The New Era of Counterforce - IA de Investigación especializada en el MIT

The New Era of Counterforce

The New Era of
Counterforce
Technological Change and the Future of
Nuclear Deterrence

Keir A. Lieber and
Daryl G. Prensa

Nuclear deterrence is
based on the threat of retaliation. A nuclear arsenal designed for deterrence
debe, por lo tanto, be able to survive an enemy ªrst strike and still inºict unac-
ceptable damage on the attacker. For most of the nuclear age, the survivability
of retaliatory forces seemed straightforward; “counterforce” attacks—those
aimed at disarming the enemy’s nuclear forces—appeared impossible because
the superpower arsenals were large and dispersed, and were considered easy
to hide and protect.1 Today, analysts tend to worry more about the dangers
of nuclear terrorism or accidents than the survivability of retaliatory arsenals.2
Nuclear deterrence appears robust.

Changes in technology, sin embargo, are eroding the foundation of nuclear de-
terrence. Rooted in the computer revolution, these advances are making nu-

Keir A. Lieber is Associate Professor in the Edmund A. Walsh School of Foreign Service and the Department
of Government at Georgetown University. Daryl G. Press is Associate Professor in the Department of Gov-
ernment at Dartmouth College.

Many more colleagues gave helpful advice and comments on earlier versions of this article than
we can list here. We especially thank Bruce Bennett, Richard Betts, Paul Davis, Geoffrey Forden,
Charles Glaser, Brendan Rittenhouse Verde, Kyudok Hong, Hunter Hustus, Chaim Kaufmann,
Scott LaFoy, Li Bin, Austin Long, John Mearsheimer, Vipin Narang, Barry Posen, Theodore Postol,
Brad Roberts, Sebastian Rosato, Joshua Rovner, Caitlin Talmadge, Tristan Volpe, and Wu Riqiang;
workshop participants at many research and government institutions; and the anonymous review-
ers. Jonathan Chipman provided invaluable assistance with the geospatial analysis. Los autores
are grateful to the Carnegie Corporation of New York for ªnancial support of this research.

1. See Robert Jervis, The Meaning of the Nuclear Revolution: Statecraft and the Prospect of Armageddon
(Ítaca, N.Y.: Prensa de la Universidad de Cornell, 1989), pag. 10; Spurgeon M. Keeny Jr. and Wolfgang K.H.
Panofsky, “Nuclear Weapons in the 1980s: MAD vs. NUTS,” Foreign Affairs, volumen. 60, No. 2 (Invierno
1981/82), páginas. 287–304; Harold Feiveson and Frank von Hippel, “The Freeze and the Counterforce
Carrera,” Physics Today, volumen. 36, No. 1 (Enero 1983), páginas. 36–49; Robert Jervis, The Illogic of American
Nuclear Strategy (Ítaca, N.Y.: Prensa de la Universidad de Cornell, 1984); Charles L. Glaser, “Why Do Strate-
gists Disagree about the Requirements of Strategic Nuclear Deterrence?” in Lynn Eden and Steven
mi. Molinero, editores., Nuclear Arguments: Understanding the Strategic Nuclear Arms and Arms Control De-
bates (Ítaca, N.Y.: Prensa de la Universidad de Cornell, 1989), páginas. 109–171, especially pp. 134–142; and Mi-
chael Salman, Kevin J. sullivan, and Stephen Van Evera, “Analysis or Propaganda? Measuring
American Strategic Nuclear Capability, 1969–88,” in Eden and Miller, Nuclear Arguments, páginas. 172–
263.
2. On accidents, see Eric Schlosser, Command and Control: Nuclear Weapons, the Damascus Accident,
and the Illusion of Safety (Nueva York: Pingüino, 2013); and Scott D. sagan, The Limits of Safety: Organi-
zaciones, Accidents, and Nuclear Weapons (Princeton, NUEVA JERSEY.: Prensa de la Universidad de Princeton, 1993). On nu-
clear terrorism, see Graham Allison, Nuclear Terrorism: The Ultimate Preventable Catastrophe (Nuevo
york: Owl, 2004); but also John Mueller, Atomic Obsession: Nuclear Alarmism from Hiroshima to Al-
Qaeda (Nueva York: prensa de la Universidad de Oxford, 2010); and Keir A. Lieber and Daryl G. Prensa, “Why
States Won’t Give Nuclear Weapons to Terrorists,” Seguridad Internacional, volumen. 38, No. 1 (Verano
2013), páginas. 80–104.

Seguridad Internacional, volumen. 41, No. 4 (Primavera 2017), páginas. 9–49, doi:10.1162/ISEC_a_00273
© 2017 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Seguridad Internacional 41:4 10

clear forces around the world far more vulnerable than before. De hecho, one of
the principal strategies that countries employ to protect their arsenals from de-
estructura, hardening, has already been largely negated by leaps in the accuracy
of nuclear delivery systems.3 A second pillar of survivability, concealment, es
being eroded by the revolution in remote sensing.4 The consequences of pin-
point accuracy and new sensing technologies are numerous, synergistic, and in
some cases nonintuitive. Tomados juntos, these developments are making the
task of securing nuclear arsenals against attack much more challenging.

To be clear, nuclear arsenals around the world are not becoming equally vul-
nerable to attack. Countries that have considerable resources can buck these
trends and keep their forces survivable, albeit with considerable cost and ef-
fuerte. Other countries, however—especially those facing wealthy, technologi-
cally advanced adversaries—will ªnd it increasingly difªcult to secure their
arsenals, as guidance systems, sensors, data processing, comunicación, arti-
ªcial intelligence, and a host of other products of the computer revolution
continue to improve.5

The growing vulnerability of nuclear forces sheds light on an enduring theo-
retical puzzle of the nuclear age. According to one of the leading theories
of geopolitics in the nuclear era, the “theory of the nuclear revolution,” nu-
clear weapons are the ultimate instruments of deterrence, protecting those
who possess them from invasion or other major attacks.6 Yet, if the theory is
correct—that is, if nuclear weapons solve countries’ most fundamental secu-
rity problems—why do nuclear-armed countries continue to perceive serious
threats from abroad and engage in intense security competition? Why have the
great powers of the nuclear era behaved in many ways like their predecessors
from previous centuries: by building alliances, engaging in arms races, com-
peting for relative gains, and seeking to control strategic territory—none of
which should matter much if nuclear weapons guarantee one’s security? Alabama-

3. “Hardening” here refers to the deployment of nuclear forces (such as delivery systems, guerra-
cabezas, and command sites) in reinforced structures that are difªcult to destroy.
4. “Concealment” here refers to efforts to prevent adversaries from identifying or locating one’s
forces (such as through the use of camouºage, decoys, and especially mobility).
5. For an earlier analysis of the consequences of technological trends in the U.S.-Russia case, ver
Keir A. Lieber and Daryl G. Prensa, “The End of MAD? The Nuclear Dimension of U.S. Primacy,"
Seguridad Internacional, volumen. 30, No. 4 (Primavera 2006), páginas. 7–44; and Keir A. Lieber and Daryl G. Prensa,
“The Rise of U.S. Nuclear Primacy,” Foreign Affairs, volumen. 85, No. 2 (March/April 2006), páginas. 42–54.
6. According to Robert Jervis, the deterrence implications of invulnerable nuclear arsenals are
“many and far-reaching”: war should not occur, crises will be rare, and the status quo will be rela-
tively easy to maintain. Jervis, The Meaning of the Nuclear Revolution, pag. 45. Two of the most impor-
tant works on the theory of the nuclear revolution are Kenneth N. Vals, “Nuclear Myths and
Political Realities,” American Political Science Review, volumen. 84, No. 3 (Septiembre 1990), páginas. 731–745;
and Jervis, The Meaning of the Nuclear Revolution.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

The New Era of Counterforce 11

though proponents of the theory of the nuclear revolution acknowledge this
anomalous behavior, they attribute it to misguided leaders, bureaucratic pa-
thologies, or dysfunctional domestic politics, not ºaws in the theory itself.7

Our analysis offers a simpler explanation for the disjuncture between the
theory of the nuclear revolution’s predictions and the foreign policy behavior
in the nuclear age because
of states: geopolitical rivalry remains logical
stalemate is reversible. For nuclear weapons to revolutionize international
politics—that is, to render countries fundamentally secure—the condition of
stalemate must be enduring. Arsenals that are survivable today, sin embargo, poder
become vulnerable in the future. Nuclear-armed states thus have good reason
to engage in intense competition, even if their own arsenals are currently se-
cure.8 Stated differently, nuclear weapons are the best tools of deterrence ever
creado, but the possibility of acquiring disarming strike capabilities—and the
fear that an opponent might do the same—explains why nuclear weapons
have not transformed international politics.9

The increasing vulnerability of nuclear forces also has several implications
for nuclear policy. Primero, if nuclear forces are becoming easier to attack, then all
else being equal, nuclear-armed states need to deploy more capable retaliatory
arsenals to counter the growing risks. Whether one believes that a deterrent
force must present potential attackers with “near-certain retaliation,” “likely
retaliation,” or some other level of risk, improvements in counterforce systems
require that retaliatory forces adapt—through better capabilities, increased
numbers, or both—to maintain the same level of deterrent threat. Además,
the rapid rate of change in counterforce technologies increases uncertainty
about adversaries’ future capabilities, suggesting that countries will need to re-
tain diverse retaliatory forces as a hedge against adversary breakthroughs.

7. Both Waltz and Jervis criticized the United States and the Soviet Union during the Cold War for
being excessively concerned with changes in the military balance of power; for building vast arse-
nals and seeking nuclear superiority; and for competing hard for allies and strategic territory in
the developing world. Ellos (y otros) attributed this behavior to the failure of policymakers to
appreciate the implications of the nuclear revolution. Jervis, Por ejemplo, titled his book The Illogic
of American Nuclear Strategy because, according to him, A NOSOTROS. policy “seeks to repeal the nuclear rev-
olution rather than coming to grips with [él].” Jervis, The Illogic of American Nuclear Strategy, pag. 147.
See also Waltz, “Nuclear Myths and Political Realities”; Charles L. Glaser, Analyzing Strategic Nu-
clear Policy (Princeton, NUEVA JERSEY.: Prensa de la Universidad de Princeton, 1990); and Stephen Van Evera, Causes of
Guerra: Power and the Roots of Conºict (Ítaca, N.Y.: Prensa de la Universidad de Cornell, 1999).
8. As John J. Mearsheimer noted about the Cold War competition: “The continuation of the arms
race was not misguided, even though nuclear superiority remained an elusive goal. De hecho, él
made good strategic sense for the United States and the Soviet Union to compete vigorously in the
nuclear realm, because military technology tends to develop rapidly and in unforeseen ways.”
mearsheimer, La tragedia de la política de las grandes potencias (Nueva York: W.W. norton, 2001), pag. 231.
9. We explore two other sources of the discrepancy between theoretical expectations and the for-
eign policies of nuclear-armed states in a forthcoming book manuscript.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Seguridad Internacional 41:4 12

Segundo, the increasing vulnerability of nuclear arsenals raises questions
about the wisdom of future nuclear arms reductions. Por décadas, engi-
neers have toiled to improve weapons accuracy and remote sensing capabili-
corbatas. Mientras tanto, arms negotiators have devised agreements to reduce nuclear
arsenals, with the consequence of reducing the number of targets an at-
tacker must destroy in a disarming strike. Either endeavor—improving weap-
ons or cutting stockpiles—can be defended as a policy for promoting strategic
stability, but taken together they are creating underrecognized vulnerabilities.
The danger of nuclear arms cuts is exacerbated by improvements in non-
nuclear means of attacking nuclear forces: Por ejemplo, through precision
conventional strike, missile defense, anti-submarine warfare (ASW), y
cyber operations.10

Tercero, the emergence of a new era of counterforce raises the question of
whether it is wise, for the United States in particular, to continue improving
nuclear and nonnuclear counterforce capabilities. Por un lado, mejorado
counterforce capabilities could be invaluable in a range of plausible scenar-
ios.11 Improved offensive capabilities could help the United States deter weak
countries from initiating conventional conºicts or from escalating in the midst
of war. Enhanced counterforce capabilities could also help protect U.S. forces,
allies, and the U.S. homeland from nuclear attack if a conventional war did es-
calate. Por otro lado, better counterforce could be a source of danger: no
only might improved disarming strike capabilities—in any country’s hands—
increase the temptation to attack, but also potential victims of disarming
strikes will seek to escape their vulnerability, thereby possibly triggering arms
racing and incentives to strike preemptively.12

Both views may be correct. The net beneªt of decisions to enhance counter-
force capabilities will therefore depend on the particular case. For countries
that perceive a highly malign threat environment, face aggressive nuclear-

10. Arms controllers may respond that one merely needs to restrict the development of advanced
counterforce systems, and then arms cuts will be safe again. The technological developments that
we describe below, sin embargo, especially in accuracy and remote sensing, are so integral to modern
conventional warfare that they will be difªcult to halt.
11. On the value of effective counterforce, see Keir A. Lieber and Daryl G. Prensa, “The Nukes We
Need: Preserving the American Deterrent,” Foreign Affairs, volumen. 88, No. 6 (November/December
2009), páginas. 39–51; and Keir A. Lieber and Daryl G. Prensa, “Coercive Nuclear Campaigns in the 21st
Century: Understanding Adversary Incentives and Options for Nuclear Escalation,” report num-
ber 2013-001 (Monterey, California: Project on Advanced Systems and Concepts for Countering
Weapons of Mass Destruction, A NOSOTROS. Naval Postgraduate School, Marzo 2013).
12. For a recent discussion of the dangers of ªrst-strike capabilities, see Charles L. Glaser and
Steve Fetter, “Should the United States Reject MAD? Damage Limitation and U.S. Nuclear Strat-
egy toward China,” Seguridad Internacional, volumen. 41, No. 1 (Verano 2016), páginas. 49–98, at pp. 52–53, 92–
97.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

The New Era of Counterforce 13

armed adversaries, or have ambitious foreign policy goals, the beneªts of de-
veloping advanced counterforce capabilities may outweigh the costs. For those
countries that face a benign environment and have more modest goals, cómo-
alguna vez, the secondary costs of enhancing counterforce may be too great. In any
caso, these contentious issues have not received sufªcient attention; analysts
and policymakers have largely overlooked the ways that rapidly changing
technologies are eroding the foundation of deterrence.

The remainder of this article is organized as follows. We ªrst discuss the key
role that arsenal survivability plays in nuclear deterrence theory. Segundo, nosotros
describe the main strategies that planners employ to ensure arsenal survivabil-
ity in practice. Próximo, we explore one of the major technological trends eroding
survivability, the great leap in weapons accuracy, and illustrate how improved
accuracy creates new possibilities for counterforce strikes. We then focus on
the second major trend, dramatic improvements in remote sensing, y
how the resulting increase in transparency threatens concealed and mobile nu-
clear forces. We conclude with a summary of our ªndings and their implica-
tions for international politics and U.S. national security.

Nuclear Survivability in Theory

At its core, nuclear deterrence theory rests on two simple propositions. Primero,
countries will not attack their adversaries if they expect the costs to exceed the
beneªts. Segundo, nuclear weapons allow countries, even relatively weak ones,
to inºict unprecedented levels of damage on those who attack them. Taken to-
juntos, these propositions suggest that nuclear weapons are the ultimate in-
struments of deterrence: no conceivable beneªt of attacking a nuclear-armed
state could be worth the cost of getting hit with nuclear weapons in retaliation.
As long as nuclear arsenals are survivable, eso es, able to withstand an en-
emy’s ªrst strike and retaliate, nuclear weapons are a tremendous force
for peace.

