North Korea’s 2009 Nuclear Test:
Containment, Monitoring, Implications
Jonathan Medalia
Specialist in Nuclear Weapons Policy
April 2, 2010
Congressional Research Service
7-5700
www.crs.gov
R41160
CRS Report for Congress
P
repared for Members and Committees of Congress
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Summary
On May 25, 2009, North Korea announced that it had conducted its second underground nuclear
test. Unlike its first test, in 2006, there is no public record that the second one released radioactive
materials indicative of a nuclear explosion. How could North Korea have contained these
materials from the May 2009 event and what are the implications?
As background, the Comprehensive Nuclear-Test-Ban Treaty (CTBT) would ban all nuclear
explosions. It was opened for signature in 1996. Entry into force requires ratification by 44 states
specified in the treaty, including the United States and North Korea. As of April 2010, 151 states,
including 35 of the 44, had ratified. North Korea has not signed the CTBT. President Clinton
signed it in 1996; in 1999, the Senate voted not to consent to its ratification. In 2009, President
Obama pledged to press for its ratification.
The treaty establishes a verification mechanism, including an International Monitoring System
(IMS) to detect nuclear tests. Three IMS technologies detect waves that pass through the oceans
(hydroacoustic), Earth (seismic), or atmosphere (infrasound); a fourth detects radioactive material
from a nuclear test. Scientists concur that only the latter proves that an explosion was nuclear.
Some believe that deep burial and other means can contain radioactive effluents. Another view is
that containment is an art as much as a science. The United States learned to improve containment
over several decades. Yet by one estimate, North Korea contained over 99.9% of the radioactive
effluents from its 2009 test. It might have done so by application of lessons learned from its 2006
test or the U.S. nuclear test experience, use of a higher-yield device, release of material below the
detection threshold, good luck, or some combination. Alternatively, the 2009 event may have
been a nonnuclear explosion designed to simulate a nuclear test.
Containment could be of value to North Korea. It could keep radioactive fallout from China,
Japan, Russia, or South Korea, averting an irritant in relations with them. It could prevent
intelligence services from gathering material that could reveal information about the weapon that
was tested. It could permit North Korea to host nuclear tests by other nations, such as Iran; while
such tests would be detected by seismic means, they could not be attributed to another nation
using technical forensic means if effluents, especially particles, were contained.
An issue for Congress is how containment could affect CTBT prospects. Supporters might argue
that explosion-like seismic signals without detected radioactive material would lead to calls for an
onsite inspection. Opponents might claim that only detection of radioactive material proves that a
nuclear explosion occurred. Both would note that inspections could not be required unless the
treaty entered into force, supporters to point to a benefit of the treaty and opponents to note that
North Korea could block inspections by not ratifying the treaty. Congress may also wish to
consider options to improve monitoring capability, such as supporting further research on test
signatures, improving the capability of monitoring systems, and deploying more monitoring
equipment. This report may be updated, especially if North Korea conducts another test.
Related CRS reports include CRS Report RL34256, North Korea’s Nuclear Weapons: Technical
Issues, which summarizes open-source information on that nation’s nuclear weapons program,
including fissile material and warhead estimates, and assesses developments toward
denuclearization; and CRS Report R40684, North Korea’s Second Nuclear Test: Implications of
U.N. Security Council Resolution 1874, which analyzes possible economic effects on North
Korea of sanctions and vessel inspections that Resolution 1874 puts in place.
Congressional Research Service
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Contents
Background ................................................................................................................................ 1
The North Korean Nuclear Tests .................................................................................................2
The 2006 Test ....................................................................................................................... 2
The 2009 Test ....................................................................................................................... 3
Monitoring and Containing Nuclear Tests.................................................................................... 5
Monitoring, Verification, Intelligence .................................................................................... 5
Monitoring Nuclear Tests ...................................................................................................... 6
What Radioactive Materials Can a Nuclear Test Release into the Atmosphere and
How Can They Be Detected at a Distance?......................................................................... 9
How Can Radioactive Material Be Contained? .................................................................... 18
Potential Value of Containment for North Korea........................................................................ 22
Issues for Congress ................................................................................................................... 23
Implications for the CTBT .................................................................................................. 24
Improving Monitoring and Verification Capability .............................................................. 25
Conduct Research to Better Characterize Nuclear Explosions and Containment............. 26
Deploy More Monitoring Equipment............................................................................. 27
Improve the Capability of Monitoring Systems.............................................................. 29
Look for New Signatures to Help Determine If a Test Is Nuclear ................................... 33
Figures
Figure 1. Gamma-ray Signatures of Four Radioactive Isotopes or Isomers of Xenon ................. 13
Figure 2. Atmospheric Transport Modeling ............................................................................... 15
Figure 3. Radionuclide Monitoring Stations of the International Monitoring System.................. 17
Figure 4. Venting of Nuclear Tests............................................................................................. 19
Figure 5. Detection “Window” for Argon-37 and Xenon-133..................................................... 30
Contacts
Author Contact Information ...................................................................................................... 34
Congressional Research Service
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Background
On May 25, 2009, North Korea announced that it had conducted a nuclear test.1 The test produced
seismic signals characteristic of an explosion, indicating that they were generated by human
activity. They were detected by at least 61 seismic stations. However, no radioactive materials
were reportedly detected, in contrast to the first North Korean test on October 9, 2006. Such
materials could confirm that the test was nuclear. Although a sample size of one is not sufficient
to draw conclusions with high confidence, the possible ability of North Korea to contain
radioactive materials from a nuclear test could be of value for that nation. This report presents
what is known publicly about the tests, discusses detection and containment of nuclear tests,
explores the possible significance of containment for North Korea, and raises, as issues for
Congress, implications for the Comprehensive Nuclear-Test-Ban Treaty (CTBT) and possible
means of improving U.S. and international ability to monitor nuclear testing.
States currently possessing nuclear weapons would probably need to conduct nuclear tests to
develop more advanced designs and, some argue, to ensure that existing weapons are safe, secure,
and reliable. States with fledgling nuclear weapon programs could design and deploy the simplest
type of nuclear weapon without testing,2 but such weapons make very inefficient use of scarce
fissile material and are heavy and bulky. To develop small, rugged, powerful warheads for long-
range missiles, these states would need to conduct nuclear tests.
Nonnuclear experiments can answer some questions important to the design of nuclear weapons,
but many processes essential to the functioning of a nuclear weapon can only be studied under the
conditions of an actual nuclear test. Each test not only shows whether a device “works” or “fails,”
but also provides much more data. Many technical disciplines contribute to a test, and each gains
data from it. Weapon designers learn how the design might be improved, physicists gain data on
the science underlying nuclear explosions, metallurgists gain data on how uranium or plutonium
deforms under pressure, engineers can discover unanticipated flaws arising from manufacturing
processes, physicists who design computer models of nuclear weapon performance gain data to
refine their models, electrical engineers gain data to improve the instrumentation for collecting
nuclear test data, radiochemists can analyze radioactive samples from the test for data on the yield
and performance of the device, and those involved in preventing radioactive material from
escaping from the test gain data to improve containment.
Because testing is crucial for developing weapons, efforts to ban nuclear tests have been
underway for decades as an arms control measure.3 The multilateral 1963 Limited Test Ban
Treaty banned atmospheric, space, and underwater tests. The U.S.-Soviet Threshold Test Ban
1 “Text of the North Korean Announcement of Nuclear Test,” Reuters, May 25, 2009, http://www.nytimes.com/2009/
05/25/world/asia/25nuke-text.html. For information on North Korea’s nuclear weapons program, see CRS Report
RL34256, North Korea’s Nuclear Weapons: Technical Issues, by Mary Beth Nikitin.
2 In a “gun assembly” weapon, one piece of uranium-235 is fired into another such piece to create a critical mass. The
Hiroshima bomb was of this design. U.S. scientists had such confidence in this design that they did not test it prior to
use. Gun-assembly weapons can only use highly enriched uranium, not plutonium.
3 For a history of nuclear test bans, see Appendix A in CRS Report RL34394, Comprehensive Nuclear-Test-Ban
Treaty: Issues and Arguments, by Jonathan Medalia.
Congressional Research Service
1
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Treaty and Peaceful Nuclear Explosions Treaty, signed in 1974 and 1976, respectively, limited
underground nuclear explosions to a yield of 150 kilotons.4 Both entered into force in 1990.
In an attempt to extend these bans to cover all nuclear tests, negotiations on the CTBT were
completed in 1996. The treaty’s basic obligation is to ban all nuclear explosions. It establishes an
International Monitoring System (IMS) to monitor signs of an explosion. The treaty contains
procedures for authorizing and conducting on-site inspections (OSIs), which would search the site
of a suspected nuclear test for evidence of the test, and permits data from national technical
means of verification as well as from the IMS to be used to support a request for an OSI.
