Earth-Penetrating Weapons

Underlying Principles and other Technical Aspects

An earth-penetrating weapon (EPW) is designed to hit the earth at high speed and penetrate into the ground before exploding. Such weapons can be delivered by short-range missiles or aircraft, and are intended primarily to attack underground targets.
An EPW only burrows a few meters into the ground before it explodes. Indeed, the earth slows the warhead so quickly on impact that it cannot penetrate very deeply. Rather, by exploding just a few meters underground instead of at or above the surface, a much larger fraction of the energy of the explosion is transmitted to the ground. The explosion creates a strong seismic shock wave that propagates and can crush or damage an underground bunker.  Even a short penetration distance accomplishes this goal of "coupling" the energy of the explosion to the ground: penetration of a few meters increases the underground destructive effects by more than a factor of twenty for a wide range of warhead yields.
For example, exploding a 10-kiloton nuclear weapon at a depth of one meter would increase the effective yield by a factor of 20, resulting in underground damage equivalent to that of a 200-kiloton weapon exploded at the surface of the ground. But increasing the penetration depth to five meters would only increase the effective yield by an additional 60%, to 320 kilotons.

Nevertheless, even nuclear weapons have limited effectiveness at destroying the deepest or widely separated underground bunkers. For example, an earth penetrating weapon using the 1.2 megaton B83 warhead—the highest yield weapon in the U.S. nuclear stockpile—could crush underground bunkers to a depth of about 1000 feet. Deeper bunkers can be constructed with modern tunneling equipment, and are essentially invulnerable to nuclear attack.

A nuclear EPW would also likely be ineffective against underground bunkers containing chemical or biological weapons. In fact, the explosion could release active agents into the environment rather than destroy or sterilize them. These lethal agents could kill thousands of unprotected civilian or military personnel in a large area downwind—in addition to expected casualties from radioactive fallout, which could number in the millions. 

Radioactive Fallout from a Nuclear EPW
A commonly held fallacy is that an earth-penetrating nuclear weapon can penetrate deep enough so that the nuclear explosion and radiation is contained underground. This is not the case.

Penetration Depth
While the penetration depth increases with higher impact velocity, the weapon casing will be crushed—destroying the warhead inside—if it strikes the ground at too high a speed.  Empirical and theoretical data show that the maximum impact velocity is roughly one kilometer per second and the maximum achievable penetration depth of such a projectile in concrete is roughly 10-20 feet.3  In practice, the weapon must impact at lower velocity to reduce the deceleration experienced by the warhead, resulting in shorter penetration depths.  Penetration depths will be larger in dry soil than concrete or rock, but one would have to expect that a hardened target would be placed below hard rock or concrete.

Containment Depth
The depth at which even a small nuclear weapon must be buried to ensure that it is "contained" —that is, that no radiation is released when it explodes—is much greater than the achievable penetration depth, so that it is impossible to prevent radioactive fallout from a nuclear EPW.

For example, the minimum burial depth to ensure containment at the US Nevada Test Site was empirically determined, and is roughly 100 meters for a one-kiloton explosion and 500 meters for a 100-kiloton explosion.4  In dry hard rock, the required depths would be roughly 60 meters for a one-kiloton explosion and 300 meters for a 100-kiloton explosion.5   These depths are far greater than what can be achieved by an earth penetrator.

Moreover, even if sufficient penetration depths could be achieved, the explosion would still not be contained since penetrating the earth (rather than burying an explosive) creates an open shaft through which radiation would leak to the surface.

Since an EPW will not penetrate enough to be fully contained, it will produce a surface crater when it explodes, and large amounts of radioactive dust and debris from the crater will be ejected into the air and surrounding region. The size of the crater—and the amount of material ejected—will depend on the local ground properties, the depth of the explosion, and the yield of the weapon. The level of fallout will also depend on the local weather conditions, such as wind and rain.

Collateral Damage
Because using a nuclear EPW will necessarily result in radioactive fallout, people in the surrounding areas will be killed or have an increased risk of cancer. The size of this area will depend on wind conditions and the size of the nuclear weapon, and the number of people affected will depend on the number and population of cities within the downwind region.  However, the number of deaths could exceed a million, and the number of people with increased cancer risks could exceed 10 million.

For example, the new nuclear earth penetrator that the United States plans to research would use a 1.2-megaton weapon. According to a simulation using software developed for the Pentagon, if one of these weapons were used against the underground nuclear facility in Esfahan, Iran, 3 million people would be killed by radiation within 2 weeks of the explosion, and 35 million people in Afghanistan, Pakistan and India would be exposed to increased levels of cancer-causing radiation.6

Attacks against Underground Chemical or Biological Storage Bunkers


Even a nuclear-armed earth penetrator would be unlikely to destroy buried stockpiles of chemical or biological agents. Despite the extremely high temperatures and radiation levels reached very near a nuclear explosion, to destroy or neutralize these agents the warhead would have to detonate very close to the actual containers—nearly inside the same underground room in which canisters of biological or chemical agents were stored. This is highly unlikely given that in most cases the bunker location and underground geometry would not be known with any precision. It would be unlikely in any case that all chemical or biological agents in an underground complex would be stored within a single room.