The theory of the nuclear revolution builds on the logic of deterrence theory
and extends its implications. Because nuclear weapons make countries funda-
mentally secure, countries can escape the most pernicious consequences of an-
archy. According to the theory of the nuclear revolution, once countries deploy
survivable arsenals they no longer need to fear conquest.13 As a result, ellos

13. The simple reason for this is that “military victory is not possible.” See Jervis, The Meaning of
the Nuclear Revolution, pag. 23. See also Kenneth N. Vals, The Spread of Nuclear Weapons: More May Be
Better, Adelphi Paper 171 (Londres: International Institute for Strategic Studies, 1981); and Van
Evera, Causes of War, páginas. 240–247.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Seguridad Internacional 41:4 14

can stop worrying about the relative balance of power;14 engaging in arms
races;15 or competing for alliance partners and strategic territory.16

Proponents of the theory of the nuclear revolution have always recognized
the discrepancy between their theory’s predictions and the actual behavior of
countries in the nuclear era. The Cold War competition between the United
States and the Soviet Union, En particular, is ªlled with empirical anomalies:
extensive arms racing, intense concerns about relative power gains and losses,
and competition for allies and control of strategic territory—all occurring at a
time when the main adversaries appeared to be invulnerable to disarming
strikes.17 World War III was averted, as nuclear deterrence theory would pre-

14. Adversaries no longer need to worry about shifts in the global economic balance of power or
the details of the military balance, because one side’s “gain” cannot undermine the fundamental
security of other nuclear-armed states. As Waltz writes, “Nuclear weapons can carry out their de-
terrent task no matter what other countries do.” See Waltz, “Nuclear Myths and Political Real-
ities," pag. 732. Elsewhere Waltz argues, “Nuclear weapons eliminate the thorny problems of
estimating the present and future strengths of competing states and of trying to anticipate their
strategies.” Kenneth N. Vals, “The Emerging Structure of International Politics,” International Se-
curity, volumen. 18, No. 2 (Caer 1993), páginas. 44–79, at p. 73. Similarmente, Stephen Van Evera writes that a con-
dition of mutual second-strike capabilities “clears the ‘fog of war’ by removing the relevance of
the clash of military machines. The calculus of relative capabilities drops from the calculus
of war.” See Van Evera, Causes of War, pag. 244. According to Robert Jervis, “If the arguments about
the nuclear revolution are correct, there should be only tenuous links between the details of the
military balance and political outcomes.” See Jervis, The Meaning of the Nuclear Revolution, pag. 42.
15. As Jervis explains, “When security comes from the absolute capability to annihilate one’s en-
emy, then each side can gain it simultaneously. Neither side need acquire more than a second-
strike capability and, if either does, the other need not respond since its security is not threatened.”
Robert Jervis, “Why Nuclear Superiority Doesn’t Matter,” Political Science Quarterly, volumen. 94, No. 4
(Invierno 1979/80), páginas. 617–633, at p. 618. According to Waltz, “Nuclear weapons make it possible
for states to escape the dynamics of arms racing.” See Waltz, “Nuclear Myths and Political Real-
ities," pag. 741. Because the ability to destroy all of an adversary’s nuclear forces in a ªrst strike is
“not technologically feasible,” states should not build counterforce weapons (or even defensive
sistemas) aimed at neutralizing each other’s retaliatory capability. See Jervis, The Meaning of the Nu-
clear Revolution, pag. 10. For Waltz, “Because thwarting a ªrst strike is easy, deterrent forces are quite
cheap to build and maintain.” See Scott D. Sagan and Kenneth N. Vals, The Spread of Nuclear
Weapons: A Debate Revisited (Nueva York: W.W. norton, 2003), pag. 30. According to their argument,
even conventional arms racing no longer makes sense once a state acquires a survivable nuclear
arsenal.
16. In the pre-nuclear era, shifting alliances could have great ramiªcations for national security.
When countries gain the security that comes with survivable arsenals, sin embargo, they no longer
need to pool resources. See Jervis, The Meaning of the Nuclear Revolution, páginas. 35–36. “Strategically,"
Waltz writes, “nuclear weapons make alliances obsolete.” Waltz, “The Emerging Structure of In-
ternational Politics," pag. 73. Similarmente, with a secure nuclear arsenal the gain or loss of territory has
little effect on the balance of power or national security. Nuclear weapons can be delivered across
vast distances, so nuclear-armed states no longer need to ªght over foreign bases, buffer zones, o
military resources derived from the possession of territory. As Van Evera writes, secure nuclear de-
terrents make “geographic assets less signiªcant.” See Van Evera, Causes of War, pag. 245. For other
discussions of the relative unimportance of territory in a nuclear world, ver Klaus Knorr, Sobre el
Uses of Military Power in the Nuclear Age (Princeton, NUEVA JERSEY.: Prensa de la Universidad de Princeton, 1966), páginas. 86–
87; and Jervis, “Was the Cold War a Security Dilemma?" pag. 54.
17. As Jervis writes, “Most of the history of American doctrine and war planning has . . . not come
to grips with the fundamental characteristics of nuclear politics.” See Jervis, The Meaning of the Nu-

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

The New Era of Counterforce 15

dict, but the transformation of international politics that advocates of the the-
ory of the nuclear revolution anticipated never materialized. Hoy, nuclear
powers still eye each other’s economic power and military capabilities warily;
strive for superiority over their adversaries in conventional and nuclear arma-
mentos; aim to control strategically relevant areas of land, aire, mar, and space;
seek to build and maintain alliances; and prepare for war.

The discrepancy between the theory of the nuclear revolution and the be-
havior of states stems from the theory’s misplaced conªdence in the surviv-
ability of nuclear arsenals.18 Proponents of the theory believe that nuclear
weapons deployed in even moderate numbers are inherently survivable.19
Además, according to the argument, survivability is a one-way street: una vez
country deploys a survivable arsenal, it will remain that way. Todavía, what if sur-
vivability is reversible?

If arsenal survivability depends on the uncertain course of technological
change and the efforts of adversaries to develop new technologies, states will
feel compelled to arms race to ensure that their deterrent forces remain surviv-
able in the face of adversary advances. They will worry about relative gains,
because a rich and powerful adversary will have more resources to invest in
technology and military forces. They will value allies, which help contribute
resources and valuable territory. Además, states may be enticed to develop
their own counterforce capabilities in order to disarm their adversaries or limit
the damage those adversaries can inºict in case of war. En breve, if nuclear
stalemate can be broken, one should expect countries to act as they always
have when faced with military threats: by trying to exploit new technologies

clear Revolution, pag. 8. More recently, Glaser and Fetter argue that U.S. nuclear strategy in the Cold
War “was overly competitive, diverging signiªcantly from the policies implied by the powerful
logic of the nuclear revolution.” See Glaser and Fetter, “Should the United States Reject MAD?"
pag. 50.
18. Other deterrence literatures that rely on the assumption that nuclear weapons inherently cre-
ate stalemate should also be reassessed in light of the technological trends beneªting counterforce.
Por ejemplo, classic works by Thomas C. Schelling and Robert Powell (and related works on re-
solve, signaling, and bargaining) equate nuclear weapons with military stalemate, and use that as-
sumption as a starting point to explore how states gain coercive leverage under such a condition.
By assuming that nuclear weapons create military stalemate, their studies overstate the role of re-
solve and credibility in deterrence outcomes, and underplay the importance of military capabili-
corbatas. Ver, Por ejemplo, Schelling, Arms and Inºuence (nuevo refugio, Conexión.: Prensa de la Universidad de Yale,
1966); and Powell, Nuclear Deterrence Theory: The Search for Credibility (Cambridge: Cambridge Uni-
versity Press, 1990).
19. Some proponents, especially Waltz, argue that retaliatory arsenals are very easy to build, de-
ploy, and maintain. Ver, Por ejemplo, Waltz and Sagan, The Spread of Nuclear Weapons, páginas. 20–23,
142–143. Others are more conservative about the requirements, but nonetheless conªdent that the
development of ªrst-strike capabilities is impossible and will be for the foreseeable future. See Jer-
vis, The Meaning of the Nuclear Revolution, pag. 10; Jervis, The Illogic of American Nuclear Strategy;
Glaser, Analyzing Strategic Nuclear Policy; and Glaser and Fetter, “Should the United States Reject
ENOJADO?"

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Seguridad Internacional 41:4 16

and strategies for destroying adversary capabilities. If arsenals have been more
vulnerable than theorists assume, or if survivability and stalemate are re-
versible, then the central puzzle of the nuclear era—continued geopolitical
competition—is no longer a puzzle.

We argue not only that stalemate is reversible in principal, but also that
changes in technology occurring today are making all countries’ arsenals less
survivable than they were in the past. The fear of suffering devastating retalia-
tion will still do much to deter counterforce attacks, but countries will increas-
ingly worry that their adversaries are trying to escape stalemate, and they will
feel pressure to do the same. Deterrence will weaken as arsenals become more
vulnerable. In extreme circumstances—for example, if an adversary threatens
escalation (or begins to escalate) during a conventional war—the temptation
to launch a disarming strike may be powerful.20 In short, in stark contrast to
the expectations of the theory of the nuclear revolution, security competition
has not only endured, but also will intensify as enhanced counterforce capabil-
ities proliferate.

Nuclear Survivability in Practice

The survivability of retaliatory arsenals has long been a crucial objective of
real-world military planning, not just a fertile topic of theoretical analysis. Mil-
itary planners have employed three basic approaches to protect their coun-
tries’ nuclear forces from attack: hardening, concealment, and redundancy. En
terms of hardening, planners deploy missiles in reinforced silos designed to re-
sist blast, calor, ground shock, and the other effects of nuclear detonations;
place aircraft in hardened shelters; create protective sites for patrolling mo-
bile missile launchers; and bury command and control sites, as well as the se-
cure means used to communicate launch orders.

Nuclear planners also rely heavily on concealment. Concealment is the
foundation of survivability for mobile delivery systems, such as ballistic mis-
sile submarines (SSBNs) or mobile missile launchers (known as “transporter
erector launchers,” or TELs), both of which hide in vast deployment areas. Air-
craft are harder to hide because they require airªelds for takeoff and landing,
but they too can employ concealment by dispersing to alternate airªelds or re-

20. Keir A. Lieber and Daryl G. Prensa, “The Next Korean War,” Foreign Affairs, Abril 1, 2013, https://
www.foreignaffairs.com/articles/north-korea/2013-04-01/next-korean-war. For a more general
discussion of wartime escalation risks, see Lieber and Press, “Coercive Nuclear Campaigns in the
21st Century”; and Caitlin Talmadge, “Would China Go Nuclear? Assessing the Risk of Chinese
Nuclear Escalation in a Conventional War with the United States,” Seguridad Internacional, volumen. 41,
No. 4 (Primavera 2017), páginas. 50–92.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

The New Era of Counterforce 17

maining airborne during alerts. Even the most difªcult facilities to hide, hard-
ened missile silos or command bunkers, can be concealed using camouºage
and decoys.

Finalmente, redundancy is used to bolster every aspect of the nuclear mission,
especially force survivability. Most nuclear-armed states use multiple types of
delivery systems and warheads to complicate enemy strike plans and protect
against warhead design ºaws. They spread their forces and warheads across
multiple bases. Además, the most powerful nuclear-weapon states employ
redundant communication networks, command and control arrangements,
and early warning systems.

No single strategy of survivability is ideal, because each entails important
trade-offs. Hardening is attractive, but it comes at the price of concealment: para
ejemplo, it is difªcult to hide the major construction entailed in building a nu-
clear silo. También, hardened sites are not mobile; once discovered, they remain
so.21 Similarly, concealment comes at the price of hardening. If mobile forces
are discovered, they tend to be easy to destroy. Concealment has another sig-
niªcant drawback: it is a “fail deadly” strategy, meaning that if an adversary
develops a way to locate one’s forces, one’s arsenal might go from highly sur-
vivable to completely vulnerable almost overnight. Even worse, one might not
know that the nuclear balance has shifted in such a calamitous manner.22 Some
countries have adopted operating doctrines that attempt to capitalize on the
advantages of both hardening and concealment: China today, Por ejemplo,
appears to plan to disperse its mobile missiles in a nuclear crisis from its
peacetime garrisons to remote protective sites.23 Such approaches capture the

21. Chinese planners assume that their underground nuclear sites have been discovered by the
United States because digging is difªcult to hide, and because Chinese planners know that
the United States is using a wide range of surveillance tools to ªnd the sites. See Wu Riqiang, “Cer-
tainty of Uncertainty: Nuclear Strategy with Chinese Characteristics,” Revista de Estudios Estratégicos,
volumen. 36, No. 4 (July/August 2013), páginas. 579–614, at pp. 586–587; Li Bin, “Tracking Chinese Strategic
Mobile Missiles,” Science and Global Security, volumen. 15, No. 1 (2007), páginas. 1–30, at p. 5; Zhang Yuliang,
ed., Zhanyi xue [The science of campaigns] (Beijing: Prensa Universitaria de la Defensa Nacional, 2006),
páginas. 635, 637; and Yu Jixun, ed., Di’er pao bing zhanyi xue [The science of Second Artillery cam-
paigns] (Beijing: PLA Press, 2004), pag. 302.
22. New historical evidence reveals that technological breakthroughs and innovative naval opera-
tions allowed the United States to trail Soviet ballistic missile submarines during periods of the
Cold War. The Soviets were unaware of the extent of the vulnerability of their submarines for sev-
eral years. See Austin Long and Brendan Rittenhouse Green, “Stalking the Secure Second Strike:
Inteligencia, Counterforce, and Nuclear Strategy,” Revista de Estudios Estratégicos, volumen. 38, Nos. 1–2
(2015), páginas. 38–73; owen r. Coté Jr., The Third Battle: Innovation in the U.S. Navy’s Silent Cold War
Struggle with Soviet Submarines (Newport, R.I.: Escuela de Guerra Naval, 2003); and Peter Sasgen, Stalking
the Red Bear: The True Story of a U.S. Cold War Submarine’s Covert Operations against the Soviet Union
(Nueva York: Calle. Martin’s, 2009).
23. Wu, “Certainty of Uncertainty,” páginas. 586–587; li, “Tracking Chinese Strategic Mobile Missiles,"
páginas. 7–11; Hans M. Kristensen, Robert S. Norris, y Mateo G.. McKinzie, Chinese Nuclear Forces

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Seguridad Internacional 41:4 18

beneªts of both strategies, but they also pay the costs. Por ejemplo, China’s
strategy leaves its forces vulnerable if an attacker has identiªed its dispersal
sites or detects mobile missiles in transit.24

Major technological trends are directly undermining these strategies of sur-
vivability. Leaps in weapons accuracy threaten nuclear forces that rely on
hardening, while an unfolding revolution in remote sensing threatens nuclear
forces that depend on concealment. (Another major change since the end of the
Cold War, far smaller nuclear arsenals among potential adversaries, weakens
the third strategy of survivability: redundancy.)25 Developing survivable
forces is not impossible, but a new age of vulnerability has begun.