As of April 2010, 182 nations had signed the treaty and 151 of them had ratified.5 To enter into
force, 44 specified nations, basically those with a nuclear reactor in 1995 or 1996, must all ratify.
As of April 2010, 35 had done so; the others are China, Egypt, India, Indonesia, Iran, Israel,
North Korea (the Democratic People’s Republic of Korea, DPRK), Pakistan, and the United
States. The U.S. Senate voted not to give its advice and consent to ratification of the treaty, 48 for,
51 against, and 1 present, in 1999. Two uncertainties that led to its defeat concerned U.S. ability
to verify compliance with the treaty and U.S. ability to maintain its nuclear stockpile without
testing. In April 2009, President Obama pledged to pursue U.S. ratification of the CTBT
“immediately and aggressively.”6
The North Korean Nuclear Tests
The 2006 Test
North Korea conducted its first nuclear test on October 9, 2006. It was clearly nuclear because it
released radioactive materials. The U.S. Office of the Director of National Intelligence (ODNI)
released this statement: “Analysis of air samples collected on October 11, 2006 detected
radioactive debris which confirms that North Korea conducted an underground nuclear explosion
in the vicinity of P’unggye on October 9, 2006. The explosion yield was less than a kiloton.”7
(ODNI declined to state whether “debris” referred to particulates, gases, or both.8) According to a
press report, “American intelligence agencies have concluded that North Korea’s test explosion
last week was powered by plutonium that North Korea harvested from its small nuclear reactor,
according to officials who have reviewed the results of atmospheric sampling since the blast.”9 In
a similar vein, the Comprehensive Nuclear-Test-Ban Treaty Organization (CTBTO) Preparatory
4 One kiloton is equivalent to the explosive force of 1,000 tons of TNT. For comparison, the explosive yield of the
Hiroshima bomb was 15 kilotons.
5 For the status of signatures and ratifications, and text of the treaty, see the website of the Comprehensive Nuclear-
Test-Ban Treaty Organization Preparatory Commission, http://www.ctbto.org. For current developments on the CTBT,
see CRS Report RL33548, Comprehensive Nuclear-Test-Ban Treaty: Background and Current Developments, by
Jonathan Medalia.
6 U.S. White House. Office of the Press Secretary. “Remarks by President Barack Obama,” Hradcany Square, Prague,
Czech Republic, April 5, 2009, http://www.whitehouse.gov/the_press_office/Remarks-By-President-Barack-Obama-In-
Prague-As-Delivered/.
7 U.S. Office of the Director of National Intelligence. Public Affairs Office. “Statement by the Office of the Director of
National Intelligence on the North Korean Nuclear Test,” news release 19-06, October 16, 2006.
8 Information provided by Office of the Director of National Intelligence, personal communication, July 17, 2009.
9 Thom Shanker and David Sanger, “North Korean Fuel Identified as Plutonium,” New York Times, October 17, 2006.
Congressional Research Service
2
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Commission (PrepCom) stated, “Two weeks after the event, the radionuclide noble gas station at
Yellowknife, Canada, registered a higher concentration of Xenon 133. Applying atmospheric
transport models to backtrack the dispersion of the gas, its registration at Yellowknife was found
to be consistent with a hypothesized release from the event in the DPRK.”10 The Swedish
Defence Research Agency (FOI) used atmospheric models at a much shorter distance. It flew
mobile xenon analysis equipment to South Korea and began collecting samples within three days
of the test. “All the samples were found to contain radioactive xenon and, in combination with
meteorological information, FOI were able to conclude that the gas did, with a relatively high
level of probability, originate from the area in North Korea where the explosion took place.”11
The 2009 Test
North Korea announced on May 25, 2009, that it had conducted a second nuclear test. ODNI
stated: “The U.S. Intelligence Community assesses that North Korea probably conducted an
underground nuclear explosion in the vicinity of P'unggye on May 25, 2009. The explosion yield
was approximately a few kilotons. Analysis of the event continues.”12 The lack of certainty as to
whether the test was nuclear arises because seismic signals, including those detected by 61
stations of the IMS,13 were consistent with a nuclear test, and seismic signals from the 2006 and
2009 events were very similar,14 but open sources did not report the detection of physical
evidence that would provide conclusive proof of a nuclear test, such as certain radioactive
isotopes of noble gases or radioactive particulates (i.e., fallout). For example, the CTBTO
PrepCom stated,
The detection of radioactive noble gas, in particular xenon, could serve to corroborate the
seismic findings. Contrary to the 2006 announced DPRK nuclear test, none of the CTBTO’s
noble gas stations have detected xenon isotopes in a characteristic way that could be
attributed to the [2009] DPRK event so far, even though the system is working well and the
network’s density in the region is considerably higher than in 2006. …
Nor have CTBTO Member States using their own national technical means reported any
such measurements. Given the relatively short half-life of radioactive xenon (between 8
10 Comprehensive Nuclear-Test-Ban Treaty Organization Preparatory Commission, “The CTBT Verification Regime
Put to the Test – The Event in the DPRK on 9 October 2006,” http://www.ctbto.org/press-centre/highlights/2007/the-
ctbt-verification-regime-put-to-the-test-the-event-in-the-dprk-on-9-october-2006/page-2/. An atmospheric transport
model, as discussed later, shows how winds move gases or particles through the atmosphere. Xenon is a noble gas, i.e.,
one that is chemically inert. Nuclear explosions create radioactive isotopes of some noble gases. The IMS has some
stations that can detect radioactive isotopes of xenon, such as xenon-133.
11 Swedish Defence Research Agency (FOI), “FOI found radioactive xenon following explosion in North Korea,” press
release, December 19, 2006, http://www.foa.se/FOI/Templates/NewsPage____5412.aspx.
12 U.S. Office of the Director of National Intelligence. Public Affairs Office. “Statement by the Office of the Director
of National Intelligence on North Korea’s Declared Nuclear Test on May 25, 2009,” ODNI News Release No. 23-09,
June 15, 2009, http://www.dni.gov/press_releases/20090615_release.pdf.
13 Comprehensive Nuclear-Test-Ban Treaty Preparatory Commission, “Homing in on the Event,” May 29, 2009,
http://www.ctbto.org/press-centre/highlights/2009/homing-in-on-the-event/.
14 For seismograms of these two events and of an earthquake from the same region, see Won-Young Kim, Paul
Richards, and Lynn Sykes, “Discrimination of Earthquakes and Explosions Near Nuclear Test Sites Using Regional
High-Frequency Data,” poster SEISMO-27J presented at the International Scientific Studies Conference, June 2009,
http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/SEISMO-27J%20%28US%29%20-
%20Won_Young_Kim%20_Paul_Richards%20and%20Lynn_Sykes.pdf.
Congressional Research Service
3
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
hours and 11 days, depending on the isotope), it is unlikely that the IMS will detect or
identify xenon from this event after several weeks.15
It would be desirable to establish if the event was nuclear because the possibility that chemical
explosives caused the seismic waves could undermine confidence in the ability to verify
compliance with the CTBT. Earthquakes can be differentiated from explosions (whether chemical
or nuclear) because their seismic waves have different characteristics. But while seismic signals
from the 2009 event were consistent with a nuclear test, it is very difficult to differentiate between
seismic signals generated by a nuclear test and a chemical explosion of comparable energy, so it
is conceivable that the test was nonnuclear. Geoffrey Forden, a scientist at MIT, posits a scenario
in which a room could be filled with 2,500 tons of TNT, enough to create an explosion within the
yield range estimated for the 2009 North Korean test, in two months using about four 10-ton
truckloads per day. He finds this scenario “quite doable and to be potentially undetectable by the
West.”16 The United States conducted large aboveground17 and underground18 tests using
chemical explosives to simulate some effects of nuclear explosions.
The CTBTO PrepCom cites analysis that rejects the chemical-explosive possibility:
Verification technology experts such as Professor Paul Richards from the Lamont-Doherty
Earth Observatory, Columbia University, USA, considered the scenario of a “bluff”, i.e. the
creation of a nuclear explosion-like seismic signal using conventional explosives. While
technically possible, he stated that it was highly implausible. As CTBTO seismic data have
clearly indicated an explosion of a yield many times greater than that of 2006, it would have
required several thousand tons of conventional explosives to be fired instantaneously.
Richards explained that such a massive logistical undertaking would have been virtually
impossible under the prevailing circumstances and would not have escaped detection.19
15 Comprehensive Nuclear-Test-Ban Treaty Organization Preparatory Commission, “Experts Sure about the Nature of
the DPRK Event,” (referring to the May 2009 North Korean test), June 12, 2009, http://www.ctbto.org/press-centre/
highlights/2009/experts-sure-about-nature-of-the-dprk-event/.
