Jeffrey LewisCan CTM Bust Russian Silos?

The Russians have made a lot of noise about the possibility that advanced conventional weapons in the US arsenal place their nuclear forces at risk and, as a corollary, that certain capabilities should be included in future arms control negotiations.

Which raises an interesting question — do they? Can the proposed Conventional Trident Modification (CTM) program — the program to put a conventional warhead on a D5 submarine-launched ballistic missile — or its likely follow-ons, bust Russian silos?

I should start by noting that CTM, as proposed by the Navy had little or no capability against hard and deeply buried targets.

However, one of the little noticed aspect of the National Academies report on U.S. Conventional Prompt Global Strike is that the Committee invented a hard-target kill capability for the Conventional Trident Modification (CTM) program — a “committee-proposed additional CPGS option” that would have have the space for an “earth-penetrator munition weighing on the order of 1,000 lb” that could “attack small, hardened buried targets…” They called this little devil the CTM-2.

This surely has to be a first in the history of the National Academies.

So, could CTM-2 bust Russian silos?

Keeping in mind that this is a paper-weapon, as it turns out the Committee on Conventional Prompt Global Strike was not the first set of smarty-pants to think about arming an SLBM with a conventional penetrator. As the slide atop this paper demonstrates, Lockheed Martin’s Nancy F. Swinford and Dean A. Kudlick were were doing similar work in the mid-1990s.

Dennis Gormley found the Swinford and Kudlick paper, and then used it to assess whether a hard-target CTM (or similar capability) could hold at risk Russian silos, in his new paper, The Path to Deep Nuclear Reductions: Dealing with American Conventional Superiority:

Tomahawk cruise missiles are surely accurate enough to hit on or very near to a Russian missile silo, but their warhead carries only 450kg of either blast fragmentation or combined-effects submunitions. The former is a mere pinprick vis-à-vis hardened missile silos; the latter is only relevant against soft targets. Indeed, even a Trident missile armed with a conventional penetrator would require Herculean accuracy and absolutely perfect targeting conditions to have any chance whatsoever of threatening silo-based missiles.76

76 Russian concrete silo covers are dome-shaped and approximately 20 feet in diameter and 5 feet high in the center. This means that they have a radius of curvature of about 12.5 feet. Employing the targeting requirement of approaching the target at less than 2 degrees from the vertical, the penetrator would have to impact less than 5 inches from the absolute center of the silo cover, or within a 10-inch diameter circle whose center is at the apex of the dome. My thanks to Dr. Gregory DeSantis, a former U.S. Department of Defense scientist, for making these calculations based on the penetrator design discussed in Nancy F. Swinford and Dean A. Kudlick, “A Hard and Deeply Buried Target Defeat Concept”, op. cit.

As you can see, Dennis is very, very skeptical that a conventional weapon will achieve the accuracy necessary for busting silos.

But perhaps this is the sort of thing we might usefully crowd-source.

The Swinford and Kudlick paper — “A Hard and Deepl Buried Target Defeat Concept”, Lockheed Martin Missiles & Space, Sunnyvale, CA 94088, Defense Technical Information Center document no. 19961213 060, January 1996, — is online and unbelievable. Take a look.

And kudos to Dennis for digging out a hard and deeply buried bit of paper!


  1. Daryl Press (History)


    Interesting stuff. I look forward to reading the original report.

    Before people begin to weigh in on “whether a conventional weapon will achieve the accuracy necessary for busting silos” we need to take a step back and ask ourselves about the scenarios — as the scenario will drive the accuracy requirements.

    a) how many silos are we trying to destroy? 1,000 Russian, or 20 Chinese? For any given level of desired target destruction (e.g, 90%), the required pK of the system goes up rapidly as the number of targets increases. Therefore, it’s possible that we can get good enough accuracy with conventional Trident to destroy 20 hardened Chinese silos but not 1,000 hardened Russian ones. Or perhaps neither. But this is a key question before one can talk about accuracy requirements.

    b) how many conventional warheads do we have per silo? Obviously, the more warheads per silo, the lower pK you need for each warhead to achieve any given level of success.

    c) what is the required level of target destruction? If this is a preemptive strike on, say, China, you might need 95% (or greater) of destroying them all (which requires 99%+ per target). On the other hand, if North Korea has fired a nuclear missile at Japan, we might not require anything close to 95% of getting them all to launch a conventional trident strike — or a nuclear counterforce strike — because reducing the number of remaining NK nuclear-armed missiles would presumably be seen as vital, even if we couldn’t get them all.