The size of the crater produced would be much larger than the area in which the agents would be destroyed, and would result in the venting and dispersal into the atmosphere of any undestroyed agents inside the crater zone. Simply blowing up a bunker filled with chemical or biological agents—even using a nuclear weapon—may thus have the undesirable effect of dispersing the agents, rather than destroying them.

If dangerous material were already stored deep underground, the most sensible strategy would be to make sure it stays there using conventional means to seal all entrances and exits to the facility and keep them sealed until the territory can be captured and the agents carefully neutralized.

Current U.S. EPW Capabilities

The United States currently deploys both conventional and nuclear earth-penetrating weapons. The two largest conventional EPWs (called GBU-28 and GBU-37) use the same body but different guidance systems.8  The bodies are long tubes, a third of a meter (14.5 inches) in diameter and four meters (12.8 feet) long, that weigh over two tons and contain nearly 300 kilograms (630 pounds) of high explosive. Both are dropped from aircraft, and tests have shown they can penetrate six meters of concrete or 30 meters of earth. The GBU-28 is laser guided and the GBU-37 is guided by the Global Positioning System, which reportedly makes it more accurate than the GBU-28 and allows it to operate under all weather conditions. Very high accuracy increases the ability of these weapons to destroy shallow hardened targets with known locations (such as missile silos) but not deeply buried targets.9

The United States reportedly stockpiles about fifty B61-11 nuclear EPWs, which—like the conventional alternatives—are dropped from aircraft. This bomb entered the stockpile in 1997, and replaced the 9-megaton B53. It is a modification of the B61-7 bomb with a new casing,10  and reportedly has a variable yield, from 0.3 to 340 kilotons.11  It is shorter than the GBU-28, with a length of 3.6 meters, and has only a quarter of the mass.12  It reportedly can penetrate two to three meters in frozen soil.13 

For a penetration depth of three meters and a yield of 0.3 kilotons, the B61-11 could destroy a target buried under roughly 15 meters of hard rock or concrete. For the same penetration depth and the maximum yield of 340 kilotons, the destruction depth would be roughly 70 meters for a hardened target.14

In the 1980s the United States developed and tested but did not deploy a nuclear EPW intended for the intermediate-range Pershing II missile. This warhead, called the W86, was also a modification of the B61.15  It was about two meters long and 0.2 meters in diameter, and could penetrate less than 10 meters of granite or hardened concrete.16  The main mission of this weapon was to crater runways rather than attack buried targets.

Plans for New Nuclear Earth-Penetrating Weapons

In principle, a new nuclear EPW that has a greater capability to destroy deep, hardened targets than does the B61-11 could be achieved by increasing the penetration depth and/or the explosive yield.  However, while a new penetrator design could increase the penetration depth somewhat, the above discussion shows that it is very unlikely that the penetration depth could be increased enough to result in a significant increase in the destructive capability of the weapon. Instead, increasing destructive capability significantly would require increasing the yield of the nuclear warhead. This appears to be what the Bush administration has in mind.  

According to Senate testimony by Everett Beckner, the Deputy Administrator for Defense Programs of the National Nuclear Security Administration, the proposed study of the new Robust Nuclear Earth Penetrator (RNEP) would include modifying two existing nuclear weapons to allow them to attack buried targets—the B61 (which is already the basis for the B61-11 earth penetrator), but also the B83 air-dropped bomb.17  The B83 warhead is currently the largest warhead the United States deploys, reportedly having a yield of one to two megatons—much larger than the maximum yield of the current B61-11 EPW.18

Besides developing new cases for these warheads, other possible modifications include improved guidance systems for higher accuracy (which would increase the destructive capability against shallow, but not deep, hardened targets), better attitude control at impact (which would help ensure that the penetrator hits the ground at the right angle so it is not simply deflected from the earth’s surface), and smarter fuzes (which would increase control over where the weapon explodes and make sure that it is not detonated by hitting structures above ground). 

The B61-11 EPW was produced without nuclear testing, since it was essentially a repackaging of an existing warhead. Similarly, it appears that other options currently being considered would use existing warheads and would therefore not require nuclear testing.

Political Implications of a New Nuclear EPW

A US decision to develop new nuclear earth-penetrating weapons would have several negative political implications internationally. First, such weapons are explicitly designed to be more "usable" and to be used in what would otherwise be a non-nuclear conflict. As a result, they blur the line between conventional and nuclear weapons and lower the threshold for nuclear use. Second, by contravening US pledges under the Nuclear Non-Proliferation Treaty (NPT) not to target non-nuclear weapon states with nuclear weapons, such weapons undermine the non-proliferation regime.