Counterforce in the Age of Accuracy

For most of the nuclear age, neither bombers nor ballistic missiles could de-
liver weapons accurately enough to reliably destroy hardened targets. Too
many variables affected the impact point of a bomb—such as the aircraft’s
speed and altitude; the air defense environment; and atmospheric conditions
including wind, temperatura, and humidity—for even highly skilled crews to
deliver bombs precisely.26 Long-range ballistic missiles were even less accu-
tasa. Although their initial deployment conjured fears of “bolt-from-the-blue”
disarming strikes, throughout the 1970s long-range missiles were not accurate
enough to destroy ªelds of hardened silos.27

Technological improvements chipped away at the sources of inaccuracy,
sin embargo. Leaps in navigation and guidance, including advanced inertial sen-

and U.S. Nuclear War Planning (Washington, CORRIENTE CONTINUA.: Federation of American Scientists/Natural Re-
sources Defense Council, Noviembre 2006), pag. 51; and Yu, Di’er pao bing zhanyi xue.
24. Redundancy involves trade-offs, también. Building additional command sites, warning systems, o
communication systems is expensive and therefore comes at the cost of deploying additional
armas. Además, redundancy may promote complacency, thus undermining survivability.
25. Both global and U.S. nuclear weapons stockpiles have declined about 80 percent since the end
of the Cold War and 85 percent from peak levels in the Cold War. See U.S. Bureau of Arms Control,
Veriªcation, and Compliance, “Fact Sheet: Transparency in the U.S. Nuclear Weapons Stockpile”
(Washington, CORRIENTE CONTINUA.: A NOSOTROS. Department of State, Abril 29, 2014); Hans M. Kristensen and Robert S.
Norris, “Nuclear Notebook: Nuclear Arsenals of the World,” Bulletin of the Atomic Scientists, n.d.,
http://thebulletin.org/nuclear-notebook-multimedia; Hans M. Kristensen and Robert S. Norris,
“United States Nuclear Forces, 2016,” Bulletin of the Atomic Scientists, volumen. 72, No. 2 (Marzo 2016),
páginas. 63–73; and Shannon N. Kile and Hans M. Kristensen, “Trends in World Nuclear Forces, 2016"
(Stockholm: Stockholm International Peace Research Institute, Junio 2016), pag. 2.
26. See Michael Russell Rip and James M. Hasik, The Precision Revolution: GPS and the Future of
Aerial Warfare (Annapolis: Prensa del Instituto Naval, 2002), páginas. 14–67.
27. Por ejemplo, the ªrst U.S. intercontinental ballistic missiles (ICBMs) had a median miss dis-
tance (called circular error probable, or CEP) that was two or three times worse than that of con-
temporary bombers. See David Miller, The Cold War: A Military History (Nueva York: Thomas Dunne,
1998), appendix 7; and Duncan Lennox, IHS Jane’s Weapons: Strategic 2012/2013 (Londres: IHS,
2012).

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

The New Era of Counterforce 19

sors with stellar updates, improved the ability of missiles to precisely deter-
mine their position in ºight and guide themselves, as needed, back on course.
Other breakthroughs allowed mobile delivery systems, such as submarines
and mobile land-based launchers, to accurately determine their own position
prior to launch, greatly improving their accuracy.28 As a result of these innova-
ciones, new missiles emerged in the mid-1980s with far better accuracy than
their predecessors, rendering hardened targets vulnerable as never before. Para
bombers, onboard computers now continuously measure the variables that
previously confounded bombardiers. Data on aircraft speed and location are
uploaded from the aircraft into the computers of “smart” bombs and cruise
missiles, which in turn automatically plot a ºight path from the release loca-
tion to the target. The weapons adjust their trajectory as they ºy to remain on
course.29 As a result, bombs and missiles can achieve levels of accuracy un-
imaginable at the start of the nuclear age.

The leap in munitions accuracy has been showcased repeatedly during con-
ventional wars: videos of missiles and bombs guiding themselves directly to
designated targets now appear mundane. Although the effects of the accuracy
revolution on nuclear delivery systems are equally dramatic, they have re-
ceived far less attention, despite huge implications for the survivability of
hardened targets.

improved missile accuracy

Cifra 1 illustrates one consequence of the accuracy revolution, as applied to
nuclear forces, by comparing the effectiveness of U.S. ballistic missiles in 1985
to those in the current U.S. arsenal.30 We use formulas, employed by nuclear
analysts for decades, to estimate the effectiveness of missile strikes against a

28. Before submarines used global positioning system (GPS) navigation, ballistic missile accuracy
was measured in kilometers. See Rip and Hasik, The Precision Revolution, páginas. 63, 66.
29. Actualmente, the two main technologies underlying smart weapons are laser- and GPS-guidance.
In the former, a laser is trained on the target, and a computer in the bomb adjusts the tail ªns to
guide the weapon toward the laser’s reºection. In a GPS-guided bomb, a computer on the muni-
tion uses GPS to repeatedly assess its location as it falls; the bomb adjusts its tail ªns to guide it to
a predetermined aimpoint.
30. We focus on U.S. capacidades, but recent versions of Israel’s Jericho, India’s Agni, Pakistan’s
Hatf, and Russia’s Iskander nuclear-capable missiles employ advanced guidance systems that
may outperform even the best contemporary U.S. ballistic missiles described here. See Hans M.
Kristensen and Robert S. Norris, “Israeli Nuclear Weapons, 2014,” Bulletin of the Atomic Scientists,
volumen. 70, No. 6 (Noviembre 2014), páginas. 97–115; “Quiet Leap,” Aviation Week & Space Technol-
ogia, volumen. 175, No. 26 (Julio 2013), pag. 1; Hans M. Kristensen and Robert S. Norris, “Indian Nuclear
Efectivo, 2015,” Bulletin of the Atomic Scientists, volumen. 71, No. 5 (Septiembre 2015), páginas. 77–83; James C.
O’Halloran, IHS Jane’s Weapons: Strategic (Londres: IHS, 2015), pag. 33; Hans M. Kristensen and Rob-
ert S. Norris, “Pakistani Nuclear Forces, 2015,” Bulletin of the Atomic Scientists, volumen. 71, No. 6 (No-
noviembre 2015), páginas. 59–66; and “SS-26 (Iskander),” MissileThreat (CSIS Missile Defense Project),
Septiembre 27, 2016, https://missilethreat.csis.org/missile/ss-26/.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Seguridad Internacional 41:4 20

Cifra 1. The Growing Vulnerability of Hard Targets, 1985–2017

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

NOTE: The calculations underlying this ªgure assume targets hardened to withstand
3,000 pounds per square inch (psi). Data for 1985 are based on the most capable U.S.
land-based intercontinental ballistic missile (ICBM) and submarine-launched ballistic mis-
sile (SLBM) at the time: the Minuteman III ICBM armed with a W78 warhead and the
Trident I C-4 SLBM armed with a W76 warhead. El 2017 ICBM data are based on
the same Minuteman III / W78, with an improved guidance system. El 2017 SLBM data
show both contemporary conªgurations of the Trident II D-5 missile: one version armed
with the W76 and the other with higher-yield W88 warheads. The data and sources for
A NOSOTROS. weapon systems are in the online appendix, http://dx.doi:10.7910/DVN/NKZJVT,
table A1.

typical hardened silo.31 The ªgure distinguishes three potential outcomes of
a missile strike: hit, miss, and fail. “Hit” means that the warhead detonates
within the lethal radius (LR) of the aimpoint, thus destroying the target.
“Miss” means that the warhead detonates outside the LR, leaving the tar-
get undamaged. “Fail” means that some element of the attacking missile sys-
tem malfunctioned, leaving the target undamaged.

31. See the online appendix at http://dx.doi.org/10.7910/DVN/NKZJVT. The seminal unclassi-
ªed work on the effects of nuclear weapons is Samuel Glasstone and Philip J. Dolan, The Effects of
Nuclear Weapons (Washington, CORRIENTE CONTINUA.: A NOSOTROS. Government Printing Ofªce, 1977). See also Lynn E. Da-
vis and Warner R. Schilling, “All You Ever Wanted to Know about MIRV and ICBM Calculations

The New Era of Counterforce 21

Cifra 1 shows that the accuracy improvements of the past three decades
have led to substantial leaps in counterforce capabilities. En 1985 a U.S. enterrar-
continental ballistic missile (ICBM) had only about a 54 percent chance of de-
stroying a missile silo hardened to withstand 3,000 pounds per square inch
(psi) overpressure. En 2017 that ªgure exceeds 74 por ciento. The improvement in
submarine-launched weapons is starker: de 9 por ciento a 80 por ciento (usando
the larger-yield W88 warhead). Cifra 1 also suggests, sin embargo, that despite
vast improvements in missile accuracy, the weapons still are not effective
enough to be employed individually against hardened targets. Even modern
ballistic missiles are expected to miss or fail 20–30 percent of the time. The sim-
ple solution to that problem, striking each target multiple times, has never
been a feasible option because of the problem of fratricide: the danger that in-
coming weapons might destroy or deºect each other.32 The accuracy revolu-
ción, sin embargo, also offers a solution to the fratricide problem, opening the door
to assigning multiple warheads against a single target, and thus paving the
way to disarming counterforce strikes.

the fading problem of fratricide

One type of fratricide occurs when the prompt effects of nuclear detonations—
radiation, calor, and overpressure—destroy or deºect nearby warheads. A
protect those warheads, targeters must separate the incoming weapons by at
least 3–5 seconds.33 A second source of fratricide is harder to overcome. De-
stroying hard targets typically requires low-altitude detonations (so-called
ground bursts), which vaporize material on the ground. When the debris be-
gins to cool, 6–8 seconds after the detonation, it solidiªes and forms a dust
cloud that envelops the target. Even small dust particles can be lethal to in-
coming warheads speeding through the cloud to the target. Particles in the de-
bris cloud take approximately 20 minutes to settle back to ground.34

Por décadas, these two sources of fratricide, acting together, posed a major

but Were Not Cleared to Ask,” Journal of Conºict Resolution, volumen. 17, No. 2 (Junio 1973), páginas. 207–242.
These formulas and calculations are also discussed in Lieber and Press, “The End of Mad?” páginas. 7–
44, and appendix 1.
32. On nuclear fratricide, see John D. Steinbruner and Thomas M. Garwin, “Strategic Vulnerabil-
idad: The Balance between Prudence and Paranoia,” Seguridad Internacional, volumen. 1, No. 1 (Verano
1976), páginas. 138–181; and Bruce W. bennett, “How to Assess the Survivability of U.S. ICBMs” (Santa
Monica, California: Corporación RAND, 1980), with appendices. For an excellent nontechnical discus-
sion of nuclear fratricide, see Andrew Cockburn and Alexander Cockburn, “The Myth of Missile
Accuracy,” New York Review of Books, Noviembre 20, 1980, http://www.nybooks.com/articles/
1980/11/20/the-myth-of-missile-accuracy/.
33. A 4-second buffer would separate warheads by approximately 10 kilometers. For an analysis
of the targeting issues involved with spacing out warheads, including time of arrival uncertainty,
see Bennett, “How to Assess the Survivability of U.S. ICBMs,” appendix E, páginas. 34–39.
34. John Steinbruner and Thomas Garwin estimate that a reentry vehicle (RV) would collide, en
promedio, con 5 a 10 particles in the range of 3 a 10 grams as it passed through a typical dust

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Seguridad Internacional 41:4 22

problem for nuclear planners.35 Multiple warheads could be aimed at a single
target if they were separated by at least 3–5 seconds (to avoid interfering with
entre sí); todavía, all inbound warheads had to arrive within 6–8 seconds of the
ªrst (before the dust cloud formed). Como resultado, assigning more than two
weapons to each target would produce only marginal gains: if the ªrst one re-
sulted in a miss, the target would likely be shielded when the third or fourth
warhead arrived.36

Improvements in accuracy, sin embargo, have greatly mitigated the problem
of fratricide. As ªgure 1 muestra, the proportion of misses—the main culprit of
fratricide—compared to hits is fading. To be clear, some weapons will still fail;
eso es, they will be prevented from destroying their targets because of mal-
functioning missile boosters, faulty guidance systems, or defective warheads.
Those kinds of failures, sin embargo, do not generally cause fratricide, because the
warheads do not detonate near the target. Only those that miss—that is, those
that travel to the target area and detonate outside the LR—will create a dust
cloud that shields the target from other incoming weapons. En breve, leaps in
accuracy are essentially reducing the set of three outcomes (hit, fail, or miss) a
just two: hit or fail. The “miss” category, the key cause of fratricide, has virtu-
ally disappeared.37

the cumulative consequences for counterforce

The end of fratricide is just one development that has helped negate hardening
and increased the vulnerability of nuclear arsenals. The computer revolution
has led to other improvements that, taken together, signiªcantly increase
counterforce capabilities.

Primero, improved accuracy has transformed the role of ballistic missile sub-
marines, turning these instruments of retaliation against population centers
into potent counterforce weapons. Recordar (from ªgure 1 arriba) that a 1985
submarine-launched ballistic missile (SLBM) had only a 9 percent chance of
destroying a hardened target. This meant that although ballistic missile sub-
marines could destroy “soft” targets (p.ej., cities), they could not destroy the
hardened sites that would be a key focus of a disarming attack. Increased

cloud—any one of which would destroy the RV. See Steinbruner and Garwin, “Strategic Vulnera-
habilidad,” appendix C, pag. 178.
35. For a Cold War example, see Salman, sullivan, and Van Evera, “Analysis or Propaganda?” For
a contemporary analysis that makes the same assumption, see Lauren Caston et al., The Future of
Estados Unidos. Intercontinental Ballistic Missile Force (Santa Mónica, California: Corporación RAND, 2014),
pag. 36, including n. 16.
36. Debris clouds around silos would not prevent missiles from launching. At the relatively slow
speeds of early “boost phase,” missiles could ascend through particles in the debris cloud.
37. The consequences of the fading problem of fratricide for counterforce are illustrated in table 1.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

The New Era of Counterforce 23

SLBM accuracy has added hundreds of SLBM warheads to the counterforce ar-
senal; it has also unlocked other advantages that submarines possess over
land-based missiles. Por ejemplo, submarines have ºexibility in ªring loca-
ción, allowing them to strike targets that are out of range of ICBMs or that are
deployed in locations that ICBMs cannot hit.38 Submarines also permit strikes
from close range, reducing an adversary’s response time. And because subma-
rines can ªre from unpredictable locations, SLBM launches are more difªcult
to detect than ICBM attacks, further reducing adversary response time be-
fore impact.