16 Geoffrey Forden, “DPRK: Drilling for Nuke Verification,” Arms Control Wonk, July 20, 2009,
http://www.armscontrolwonk.com/2392/dprk-nuke-verification-will-require-drilling.
The amount of chemical explosive needed to implement this scenario may be less than it would appear. For example,
one study observes a range of calculations in which it takes a nuclear explosive with a yield of 1 to 2 kilotons to
produce the seismic signature of 1 kiloton of chemical explosive. That study calculates that 1.25 kilotons of nuclear
explosives produces the seismic signature of 1 kiloton of chemical explosive. James Kamm and Randy Bos,
“Comparison of Chemical and Nuclear Explosions: Numerical Simulations of the Non-Proliferation Experiment,” Los
Alamos National Laboratory report LA-12942-MS, UC-700 and UC-703, June 1995, pp. 89-92, http://www.osti.gov/
bridge/servlets/purl/72900-YlaqIV/webviewable/72900.pdf. Another study finds, “The basic results from the U.S.
Department of Energy’s (DOE’s) Non-Proliferation Experiment (NPE) for seismic signal generation are that the source
function for a chemical explosion is equivalent to that of a nuclear explosion of about twice the yield.…” Marvin
Denny et al., “Seismic Results from DOE’s Non-Proliferation Experiment: A Comparison of Chemical and Nuclear
Explosions,” Lawrence Livermore National Laboratory, UCRL-JC-119214 preprint, January 1995, p. 1,
http://www.osti.gov/energycitations/servlets/purl/93630-EAQRwr/webviewable/93630.pdf. (The NPE was an
underground nonnuclear explosion of about 1 kiloton yield conducted in 1993 to simulate a nuclear explosion.)
17 In 1985, the Defense Nuclear Agency conducted a test, “Minor Scale,” using 4,800 tons of high explosive. U.S.
Department of Defense. Defense Threat Reduction Agency. Defense’s Nuclear Agency, 1947-1997, Washington, 2002,
pp. 269-270.
18 See “Nonproliferation, Arms Control, and International Security,” Lawrence Livermore National Laboratory, c.
1995, p. 8, https://www.llnl.gov/etr/pdfs/01_95.02.pdf; and “Non-Proliferation Experiment (NPE),” Globalsecurity.org,
http://www.globalsecurity.org/wmd/ops/npe.htm.
19 Comprehensive Nuclear-Test-Ban Treaty Organization Preparatory Commission, “Experts Sure about the Nature of
the DPRK Event.”
Congressional Research Service
4
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Specifically, satellite imagery might have detected preparations for a large chemical explosion,
and if the explosives did not detonate instantaneously, they could have caused a seismic signal
with characteristics of a chemical explosion.20
Questions about whether the May 25 event was nuclear or nonnuclear remain unresolved. Onsite
inspections could prove conclusively that a test was nuclear, but they could only be conducted if
the CTBT were to enter into force, or if North Korea gave its permission outside the treaty. Other
ways to establish (but not necessarily prove) that an explosion was nuclear are nonseismic means,
such as communications intercepts and satellite imagery; note that high-quality commercial
satellite imagery is available for purchase. The CTBT envisions that some monitoring
technologies not part of the IMS could be added (Article IV, paragraphs 11, 23) to that system if
agreed pursuant to the treaty’s amendment process (Article VII).
The apparent absence of radioactive material released from the May 25 event raises several
questions: How can such material be detected? How might North Korea have contained its second
test? What are some implications of successful containment for North Korea? What issues do
detection and containment raise for Congress? This report now turns to these questions.
Monitoring and Containing Nuclear Tests
Monitoring, Verification, Intelligence
Central concerns in negotiating an arms control agreement are to establish a regime that facilitates
detection of cheating and to ensure, insofar as possible, that a state party to the treaty cannot gain
an advantage by cheating. To this end, the CTBT, the CTBTO, and individual nations would take
several interlocking steps. The first is monitoring, which provides technical data on suspicious
events. The treaty establishes the IMS, which is one of several components in the verification
regime established by Article IV of the treaty. Verification refers to determining whether a nation
is in compliance with its treaty obligations, which in this case means determining whether a
suspicious event was a nuclear test. The treaty establishes the verification regime in great detail:
Article IV takes up nearly half the treaty, and the Protocol, which provides details on the
verification regime, is nearly as long as the treaty itself. The verification regime, in addition to the
IMS, includes provisions for consultation and clarification of suspicious events; an International
Data Center (IDC) to analyze IMS data and distribute the results to states parties to the treaty;
detailed provisions for on-site inspections; and confidence-building measures. As one of its
functions, the Provisional Technical Secretariat of the CTBTO PrepCom operates the verification
regime.21
20 A large chemical explosion designed to mimic a nuclear explosion but that is not detonated all at once could generate
signals that can be differentiated from a nuclear explosion. Similarly, mining explosions are often ripple-fired and, as a
result, generate a different seismic signal than would a nuclear explosion.
21 The CTBTO and Technical Secretariat would come into existence only upon entry into force of the CTBT. As an
interim measure, the states that had signed the CTBT in 1996 adopted a resolution establishing the CTBTO Preparatory
Commission as a means “to ensure the rapid and effective establishment of the future Comprehensive Nuclear-Test-
Ban Treaty Organization.” “Resolution Establishing the Preparatory Commission for the Comprehensive Nuclear-Test-
Ban Treaty Organization,” adopted November 19, 1996, U.N. document CTBT/MSS/RES/1. Upon entry into force of
the CTBT, the CTBTO Preparatory Commission would become the CTBTO and the Provisional Technical Secretariat
would become the Technical Secretariat.
Congressional Research Service
5
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
The treaty (Article IV, paragraph 14) does not direct the Technical Secretariat to determine that a
particular nation has cheated. But while some IMS sensors (e.g., seismic) can only provide
evidence that a nuclear test may have occurred, IMS radionuclide sensors would prove that a
nuclear test had occurred if they found certain types of radioactive debris, and an OSI would
prove that a nuclear test had occurred at a particular location if it found the radioactive cavity left
by an underground nuclear test. Once the verification regime provides data and analysis of
suspected or actual nuclear tests to the states parties, the role of the Technical Secretariat would
end. In the event of a suspected violation, the Conference of the States Parties, pursuant to Article
V, paragraphs 3 and 4, “may recommend to States Parties collective measures which are in
conformity with international law,” or the conference or the Executive Council “may bring the
issue … to the attention of the United Nations.”22
In judging how to respond, nations would use various means to determine whether the event was
a nuclear test, combining data and analysis from multiple sources, such as the International Data
Center; national technical means of verification; non-technical means, such as information from
people and open sources; and other governments and nongovernmental organizations. But
because of background noise, limitations of detectors, etc., it is almost a truism that there will
always be some threshold below which a nuclear test cannot be unambiguously identified by
seismic and other IMS technologies. As a result, a decision on a nation’s response to a suspected
nuclear test would depend not on perfect verification but on effective verification. In 1988, Paul
Nitze offered a widely-used definition: by effective verification, “[w]e mean that we want to be
sure that, if the other side moves beyond the limits of the treaty in any militarily significant way,
we would be able to detect such violation in time to respond effectively, and thereby deny the
other side the benefit of the violation.”23 Judgments on the effectiveness of verification have been
crucial in consideration of past nuclear testing treaties; this is likely to be the case for any future
CTBT debate as well.
Beyond that, some nations can be expected to use their intelligence capabilities to learn more than
whether an event was a nuclear test. They will want to know such details as weapon yield,
weapon fuel (uranium or plutonium), and weapon design to understand how quickly a nation’s
weapons program is advancing, what problems it is encountering, and what development path it is
following. In some cases, such as a nuclear detonation in a remote ocean area or by terrorists,
technical analysis may support attribution of the detonation to a specific nation.
Monitoring Nuclear Tests
Because a nuclear test generates immense amounts of energy and radioactive material, it presents
many signatures by which it can be detected, some at a distance of thousands of miles, others
only on site. Atmospheric tests are easy to detect because of their radioactive fallout. China
conducted the most recent atmospheric nuclear test in October 1980.24 Underwater tests are also
easy to detect, though attribution may be a problem. Satellites (not part of the IMS) can detect
nuclear tests in space, though some evasion scenarios have been suggested. A particularly
22 Pursuant to Article II of the treaty, the Conference of the States Parties is composed of all states that are parties to the
treaty. The Executive Council has 51 member states and is the executive organ of the CTBTO.
