. This was demonstrated time and again in WWII and Korea where munition expenditures far exceeded predictions. We’ve become so used to the small Tomahawk strikes against unresisting targets that we’ve come to believe that peer warfare will involve the same minimal usage of weapons. Nothing could be further from the truth. We’ll see unimaginably massive expenditures of weapons against us and unbelievable salvos launched against our fleets. We absolutely must come to terms with this reality because it drives our ship design sensor and weapons density, fire control capacities, sensor design, armor considerations, etc. Ships with one CIWS will not survive saturation attacks. We must heavily arm our ships – far beyond anything imagined by today’s designers.
Defensive Guidance/Sensors - Defending ships at Okinawa did not possess any weapon guidance comparable to the Kamikaze pilots and this put them at a huge disadvantage. The defending weapons were optically (and radar, to a degree) aimed and fuzed to a marginally successful degree. The mismatch in technology between the advanced [human] guidance of the Kamikazes and the unguided defensive weapons mimics and demonstrates the consequences of a loss of defensive sensors and fire control in modern engagements. Given the very limited number of sensors and fire controls on modern ships, it is all too easy to imagine a ship being blinded early in an engagement and being unable to continue fighting even though the weapons, themselves, might still be available. We desperately need to increase the number of sensors (redundancy) and types of sensors on our ships. For example, we should have much more extensive, physically distributed EO/IR sensors tied into the fire control system as well as a separate, technologically dissimilar type of radar as a backup to the Aegis arrays. The Aegis arrays are large, exposed targets and likely to be seriously damaged and degraded from almost any hit. Consider the Burke destroyer that was involved in the collision with a commercial ship. One of its radar arrays was, apparently, rendered completely inoperative and that was from a waterline collision, not a missile hit. In like fashion, the Port Royal’s arrays were reportedly rendered inoperative when it gently nosed aground off Pearl Harbor. That doesn’t bode well for the combat resilience of the Aegis system. We need sensor redundancy and backups.
Armament – The use of picket ships mimics the Navy’s desire for advanced screens of unmanned vessels. The pickets succeeded in their mission but were devastated – what does that suggest for today’s unmanned vessel screens? The picket ships were heavily armed and armored but were still devastated. The Navy, in contrast, envisions the unmanned escorts being unarmed. They’ll be quickly eliminated in any combat which will transfer the burden of their functions back to the manned escorts who won’t be trained or proficient at the functions and certainly won’t be properly positioned.
Armor – The Okinawa picket ships were all armored to varying degrees. Again, the Navy envisions completely unarmored, unmanned vessels as pickets. Not only will the unmanned vessels be quickly eliminated but the absence of armor ensures that weapon expenditure by the enemy to do so will be absolutely minimal. One of the major benefits of the Okinawa pickets was that they soaked up so many of the Kamikazes. Imagine if each picket had instantly sunk from a single hit. The remaining Kamikazes would have been able to continue on to the amphibious ships, the true targets of the Kamikazes, instead of being wasted against the pickets.
Weapon Density – The number of weapons on the picket ships was incredible and that redundancy allowed the pickets to keep fighting even after taking multiple hits. The USS Aaron Ward is an outstanding example of a ship that was able to keep firing despite taking half a dozen or so Kamikaze strikes and a couple of bomb hits. Even the ships that would make up today’s core group have very limited numbers of defensive weapons. While the VLS numbers are large, and quad packing makes the missile inventory numbers even larger, we’ve shown that the number of VLS weapons that are actually usable in an engagement is limited to around four. Beyond that, the number of close in weapons is nearly non-existent. Burkes have a single CIWS. Many ships have a single RAM/SeaRAM. We need to greatly increase the number of defensive weapons installed on our ships.
Picket Spacing – One of the aspects that jumps out from looking at the diagrams of the picket locations is the distances involved. For those of us who have grown up looking at Navy PR photos of ships sailing side by side, the idea of spacing is foreign to us. The Okinawa pickets were located 20-80 miles (mostly 50-80 miles) from the center of the defended area. Translating that to modern terms is difficult but one way to sort of get a handle on it is to compare the Okinawa distancing to the speed of the incoming attackers. Obviously, the faster the attacker, the farther out the picket has to be located in order to provide sufficient warning. At Okinawa, the pickets were, generally, 50-80 miles from the center point of the defended area which puts them at distance equivalent to 25% - 40% of the attacking aircraft’s speed (assuming 200 mph). For a modern high subsonic (assume 500 mph), anti-ship missile that would, proportionally, place pickets at 125 miles – 200 miles. That seems unbelievable to us, today, but facing supersonic or high subsonic missiles, those are the kinds of distances required to provide sufficient early warning and engagement.