Lowering the Threshold for Nuclear Use
The Bush administration's Nuclear Posture Review (NPR), which was completed in 2002, explicitly calls for US nuclear weapons to deter and respond to a "wide range of threats," including attacks by conventional, chemical, or biological weapons as well as "surprising military developments."19  In keeping with its stated interest in increasing the roles for US nuclear weapons, the NPR also calls for "improved earth penetrating weapons (EPWs) to counter the increased use by potential adversaries of hardened and deeply buried facilities."  While the United States has never forsworn the first use of nuclear weapons, the Bush NPR carries this policy further and makes it more explicit. The proposed study of the new Robust Nuclear Earth Penetrator would be part of this effort to increase US nuclear attack options and capabilities.

However, this policy is counter to US and international security interests. Maintaining and strengthening the firebreak against the use of nuclear weapons by all countries should be a paramount concern for US national security. Thus, the sole purpose for US nuclear weapons should be to deter the use of nuclear weapons and, if necessary, to respond to nuclear attacks. The additional roles for nuclear weapons called for by the 2002 Nuclear PostureReview undermine the overriding goal of preventing the proliferation and use of nuclear weapons. If the United States, with unquestioned conventional superiority, chooses to rely on nuclear weapons, then weaker states—particularly those not covered by US security guarantees—would apparently have a far greater need for nuclear weapons. Ultimately, this policy of first use will encourage the proliferation of nuclear weapons.

Breaking U.S. Pledges under the NPT
The Bush NPR explicitly calls for targeting nuclear weapons against several non-nuclear weapon state signatories to the NPT, which is contrary to previous US pledges to not do so.  Such pledges were made by all the nuclear weapon state signatories to the NPT as an incentive for other countries to renounce nuclear weapons. By developing new nuclear weapons expressly to target non-nuclear weapon states, the United States would undermine the continued viability of the NPT. In short, the security costs of developing such weapons outweigh any conceivable security benefits.


1.  Lisbeth Gronlund and David Wright are Co-Directors of the Global Security Program and Senior Staff Scientists at the Union of Concerned Scientists and Research Scientists at the MIT Security Studies Program.

2.  Robert Nelson is a Senior Staff Scientist at the Union of Concerned Scientists and a visiting Research Scientist at the Princeton University Program on Science and Global Security.

3.   R. W. Nelson, "Low-Yield Earth-Penetrating Nuclear Weapons," Science and Global Security, 2002, 10, p. 1-20.

4.  Nelson, "Low Yield."

5.  Glasstone and Dolan, The Effects of Nuclear Weapons, 3rd Edition (Washington, DC: U.S. Government Printing Office, 1977), p. 255, Figure 6.72a.

6.  Peter Wilk MD, et al., "Projected Casualties Among U.S. Military Personnel and Civilian Populations from the Use of Nuclear Weapons Against Hard and Deeply Buried Targets," 2005, Physicians for Social Responsibility. Available at:
, accessed May 2005.

7.  R. W. Nelson, "Nuclear 'Bunker Busters' Would More Likely Disperse Than Destroy Buried Stockpiles of Biological and Chemical Agents," Science and Global Security, 2004, 12, p.69-89.

8.  See and

9.  The United States has or is developing a number of other conventional penetrators for delivery by aircraft and cruise missiles. See, for example, "Report to Congress on the Defeat of Hard and Deeply Buried Targets," July 2001, pp. 16-18, available at

10.  For information on the modifications made to the B61, see the statements by then Sandia Laboratory Vice President for National Security Programs Roger Hagengruber in "B61-11 a Modification, Not a New Weapon," Sandia Lab News, 1 August 1997, available at

11.  Greg Mello, "New Bomb, No Mission," Bulletin of the Atomic Scientists, May/June 1997, pp. 28-32.

12.  T. Cochran, W. Arkin, and M. Hoenig, Nuclear Weapons Databook, Volume I: U.S. Nuclear Forces and Capabilities (Cambridge, MA: Ballinger, 1984), p. 199, Table 7.1; Mello, "New Bomb."

13.  Nelson, "Low Yield."

14.  Glasstone and Dolan, Effects, p. 256, Figure 6.72b. This table gives values for the apparent crater depth of a nuclear explosion. We assume a target could be destroyed to a depth 50% greater than this.

15.  Mello, "New Bomb."

16. Bob Peurifoy, "For-the-Record Statement Concerning Nuclear Weapon Earth Penetrators," 15 May 2002.

17.  Philipp C. Bleek, "Energy Department to Study Modifying Nuclear Weapons," Arms Control Today, April 2002.

18.  Chuck Hansen, U.S. Nuclear Weapons: The Secret Story (New York: Orion, 1988), p. 171.

19.  The NPR is classified, but portions of it were leaked and are available at:

Last revised date: June 6, 2005

We Need Your Support
to Make Change Happen

We can increase global security and take U.S. nuclear weapons off hair-trigger alert—but not without you. Your generous support helps develop science-based solutions for a healthy, safe, and sustainable future.