Segundo, upgraded fuses are making ballistic missiles even more capable
than ªgure 1 reports. In a compelling new analysis, Theodore Postol explores
the implications of new “compensating” fuses that exist on most U.S. SLBMs
and that will soon be deployed on the entire force.39 Reentry vehicles equipped
with this fusing system use an altimeter to measure the difference between the
actual and expected trajectory of the reentry vehicle, and then compensate for
inaccuracies by adjusting the warhead’s height of burst.40 Speciªcally, if the
altimeter reveals that the warhead is off track and will detonate “short”
of the target, the fusing system lowers the height of burst, allowing the
weapon to travel farther (hence, closer to the aimpoint) before detonation. Alabama-
ternatively, if the reentry vehicle is going to detonate beyond the target, el
height of burst is adjusted upward to allow the weapon to detonate before it
travels too far.41 Without this technology, as ªgure 1 muestra, the lower-yield
W76 warheads are much less effective against hardened targets than their
higher-yield cousins, the W88s. The improved fuse cuts the effectiveness gap
roughly in half, making the hundreds of W76s in the U.S. arsenal potent
counterforce weapons for the ªrst time.42 The consequences of the new fuse

38. A NOSOTROS. ICBMs launched at Russia or China—or vice versa—would take a polar route to their tar-
gets. Como resultado, critical sites could be shielded from ICBMs by locating them on the south side of
steep mountains. SLBMs can strike targets from a wide range of launch locations, thwarting efforts
to shield them.
39. Theodore Postol, “Monte Carlo Simulations of Burst-Height Fuse Kill Probabilities,” unpub-
lished presentation, Julio 28, 2015. The compensating fuse is reportedly deployed on all SLBMs
with Mk-5 RVs (es decir., those armed with W88 warheads) and Mk-4A RVs (es decir., those armed with the
recently upgraded W76-1 warheads). The remaining SLBMs with older Mk-4 RVs will be up-
graded by 2019. See Kristensen and Norris, “United States Nuclear Forces, 2016,” páginas. 64, 66, 68;
and Sandia National Laboratories, “Defense Programs,” Sandia Weapon Review Bulletin, Otoño
1992, páginas. 3–4.
40. See Postol, “Monte Carlo Simulations of Burst-Height Fuse Kill Probabilities”; Donald A. Precio
and Charles A. luis, “Burst Height Compensation,” United States Patent US4456202 A, Junio 26,
1984; and John Ainsle, “Sharpening Trident,” swordofdamocles.org, 2009, and published sources
cited therein.
41. See the online appendix.
42. See ibid. See also Hans M. Kristensen, “Small Fuze, Big Effect,” Strategic Security blog (Federa-

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Seguridad Internacional 41:4 24

son, por lo tanto, profound, essentially tripling the size of the U.S. submarine-
based arsenal against hard targets.43 More broadly, the technology at the core
of compensating fuses is available to any state capable of building modern
multistage ballistic missiles.44

A third key improvement, rapid missile retargeting, increases the effective-
ness of ballistic missiles by reducing the consequence of malfunctions. Como
ªgure 1 ilustra, when accuracy increases, missile reliability becomes the
main hurdle to attacks on hardened targets. For decades analysts have recog-
nized a solution to this problem: if missile failures can be detected, the targets
assigned to the malfunctioning missiles can be rapidly reassigned to other mis-
siles held in reserve.45 The capability to retarget missiles in a matter of minutes
was installed at U.S. ICBM launch control centers in the 1990s and on U.S. sub-
marines in the early 2000s, and both systems have since been upgraded.46 We
do not know if the United States has adopted war plans that fully exploit rapid
reprogramming to minimize the effects of missile failures.47 Nevertheless,
such a targeting approach is within the technical capabilities of the United
States and other major nuclear powers and may already be incorporated into
war plans.48

tion of American Scientists), Marzo 14, 2007, https://fas.org/blogs/security/2007/03/small_fuze
_-_big_effect/.
43. En 2016 the United States reportedly had 768 W76 warheads and 384 W88s. See Kristensen and
Norris, “United States Nuclear Forces, 2016," pag. 64.
44. Compensating fuses may also enhance the capability of SLBMs to conduct “depressed trajec-
tory” strikes, in which a missile ºies along a ºatter trajectory, thereby reducing its ºight time (y
hence the target’s warning). In the past, the beneªt of depressed trajectory for counterforce strikes
was mitigated because ºat trajectories eroded accuracy. Compensating fuses, sin embargo, allow plan-
ners to minimize the deleterious effects of depressed trajectories and thus allow SLBMs to strike
hard targets with little warning.
45. Writing in 1976, Steinbrunner and Garwin argued that the threats to U.S. ICBMs were over-
blown, but they cautioned that if the Soviets developed highly accurate delivery systems (es decir., sys-
tems less accurate than U.S. missiles today) and utilized reprogramming, ICBM ªelds would be
highly vulnerable. The technological conditions that they feared have come to pass. Steinbrunner
and Garwin, “Strategic Vulnerability,” páginas. 151–155, 159–168.
46. On the Air Force’s Rapid Execution and Combat Targeting (REACT) system and the Navy’s
SLBM Retargeting System (SRS), see Amy E. Woolf, “U.S. Strategic Nuclear Forces: Fondo,
Desarrollo, and Issues” (Washington, CORRIENTE CONTINUA.: Servicio de investigación del Congreso, Marzo 10, 2016),
pag. 14; Hans M. Kristensen, “U.S. Strategic War Planning after 9/11,” Nonproliferation Review,
volumen. 14, No. 2 (Julio 2007), páginas. 373–390, at p. 382; Andrew S. Kovich, “ICBM Strike Planning,” Asso-
ciation of Air Force Missileers (AAFM) Newsletter, volumen. 15, No. 2 (Junio 2007), páginas. 6–11; and William M.
Arkin, “The Six-Hundred Million Dollar Mouse,” Bulletin of the Atomic Scientists, volumen. 52, No. 6
(November/December 1996), pag. 68.
47. Reprogramming creates complications for war planners. Por ejemplo, ballistic missile strike
plans are orchestrated to prevent incoming weapons from interfering with each other. A plan that
fully employed reprogramming to negate missile failures would need to establish two (or more)
temporal windows for reentry vehicles to safely approach their targets—one for the warheads on
the initial missile assigned to a target and one for the warheads on reserve missiles if the initial
missile failed. Planners might also need to employ lofted trajectories for reserve missiles to clear
the dust clouds shielding targets that were already struck.
48. Bruce Blair, a former missile launch control ofªcer, testiªed to the U.S. Congress nearly two de-

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

The New Era of Counterforce 25

Mesa 1 illustrates the consequences of these improvements against two
hypothetical target sets: 100 moderately hard mobile missile shelters and
200 hardened missile silos.49 Row 1 shows the approximate counterforce capa-
bilities of a 1985-era U.S. Minuteman III ICBM strike; a 2-on-1 attack would
have been expected to leave 8 mobile missile shelters intact. A strike against
200 hardened silos would fare worse, con 42 targets expected to survive.

The remaining rows in table 1 highlight the implications of the changes that
have occurred from 1985 a 2017. Fila 2 illustrates the impact of improved
Minuteman III guidance, which reportedly reduced circular error probable
(CEP) de 183 a 120 meters. Fila 3 employs the most capable missile and
warhead combination in the current U.S. arsenal: the Trident II armed with a
high-yield W88 warhead. As the results in both rows show, upgraded missiles
perform better than their predecessor, but not well enough to conduct effective
disarming strikes against large target sets.

Rows 4–7 demonstrate how the various improvements in missile technology
have combined to create transformative counterforce capabilities. In row 4,
we use a more realistic ªgure for missile system reliability. A pesar de 80 por-
cent missile reliability is traditionally used as a baseline, much evidence sug-
gests that the actual reliability of modern missiles exceeds 90 percent.50 Row 4
shows attack outcomes for a Trident
II/W88 with 90 percent reliabil-
idad. Fila 5 shows the consequences if the United States can reprogram its mis-

cades ago that Russia could reprogram its silo-based missiles in 10 artículos de segunda clase. See Blair, testimony
before the House Committee on National Security, Subcommittee on Military Research and Devel-
opment, 105th Cong., 1st sess., Marzo 17, 1997.
49. One hundred moderately hard targets is a plausible estimate of the number of targets that the
United States might strike in a disarming attack against North Korea (to hit possible missile shel-
ters, weapon storage, and command and control sites) or in a limited strike on China. Two hun-
dred hardened silos roughly correspond to Russia’s ªxed ICBM force. See Hans M. Kristensen and
Robert S. Norris, “Russian Nuclear Forces, 2015,” Bulletin of the Atomic Scientists, volumen. 71, No. 3
(Puede 2015), páginas. 84–97; Hans M. Kristensen and Robert S. Norris, “Chinese Nuclear Forces, 2015,"
Bulletin of the Atomic Scientists, volumen. 71. No. 4 (July/August 2015), páginas. 77–84; and Ofªce of the Sec-
retary of Defense, Annual Report to Congress: Military and Security Developments Involving the People’s
República de China (Washington, CORRIENTE CONTINUA.: A NOSOTROS. Department of Defense, 2015).
50. A group of engineers and scientists with relevant expertise advised us that our 80 por ciento
baseline ªgure for missile reliability is too low. En efecto, the Trident II boasts a 96 percent success
tasa. See “Navy’s Trident II D5 Missile Marks 155 Successful Test Flights,” press release (Bethesda,
Maryland.: Lockheed Martin, Febrero 23, 2015); and “Trident D-5,” Encyclopedia Astronautica, http://
www.astronautix.com/t/tridentd-5.html. The former commander of Russia’s Strategic Rocket
Forces claims a 92 percent launch success rate for Russian missiles throughout the Cold War; el
A NOSOTROS. Central Intelligence Agency (CIA) estimates the success rate to be slightly higher. On Russia
claims and tests, see Pavel Podvig, “History of Missile Launches and Reliability,” Russianforces.org,
Enero 6, 2005, http://russianforces.org/blog/2005/01/history_of_missile_launches_an.shtml.
For CIA estimates, see CIA, “Russian Expectations,” February 1, 1993, documento 0000382541,
CIA FOIA Electronic Reading Room, páginas. 9–10, https://www.cia.gov/library/readingroom/
document/estimated-pub-date-russian-expectations. Tests cannot simulate the conditions of a
wartime missile launch, pero un 90 percent reliability ªgure—which we employ in rows 4–7 of
table 1—seems reasonable given this evidence and expert opinion.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

d
mi
norte
mi
d
r
a
h

0
0
2

s
oh

yo
i

mi
yo
i
s
s
i
METRO

mi
yo
i

b
oh
METRO
0
0
1

s
r
mi
t
yo
mi
h
S

yo
i
s
s
i
METRO

s
mi
v
i
v
r
tu
S

2
4

3
1

)

k
(
pag

9
7
.
0

3
9
.
0

6
9
.
0

9
9
.
0

(cid:2)
9
9
.
0

9
9
.
0

s
mi
v
i
v
r
tu
S

)

k
(
pag

2
9
.
0

6
9
.
0

9
9
.
0

(cid:2)
9
9
.
0

s
C
i
t
s
i
r
mi
t
a
r
a
h
C

norte
oh
pag
a
mi
W.

k
C
a
t
t

norte
a
yo
PAG

y
t
i
yo
i

b
a
i
yo
mi
R

1
:
2

R
1
:
2

1
:
3

8
.
0

9
.
0

PAG
mi
C

)

metro

(

3
8
1

0
2
1

0
9

d
yo
mi
i
Y

)
t
k
(

5
3
3

5
5
4

8
7
W.
norte
a
metro
mi
t
tu
norte
METRO

5
8
9
1

mi
norte

i
yo

mi
s
a
B

norte
oh
pag
a
mi
W.

norte
oh
i
t
pag
i
r
C
s
mi
D

8
7
W.
norte
a
metro
mi
t
tu
norte
METRO

8
8
W.

I
I

t
norte
mi
d
i
r
t

y
t
i
yo
i

b
a
i
yo

mi
r

C
i
t
s
i
yo

a
mi
R

7
1
0
2

mi
norte

i
yo

mi
s
a
B

5
5
4

8
8
W.

I
I

t
norte
mi
d
i
r
t

gramo
norte
metro
metro
a
r
gramo
oh
r
pag
mi
R

0
0
1

6
7
W.

I
I

t
norte
mi
d
i
r
t

mi
s
tu
F

gramo
norte
i
t
a
s
norte
mi
pag
metro
oh
C

norte
oh

y
norte
a
METRO

h
C
a
mi
oh
t
d
mi
norte
gramo
i
s
s
a

s
d
a
mi
h
r
a
w

F
oh
r
mi
b
metro
tu
norte
mi
h
t

s
mi
t
a
C
i
d
norte

"
norte
a
yo
PAG
k
C
a
t
t

A
“
norte
metro
tu
oh
C

mi
h
t

.
s
r
mi
t
mi
metro
norte

s
i

)
PAG
mi
C

(

mi
yo
b
a
b
oh
r
pag
r
oh
r
r
mi

r
a
yo
tu
C
r
i
C
d
norte
a

s
norte
oh
t
oh

yo
i
k
norte

s
i

d
yo
mi
i
Y

.
i
s
pag
0
0
0

oh
t
d
mi
norte
mi
d
r
a
h
s
oh

yo
i
s

mi
yo
i
s
s
i
metro
0
0
2

r
oh
)
i
s
pag
(
h
C
norte

mi
r
a
tu
q
s

r
mi
pag
s
d
norte
tu
oh
pag
0
0
0
,
1
oh
t
pag
tu
d
norte
a
t
s
h
t
i

w
oh
t
d
mi
norte
mi
d
r
a
h
s
r
mi
t
yo
mi
h
s

mi
yo
i
s
s
i
metro
mi
yo
i

b
oh
metro
0
0
1

r
oh
F
d
mi
y
a
yo
pag
s
i
d
mi
r
a

s
t
yo
tu
s
mi
R

:
mi
t
oh
norte

y
t
i
yo
i

b
a
v
i
v
r
tu
S

t
mi
gramo
r
a
t

d
r
a
h

F
oh

mi
s
i
metro
mi
D

mi
h
t

mi
b
a
t

mi
h
t

t
s
i
yo

"
)
k
(
pag
“

d
mi
yo
t
i
t

s
norte
metro
tu
oh
C

mi
h
t

.
s
norte
oh
i
t
C
norte
tu
F
yo
a
metro
mi
s
a
h
pag

t
s
oh
oh
b

mi
C
a
yo
pag
mi
r

oh
t

s
mi
yo
i
s
s
i
metro
mi
v
r
mi
s
mi
r

s
mi
s
tu

r
mi
k
C
a
t
t
a

mi
h
t

t
a
h
t

s
norte
a
mi
metro

)
gramo
norte
metro
metro
a
r
gramo
oh
r
pag
mi
r

r
oh
F
(

"
R
“

;
t
mi
gramo
r
a
t

–
norte
tu
"
(cid:2)
9
9
.
0
“
F
oh
norte
oh
i
t
a
norte
gramo
i
s
mi
d
mi
h
t

.
k
C
a
t
t
a
mi
h
t
gramo
norte
i
v
i
v
r
tu
s

s
t
mi
gramo
r
a
t

F
oh
r
mi
b
metro
tu
norte
d
mi
t
C
mi
pag
X
mi
mi
h
t

s
i
"
s
mi
v
i
v
r
tu
S
“
d
norte
a
,

d
mi
y
oh
r
t
s
mi
d
s
i

t
mi
gramo
r
a
t

yo
a
tu
d
i
v
i
d
norte

h
C
a
mi
t
a
h
t
y
t
i
yo
i

b
a
b
oh
r
pag

s
yo
yo
mi
C

r
mi
k
r
a
d

;
s
k
C
a
t
t
a
gramo
norte
metro
r
a
s
i
d

tu
F
s
s
mi
C
C
tu
s
mi
t
a
C
i
d
norte

s
yo
yo
mi
C
d
mi
d
a
h
s

t
h
gramo
i
l
.
t
mi
gramo
r
a
t

yo
a
tu
d
i
v
i
d
norte

h
C
a
mi
gramo
norte
i
y
oh
r
t
s
mi
d
F
oh
mi
C
norte
a
h
C

r
mi
t
a
mi
r
gramo
r
oh
t
norte
mi
C
r
mi
pag
9
.
9
9
s
mi
t
a
C
i
d
norte

)

k
(
pag
r
mi
d

s
a
mi
r
mi
h
w

d
mi
v
i
v
r
tu
s

mi
yo
i
s
s
i
metro
mi
yo
i

b
oh
metro
a

t
a
h
t

y
yo
pag
metro

y
yo
i
r
a
s
s
mi
C
mi
norte

t
oh
norte

s
mi
oh
d

r
mi
t
yo
mi
h
s

mi
yo
i
s
s
i
metro
mi
yo
i

b
oh
metro
gramo
norte
i
v
i
v
r
tu
s

mi
yo
gramo
norte
i
s

t
a
h
t

mi
t
oh
norte

.
s
mi
k
i
r
t
s

tu
F
s
s
mi
C
C
tu
s

y
r
mi
v

mi
t
a
C
i
d
norte

.
mi
yo
i
s
s
i
metro
gramo
norte
i
v
i
v
r
tu
s

s
t
s
mi
gramo
gramo
tu
s

yo
i
s

gramo
norte
v

v
r
tu
s

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

The New Era of Counterforce 27

siles to replace boost-phase failures. As row 5 reveals, a 2-on-1 attack with
reprogramming would be expected to destroy every hardened shelter or silo.
Fila 6 omits reprogramming, but it demonstrates the impact of the decline in
fratricide by adding a third warhead to each target, resulting again in the de-
struction of either target set.