    This isn’t a criticism of the question you raise AT ALL. I’m just hoping to frame the debate that I hope comes in response to your post.

    For what it’s worth, Keir Lieber and I have a Foreign Affairs article coming out in about two weeks that, among other things, argues that even highly-precise conventional warheads are unlikely to be sufficiently effective against hardened silos to obviate the need for nuclear counterforce capabilities.

    — Daryl

  2. Pavel (History)

    Eugene Miasnikov wrote a paper on counterforce capabilities of U.S. conventional weapons back in 2000. It is available in Russian, but he may have an English version as well.

  3. Andrew Tubbiolo (History)

    I would believe this were a real issue if we saw a global real time surveillance capability to act as a means of verification of a hit or a miss. Could the newest version of DSP verify placement of a KE silo/mobile buseter? My background in CCD cameras and astronomical observations tells me that DSP could tell which silo a missile was launched from, but not confirm a hit or miss by a conventional munition. I’ve been playing with the requirements for a system that would allow the US, to monitor real time the position of mobile launchers to allow for conventional first strike against mobile launch platforms. So when I heard about the Chinese mobile ballistic anti-shipping system I thought this might push the US toward the capability of global mobile missile monitoring and strike. My, my, my, conventional weapons might start to drive nuclear systems in a more direct manner again.


  4. Greenish

    Reading the Swinford & Kudlick paper is, I agree, fascinating… never read this stuff publicly before.

    But I’m pretty sure they cite the AoO (Angle of Obliquity [to the target perpendicular]) limit as 20 degrees, in fact using a 70-degree impact angle as an example on pg 4. The two degrees is AoA (angle of attack, or alpha in the aero literature) which, as I’m sure you know, is the angle between the axis of the reentry body and its flight path. This much smaller angle is what causes maneuvering in response to control inputs.

  5. Azr@el (History)

    A properly designed 1-2 ton unitary KEP could punch thru a Russian silo, not thru the cap but sideways digging under the ground, but that’s besides the point. The second a trident flight is detected, may be a awhile given the shabby disrepair their radars are in, the Ruskies may retaliate; disproportionately. I know the Navy and Air Force majors and commanders bucking for meritorious promotions are trying to get us all to drink the kool-aid and believe that giving the Russians advance warning will cool any chance of them viewing this as a threat or convincing us that the Russians won’t view the launch of 2-4 tridents as anything to make much ado about. Of course this is insane; the Russians would view 2-4 tridents, potentially carrying 16 to 32 half megaton w88s capable of decapitating their C&C and blinding their sensors with EMP as part of a larger unconventional campaign. This may give them the urge to launch while they can. Remember the Russians have been on the wrong side of strategic surprise quite a bit the last century or so and are little bit paranoid about getting caught with their trousers down again.

  6. Carey Sublette

    Even if the system would be ineffective in practice it helps to focus on the right objections to it.

    Criticizing the accuracy problem may be a weak reed. Is striking within a 10 inch circle really out of the question?

    Certainly an INS/GPS only navigation system, which would be reasonable for a general-purpose hard-target kill system, would seem incapable of such extreme accuracy. But would it be implausible if the system were designed to strike Russian silos and had a terminal guidance package?

    The desired terminal velocity of 4000 FPS would provide 1 second of acquisition and maneuver time from an altitude of 4000 feet. Certainly an imaging sensor of some kind should be able to find the center of a 20 feet dome with high precision from that distance. Whether there are practical limits to terminal maneuvering that might limit achievable accuracy I can’t say.

    But a crucial issue not addressed by either paper is whether the penetrator might actually penetrate the silo dome once it struck.

    Even if the answer is “yes” in the basic situation the usual anti-armor-penetrator strategies can be applied to protect the silo: e.g. a sloped stand-off plate above the center of the silo that will fracture the penetrator.

    I find it hard to imagine that a Trident-carriable penetrator can be designed that would defeat all countermeasures that could be installed on a fixed high value point target at comparatively modest cost.

  7. Yale Simkin (History)

    I think DeSantis completely misread the angle requirements for the impactor.

    It is up to 20 degrees, not 2 degrees. The 2 degrees is for how lined up the impactor must be with it’s flight path.

    The conclusion that it must hit some tiny spot dead center and within 2% of vertical is completely incorrect.

  8. MarkoB

    I notice that Gormley states that the Tomahawk is more problematical than MM3 or Trident from the perspective of conventional counterforce.

    There is a good report on this in Russian by Eugene Miasnikov I think; that should be considered.