23 U.S. Congress. Senate. Committee on Foreign Relations. The INF Treaty. S.Hrg. 100-522, 100th Congress, 2nd
Session, 1988, part 1, p. 289.
24 Nuclear Threat Initiative, “China’s Nuclear Tests: Dates, Yields, Types, Methods, and Comments,”
http://www.nti.org/db/china/testlist.htm.
Congressional Research Service
6
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
difficult environment in which to detect clandestine tests is underground, and is the one relevant
to North Korean nuclear testing, since both of its tests were conducted in that manner.
When completed,25 the IMS will have 321 monitoring stations, 16 laboratories, and an
International Data Center to process data. A Global Communications Infrastructure will link IMS
facilities, the data center, and member states. As of April 2010, 245 monitoring stations and 10
laboratories were operational.26 The IMS uses four technologies to detect nuclear tests:27
Seismic: Seismographs detect various types of waves (e.g., pressure waves) moving through the
Earth.28 The science of seismology has made great progress over the last half-century in filtering
out seismic signals characteristic of explosions from other seismic signals, such as by using more
sensitive instruments and more elaborate data-processing algorithms. However, there are some
scenarios for concealing the seismic signals from low-yield tests that have been hotly debated for
many years, as discussed later. Analysis of seismic and other waves cannot by itself prove that an
explosion was nuclear; for example, it is difficult if not impossible to distinguish by seismic
means between a nuclear explosion and a large (e.g., 1,000-ton) chemical explosion that is
designed to mimic a nuclear explosion and is successfully conducted.
Hydroacoustic: Hydrophones can detect a very small underwater chemical explosion at distances
of thousands of miles as pressure waves generated by the explosion move through the water, so
an underwater nuclear explosion would be readily detected.29 While this method might detect
some nuclear tests conducted on land, such as on a small island, there were no reports that it was
of use in detecting the 2009 North Korean test.
Infrasound: Sensors measure very small changes in atmospheric pressure caused by very low
frequency acoustic waves. Infrasound sensors are not at present intended for monitoring of
underground nuclear explosions, though they did detect the 2009 North Korean test.30 The
observed magnitude of the infrasound signal, approximately three tons of TNT equivalent31 vs.
several kilotons for the seismic signal,32 indicated the explosion was not at the Earth’s surface.33
25 While the IMS is to be completed by the time the CTBT enters into force, it could be completed sooner. CTBTO
PrepCom believes it can project 90% completion, but the remainder depends on political, financial, and environmental
factors. The support from Member States and countries hosting stations is necessary for completing the IMS.
Information provided by Comprehensive Nuclear-Test-Ban Treaty Preparatory Commission, March 3, 2010.
26 For an up-to-date list of these facilities, see http://www.ctbto.org/map/#ims.
27 For details on the IMS, see http://www.ctbto.org/verification-regime/.
28 The instrument that records seismic signals is a “seismograph” or “seismometer”; the visual record of these signals is
a “seismogram.” For an old but useful description of the science of seismic detection, see U.S. Congress, Office of
Technology Assessment, Seismic Verification of Nuclear Testing Treaties, OTA-ISC-361, 139 pages, 1988,
http://www.fas.org/ota/reports/8838.pdf.
29 For example, IMS hydrophones near the coast of Chile detected signals from an underwater detonation of 20
kilograms of TNT, a tiny fraction of the yield of a nuclear weapon, off the coast of Japan, 16,300 km away.
International Scientific Studies Conference (summary brochure), Vienna, Austria, June 10-12, 2009, p. 3,
http://www.ctbto.org/fileadmin/user_upload/pdf/ISSAFC2_Web.pdf.
30 Il-Young Che et al., “Infrasound Observation of the Apparent North Korean Nuclear Test of 25 May 2009,”
Geophysical Research Letters, vol. 36, L22802, doi: 10.1029/2009GL041017, 2009, 5 p.
31 Ibid.
32 Vitaly Fedchenko, “North Korea’s Nuclear Test Explosion, 2009,” Stockholm International Peace Research Institute
(SIPRI) fact sheet, December 2009, pp. 3-4, http://books.sipri.org/files/FS/SIPRIFS0912.pdf.
33 The difference in apparent magnitude (or yield) of the infrasound and seismic signals results from the great
difference in density and compressibility between air and rock. Because the Earth is so stiff, even the relatively high
(continued...)
Congressional Research Service
7
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Radionuclides: To gain direct physical evidence that an explosion was nuclear, the IMS monitors
for radioactive particles and gases. When complete, it will have 80 stations that can detect
radioactive particulates, of which 40 will also have equipment to detect radioactive forms of
xenon, which are of particular value for detecting a nuclear explosion. The resulting data are sent
to the IDC and national data centers for analysis. As of April 2010, 58 particulate stations and 20
radioxenon stations were operational.
Many technological advances made in recent years improve the ability to detect nuclear
explosions. These include techniques to image the Earth’s inner structure to better understand
how that structure affects seismic waves,34 use of satellite-borne “radar imaging technology to
detect near-vertical surface deformations measuring less than 1 centimeter caused by underground
disturbances,”35 development of equipment to detect extremely low levels of radioactive noble
gases,36 development of computer models of wind patterns, use of seismic waves detected at
regional as well as longer distances to improve the ability to discriminate between earthquakes
and explosions,37 and the rollout of the IMS.38
To resolve uncertainties over whether suspicious events—such as any that generate explosion-like
seismic signals but do not release radioactive material—are nuclear, the CTBT provides for onsite
inspections (OSIs). OSIs would search for signatures that can only be detected at a test site, such
as small amounts of several radioactive noble gases, certain non-gaseous radioactive materials,39
physical signs of a test (e.g., melted snow, changes to vegetation, pebbles thrown in bushes by
ground shock), and the underground cavity formed by a nuclear test (found by drilling) that
would have tell-tale radioactive debris. The treaty and its Protocol include great detail on
authorization and conduct of an inspection. Article IV (verification), paragraph 56, of the treaty
requires each state party to permit OSIs.40 Of course, OSIs pursuant to the treaty could only occur
after the treaty had entered into force, though it is possible that OSIs could be done outside the
treaty regime, such as if requested by one country before the treaty enters into force to prove that
it had not conducted a nuclear test, or pursuant to a bilateral agreement permitting one state to
monitor another state’s nuclear test site.
(...continued)
pressure caused by an underground nuclear explosion moves the surface only a little, while air is so compressible that
the small upward motion of the Earth’s surface caused by the explosion generates only a small atmospheric pressure
wave. Information provided by Raymond Jeanloz, Professor of Earth and Planetary Sciences, University of California,
Berkeley, e-mail, January 5, 2010.
34 Katie Walter, “Sleuthing Seismic Signals,” Science & Technology Review, March 2009, pp. 4-12.
35 Gabriele Rennie, “Monitoring Earth’s Subsurface from Space,” Science & Technology Review, April 2005, p. 5.
36 Paul R.J. Saey, “Ultra-Low-Level Measurements of Argon, Krypton and Radioxenon for Treaty Verification
Purposes,” ESARDA Bulletin, no. 36 (July 2007), pp. 42-56.
37 David Hafemeister, “Progress in CTBT Monitoring Since Its 1999 Senate Defeat,” Science and Global Security, vol.
15 (2007), pp. 159-160.
38 For 236 scientific posters from 2009 detailing various aspects of the IMS and technologies for monitoring nuclear
explosions, see International Scientific Studies, “Scientific Contributions,” http://www.ctbto.org/specials/the-
international-scientific-studies-project-iss/scientific-contribtutions/.
39 See James Ely et al., “Estimation of Ground-Level Radioisotope Distributions for Underground Nuclear Test
Leakage,” poster, International Scientific Study, http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/OSI-
17B%20%28US%29%20-%20James_Ely%20etal.pdf.
40 If a state party to the treaty refused to permit an OSI, Article V (“Measures to redress a situation and to ensure
compliance, including sanctions”) would presumably come into play. It provides for actions to “redress a situation,”
including collective measures and bringing the matter to the attention of the United Nations.
Congressional Research Service
8
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
The United States has its own technical means of detecting nuclear explosions, which the Air
Force Technical Applications Center (AFTAC) operates. AFTAC “operates and maintains a global
network of nuclear event detection sensors called the U.S. Atomic Energy Detection System.
Once the USAEDS senses a disturbance underground, underwater, in the atmosphere or in space,
the event is analyzed for nuclear identification and findings are reported to national command
authorities through Headquarters U.S. Air Force.”41 USAEDS predates the CTBT, and in the
course of negotiations for that treaty, some USAEDS monitoring stations were included as
contributing stations to the IMS. Similarly, in addition to performing independent analyses of
events as the operator of USAEDS, AFTAC has a formal role under the CTBT as the U.S.