Fila 7 illustrates the impact of compensating fuses. This row, unlike the oth-
ers, employs the lower-yield warhead on the Trident II missiles (the W76).
With the compensating fuse, a 2-on-1 attack using W76s would be expected
to destroy all the mobile missile shelters and all but one of the hardened si-
los. (An attack that mixed W88s and W76s could destroy the entire hard-
ened silo force.)

The results in table 1 are simply the output of a model. In the real world, el
effectiveness of any strike would depend on many factors not modeled
aquí, including the skill of the attacking forces, the accuracy of target intelli-
gence, the ability of the targeted country to detect an inbound strike and
“launch on warning,” and other factors that depend on the political and strate-
gic context. Como resultado, these calculations tell us less about the precise vulnera-
bility of a given arsenal at a given time—though one can reach arresting
conclusions based on the evidence—and more about trends in how technology
is undermining survivability.51

One crucial consequence of the accuracy revolution is not captured in the
above results. Todavía, its impact on the vulnerability of nuclear arsenals may be
just as profound. The accuracy revolution has rendered low-casualty counter-
force attacks plausible for the ªrst time.

the dawn of low-casualty counterforce

In nuclear deterrence theory, the primary factor preventing nuclear attack is
the attacker’s fear of retaliation. En realidad, sin embargo, additional sources of in-
hibition exist, including the terrible civilian consequences of an attempted
counterforce strike. If a leader contemplating a disarming strike knows that
such an attack will inºict massive casualties on the enemy, that leader will also
understand that the failure to disarm the enemy will provoke a massive pun-
itive response, foreclosing the possibility of a limited nuclear exchange. Fur-
thermore, if a disarming strike would cause enormous civilian casualties in the
target country, but also possibly in allied and neutral neighboring countries,
leaders who value human life or the fate of allies would contemplate such an

51. In an important respect, our model substantially understates the vulnerability of hard targets,
because it does not capture the growing contribution of nonnuclear forces to counterforce
missions.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Seguridad Internacional 41:4 28

attack in only the direst circumstances. The link between civilian casualties
and nuclear inhibition explains why many arms control advocates oppose the
development of less destructive nuclear weapons; they worry that such weap-
ons are more “usable.”52

Counterforce was tantamount to mass casualties throughout the nuclear
edad, but the accuracy revolution is severing that link. In the past, the main im-
pediment to low-casualty nuclear counterforce strikes has been radioactive
fallout. Targeters would have had to rely on ground bursts to maximize de-
structive effects against hardened facilities such as silos and storage sites. Det-
onations close to the ground have a major drawback, sin embargo: debris is
sucked up into the ªreball, where it mixes with radioactive material, spreading
radiation wherever it settles. Although the other effects of nuclear detonations
(p.ej., blast and ªre) can have large-scale consequences for civilians, in many
circumstances those effects can be minimized.53 If a strike produces fallout,
sin embargo, the consequences are potentially vast and difªcult to predict.54

In theory, it has always been possible to employ nuclear weapons without
creating much fallout. If weapons are detonated at high altitude (above the
“fallout threshold”), very little debris from the ground will be drawn up into
the ªreball, greatly reducing fallout.55 In practice, sin embargo, this targeting strat-
egy has never been feasible against hardened sites. The problem is that any
high-yield weapon that detonates low enough to destroy a hardened target
will also be low enough to create fallout. Low-yield weapons could do the job
and remain above the fallout threshold, but that has always been impractical
because low-yield weapons would need to be delivered with great precision to
destroy hardened sites, which was previously impossible.56

52. Ver, Por ejemplo, Julian Borger, “America’s New, More ‘Usable,’ Nuclear Bomb in Europe,"
Guardian, Noviembre 10, 2015. As U.S. Senator Dianne Feinstein recently stated, “The so-called im-
provements to this weapon [the long-range nuclear cruise missile] seemed to be designed can-
didly to make it more usable, to help us ªght and win a limited nuclear war. I ªnd that a shocking
concepto. I think this is really unthinkable.” Cited in Hans M. Kristensen, “Flawed Pentagon
Nuclear Cruise Missile Advocacy,” Strategic Security blog, Junio 10, 2016, https://fas.org/blogs/
security/2016/06/dod-lrso-letter/.
53. If counterforce targets are located outside cities (as most are), targeters can select aimpoints,
yields, and heights of burst to minimize the ªre and overpressure consequences for civilians. Para
study that illustrates the potentially vast consequences of fallout in strategic strikes, ver Guillermo
Daugherty, Barbara Levi, and Frank von Hippel, “The Consequences of ‘Limited’ Nuclear Attacks
on the United States,” Seguridad Internacional, volumen. 10, No. 4 (Primavera 1986), páginas. 3–45. On the large-
scale consequences of ªre and blast, see Lynn Eden, Whole World on Fire: Organizations, Knowledge,
and Nuclear Weapons Devastation (Ítaca, N.Y.: Prensa de la Universidad de Cornell, 2004).
54. Por ejemplo, targeters cannot reliably predict whether wind speed and direction will blow
fallout over an unpopulated region or a city.
55. Glasstone and Dolan, The Effects of Nuclear Weapons, páginas. 36–38; and Ofªce of Technology As-
sessment, The Effects of Nuclear War (Washington, CORRIENTE CONTINUA.: A NOSOTROS. Government Printing Ofªce, 1979),
páginas. 18, 22–24, 35.
56. See the online appendix for the calculations underpinning these claims and ªgure 2.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

The New Era of Counterforce 29

Cifra 2. The Potential for Low-Fallout Nuclear Counterforce

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

NOTE: “Target hardness” (the horizontal axis) is measured in pounds per square inch (psi),
with a typical range of psi for hardened mobile missile shelters and missile silos noted.
“Yield” (the vertical axis) is measured in kilotons and plotted on a logarithmic scale. El
curve depicts the maximum weapon yield that can destroy a given target from above the
fallout threshold. Any weapon yield/target hardness combination above the line that is ef-
fective enough to destroy the target will necessarily result in fallout. Points below the line
indicate that weapons can be detonated at an altitude that will destroy the target yet pro-
duce little or no fallout. See the online appendix for calculations.

Cifra 2 illustrates why high-yield strikes against hard targets inevitably
create fallout, and it highlights the potential low-yield solution to the fallout
problema. The vertical axis reºects weapon yield, and the horizontal axis de-
picts the hardness of potential targets—with the approximate values for mo-
bile missile shelters and missile silos indicated. The solid black line shows the
maximum yield of a weapon that can generate enough overpressure to destroy
a target from above the fallout threshold. Por ejemplo, ªgure 2 shows that
for a 3,000 psi target, the highest-yield weapon that can destroy it while re-
maining above the fallout threshold is 0.35 kilotons. A larger-yield weapon
will necessarily cause fallout if it destroys the target. A low-fallout strike
against a 1,000 psi mobile missile shelter would require a weapon with 50 kilo-

Seguridad Internacional 41:4 30

tons yield, or less. En breve, low-fatality nuclear counterforce is possible, pero
requires low-yield weapons, and hence very accurate delivery.

The accuracy of nuclear delivery systems is now to the point that low-
casualty disarming strikes are possible. Por ejemplo, a 0.3 kiloton bomb would
require a CEP of 10–15 meters to be highly effective against hard targets;57
that level of accuracy is likely within the reach of the new guided B61-12,
which is slated to replace all nuclear gravity bombs in the U.S. arsenal.58 Simi-
mucho, a 5-kiloton missile warhead, which may approximate the yield of the ªs-
sion primary on many existing ballistic missiles, could destroy a hardened
target if its CEP was approximately 50 meters.59 That level of accuracy was
implausible for most of the Cold War, yet it is within reach of many coun-
tries today.60

By detonating weapons above the fallout threshold, targeters can greatly re-
duce fallout relative to ground bursts. But how signiªcant are these reduc-
ciones? How many fewer deaths would be caused in comparison with ground
burst strikes?

To compare the fallout and potential fatalities from high-yield and low-yield
counterforce operations, we used unclassiªed U.S. Defense Department soft-
mercancía, called Hazard Prediction and Assessment Capability (HPAC).61 Nosotros
modeled two different counterforce strikes, one using a “traditional” high-
yield approach and one employing low-yield airbursts, against ªve hardened
targets in North Korea (p.ej., nuclear storage sites or hardened mobile missile
shelters). Because there is no available unclassiªed information about the lo-
cation of North Korea’s nuclear storage sites, we modeled strikes against
notional locations around the DPRK’s periphery.

57. See ibid.
58. The B61-12 is a guided munition that is said to be similar to a conventional Joint Direct Attack
Munition (JDAM). JDAMs use inertial navigation and GPS in tandem to guide the bomb to the tar-
conseguir. If a JDAM has a clear GPS signal all the way to the target, the CEP is approximately 5 meters.
If the GPS signal is not available, accuracy is approximately 30 meters. If the B61-12 uses inertial
navigation with in-ºight updates from some external source (perhaps GPS), it should have accu-
racy comparable to that of the JDAM. De hecho, in a recent test drop the B61-12 appears to have
landed within 15 meters of the aimpoint. See Hans M. Kristensen and Matthew McKinzie, “Video
Shows Earth-Penetrating Capability of B61-12 Nuclear Bomb,” Strategic Security blog, Enero 14,
2016, https://fas.org/blogs/security/2016/01/b61-12_earth-penetration/; and Hans M. Kristen-
sen, “B61 LEP: Increasing NATO Capability and Precision Low-Yield Strikes,” Strategic Security
blog, Junio 15, 2011, https://fas.org/blogs/security/2011/06/b61-12/.
59. See the online appendix for these calculations.
60. Several nuclear-armed countries have deployed short- and medium-range ballistic missiles
with approximately 50-meter CEP. A pesar de (according to open source data) no intercontinental-
range ballistic missiles can achieve 50-meter CEP, compensating fuses may allow existing missiles
(with primary-only options) to destroy hardened sites from above the fallout threshold.
61. HPAC allows the user to select the number, producir, altitude, ubicación, fecha, and time of simu-
lated nuclear detonations, then estimates the amount and pattern of fallout that would likely be
generated by the strikes.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

The New Era of Counterforce 31

Cifra 3. Low-Fallout Counterforce Option against North Korea

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

NOTE: The ªgure illustrates the potential fallout consequences of two alternative counter-
force strikes against ªve notional North Korean hardened nuclear sites. In both strike
opciones, each target is destroyed with greater than 95 percent probability. The high-
yield attack employs ten W88 warheads (455-kiloton yield), with two warheads against
each target. Because high-yield weapons cannot destroy hardened sites from above the
fallout
the W88s are ground bursts. The low-yield attack uses twenty
B61 bombs (0.3-kiloton yield), set to detonate at an altitude that maximizes effectiveness
while minimizing fallout. The fallout patterns and casualty ªgures were generated using
unclassiªed U.S. Defense Department software, called Hazard Prediction and Assessment
Capability.

límite,

The results of the two strikes, illustrated in ªgure 3, are starkly different. El
traditional approach (on the left side) would likely destroy the targets, but at
a terrible price: millions of fatalities across the Korean Peninsula. The low-
yield option, por el contrario, would produce vastly fewer deaths. As long as the
targets were located outside North Korean cities, the number of Korean fatali-
ties from a low-yield strike would be comparable to the human losses from
conventional operations. De hecho, the fallout contours that are visible in ªgure 3
for the low-yield scenario correspond to annual radiation levels deemed ac-
ceptable by the U.S. Occupational Safety and Health Administration.

The precise results of the HPAC simulation should be treated with skepti-
cism: wind speed and direction change constantly, altering fallout patterns.
The amount of fallout generated in the low-yield scenario is so low, sin embargo,
that the results of ªgure 3 are robust regardless of which way the wind blows:

Seguridad Internacional 41:4 32

few people located away from the actual targets would be killed. The point of
ªgure 3 is not to predict the outcome of a counterforce strike on North Korea,
but to reveal the relationship between accuracy and fallout. When accuracy
was poor, the only approach to nuclear counterforce was high-yield strikes,
which would create catastrophic results such as the one depicted above. El
accuracy revolution has changed the calculus, sin embargo; low-fatality nuclear
strikes are now possible.62

The accuracy revolution is ongoing. As accuracy continues to improve, el
effectiveness of conventional attacks on hard targets will continue to increase.
Hoy, low-yield nuclear weapons can destroy targets that once required very
large yield detonations. En el futuro, many of those targets will be vulnerable
to conventional attacks.

En suma, from the start of the nuclear age to the present, force planners have
relied on hardening as a key strategy for ensuring the survivability of their ar-
senals. That strategy made sense, and until recently ensured that disarming
strikes would not only fail, but also kill millions of civilians in the process.
Technology never stands still, sin embargo, and the technical foundations of deter-
rence, particularly for the strategy of hardening, have been greatly under-
mined by leaps in accuracy.

Counterforce in the Age of Transparency

While advances in accuracy are negating hardening as a strategy for protecting
nuclear forces, leaps in remote sensing are undermining the other main ap-
proach: concealment. Finding concealed forces, particularly mobile ones, re-
mains a major challenge. Trends in technology, sin embargo, are eroding the
security that mobility once provided. In the ongoing competition between
“hiders” and “seekers,” waged by ballistic missile submarines, mobile land-
based missiles, and the forces that seek to track them, the hider’s job is grow-
ing more difªcult than ever before.