    But anyway, I notice two features of the Tomahawk that Gormley see’s as being potentially useful are its retargeting and loitering capacity; that doesn’t apply for ballistic RV’s. But, could that be false?

    You need to consider the TSV re-entry vehicle which is envisaged, on my understanding, to have both a loitering and a retargeting capacity. The idea is to enable RV’s to hit targets behind mountains and road mobile targets. A conventionally armed Trident armed with a TSV reentry vehicle (GPS accuracy as well) might be a problem; the TSV also is supposed to have an earth penetrating capacity.

  9. Scott Monje (History)

    Here is an English translation of a 2008 paper by Miasnikov on the topic.

  10. Allen Thomson (History)

    > Eugene Miasnikov wrote a paper on counterforce capabilities of U.S. conventional weapons back in 2000. It is available in Russian, but he may have an English version as well.

    Thanks for that. I had to hunt for the right character encoding to make it come out right, but on my Vista system, it’s “Cyrillic(Windows-1251).”

  11. GR (History)

    Just running through the equations given in this paper for a 4” diameter, 30” long, ogive-nose tungsten penetrator [which should weigh about the same as a Mk 5 warhead leaving another ~800lbs for guidance and RB, given that the paper says that they would have to cut the number of warheads in half], I get 22 feet of penetration in 10,000 psi concrete…

  12. George William Herbert (History)

    The penetrator they proposed appears to be based on the aircraft bomb penetrators LM has designed.

    As an alternative, the KE penetrators off tank guns (M829 etc; ) are designed for high velocity impacts (1,600 to 1,800 m/s muzzle velocity, and they lose around 50 m/s/km of travel). They’ll go through reinforced concrete just fine.

    They don’t detonate per se – so a direct hit on the missile under the cap is necessary for a hard kill – but the penetrators only weigh 10-12 pounds. A 500 lb RV payload translates into 40 to 50 projectiles, and impact at 2 km/s (6,600 fps) is within the safe velocity range.

    You still need to target within about 3-4 meters of centerline, for the shotgun effect to probably drop one of the penetrators center on the missile. So accuracy requirements are pretty tight.

  13. George William Herbert (History)

    Also, on the conventional warhead front …

    A Tomhawk’s bunker/silo buster warhead solution properly is a custom engineered EFP warhead, a jumbo version of the downwards firing multiple warhead EFPs used in Javelin and the latest TOW missiles.

    A 20” notional diameter EFP can punch through about eight feet of RHA steel, and using the normal scaling something like thirty feet of reinforced concrete. You can cluster several smaller ones in the Tomhawk warhead bay and hit 3-4 ground hit points – as with the KE warheads, those don’t detonate in the silo, but if you punch down into the ICBM itself there’s a hard kill.

  14. Azr@el (History)

    “Russian concrete silo covers are dome-shaped and approximately 20 feet in diameter and 5 feet high in the center”

    Gormley places himself in a bit of a time warp with his reference to SS-4 dome shaped cupola style silos. I believe they’re using 120 ton flushed armoured doors last I checked.

    Daryl Press seems to be jumping on to this time travel bandwagon with his reference to a “1,000 hardened Russian ones”

    The current Russian count is less than 200 silo based ICBMs left, afterall it’s not 1991 anymore and even 1991 was only like 400 silos, less than half hardened.

  15. tbaum (History)

    Another question that needs addressing: is the purpose of the CTM to destroy the missile, to destroy the missile silo, or to prevent the missile from being used (perhaps because the door can’t be opened) for some period of time? I’d think that would impact the capabilities, too…

  16. Scott Monje (History)

    The link to the English version of the Miasnikov paper (above) doesn’t work because I put an extra / at the end. The site is

  17. Carey Sublette

    Thanks Azr@el, that’s a great document.

    Straightforward (though possibly too simplistic) application of anti-armor scaling laws suggests that 10 tons of armor per square meter would defeat the heaviest penetrator that could be delivered by U.S. RV. The square door pocket of the SS-18 silo is 12m across, and thus up-armoring to this level would take 1440 tons of steel. The door itself looks like it is 8 meters or so, and thus would weigh 640 tons (the actual existing door weighs 120 tons).

    So it looks like just a moderate amount of up-armoring of the existing door pocket structure on top of the silo, plus a much heavier door system, should defeat a scheme like this at low risk and comparatively moderate cost.

    On the other hand the comparatively slow terminal velocity of a kinetic energy penetrator (max 2 km/sec), plus the fact that the keep-out height is very low (it does not explode) suggests that ordinary air-defense missiles could just shoot these down before they become effective.

    It doesn’t look like the Russians have a lot to worry about.