National Data Center to receive data from the International Data Center.
What Radioactive Materials Can a Nuclear Test Release into the
Atmosphere and How Can They Be Detected at a Distance?
An underground nuclear test may be fully contained, or it may release two types of material,
particulates and gases, into the atmosphere. Either may prove that a nuclear test occurred. Gases
include radioactive isotopes of noble gases (gases that are chemically inert), such as krypton-85,
argon-37, and several xenon isotopes.
Krypton-85 is of little value for nuclear detection at a distance because a substantial background
of this isotope is present in the atmosphere. Most of it is generated by nuclear power plants and is
released when the spent fuel is reprocessed, but since its half-life is 10.76 years, some remains
from past atmospheric nuclear tests.42
Argon-37 (half-life 35.04 days) is of value for onsite inspections and is not—but perhaps could
be—used for long-range detection. It is produced when neutrons interact with calcium-40 in soil
or rock.43 While naturally-occurring neutrons produce a background of argon-37,44 that
mechanism would not produce a local concentration of the isotope. Therefore, finding a local
concentration of it at the site of a suspected underground nuclear explosion would be an indicator
of a nuclear explosion, making it of value for OSIs; one source calls it “a definitive and
unambiguous indicator of a nuclear underground explosion.”45 Some believe that argon-37 could
be detected at long range. Professor Roland Purtschert, Department of Climate and
Environmental Physics, University of Bern, Switzerland, an expert on argon-37, states,
I am very confident that instruments could be developed that could operate automatically at
remote locations to detect argon-37 from underground nuclear tests. At present, radioxenon
detection equipment used by the International Monitoring System concentrates xenon
41 U.S. Air Force. “Air Force Technical Applications Center,” http://www.afisr.af.mil/library/factsheets/factsheet.asp?
id=10309.
42 Argonne National Laboratory, “Krypton,” Human Health Fact Sheet, August 2005, 2 p., http://www.ead.anl.gov/pub/
doc/krypton.pdf.
43 The interaction is that when a neutron strikes the nucleus of a calcium-40 atom, the nucleus immediately emits an
alpha particle (two neutrons and two protons), producing argon-37.
44 These neutrons are generated by cosmic rays, naturally-occurring uranium, and other sources.
45 R. Purtschert, R. Riedmann, and H.H. Loosli, “Evaluation of Argon-37 as a means for identifying clandestine
subsurface nuclear tests,” Proceedings of the 4th Mini Conference on Noble Gases in the Hydrosphere and in Natural
Gas Reservoirs, Potsdam, Germany, February 28-March 2, 2007, p. 1, http://bib.gfz-potsdam.de/pub/minoga/
minoga_purtschert-r.pdf.
Congressional Research Service
9
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
isotopes from a large volume of air. It should be possible to separate argon from this same air
sample, keep the argon sample as a backup, and measure the fraction of argon-37 when
xenon isotopes indicate a possible nuclear explosion. The amount of argon-37 from a nuclear
test that is detected would depend on the bomb yield, the rate at which the isotope is released
from the underground explosion cavity to the atmosphere, and the sensitivity of the detection
system.
I also think that it would be desirable to develop such instruments and deploy them as part of
the International Monitoring System. The atmospheric background for argon-37 is very low
and constant (about 0.5 to 1 nuclei decaying per second per 1000 cubic meters of air).
Civilian sources generate large quantities of xenon and krypton isotopes. As a result,
elevated xenon concentrations, for example, become unambiguous indicators of a nuclear
explosion only in combination with atmospheric transport modeling. In contrast, there are
virtually no civilian sources for argon-37, and the background is low due to argon-37’s short
half-life. At the same time, the half-life is long enough to allow for the isotope to be
transported from the cavity of a nuclear explosion to the atmosphere.46
Since argon-37 is produced from neutron reactions on calcium in the soil, neutrons from a nuclear
explosion will produce that isotope if calcium is present. It appears that calcium is present almost
everywhere. For example, one text states that calcium “is the fifth most abundant element in the
earth’s crust.... Vast sedimentary deposits of [calcium carbonate], which represent the fossilized
remains of earlier marine life, occur over large parts of the earth’s surface.”47 It makes up some
4% of the Earth’s crust.48 Even if there were a potential test site devoid of calcium, it would be
difficult for a would-be evader to find that site and then to be certain that no calcium was present.
In addition, the soil would need to be free of potassium as well because a reaction of a neutron
with a potassium atom can produce argon-39, which would be detected in the same way as argon-
37.
One value of long-range detection of argon-37 from a nuclear test is that it has a longer half-life
than radioxenons, enabling detection at a distance for a longer time. A second value is that
ensuring that no argon-37 seeps out, in addition to making sure no other effluents leak out, would
increase the difficulty of conducting a nuclear test clandestinely. A third value, pointed out by
Charles Carrigan, a geophysicist at Lawrence Livermore National Laboratory, is that detection of
a seismic signal characteristic of an explosion followed by simultaneous detection of a spike in
radioxenons and in argon-37 would be a compelling indicator of a nuclear explosion.49
While the radiation emitted when argon-37 decays has a much lower energy than gamma rays
produced by radioactive decay of many other elements, that energy “can be detected with special
techniques with relative ease,” according to Ted Bowyer, a physicist at Pacific Northwest
National Laboratory who specializes in atmospheric detection of radioactive isotopes of noble
gases.50
46 Personal communication, February 5, 2010. These are Professor Purtschert’s personal views and not necessarily
those of any institution.
47 N.N. Greenwood and A. Earnshaw, Chemistry of the Elements, Oxford, England, Butterworth Heinemann, Publisher,
1998, p. 109.
48 U.S. National Aeronautics and Space Administration. “World Book at NASA,” http://www.nasa.gov/worldbook/
earth_worldbook.html.
49 Personal communication, January 29, 2010.
50 Personal communication, February 2, 2010.
Congressional Research Service
10
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
There are several uncertainties regarding the use of argon-37 for long-range detection of nuclear
explosions. First, what is the background level of that isotope from natural and human sources?
While the background appears to be low, a definitive conclusion would require further study.
Second, can an automated system for detecting this isotope be designed and fielded? While it can
be detected in the laboratory, or in the field using manual equipment, an automated system would
be needed if detectors are to be placed at remote locations, such as IMS radionuclide stations.
Carrigan notes a third uncertainty: the detectability of argon-37 would depend on the rate at
which it reaches the surface. If a nuclear test released a large quantity promptly, the isotope would
be much easier to detect at long range than if it were released over days or weeks.
Radioactive isotopes of xenon (“radioxenons”) are of great value for long-range detection, and
the noble gas detection equipment deployed at some IMS radionuclide stations monitors only for
them.51 They are produced by nuclear explosions and nuclear reactors. Nuclear explosions also
generate iodine-133 (half-life, 20.8 hours) and iodine-135 (half-life, 6.6 hours), which decay into
xenon-133 and xenon-135, respectively. Radioxenons can be detected in minute quantities at
great distances, but such detection must be accomplished soon after a nuclear test because of
short half-lives. The half-life of xenon-133, an isotope of particular value for identifying nuclear
explosions, is 5.24 days, so long-range detection can only be done within about 3 weeks of a
test.52 The other radioxenons of use for monitoring nuclear tests are xenon-135 (half-life, 9.14
hours), xenon-133m (half-life, 2.19 days), and xenon-131m (half-life, 11.84 days).53
Several techniques and technologies have greatly improved the ability to detect radioxenons
worldwide over the past several decades. As a result, it is possible to detect and identify a
particular form of radioxenon thousands of miles from its source within a few weeks of a nuclear
test, during which time it will have been reduced to a minute concentration through radioactive
decay and mixing with air. IMS equipment takes in large quantities of air, separates and collects
any xenon, and compresses the latter to a small volume. Various techniques then acquire data for
transmission to the IDC for analysis. One is gamma-ray spectroscopy.54 Gamma rays are high-
energy photons emitted by atomic nuclei when they undergo radioactive decay. Each radioxenon
emits gamma rays in a pattern, or spectrum, of energies that uniquely identifies its source. Figure
1 illustrates the combined spectra of four radioxenons. The horizontal axis has a range of energies
measured in keV, or thousands of electron volts; the vertical axis records the number of gamma
51 Information provided by Comprehensive Nuclear-Test-Ban Treaty Preparatory Commission, personal
communication, July 30, 2009. For a description of how the radioxenon equipment works, see Kalinowski et al., “The
Complexity of CTBT Verification. Taking Noble Gas Monitoring as an Example,” Complexity, vol. 14, no. 1,
published online July 14, 2008, pp. 92-93.