Five trends are ushering in an age of unprecedented transparency.63 First,

62. Countries may respond to these advances by putting some counterforce targets in urban areas,
denying their adversaries the option of low-fatality nuclear strikes. Haciéndolo, sin embargo, would ex-
acerbate the vulnerability of those targets in other ways. Por ejemplo, mobile missiles deployed in
or near cities would be exposed to surveillance techniques that would be more difªcult to employ
if the launchers were deployed in rural areas. It is easier to surreptitiously emplace sensors and
tracking systems in urban areas. Protection from low-fatality nuclear strikes would thus come at
the cost of concealment, y, if target intelligence improved sufªciently, those city-based weapons
would be more vulnerable to conventional strikes.
63. For an overview of modern remote sensing capabilities, see Thomas L. Lillesand, Ralph W.
Kiefer, and Jonathan W. Chipman, Remote Sensing and Image Interpretation, 7ª edición. (Hoboken, NUEVA JERSEY.:

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

The New Era of Counterforce 33

sensor platforms have become more diverse. The mainstays of Cold War tech-
nical intelligence—satellites, submarines, and piloted aircraft—continue to
play a vital role, and they are being supplemented by new platforms. por ejemplo-
amplio, remotely piloted aircraft and underwater drones now gather intelli-
gence during peacetime and war. Autonomous sensors, hidden on the ground
or tethered to the seabed, monitor adversary facilities, forces, y operaciones.
Además, the past two decades have witnessed the development of a new
“virtual” sensing platform: cyberspying.64

Segundo, sensors are collecting a widening array of signals for analysis using
a growing list of techniques. Early Cold War strategic intelligence relied
heavily on photoreconnaissance, underwater acoustics, and the collection of
adversary communications—all of which remain important. Ahora, modern
sensors gather data from across the entire electromagnetic spectrum; they em-
ploy seismic and acoustic sensors in tandem; and they emit radar at various
frequencies depending on their purpose, Por ejemplo, to maximize resolution
or to penetrate foliage. Modern remote sensing exploits an increasing number
of analytic techniques, including spectroscopy to identify the vapors leaking
from faraway facilities, interferometry to discover underground structures,
and signals processing techniques (such as those underpinning synthetic aper-
ture radars) that allow radars to perform better than their antenna size would
seem to permit.65

Tercero, remote sensing platforms increasingly provide persistent observa-
ción. At the beginning of the Cold War, strategic intelligence was hobbled by
sensors that collected snapshots rather than streams of data. Spy planes
sprinted past targets, and satellites passed overhead and then disappeared
over the horizon. Over time those sensors were supplemented with platforms
that remained in place and soaked up data, such as signals intelligence an-
tennas, undersea hydrophones, and geostationary satellites. The trend toward
persistence is continuing. Hoy, remotely piloted vehicles can loiter near en-
emy targets, and autonomous sensors can monitor critical road junctures for
months or years. Persistent observation is essential if the goal is not merely to
count enemy weapons, but also to track their movement.

wiley, 2015). For an excellent discussion of the military implications of advanced remote sensing
technology—written in the context of capabilities as they were in 2001—see Alan J. Vick et al.,
Aerospace Operations against Elusive Ground Targets (Santa Mónica, California: Corporación RAND,
2001).
64. See Vick et al., Aerospace Operations against Elusive Ground Targets, appendix A; and Gordon
Corera, Cyberspies: The Secret History of Surveillance, Hacking, and Digital Espionage (Nueva York: Pega-
sus, 2015).
65. For a discussion of some of these advances, see Defense Advanced Research Projects Agency
(DARPA), Breakthrough Technologies for National Security (Arlington, Va.: DARPA, 2015).

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Seguridad Internacional 41:4 34

The fourth factor in the ongoing remote sensing revolution is the steady im-
provement in sensor resolution. In every ªeld that employs remote sensing
tecnología, including medicine, geology, y astronomía, improved sensors
and advanced data processing are permitting more accurate measures and
fainter signals to be discerned from background noise. The leap in satellite im-
age resolution is but one example: the ªrst U.S. reconnaissance satellite
(Corona) could detect objects as small as 25 feet across. Hoy, even commer-
cial satellites (p.ej., DigitalGlobe’s WorldView-3 and WorldView-4) can collect
images with 1-foot resolution, and U.S. spy satellites are reportedly capable of
resolutions less than 4 inches.66 Advances in resolution are not merely trans-
forming optical remote sensing systems; they are extending what can be seen
by infrared sensors, advanced radars, interferometers and spectrographs, y
many other sensors.

The ªfth key trend is the huge increase in data transmission speed. During
the ªrst decades of the Cold War, it took days or longer to transmit information
from sensors to analysts. At least a full day passed before the photographs
snapped by U-2 aircraft were developed and analyzed. Early satellites were
slower: the satellite had to ªnish its roll of ªlm, and then eject the canister,
which would be caught midair and ºown to a facility for development and
análisis. All told, images collected at the beginning of a satellite mission might
take weeks before they arrived at an analyst’s desk. Hoy, por el contrario, intelli-
gence gathered by aircraft, satellites, and drones can be transmitted in nearly
tiempo real. The data can be transmitted to intelligence analysts, political leaders,
and in some cases directly to military commanders conducting operations.

None of these technological trends alone is transformative. Tomados juntos,
sin embargo, they are creating a degree of transparency that was unimaginable
even two decades ago. These new remote sensing technologies are not prolifer-
ating around the world evenly; the United States, Por ejemplo, seems to have
exploited new sensing technologies more intensively than other countries.
Many countries are developing expertise in advanced sensing, sin embargo. El
sensing revolution is a global phenomenon, with implications for the surviv-
ability of all countries’ nuclear arsenals.

Remote sensing technologies have improved greatly, but the crucial ques-
tion is whether these advances have meaningfully increased the vulnerability
of the two most elusive types of nuclear delivery systems: SSBNs and mobile
land-based missiles. If the ability to track submarines at sea or mobile missiles

66. Richard Hollingham, “Inside the Google Earth Satellite Factory,” BBC News, http://www
.bbc.com/future/story/20140211-inside-the-google-earth-sat-lab; and “WorldView-4 Launch Page”
(Bethesda, Maryland.: Lockheed Martin, n.d.), http://www.lockheedmartin.com/worldview4.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

The New Era of Counterforce 35

on patrol remains out of reach, then the counterforce improvements we identify
are less signiªcant, at least for now. De hecho, SSBNs have never been as invulnera-
ble as analysts typically assume, and advances in remote sensing appear to be
reducing the survivability of both submarines and mobile missiles.

remote sensing and tracking submarines

During the Cold War,
the competition between submariners and anti-
submarine warfare operators was shrouded in secrecy, but that history is
ªnally being revealed. We now know that the United States was able to locate,
and even track, Soviet SSBNs during extended periods of the Cold War.67

The core of U.S. ASW efforts against the Soviet Union lay in a series of
breakthroughs in passive sonar and signals processing, as well as doctrine and
tactics to exploit those advances. Starting in the 1950s, the United States de-
ployed an expanding network of underwater hydrophones designed to iden-
tify and locate adversary submarines. Data from the hydrophones were
transmitted across undersea cables to onshore computing facilities, dónde
powerful computers discerned the faint sounds of submarines from ocean
ruido. Potential targets were then passed along to aircraft and attack subma-
rines (SSNs) for further location and tracking. A NOSOTROS. capabilities to track Soviet
submarines leapt forward in the late 1960s and 1970s, as the United States de-
ployed new attack submarines, which were equipped with powerful sonars in
their bows, towed sonar arrays, and improved on-ship computing power, giv-
ing U.S. SSNs an unprecedented combination of acoustic gathering and data
processing capabilities.68

The competition between Soviet SSBNs and the pack of U.S. submarines, aire-
craft, and surface ships hunting them varied throughout the Cold War. Allá
were periods in which U.S. forces were winning, trailing every Soviet SSBN on
patrol, from port to sea and back. In later periods, after discovering their vul-
nerability, the Russians pulled their forces into protected “bastions” near
Soviet territory to counter the U.S. ASW strategy. The United States did not
give up, and worked until the end of the Cold War (and beyond) to regain
undersea superiority.

67. Among journalistic accounts, which are based on interviews with ASW operators, see Sasgen,
Stalking the Red Bear; and Sherry Sontag and Christopher Drew, with Annette Lawrence Drew,
Blind Man’s Bluff: The Untold Story of American Submarine Espionage (Nueva York: Asuntos Públicos, 1998).
For academic analyses, see Coté, The Third Battle; Long and Green, “Stalking the Secure Second
Strike”; and Brendan Rittenhouse Green and Austin Long, “The Role of Clandestine Capabilities
in World Politics,” paper presented at the annual meeting of the American Political Science Associ-
ación, Filadelfia, Pensilvania, September 1–4, 2016.
68. Coté, The Third Battle; Long and Green, “Stalking the Secure Second Strike”; and Green and
Largo, “The Role of Clandestine Capabilities in World Politics.”

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Seguridad Internacional 41:4 36

The duration of U.S. Cold War ASW superiority cannot be accurately as-
sessed today because of enduring classiªcation constraints. But for periods of
the superpower competition, A NOSOTROS. naval leaders believed they had the ASW
problem well in hand. As the former commander of the U.S. Paciªc Fleet in the
mid-1980s remarked, the United States was able to “identify by hull number
the identity of Soviet subs . . . and know exactly where they were. In port or at
mar. If they were at sea, N3 [director for operations] had an SSN [on them].”69

There are three key lessons to draw from the Cold War ASW competition.
Primero, previous advances in remote sensing greatly increased the vulnerability
of deployed submarines.70 Second, escaping vulnerability was no easy task. En
the late 1960s, the Soviet Union learned that its submarines were vulnerable.
But despite Moscow’s signiªcant economic and technological resources, él
took the Soviet navy more than a decade to develop good countermeasures
against the evolving U.S. ASW capabilities.71

Tercero, and most broadly, the Cold War ASW competition demonstrates that
the deployment of ballistic missile submarines neither ended the Cold War nu-
clear competition nor negated hopes on either side of attaining military superi-
ority. The United States led the undersea competition for a time because of its
superior technology and tactics; the Soviet Union developed countermeasures
because it discovered its vulnerabilities and innovated. This back-and-forth
struggle between hiders and seekers looks more like a traditional struggle for
naval superiority than the common depiction of invulnerable submarines.

Today’s technological advances in remote sensing, data processing, y
communication are occurring at a rapid pace, and their ultimate impact on the
submarine competition is too uncertain to predict with conªdence (especially
given the tight controls over information on contemporary ASW capabilities).
Todavía, there are good reasons to suspect that the dramatic leaps in remote sensing
are increasing the transparency of the seas and undermining the ability of sub-
marines to remain concealed.72 Some of the promising new anti-submarine

69. As quoted in Long and Green, “Stalking the Secure Second Strike," pag. 51.
70. Although U.S. Cold War ASW successes rested heavily on technical breakthroughs in acous-
tics and data processing, well-trained operators and intelligence analysts were essential to the
success.
71. The main Soviet countermeasures included deploying longer-range SLBMs, which permitted
Soviet submarines to target the U.S. homeland from well-defended bastions near the Soviet coast,
and developing quieter Soviet SSBNs and SSNs to elude detection and threaten the submarine
hunters. It appears that Soviet countermeasures signiªcantly reduced the U.S. undersea advan-
llevar, but this judgment (like the Cold War assumptions that the United States was not tracking So-
viet SSBNs) is tentative given that relevant documents remain classiªed.
72. See Bryan Clark, The Emerging Era in Undersea Warfare (Washington, CORRIENTE CONTINUA.: Center for Strategic
and Budgetary Affairs, 2015); James Holmes, “Sea Changes: The Future of Nuclear Deterrence,"
Bulletin of the Atomic Scientists, volumen. 72, No. 4 (Julio 2016), páginas. 228–233; and Bryan Clark, “Undersea

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

The New Era of Counterforce 37

technologies include improved acoustic sensors (including low-frequency ac-
tive sonars and new networks of seabed passive sonars); non-acoustic tech-
niques (such as laser detection); sophisticated “big data” analysis (cual
exploits leaps in processor speed to sift vast quantities of sensor data); y un
variety of unmanned and autonomous undersea vehicles (including those de-
signed to ªnd and shadow adversary submarines for weeks or months).73

The point is not that submarines are now easy to locate or that the chal-
lenges of ASW have been solved. Locating technologically sophisticated,
well-operated submarines in vast ocean sanctuaries remains a substantial
challenge. Bastante, the key point is that even the nuclear delivery system some-
times touted as the most survivable has been vulnerable in the past and ap-
pears to be increasingly vulnerable today, as ASW efforts and capabilities
rapidly improve.

What about mobile land-based missiles? Are breakthroughs in sensing tech-

nology increasing their vulnerability as well?

remote sensing and hunting mobile missiles

We illustrate the impact of two advanced surveillance systems, radar satellites
and remotely piloted aircraft, on the survivability of mobile land-based nu-
clear missiles. The effectiveness of sensing systems depends on the characteris-
tics of the target country—for example, its size, ubicación, topography, y
defensas. Tal como, their impact is difªcult to quantify in the abstract. En cambio,
we explore the potential contributions of two advanced sensor systems in a
hypothetical case: a U.S.-led operation to destroy a small arsenal of North
Korean nuclear-tipped mobile missiles.74 We assume that North Korea’s TELs
are postured like most other countries’ mobile missiles; they remain in hard-
ened shelters during peacetime, with plans to disperse a portion of the force
during a conºict.75

A NOSOTROS. and allied strategic intelligence would have at least three critical roles in

Cables and the Future of Submarine Competition,” Bulletin of the Atomic Scientists, volumen. 72, No. 4
(Julio 2016), páginas. 234–237.
73. clark, The Emerging Era in Undersea Warfare, páginas. 8–17; holmes, “Sea Changes,” páginas. 229–230;
and Clark, “Undersea Cables and the Future of Submarine Competition,” páginas. 235–237.
74. This scenario is salient because if conventional war erupts on the Korean Peninsula, los unidos
States and South Korea may feel compelled to destroy North Korea’s nuclear capabilities. Ver
Lieber and Press, “Coercive Nuclear Campaigns in the 21st Century”; Lieber and Press, “The Next
Korean War”; and Talmadge, “Would China Go Nuclear?"
75. Unlike the Soviet Union, which conducted armed peacetime deterrent patrols, most countries
with nuclear-armed mobile forces appear to keep their TELs in hardened facilities during peace-
tiempo (and even during crises). On China’s doctrine, see Li, “Tracking Chinese Strategic Mobile
Missiles,” páginas. 7–11; and Wu, “Certainty of Uncertainty,” páginas. 586–587. On Israeli basing, see “Beit
Zachariah/Zekharyeh,” Globalsecurity.org, n.d., http://www.globalsecurity.org/wmd/world/
israel/sedot_mikha.htm.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Seguridad Internacional 41:4 38

support of a military operation against North Korean TELs. The ªrst, a peace-
time mission called “intelligence preparation of the battleªeld” (IPB), involves
locating North Korea’s nuclear and missile facilities, identifying the patrol
routes utilized by its missile forces, learning its organizational routines, y
mapping its command and communication network. The other two roles are
principally wartime missions. “Detection” refers to sensing possible targets; él
typically involves sensors that can monitor large areas, but that have inade-
quate resolution for positive identiªcation or targeting. “Identiªcation” is the
next step; once a possible target is detected, other platforms (often with higher-
resolution sensors) are cued to identify and precisely locate the target.76

satellites/sar sensors. A core element of U.S. surveillance capabilities
lies in a constellation of satellites that use synthetic aperture radar to image
targets on the ground. Satellites provide a unique capability to peer deep into
adversary territory, and they are especially useful for missions that require fre-
quent observations of critical facilities. Whereas manned aircraft and un-
manned aerial vehicles (UAVs) are often restricted from adversary airspace,
satellites routinely overºy adversary territory. Además, unlike satellites with
optical or infrared sensors, radar satellites can image targets at night and
through cloudy weather.