If the Navy intends, as they say, to place unmanned vessels as escort pickets for the main groups, the pickets will need to be 50-200 miles out which places them well beyond any AAW support from the core group. As we stated earlier, being unarmed and unarmored, they’ll die quickly and easily. I’m pretty sure the Navy hasn’t thought this through.
The Okinawa picket stations were positioned close enough to allow continuous tracking of attacking aircraft but were too far apart to provide mutual gun support. Given today’s longer ranged anti-air missiles, mutual support may be possible but only if many, many more pickets are used due to the greater required distancing from the escorted group and only if the pickets are armed. Again, this reminds us that we’ve completely forgotten just how many ships are required to form a survivable group. We’ve grown up seeing a carrier escorted by three ships when the combat-reality is that we will need 30+ ships and that’s before we factor in any distant picket requirements.
Countermeasures - Japan did attempt radar countermeasures, employing chaff and radar reflective kites, though with limited success. Today, sophisticated radar countermeasures would, undoubtedly, be employed and would greatly decrease the effectiveness of radar pickets.
Expendability - It was understood that the pickets would be spotted and attacked. Recommendations were made that the picket ships be the smallest possible ship that could perform the function so as to make losses ‘acceptable’.(4) This is a concern for us, today, given that our smallest surface ship is the multi-billion dollar Burke. Even the future frigate is a billion-plus dollar ship and cannot be considered expendable. The Navy’s vision of small unmanned vessels may be appropriate in terms of cost, if they can resist the temptation to gold plate them.
Future naval warfare will, without a doubt, feature massive, saturation missile attacks and the US Navy has not devoted any attention to the problem. The Chinese Type 055 destroyer/cruiser, for example, has 112 VLS cells that can be loaded with anti-ship missiles. Okinawa offers historical lessons that we can apply to our defensive efforts. The Navy plans to employ unmanned picket vessels to accompany and escort carriers and surface groups but the pickets are going to be unarmed and unarmored. A peer enemy will have hundreds or thousands of missiles available for attacks and unarmed/unarmored pickets won’t stand a chance and will be quickly eliminated leaving the core group with no early warning and no early engagement.
The Okinawa pickets provided early warning but also early engagement and fighter direction assistance. In other words, the pickets were not just passive observers, they were active combatants and, as such, managed to tie up many of Kamikaze aircraft that penetrated the CAP screen. We need to give serious thought to reconfiguring our pickets beyond their purely passive sensing role and make them combatants. That requires arming them with short/medium AAW weapons and building them with an appropriate degree of armor. The Okinawa pickets clearly demonstrated the value of armor.
Given the relatively small number of kamikaze aircraft that penetrated the CAP, the damage and destruction they wrought was stunning and modern anti-ship missiles are likely to be even more destructive given the unarmored and weakly built ships that make up today’s fleet. We need to alter our ship design philosophy and start designing ships for combat, not peacetime cruises.
The Okinawa example pointed up the need for massive numbers of ships to stand up to high end saturation attacks and to compensate for sunk and damaged ships. Okinawa, alone, involved over 600 ships, not counting hundreds of additional, lesser craft such as LCIs, LSMs, LSTs, LSVs, etc. This one operation used 2-3 times more ships than the entire current US Navy. We’ve forgotten what is required to wage high end war.
Frankly, the Navy’s vision of unmanned, unarmed, unarmored picket/escort ships is ludicrous and combat-useless. They’ll be instantly eliminated in any attack without accomplishing anything. Only if we can make them powerful enough and tough enough to survive long enough to accomplish their purpose will they be combat-useful. However, this requires a complete rethink of the entire concept. Unfortunately, just like the LCS, the Navy has already committed to the design and acquisition of a fleet of unmanned vessels without any understanding of their capabilities and vulnerabilities. As with the LCS, we’re committed to buying a fleet of worthless vessels. Is the Navy truly incapable of learning from their mistakes? It would seem so.
Related side note:
The radar picket system was established to provide early warning and early defense against the kamikaze saturation attacks. The ultimate development of radar picket ships was the high speed, nuclear powered submarine USS Triton which could perform picket duty and dive when threatened. The obvious problem with this tactic is that the pickets could be kept underwater and ‘mission killed’ by a single aircraft. In addition, a submarine has no anti-air capability and cannot engage the attack, only warn of its approach.
(2)History News Network website, “Kamikazes at the Battle of Okinawa”, Joseph Wheelan, 6-Mar-2020,
(4)Naval History and Heritage Command website, “Battle Experience Radar Pickets and Methods of Combating Suicide Attacks Off Okinawa”,