52 Comprehensive Nuclear-Test-Ban Treaty Organization Preparatory Commission, “Experts Sure about Nature of the
DPRK Event.”
53 Half-life data are from Lawrence Berkeley Laboratory, “Exploring the Table of Isotopes: Isotopes of Xenon (Z=54),”
http://ie.lbl.gov/education/parent/Xe_iso.htm. Xenon-131m is of limited value for detecting nuclear explosions because
they generate very little of it, and because, given its longer half-life, it is often in the background, at least regionally,
generated by nuclear reactors or medical isotope production reactors. Lars-Erik De Geer, “Radioxenon signatures from
underground nuclear explosions,” poster for the International Scientific Studies Project, Vienna, Austria, June 10-12,
2009, http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/RN-22D%20%28Sweden%29%20-%20Lars-
Erik_DeGeer.pdf.
The “m” in xenon-131m and xenon-133m refers to a metastable isomer, which has the same nucleus as xenon-131 or
xenon-133, respectively, but in a higher-energy state. In contrast, xenon-133 refers to that isotope in its lower-energy,
or ground, state.
54 For further discussion of gamma-ray spectra, see CRS Report R40154, Detection of Nuclear Weapons and Materials:
Science, Technologies, Observations, by Jonathan Medalia, Chapter 1 and Appendix.
Congressional Research Service
11
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
rays detected at each energy from a sample of a specified mass in a specified time. For example,
the spectrum for xenon-133m in Figure 1 has a peak at 233 keV. Another technique, known as
beta-gamma coincidence counting, relies on the simultaneous emission of a beta particle (an
electron or positron) and an 81-keV gamma ray when an atom of xenon-133 decays. This
signature is unique to xenon-133 and is insensitive to background radiation, so it can be detected
even in minute concentrations of xenon-133.55 Improved sensitivity of radioxenon detection
instruments enhances such techniques.
Once radioxenons are detected and identified, the data can be used for long-range detection of
nuclear tests in at least three ways. One is to use atmospheric transport modeling (ATM). ATM
can help determine the region (as opposed to a precise location) where a nuclear test was
conducted by calculating the path of air masses that may be carrying radioactive materials. It uses
a computer model to assemble millions of pieces of data collected in near real time by weather
satellites, World Meteorological Organization stations,56 and the IMS network. The model then
generates wind speed and direction at many points across a wide area. It can then be “put in
motion.” Like a movie, the model can be run forward to show the movement of air masses in
order to predict the future path of radionuclides released at a precise location. Alternatively, it can
be run in reverse to backtrack radionuclides from a specific point (e.g., an IMS radionuclide
station) to a region where they may have originated, a process called source region attribution.
Source region data may be fused with other data, e.g., IMS seismic observations, using a software
graphics tool to narrow the location of a suspected nuclear explosion.
In 2006, IMS noble gas equipment at Yellowknife, Canada, detected xenon-133 some two weeks
after the North Korean nuclear test. The measurement could not be traced back to known releases
at nuclear facilities (e.g., a nuclear reactor at Chalk River, Canada, used in part to produce
radiopharmaceuticals57). Instead, ATM showed that the detection was consistent with a
hypothesized release of radionuclides taking place in North Korea at the place and time of the
event.58
Following the 2009 DPRK event, studies undertaken using ATM tracked the path of the air
masses that for several days passed over the site of the event. Figure 2 shows that if the event had
been accompanied by a radioactive release, a plume of radioactive noble gases would have
subsequently passed over three IMS noble gas stations in the region of the event (in China, Japan,
and Russia) that had become operational since 2006. These three stations were operating
continuously and at their full capacity in May 2009, but reported no detections of radioxenons.
55 The same approach is used to detect xenon-135, which emits a beta particle and a 250-keV gamma ray
simultaneously. Information provided by Joseph Sanders, Sandia National Laboratories, personal communications,
September 16 and October 20, 2009.
56 Comprehensive Nuclear-Test-Ban Treaty Preparatory Commission, “Major Step Forward in Detecting Nuclear
Explosions: CTBTO-WMO [World Meteorological Organization] Cooperation Enhances Nuclear Test-Ban
Verification,” press release, September 1, 2008, http://www.ctbto.org/press-centre/press-releases/2008/major-step-
forward-in-detectingnuclear-explosions/.
57 Atomic Energy of Canada Limited, “Chalk River Laboratories,” http://www.aecl.ca/Science/CRL.htm.
58 P.R.J. Saey et al., “A long distance measurement of radioxenon in Yellowknife, Canada, in late October 2006,”
Geophysical Research Letters, vol. 34 (October 2007), http://www.agu.org/pubs/crossref/2007/2007GL030611.shtml.
Congressional Research Service
12
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Figure 1. Gamma-ray Signatures of Four Radioactive Isotopes or Isomers of Xenon
Source: Provided by Scott Garner, Technical Staff Member, Los Alamos National Laboratory, October 15,
2009.
Notes: Radioactive isotopes of xenon emit gamma rays when they decay. Each of these isotopes emits gamma
rays in a particular pattern, or spectrum, with a peak at a particular energy. The gamma rays can be counted, and
the number of counts at each energy plotted on a graph. This graph and its peak can be used to differentiate one
xenon isotope from another.
Figure 1 shows the lower-energy part of a gamma-ray spectrum taken with a “Detective,” a small, commercially-
available high-purity germanium detector. The source is a small sample (less than a billionth of a gram each) of
xenon-131m, -133, -133m, and -135. According to Garner, “The relative quantities I have displayed here do not
even remotely represent the relative quantities seen in actual samples, but were chosen to make it obvious that
the different isotopes are easy to tell apart from each other when a strong enough signal is present.”
Technical details: The sample is composed of 2.45E-10 grams (7.66E+05 Becquerels) of Xe-131m, 1.50E-11
grams (1.0E+05 Becquerels) of Xe-133, 6.22E-12 grams (1.0E+05 Becquerels) of Xe-133m, and 1.07E-12 grams
(1.0E+05 Becquerels) of Xe-135. (“E” is an abbreviation for exponent; one Becquerel is one decay per second.)
The source is located 25 centimeters in front of the detector and the data were taken for 300 seconds.
Combining this fact with the level of sensitivity of the detectors,59 IDC staff concluded that “the
containment of any generated xenon (under the hypothesis that this was a nuclear test) was above
99.9 percent” for the 2009 event.60
59 The detectors can detect one radioactive disintegration per second in 5,000 cubic meters of air. Robert Pearce et al.,
“The Announced Nuclear Test in the DPRK on 25 May 2009,” CTBTO Spectrum, September 2009, p. 27,
http://www.ctbto.org/fileadmin/user_upload/pdf/Spectrum/09_2009/Spectrum13_dprk2_p26-29.pdf.
Congressional Research Service
13
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Second, “most of the environmentally occurring radio-xenon has been produced with nuclear
reactors producing radiopharmaceutical materials.”61 Monitoring releases of such reactors can
rule out particular reactors as the source of radioxenons at a specific time, as was the case in
analyzing the Yellowknife data following the 2006 test. The ratio of certain radioactive xenon
isotopes may also prove whether the xenon came from a nuclear test or a nuclear reactor.62
Third, if samples are collected and analyzed within hours of a nuclear test, the ratio of xenon-135
to xenon-133 may indicate whether the nuclear explosive was fueled by uranium or plutonium.63
This information is of interest for analyzing characteristics of the nuclear device, but is not
needed to determine whether the test would violate the CTBT if that treaty were to enter into
force, which requires only knowing if the test was nuclear. Don Barr, a retired Los Alamos
radiochemist with over 50 years of nuclear testing and related experience, calculates that the
window for such determination is only an hour or two.64 Another ratio, xenon-133m/xenon-131m,
may enable differentiation between these fuels for a longer time.65 The IMS could not perform
ratio analysis for the 2006 North Korean test because the station at Yellowknife detected only
xenon-133 two weeks after the event; xenon-135 was presumably not detected because of its
shorter half-life. Jungmin Kang, Frank von Hippel, and Hui Zhang explain a further reason why
ratio analysis is of value for only a short time:
Most of the xenon isotopes released into the atmosphere during the first few hours after a test
would have been produced directly from the nuclear fission. In this period, therefore, the
ratios of different xenon isotopes could be used to discriminate between plutonium and HEU
explosives. Within two days after an explosion, however, most of the xenon isotopes would
come indirectly from the decay of radioactive iodines that are produced in almost the same
ratio from plutonium-239 and uranium-235 fission.66
(...continued)
60 Ibid., p. 28. For further information on ATM, see Comprehensive Nuclear-Test-Ban Treaty Organization Preparatory
Commission, “Atmospheric Transport Modelling and Data Fusion,” http://www.ctbto.org/verification-regime/the-
international-data-centre/atmospheric-transport-modellingand-data-fusion/page-1-atmospheric-transportmodelling-and-
data-fusion/; Comprehensive Nuclear-Test-Ban Treaty Organization Preparatory Commission, “Major Step Forward in
Detecting Nuclear Explosions,” press release, 01 September 2008, http://www.ctbto.org/press-centre/press-releases/
2008/major-step-forward-in-detectingnuclear-explosions/; and Tibor Toth, “Building Up the Regime for Verifying the
CTBT,” Arms Control Today, September 2009, p. 10. For several scientific posters on atmospheric transport modeling,
see “Atmospheric Transport Modeling/Posters,” in the International Scientific Studies website, http://www.ctbto.org/
specials/the-international-scientific-studies-project-iss/scientific-contribtutions/atmospheric-transport-modelingposters.