Until recently, the type of radar employed on most satellites—synthetic ap-
erture radar (SAR)—could not image moving targets, limiting the effectiveness
of space-based sensors for hunting mobile missiles.77 But over the past two de-
cades, engineers have developed data-processing techniques that enable SAR
systems to detect moving targets and determine their speed and direction of
travel.78 Although the precise capabilities of intelligence satellites are class-

76. The IPB framework for elusive targets is reºected in Vick et al., Aerospace Operations against
Elusive Ground Targets, especially chap. 4 and appendix B. Some sensing platforms carry sensors
for both detection and identiªcation. Por ejemplo, the RQ-4 Global Hawk drone has a ground
moving target indicator (GMTI) radar that can scan a wide area and then switch to “spot” mode to
look more closely at an identiªed target.
77. For natural aperture radars, image resolution is constrained by the size of the antenna, and op-
erating very large antennas in space is currently impractical. Como resultado, satellites employ syn-
thetic aperture radars, which use the movement of the satellite to simulate the function of a larger
antenna, allowing satellites to generate images at higher resolution than their antenna size would
normally permit. Until recently, sin embargo, SAR systems could not image moving targets. See Jo-
seph Post and Michael Bennett, “Alternatives for Military Space Radar” (Washington, CORRIENTE CONTINUA.: Estafa-
gressional Budget Ofªce, Enero 2007).
78. For one of the earliest papers on using SAR for tracking mobile targets, see R.P. Perry, R.C.
DiPietro, and R.L. Fante, “SAR Imaging of Moving Targets,” IIEEE Transactions on Aerospace and
Electronic Systems, volumen. 35, No. 1 (Enero 1999), páginas. 188–200. For a recent study that employs a ci-
vilian radar satellite to identify the location and velocity of cars and trucks, see Christoph H.
Gierull, Ishuwa Sikaneta, and Delphine Cerutti-Maori, “Two-Step Detector for RADARSAT-2’s Ex-
perimental GMTI Mode,” IEEE Transactions on Geoscience and Remote Sensing, volumen. 51, No. 1 (Janu-
ary 2013), páginas. 436–454.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

The New Era of Counterforce 39

iªed, civilian radar satellites can scan approximately 150-kilometer-wide
swaths along the ground as they pass overhead with sufªcient resolution to
detect truck-sized moving vehicles.79 New techniques are being developed
that may soon double or triple the width of the swath that can be scanned on
each pass.80

SAR-equipped satellites, now able to ªnd mobile targets, have the potential
to transform counter-TEL operations. If U.S. intelligence satellites can detect
moving vehicles within a 150-kilometer-wide swath along the ground, a con-
servative assumption given that a civilian satellite launched nearly a decade
ago can do so, then centering the radar on a mobile missile garrison would put
all the roads within two hours’ drive-time of that facility within the radar’s
swath width.81 A single satellite can generate up to twelve 150 kilometer x
150 kilometer swaths in a single pass over North Korea, enough to image all
the country’s roads more than once—and key sections multiple times—before
passing over the horizon.82

Although SAR satellites have become powerful tools for hunting TELs, ellos
have important limitations. Surveillance satellites provide only intermittent
coverage of key areas, passing overhead and then descending over the hori-
zon. De este modo, even if a constellation of satellites could image the entire road net-
work in North Korea every hour, North Korean TELs might be able to disperse
without being observed, by seeking shelter whenever a satellite approaches.
Además, if many of North Korea’s critical facilities are located in its
mountainous regions, topography may block the satellite’s line-of-sight, cual
would allow targets within the swath to be hidden from the radar. The poten-
tial effectiveness of radar satellites for hunting mobile missiles, por lo tanto, de-

79. Uno 2015 study used RADARSAT-2 data to scour 150-kilometer-wide swaths of sea, locating
and characterizing the velocity of moving vessels. See Louis-Philippe Rousseau, Christoph
Gierull, and Jean-Yves Chouinard, “First Results from an Experimental ScanSAR-GMTI Mode on
RADARSAT-2,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,
volumen. 8, No. 11 (Noviembre 2015), páginas. 1–13. One of the authors of that paper conªrmed via personal
communication that the technique that the authors employed and the large swath widths that they
used should be applicable to truck-sized vehicles on the ground.
80. Using the experimental “High-Resolution-Wide-Swath” mode, SAR satellites may be able to
generate 300 to 500–kilometer-wide swaths with sufªcient resolution to track truck-sized moving
objetivos. See Stefan V. Baumgartner and Gerhard Krieger, “Simultaneous High-Resolution Wide-
Swath SAR Imaging and Ground Moving Target Indication: Processing Approaches and System
Conceptos,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, volumen. 8,
No. 11 (Noviembre 2015), páginas. 1-15.
81. With a 150-kilometer-wide swath, the terrain within 75 kilometers of either side of the garrison
would be within the zone covered by the radar. A typical TEL can move approximately 40 kilome-
ters per hour. Given that roads in North Korea are not perfectly straight, the actual driving time to
escape a 75-kilometer-wide swath (on either side of the garrison) would be at least two hours.
82. See the online appendix.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Seguridad Internacional 41:4 40

pends on two key factors: the time interval between satellite passes and the
percentage of road network that is observable in a given pass.83

To assess the effectiveness of SAR satellites for hunting North Korean mo-
bile missiles, we conducted an analysis with three key steps. Primero, we created
a digital map of North Korea’s roads. Segundo, we used geospatial analysis
software to determine the visible portion of those roads as a function of a
satellite’s position. Tercero, we calculated the frequency with which satellites
pass within an orbital band that provides high levels of visibility of the
road network.84

Our analysis of satellite orbits and North Korean topography reveals
that satellites passing through an orbital band that stretches as far as 1,500-
kilometer lateral distance from the Korean Peninsula can view, on average,
90 percent of North Korean roads. A typical radar satellite (which operates in
low earth orbit) will pass through such a band, what we call a “usable pass,"
apenas 2.5 times per day. The total number of usable passes per day thus de-
pends on the number of SAR satellites in orbit that are available for hunting
mobile missiles. The number of available satellites, Sucesivamente, depends on the
willingness of the United States and its close allies to share sensitive satellite
imagery, the technical preparations that have been undertaken to facilitate that
sharing, and the precise technical capabilities of the satellites.

Mesa 2 shows the implications of different assumptions about those uncer-
tainties. If the United States and key allies create the political and technical ar-
rangements to share satellite data during wartime, North Korean TEL
commanders would have little time between passes—speciªcally, as few as
24 minutes.85

Twenty-four minutes between satellite passes could provide enough time
for TELs or other vehicles to move quickly from shelter to shelter, but that
strategy requires precise information on satellite orbits, and the short time in-
terval between passes leaves little margin for error for vehicles racing for
cover. Además, the challenge for TEL operators is more serious than the data
sugerir. The analysis here focuses on the twenty military and intelligence SAR

83. These two factors are linked. If North Korea’s topography is sufªciently problematic that only
satellites in a narrow orbital band (es decir., almost directly overhead) can see over the mountains, entonces
there will be fewer usable passes each day (and hence a longer interval between passes). Sobre el
other hand, if even satellite passes that are far from North Korea can see the roads, then there will
be many usable passes per day.
84. See the online appendix for details on all three steps in the analysis.
85. The United States may or may not have agreements and technology in place to rapidly share
sensitive satellite imagery, even with its closest military partners. From the perspective of North
Korean missile commanders, sin embargo, a German or Japanese SAR satellite pass overhead poses a
major threat.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

The New Era of Counterforce 41

Mesa 2. Synthetic Aperture Radar (SAR) Satellites and Frequency of Usable Passes

Número de
SAR Satellites

Disponible
Satellites

Usable Passes
per Day

Minutes between
Passes

Cumulative

United States
Other NATO
Japón
Israel

6
9
3
2

6
15
18
20

15
37
44
50

91
34
27
24

NOTE: The category “Number of SAR Satellites” counts major military and intelligence SAR
satellites operated by the United States and key allies. The other columns are cumulative
and show how satellite coverage grows when one adds the assets of various U.S. part-
ners. “Usable Passes per Day” indicates the daily satellite overºights that pass through an
orbital band that offers, on average, 90 percent coverage of North Korean roads.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

satellites, not the half dozen or more U.S. and allied civilian platforms that
might be pressed into service in wartime.86 Nor does the analysis count the op-
tical and infrared satellites that supplement SAR coverage. Finalmente, the number
and capability of radar satellites available to the United States is growing.87 As
that number increases, the window for mobile missiles to scoot away without
being observed will narrow further.

SAR satellites do not solve the problem of locating mobile targets. For one
thing, Russia and China are improving their ASAT capabilities, partly in re-
sponse to U.S. capabilities.88 Furthermore, adversaries will seek to place mis-
sile garrisons and conduct deterrent patrols in locations that are difªcult to
observe.89 Those choices, sin embargo, force adversaries into ever-narrower zones,
which then become the focus of other surveillance tools—for example, stealthy
penetrating UAVs and unattended ground sensors.

86. The major military and intelligence radar satellites operated by the United States and key al-
lies include Lacrosse 3–5 and Topaz 1–3 (United States); SAR-Lupe 1–5 (Alemania); COSMO A–D
(Italia); IGS 7a, IGS 8a, and a third satellite with name unclear (Japón); and OFEQ 8 and OFEQ 10
(Israel). Civilian radar satellites operated by allied countries include TerraSAR-X and TanDEM-X
(Alemania); Copernicus Sentinal-1 and -3 (European Union); KOMPSat-5 (South Korea); ALOS-2
(Japón); PAZ (España); and RADARSAT-2 (Canada). See “Synthetic Aperture Radar (SAR) Satel-
lites” (Roca, Colo.: UNAVCO, Agosto 7, 2015), https://www.unavco.org/instrumentation/
geophysical/imaging/sar-satellites/sar-satellites.html.
87. Although some older satellites on our list may have limited capabilities for this mission (p.ej.,
they may have single-channel receivers), new satellites currently being deployed may supplement
those capabilities (p.ej., small cube satellites may orbit near an older radar satellite and serve as a
second receiver). Most important, as the number of radar satellites continues to grow, A NOSOTROS. and al-
lied capabilities will exceed those described in table 2.
88. See Eric Heginbotham et al., The U.S.-China Military Scorecard: Efectivo, Geography, y el
Evolving Balance of Power, 1996–2017 (Santa Mónica, California: Corporación RAND, 2015), cap. 10. De
curso, the United States is seeking to counter anti-satellite weapon technologies.
89. li, “Tracking Chinese Strategic Mobile Missiles,” páginas. 7–11, 15–25.

Seguridad Internacional 41:4 42

In terms of the three key sensing missions (IPB, detección, and identiªca-
ción), SAR-equipped satellites offer a high level of capability for the IPB mis-
sión, because they can repeatedly image stationary or moving targets in
peacetime. They also contribute a high level of capability to detection, by offer-
ing frequent wide-area coverage of North Korean roads. Finalmente, SAR satellites
offer fairly good capability for the identiªcation mission: they can produce
high-resolution images of stationary TELs and enough resolution of moving
vehicles to determine that a target is “truck-sized.”90

uavs/sar sensors. A second set of sensing capabilities lies in a ºeet of air-
craft, including manned and remotely piloted vehicles, that use powerful ra-
dars to scan adversary territory. These aircraft carry SARs, and many are
equipped with Ground Moving Target Indicator (GMTI) radars, allowing them
to create high-resolution images of stationary targets or track a large number
of moving vehicles. Most surveillance aircraft must operate from “standoff”
distances to reduce their vulnerability to air defenses. Some drones, sin embargo,
are stealthy and can penetrate adversary airspace. Below we illustrate the ca-
pabilities of standoff SAR/GMTI platforms and penetrating UAVs in the
context of a U.S. and allied operation against North Korean mobile missiles.
for standoff radar-
reconnaissance missions; we base our model on one of them: the remotely
piloted RQ-4 Global Hawk. We explore the potential effectiveness of radar sur-
veillance from four continuous orbits 80 kilometers outside North Korean ter-
ritory.91 ArcGIS software allows us to identify orbital locations that maximize
coverage of North Korean roads, as well as calculate the visible percentage of
the road network from those locations.92 Figure 4 shows the results.

The United States uses several

types of aircraft

Cifra 4 reveals that even against a small country such as North Korea,
standoff airborne radars cannot, by themselves, provide complete coverage of
key roads and regions. Four orbits can observe 54 percent of North Korea’s
caminos; the remainder is out of sensor range or shielded by mountainous ter-
rain. These results also suggest, sin embargo, that standoff UAVs could play a cru-
cial role in a sensing operation; eso es, the ability to continuously monitor

90. Identifying a vehicle as truck-sized might be sufªcient for a strike, depending on the context of
the conºict. Si, during a war, North Korea has employed nuclear weapons, detecting a large vehi-
cle on a road near a North Korean missile garrison may be enough to trigger a strike against the
vehicle. The level of identiªcation required to trigger a strike presumably depends on the weapons
available to the United States for the strike: the threshold for launching a conventional weapon at
a suspected TEL is presumably lower than the threshold for employing nuclear weapons.
91. A standoff distance of 80 kilometers puts the aircraft outside the range of most North Korean
air defenses. We assume a sensor range of 240 kilometers, which was the reported GMTI range of
joint surveillance target attack radars during the 1991 Persian Gulf War. Our results are not sensi-
tive to sensor-range assumptions because even at 200 kilometers, the beam is grazing the ground
at a shallow angle, allowing the mountainous topography to constrain line of sight.
92. See the online appendix for calculations.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

The New Era of Counterforce 43

Cifra 4. Coverage of North Korea with Standoff Unmanned Aerial Vehicles (UAVs)

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

NOTE: The white circles depict potential orbital locations for four UAVs; the locations were
selected to maximize surveillance of North Korea’s road network. The orbits are located
80 kilometers from North Korea’s territory at an altitude of 60,000 pies, which reºect plaus-
ible operations for RQ-4 Global Hawks. White road segments are observable from at least
one of the locations. For additional discussion of the underlying analysis, see the online
appendix at http://dx.doi:10.7910/DVN/NKZJVT. The image was created using ArcGIS and
road data from OpenStreetMap and DIVA-GIS.

roughly half of North Korea’s road network during a conºict would compel
North Korea to constrain its mobile missile operations to the north-central
region of the peninsula.

In addition to standoff UAVs, the United States has developed drones for so-
called penetrating operations.93 These UAVs reduce their visibility to enemy ra-

93. Two such stealthy drones are the RQ-170 and RQ-180. See Amy Butler and Bill Sweetman, “Se-
cret New UAS Shows Stealth, Efªciency Advances,” Aviation Week, December 6, 2013, http://
aviationweek.com/defense/secret-new-uas-shows-stealth-efªciency-advances; and John A.
Tirpak, “For Those Hard-to-Reach Areas,” Air Force Magazine, Junio 10, 2014, http://www
.airforcemag.com/DRArchive/Pages/2014/June%202014/June%2010%202014/For-Those-Hard-
to-Reach-Areas.aspx.

Seguridad Internacional 41:4 44

dar by utilizing stealth technologies and a combination of passive sensors and
“low-probability of intercept” (LPI) radars to observe targets on the ground.94
Even sophisticated, stealthy UAVs are vulnerable to air defenses. To some
extent their vulnerability depends on technical questions, Por ejemplo, el
state of competition between radar engineers and designers of stealth technol-
ogia. The vulnerability of penetrating drones, sin embargo, depends greatly on
their mission. Of the two critical wartime missions, “detection” is likely more
dangerous than “identiªcation.” The detection mission—continuously moni-
toring a large area to detect possible targets—would require a drone to remain
within the line-of-sight of a large portion of adversary territory. The mission
would, por lo tanto, require the drone to ºy at high altitude (to maximize line-of-
sight) and possibly use active sensors (to maximize the drone’s sensor range).
The identiªcation mission, por otro lado, would allow penetrating drones
to protect themselves better: to operate at lower altitude so that terrain would
shield them from enemy sensors, and ºy (when cued by detection systems) a
investigate a possible TEL. Only then would the penetrating UAV employ LPI
or passive sensors to examine the potential target.

We used ArcGIS to explore the potential capability of penetrating drones in
the identiªcation mission by determining the percentage of the North Korean
road network that would be visible using four UAV orbits. Because the pene-
trating UAVs would need to rapidly identify the vehicles detected by other
sensors, we restricted the UAVs to 5 minutes of ºight time to maneuver into
position to observe the suspected TEL.95 Furthermore, because LPI radars and
passive sensors have shorter range than the powerful radars on standoff plat-
formas, we limit the sensor range to 50 kilometers.96

Our analysis reveals that four penetrating drones, operating as we describe
arriba, can identify targets along 84 percent of North Korea’s roads.97 As ªg-

94. Passive sensors “look” but do not emit. Active sensors (those that emit) generally have longer
range, but increase the risk that the UAV will be detected. LPI radars emit but are designed to hin-
der adversary efforts to detect and locate the emitter.
95. RQ-170s (and probably RQ-180s) are equipped with turbofan engines, implying a ºight speed
of approximately 800 kilometers per hour. We assume that once cued to identify a target, the UAVs
gain altitude, if necessary, to attain line-of-sight to the target.
96. En el 5 minutes we allow the UAVs to maneuver, they can ºy approximately 67 kilometers
given 800 kilometer per hour ºight speed. Some sources suggest that the Lynx Multi-mode Radar,
currently deployed on some UAVs, can sense a TEL-sized target moving at 11 kilometers per hour
out to 25 kilometers. See Sherrill Lingel et al., “Methodologies for Analyzing Remotely Piloted Air-
craft in Future Roles and Missions” (Santa Mónica, California: Corporación RAND, 2012), pag. 73. (A TEL
moving more quickly would be observable at a longer range.) Our results are not sensitive to mod-
est deviations in sensor range. If the sensor range on U.S. penetrating UAVs were 25 percent less
than we estimate (35 kilometers instead of 50 kilometers), one additional minute of ºight time
would compensate. See ibid., páginas. 25–27, 73–74.
97. Maintaining four continuous orbits during a conºict would likely require at least 12 UAVs—
not accounting for losses from air defenses.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

The New Era of Counterforce 45

Cifra 5. Coverage of N. Korea with Standoff and Penetrating Unmanned Aerial Vehicles

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

NOTE: The white circles depict potential orbital locations for four UAVs operating 80 kilo-
meters outside North Korea’s territory. The black circles depict the area over North Korea
that four penetrating UAVs can overºy within ªve minutes of ºight time starting from
the center of each circle. Road segments are coded as visible (white) if they are observ-
able from either a standoff or penetrating UAV. For discussion of the underlying analysis,
see the online appendix at http://dx.doi:10.7910/DVN/NKZJVT. The image was created us-
ing ArcGIS and road data from OpenStreetMap and DIVA-GIS.

ura 5 muestra, penetrating and standoff systems would be particularly effective
in combination, increasing the road network coverage to 97 por ciento. Assuming
that penetrating UAVs can be cued by other reconnaissance systems, semejante
as satellites, unattended ground sensors, o (near the coast) standoff drones,
North Korean TEL operators would have great difªculty moving safely along
the country’s road network without being detected. If U.S. and South Korean
intelligence had identiªed mobile missile garrisons and operating areas before
the conºict, the regions surrounding those zones might be fully covered by
only one or two drone orbits.98

98. The results in the North Korea scenario should not be applied directly to other potential U.S.
adversaries—such as China or Russia—because those countries are larger, with different topogra-

Seguridad Internacional 41:4 46

Each of the sensing systems explored here has important limitations. por ejemplo-
amplio, radar satellites provide wide-area coverage, but do so intermittently
and at only moderate resolution. Standoff drones provide persistent coverage,
but only near the coast. Penetrating drones can provide persistent coverage in-
land (at the cost of increased risk to the aircraft) or intermittent inland cover-
age at lower risk. In many cases, sin embargo, the capabilities of one system can
offset the limits of another. Además, this analysis merely scratches the surface
in terms of new sensing platforms (p.ej., unattended ground and seabed sys-
tems), signals (p.ej., high-resolution spectroscopy), and approaches (p.ej., cyber
intrusions), many of which would be employed together for the same mission.
Old assumptions about the survivability of mobile forces need to be revised
in light of new sensing technologies and capabilities. Concealment is not im-
posible, por supuesto. An adversary’s mobile delivery systems can remain secure
if its air defenses can keep UAVs at bay, its navy can keep enemy ASW forces
from its coastal waters, and anti-satellite technology can blind satellites. Pero
in this new era of transparency, whether concealed forces are survivable or
not depends on the state of competition between opposing intelligence and
military organizations. Survivability through concealment can no longer
be assumed.

What About Countermeasures?

Countries will surely address the growing vulnerability of their nuclear arse-
nals by trying to develop countermeasures to thwart advanced sensor and
strike systems. They will seek to deploy radar jammers, anti-satellite weapons,
and decoys. They will try to adapt mobile missile doctrines to reduce vulnera-
habilidad, Por ejemplo, by timing movements to elude satellites and minimizing
communications to thwart signals intelligence efforts. The new era of counter-
force will not be static; it will be characterized by vigorous efforts to develop
countermeasures, as well as equally vigorous efforts to overcome them.

Todavía, there are good reasons to expect that the net result of these efforts will
leave nuclear delivery systems more vulnerable than they have been in the re-
cent past. Primero, hunters are poised to do well in the back-and-forth battle of
countermeasures. Counterforce is the domain of the powerful; those that are
seeking to track enemy nuclear forces typically have greater resources than
their rivals.99 Additionally, the countries that are leaders in sensing technology

phy, bigger arsenals, and more modern defenses. The sensor revolution is signiªcantly increasing
the vulnerability of the Chinese and Russian arsenals as well, but not as quickly as it is undermin-
ing North Korea’s force.
99. Nuclear weapon states that are weaker than their main rivals typically strive to develop sur-
vivable nuclear forces in order to stalemate more powerful foes. Counterforce tends to be the do-

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

The New Era of Counterforce 47

have an advantage in the race to build (and thwart) countermeasures. Como
Brendan Green and Austin Long observe about the Cold War ASW competi-
ción, A NOSOTROS. superiority in passive acoustics helped the United States quiet its
own SSBNs, which in turn allowed it to practice and hone its tracking capabili-
ties.100 Expertise in sensors and countermeasures go hand in hand. Tal vez
most importantly, many countermeasures reduce one vulnerability at the cost
of exacerbating others. Por ejemplo, limiting communications between mobile
missiles or submarines and their command authorities reduces vulnerability to
signals intercepts, but it increases vulnerability to attacks designed to sever (o
simular) their command and control.101 Avoiding coastal roads neutralizes
offshore sensors, but it channels forces into a smaller zone, easing the search
problema. Even the simplest countermeasures, such as increasing security near
sensitive facilities to prevent the emplacement of unattended ground sensors
or improving air defenses around key sites to thwart UAVs, may cue hunters
to the presence of high-value sites.

Segundo, the potential targets of disarming strikes cannot merely respond to a
single counterforce technology; they must respond to a daunting list of them.
The revolutions in accuracy and sensing have had multiple, synergistic effects
in bolstering counterforce. The task for hiders is not simply to thwart a single
platform, such as SAR satellites, but rather to develop countermeasures to the
entire array of (conocido) capabilities deployed by the hunters. Por ejemplo,
North Korea may ªnd ways to interfere with U.S. radar satellites, but that
still leaves its missiles vulnerable to detection by optical satellites; UAVs;
unattended ground sensors; and a variety of tagging, tracking, and locat-
ing capabilities.

Tercero, some vulnerabilities are difªcult to ªx. In the late 1960s, the Soviet
Union learned that its SSBNs were being tracked by the United States, pero
took more than a decade to counter this U.S. capacidad. Consider the challenge
faced by China today in building a survivable ballistic missile submarine force;
China deployed its ªrst submarines in the 1960s, but more than half a century
later Chinese submarines are still so noisy that experts predict it will be de-
cades before Beijing can ªeld survivable submarines.102

main of the strong, which have the resources to pursue it and the incentive to negate the
stalemating forces of their weaker enemies.
100. Green and Long, “The Role of Clandestine Capabilities in World Politics.”
101. The United States had Cold War programs to jam and simulate communications linking So-
viet nuclear command and control with deployed forces. See Benjamin B. pescador, “CANOPY
WING: Estados Unidos. War Plan That Gave the East Germans Goose Bumps,” International Journal of In-
telligence and Counterintelligence, volumen. 27, No. 3 (2014), páginas. 431–464.
102. See Wu Riqiang, “Survivability of China’s Sea-Based Nuclear Forces," Ciencia & Global
Security, volumen. 19, No. 2 (2011), páginas. 91–120; and comments by Chinese analysts in “Why Is
China Modernizing Its Nuclear Arsenal?” transcript, Carnegie International Nuclear Policy Con-

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

Seguridad Internacional 41:4 48

The battle between countermeasures and corresponding attempts to defeat
them is under way, and its outcome will likely depend on the strategic context.
Rich countries with advanced research and development infrastructure are de-
veloping technology and doctrine to protect their nuclear forces in the face of
improvements in weapons accuracy and remote sensing. Weaker countries
with modest resources, sin embargo, will be hard pressed to develop effective
countermeasures to the full spectrum of emerging means of counterforce.

Conclusión

For most of the nuclear age, there were many impediments to effective
counterforce. Weapons were too inaccurate to reliably destroy hardened tar-
gets; fratricide prevented many-on-one targeting; the number of targets to
strike was huge; target intelligence was poor; conventional weapons were of
limited use; and any attempt at disarming an adversary would be expected to
kill vast numbers of people. Hoy, in stark contrast, highly accurate weapons
aim at shrinking enemy target sets. The fratricide problem has been swept
away. Conventional weapons can destroy most types of counterforce targets,
and low-fatality nuclear strikes can be employed against others. Target intelli-
gence, especially against mobile targets, remains the biggest obstacle to effec-
tive counterforce, but the technological changes under way in that domain are
revolucionario. Of the two key strategies that countries have employed since
the start of the nuclear age to keep their arsenals safe, hardening has been ne-
gated, and concealment is under great duress.

The new era of counterforce helps solve one of the enduring theoretical
puzzles of the nuclear age. Por décadas, scholars of the theory of the nuclear
revolution wondered why leaders seemed to be ignoring the profound impli-
cations of nuclear weapons for international politics. In theory, nuclear weap-
ons make states that possess them so secure that they need not engage in
traditional forms of competition with adversaries, such as arms racing, alliance
building, relative gains competition, and rivalry over strategic territory. En
práctica, all those behaviors have endured. Scholars blame the persistent dis-
crepancy between theory and practice on misperception, illogic, or other deci-
sionmaking pathologies. The new era of counterforce suggests, sin embargo, eso
leaders have been correct to perceive that stalemate can be broken, and that the
nuclear balance can vary dramatically across cases. If today’s secure arsenal
can become tomorrow’s ªrst-strike target, then there is little reason to expect

ference, Marzo 24, 2015, http://carnegieendowment.org/ªles/12-chinanucleararsenal240315w
intro-formatted.pdf.

D
oh
w
norte
oh
a
d
mi
d

F
r
oh
metro
h

t
t

pag

:
/
/

d
i
r
mi
C
t
.

metro

i
t
.

mi
d
tu

/
i
s
mi
C
/
a
r
t
i
C
mi
–
pag
d

F
/

4
1
4
9
2
0
7
9
7
3
4

/
i
s
mi
C
_
a
_
0
0
2
7
3
pag
d

b
y
gramo
tu
mi
s
t

oh
norte
0
8
S
mi
pag
mi
metro
b
mi
r
2
0
2
3

The New Era of Counterforce 49

the geopolitical competition between countries to end with the deployment of
seemingly secure nuclear weapons.

The policy implications of the new era of counterforce are also important.
Primero, if nuclear forces are becoming increasingly vulnerable to counterforce,
then states need to improve their retaliatory arsenals just to maintain the same
level of deterrence. Given that nuclear delivery systems are expensive and
must last for decades, the challenge for force planners is extraordinary: deploy
weapon systems that will remain survivable for multiple generations, even as
technology improves at an ever-increasing pace. Segundo, the growing threat to
nuclear arsenals (from nuclear strikes, conventional attacks, missile defenses,
ASW, and cyber operations) raises major questions about the wisdom of cut-
ting the size of nuclear arsenals. In the past, many arms control advocates be-
lieved that arms cuts reduced the incentives for disarming strikes; si
right or wrong in the past, that assumption is increasingly dubious as a recipe
for deterrence stability today.

Finalmente, leaps in accuracy and remote sensing should reopen debates in the
United States about the wisdom of pursuing effective counterforce systems.
Fielding those capabilities—nuclear, conventional, and other—may prove in-
valuable: enhancing deterrence during conventional wars and, if deterrence
fails, allowing the United States to defend itself and its allies. Mejorando
counterforce capabilities, sin embargo, may trigger arms races and other dynamics
that exacerbate political and military conditions. In the past, technological con-
ditions bolstered those who favored restraint: disarming strikes seemed im-
posible, so enhancing counterforce would likely trigger arms racing without
much strategic beneªt. Hoy, technological trends appear to validate the ad-
vocates of counterforce: remote sensing, conventional strike capabilities, ASW,
and cyberattack techniques will continue to improve and increasingly threaten
strategic forces whether or not the United States seeks to maximize its counter-
force capabilities. In this new era of counterforce, technological arms racing
seems inevitable, so exercising restraint may limit options without yielding
much beneªt.

Nuclear deterrence can be robust, but nothing about it is automatic or ever-
lasting. Nuclear stalemate might endure among some pairs of states, and tech-
nology could someday reestablish the ease of deploying survivable arsenals.
Hoy, sin embargo, survivability is eroding, and it will continue to do so in the
futuro previsible. Weapons will grow even more accurate. Sensors will im-
prove. The new era of counterforce will likely yield beneªts to those countries
that best adapt to the new landscape, and costs to those that fall behind. El
ªrst step in understanding these dynamics is to recognize the new strategic re-
ality confronting nuclear powers today.