61 Personal communication, Comprehensive Nuclear-Test-Ban Treaty Organization Preparatory Commission,
December 8, 2009.
62 See, for example, De Geer, “Radioxenon signatures from underground nuclear explosions,” and Kalinowski et al.,
“The Complexity of CTBT Verification. Taking Noble Gas Monitoring as an Example,” p. 94.
63 Ibid., and Hui Zhang, “Off-Site Air Sampling Analysis and North Korean Nuclear Test,” (deals with the 2006 test),
2007, p. 7, paper presented at the 2007 meeting of the Institute of Nuclear Materials Management,
http://belfercenter.ksg.harvard.edu/files/NKSampling_INMM07_Hui.pdf.
64 Personal communications, October 20 and 28, 2009.
65 De Geer, “Radioxenon signatures from underground nuclear explosions.”
66 Jungmin Kang, Frank von Hippel, and Hui Zhang, “The North Korean Test and the Limits of Nuclear Forensics,”
letter to the editor, Arms Control Today, January/February 2007, p. 42.
Congressional Research Service
14
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Figure 2. Atmospheric Transport Modeling
Basis for Concluding 2009 North Korean Test Was >99.9% Contained
Source: Robert Pearce et al., “The Announced Nuclear Test in the DPRK on 25 May 2009,” CTBTO Spectrum, September 2009, p. 28.
Notes: (modified from text of source report) This figure shows the distribution of a hypothetical radioactive xenon plume at the time of its highest concentration at the
IMS radioxenon station indicated in each image. These stations were operational at the time of the 2009 North Korean test. Only those parts of the plume above the
minimum detectable concentration are shown. The plume was calculated assuming (1) immediate venting at the time and place of the 2009 North Korean test and (2) zero
containment corresponds to the ful release of the xenon-133 generated by a four-kiloton nuclear explosion. The key on the left shows the degree of containment of the
test. (The online version of this report shows the graphic in color.) For 90% containment, the detectable plume would cover the areas in green, yellow, and orange. For
99.9% containment, the detectable plume would cover only the areas in orange. The fact that these stations did not record xenon-133 signals at the time each would have
experienced the maximum concentration of that isotope is the basis on which the authors, who are past or current employees of the International Data Center,
“concluded that the containment of any generated xenon (under the hypothesis that this was a nuclear test) was above 99.9 percent.” The yellow circle at the upper left of
each image is a radioxenon station in Mongolia, the only such station operating in this region at the time of the 2006 North Korean nuclear test.
CRS-15
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Thus “while it is true that there is some possibility of determining fuel type from xenon ratios, it
would likely be a slim chance.”67
Nuclear tests may also release particulates, which may contain fission products from the
weapon,68 unfissioned atoms of uranium and plutonium, and melted bits of soil or rock. They
range in size from a centimeter in diameter or larger to 0.1 mm or smaller,69 often less than 0.001
mm. The smaller they are, the farther they can travel before “falling out” to the ground. For
decades, particulates have provided not only evidence of a test but other details as well. For
example, in 1949, U.S. Air Force “sniffer” aircraft flying over the Pacific Ocean collected
particulate samples on filters. A commercial laboratory “dissolved [the filters], chemically
separated a selection of fission products such as radioactive isotopes of barium, cerium,
molybdenum, zirconium and lead, carefully measured the rates of radioactive decay of the
isotopes and counted back to establish when each isotope had been created—its radioactive
birthday. Only if all the birthdays were identical could the isotopes have been created in an
atomic bomb.”70 Analysis of this sort enabled the United States to conclude that the Soviet Union
had conducted its first atomic bomb test on August 29, 1949. According to one report, “During
the first 50 years of the nuclear weapons era, radiochemistry techniques were developed and used
to determine the characteristics (such as yield, materials used, and design details) of nuclear
explosions carried out by the United States and by other countries.”71 IMS radionuclide stations
collect particles on a filter, analyze the gamma-ray spectra on location, and transmit the results to
the International Data Center. If a station’s filter collects two or more types of particulates that are
relevant to CTBT verification, the sample would be shipped to an IMS laboratory to confirm
detection and to conduct more detailed investigation. Figure 3 shows the IMS radionuclide
stations closest to North Korea. As it shows, some are not yet in service, some are planned to
have radionuclide monitoring stations but not noble gas monitoring equipment prior to entry into
force of the CTBT, and there are gaps of many hundreds of kilometers or more between stations.
Background levels of radionuclides in the atmosphere can vary at radionuclide stations due to
patterns of weather, season, or climate, and to sources of radioactive material other than current
nuclear tests, such as iodine-131 and technetium-99m from hospitals and cesium-137 from past
atmospheric nuclear tests. Because of this background variation from place to place and time to
time, a single measurement may not mean anything unless placed in context with this variation.
Accordingly, the Provisional Technical Secretariat, in collaboration with other organizations,
measures and characterizes background levels at each radionuclide station to help determine
whether an elevated level of radioxenons may have come from a nuclear test.
67 “Responses to Jonathan Medalia (Congressional Research Service), Questions related to the Comprehensive Nuclear
Test Ban Treaty,” information provided by DOE and NNSA laboratories, August 2009, p. 2.
68 Fission products are atoms, usually radioactive, of elements lighter than uranium or plutonium that are produced
when uranium or plutonium atoms fission, or split.
69 U.S. Department of Defense and Department of Energy. The Effects of Nuclear Weapons, third edition, compiled and
edited by Samuel Glasstone and Philip Dolan, Washington, U.S. Govt. Print. Off., 1977, p. 37.
70 Richard Rhodes, Dark Sun: The Making of the Hydrogen Bomb (New York, Simon & Schuster, 1995), p. 371. See
also Charles Ziegler, “Waiting for Joe-1: Decisions Leading to the Detection of Russia’s First Atomic Bomb Test,”
Social Studies of Science, May 1988, pp. 197-229. Radioactive isotopes of lead would be created by neutron
bombardment (“activation”) of materials in the ground, not by fission of uranium or plutonium in a weapon.
71 Joint Working Group of the American Physical Society and the American Association for the Advancement of
Science, Nuclear Forensics: Role, State of the Art, and Program Needs, 2008, p. 3, http://iis-db.stanford.edu/pubs/
22126/APS_AAAS_2008.pdf.
Congressional Research Service
16
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Releases of radionuclides from an underground nuclear test can occur in any of three ways. One
is referred to as “vents.” They generally occur promptly, and vary in size from very small (and
undetectable at long distances) to large (and easily detectable at long distances). Figure 4 depicts
large vents from two U.S. tests. Vents occur when high-pressure gas generated by an explosion
finds a path to the surface, often by a leak in the sealing, or “stemming,” of the excavated hole or
tunnel, or by a natural fracture in the rock that reaches the ground surface. Unfavorable geology
of the test site can contribute to venting in other ways. Certain geologic formations tend to have
more preexisting fractures, raising the probability of vent paths. Carbonates in rock produce
carbon dioxide when heated by an explosion, raising the pressure of gas in the cavity left by the
explosion and thus the probability of venting.
Figure 3. Radionuclide Monitoring Stations of the International Monitoring System
As of February 2010
Source: Map with station locations from Comprehensive Nuclear-Test-Ban Treaty Organization Preparatory
Commission, station names and symbols by CRS.
Notes: Numbers are those assigned to the stations by Annex 1 to the Protocol of the Comprehensive Nuclear-
Test-Ban Treaty. “Certified” stations have been certified by the Provisional Technical Secretariat as meeting its
technical standards and are fully operational.
Congressional Research Service
17
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
A nuclear explosion turns water in rocks into steam, adding pressure in the explosion cavity that
can lead to venting. The Soviet Union found that, at least in some instances, a significant amount
of water and carbonate rock led to seeps (discussed next) and venting at its Novaya Zemlya test
site, while the opposite conditions at its Semipalantinsk (Kazakhstan) test site did not.72 Noble
gases, being less chemically reactive than other gases and particulates, will always be released by
a large vent; seeps and smaller vents release predominantly noble gases, with very little or no
particulate matter.73 It is likely that the first North Korean test vented, since it apparently released
enough radioxenon to be detected two weeks later at Yellowknife.
A second class of release from an underground nuclear test is known as a “seep.” Seeps tend to
release much smaller amounts of radioactive material, and the released material is generally
limited to noble gases and possibly volatile elements, notably radioactive iodine. Seeps do not
occur promptly, but instead release at much lower rates, potentially over periods of weeks to
months. Seepage can occur through porous rock or small fractures in the rock. Seeps are
potentially detectable at the test site by an OSI, offering confirmation of a test. However, seeps
occurring more than a few weeks after the detonation are unlikely to release amounts in quantities
that could be detected hundreds of miles away by IMS or other monitoring systems.
Seeps and vents both result when high-pressure gases generated by a nuclear explosion escape
from the explosion-generated cavity by way of pathways in the surrounding geology or stemming
material. When pressure drops sufficiently, seeps and vents cease.
A third class of release, “barometric pumping,” relies on a different mechanism that enables gases
to reach the surface even after gas pressure in the cavity has dropped to a level of equilibrium
with its surroundings. A decrease in atmospheric pressure, such as occurs during a storm, lowers
the pressure in fractures terminating near the surface so that the relative pressure in the cavity end
of a fracture becomes greater. This pressure differential draws noble gases upward toward the
Earth’s surface. As gases flow upward in a fracture, they also diffuse into the porous walls of the
fracture and are temporarily stored there even when increased atmospheric pressure causes gases
in a fracture to flow downward. The stored gases are available to diffuse from the porous walls
into the next upward flow of gases in the fracture. Thus barometric pumping creates a ratcheting
effect that eventually can transport noble gases to the surface after seeps and vents induced by
pressure within the cavity have ceased.74 Barometric pumping would probably not aid long-range
detection because quantities of radioactive noble gases released through this mechanism are small
and can take many weeks to reach the surface, during which time most of these gases will have
undergone radioactive decay, but it can produce enough of these gases to be detected by an OSI.
How Can Radioactive Material Be Contained?
Containment is no simple matter. According to a National Academy of Sciences report, “Recent
Russian papers documenting Soviet nuclear testing state that all underground tests at Novaya
72 Vitaly Khalturin et al., “A Review of Nuclear Testing by the Soviet Union at Novaya Zemlya, 1955-1990,” Science
and Global Security, vol. 13 (2005), p. 21.
73 Information provided by Joseph Sanders, Sandia National Laboratories, personal communication, September 11,
2009.
74 See C.R. Carrigan et al., “Trace Gas Emissions on Geological Faults as Indicators of Underground Nuclear Testing,”
Nature, vol. 382, August 8, 1996, pp. 528-531; and Lars-Erik De Geer, “Sniffing out Clandestine Tests,” Nature, vol.
382, August 8, 1996, pp. 491-492.
Congressional Research Service
18
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Zemlya and about half the underground tests at the Semipalatinsk test site in Kazakhstan resulted
in release of radioactivity.”75 While fewer U.S. nuclear tests released radioactivity, containment
failures did occur. For example, Figure 4 shows the “Des Moines” nuclear test. It was conducted
in June 1962 at the Nevada Test Site, and had an explosive yield of 2.9 kilotons, comparable to
the “few kilotons” of yield that the U.S. Office of the Director of National Intelligence assessed
for the North Korean event of May 2009. Figure 4 also shows “Baneberry,” a 10-kiloton test
conducted at the Nevada Test Site in December 1970.
Figure 4. Venting of Nuclear Tests
“Des Moines” 1962 (left), “Baneberry” 1970 (right)
Source: ”Des Moines,” Lawrence Livermore National Laboratory; “Baneberry,” U.S. Department of Energy
The United States went to great lengths to contain nuclear tests. Containment relies on a detailed
understanding of how well the geology around the nuclear device may contain the explosion and
an ability to engineer containment, such as by sealing the test shaft. Barr said, “Deep burial of a
nuclear device, combined with gas blocking techniques, virtually eliminates the seepage of noble
gases to the surface, though some such gases might occasionally be detected, but only at the
surface above the detonation point.”76 On the other hand, knowledge of the geology surrounding a
nuclear explosive is imperfect, so there may be hidden pathways for vents and seeps. As a result,
there is an element of art and chance to containment:
75 John Holdren (chair) et al., Technical Issues Related to Ratification of the Comprehensive Test Ban Treaty,
Committee on Technical Issues Related to Ratification of the Comprehensive Test Ban Treaty, National Academy of
Sciences, National Academy Press, Washington, 2002, p. 45. The source cited is V.N. Mikhailov et al., Northern Test
Site: Chronology and Phenomenology of Nuclear Tests at the Novaya Zemlya Test Site, July 1992.
76 Personal communication, November 21, 2007.
Congressional Research Service
19
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
The earth, from the surface to the mile or so in depth that has been used in underground
nuclear testing is an inhomogeneous body of materials ... it is not possible to know all, or
even most, of the details of the medium where the detonation takes place.
So, empirical rules are developed, approximations are made and are used in computer codes
to model the behavior of the earth materials following a detonation, but there is a further
complication. Important processes occur during a time span that ranges from fractions of a
microsecond to hours....
In such a situation experience and empirical evidence from previous detonations assumes a
considerable importance when trying to judge what will happen when a particular detonation
takes place in some specific location. The experience and evidence that there is has been
gathered over the years, sometimes in a costly fashion.77
The unclassified U.S. literature contains information on containing a test. A 1977 publication by
the Departments of Defense and Energy provides data on how deeply to bury a nuclear explosive
device to contain radioactive gases.78 A 1989 Office of Technology Assessment (OTA) report
provides technical details. It notes that containment properties of rock depend on its type,
structure, and water content; lists some U.S. procedures used to evaluate containment; and
provides diagrams of mechanisms used to contain various types of nuclear tests.79 A 1995 report
sponsored by Lawrence Livermore National Laboratory and the Defense Nuclear Agency
provides further details.80
Better techniques can greatly improve containment. For the United States, improvement occurred
in two major steps. The Limited Test Ban Treaty, which bans nuclear tests in the atmosphere, in
space, and under water, was signed on August 5, 1963. Before that date, “no specific test
containment design criteria existed. Therefore, while radioactive effluents released from
underground tests conducted during this period [September 15, 1961, to August 5, 1963] were not
always expected, any effluent releases that did occur were not considered accidental, or even
unexpected.” 81 After August 5, 1963, “all tests (except four Plowshare cratering tests) were
designed to be completely contained underground.”82 Following Baneberry, the Atomic Energy
Commission instituted new containment practices.83 In consequence, while the Department of
Energy reported that 101 of 335 U.S. nuclear tests conducted from August 5, 1963, through 1970
accidentally released radiation, it reported that 6 of 388 tests conducted from 1971 through the
77 Lawrence Livermore National Laboratory and U.S. Department of Defense. Defense Nuclear Agency, Caging the
Dragon: The Containment of Underground Nuclear Explosions, DOE/NV-388 and DNA TR 95-74, 1995, by James
Carrothers, p. 1, http://www.scribd.com/doc/6602337/Caging-the-Dragon-The-Containment-of-Underground-Nuclear-
Explosions. (On the cover: “Distribution of this document is unlimited.”)
78 U.S. Department of Defense and Department of Energy. The Effects of Nuclear Weapons, third edition, p. 261.
79 U.S. Congress. Office of Technology Assessment. The Containment of Underground Nuclear Explosions. OTA-ISC-
414, October 1989, pp. 31-55, available at http://www.nv.doe.gov/library/publications/historical/OTA-ISC-414.pdf.
80 Carrothers, Caging the Dragon, 726 p.
81 U.S. Department of Energy. Nevada Operations Office. Radiological Effluents Released from U.S. Continental Tests,
1961 through 1992, DOE/NV-317 Rev. 1, UC-702, August 1996, p. 2, http://www.nv.doe.gov/library/publications/
historical/DOENV_317.pdf.
82 Ibid. “Plowshare” tests explored peaceful uses of nuclear explosions, such as digging canals or harbors, in the 1960s
and 1970s. Tests intended to create large craters for such purposes of course could not be contained.
83 Office of Technology Assessment. The Containment of Underground Nuclear Explosions, p. 32. The Atomic Energy
Commission was a predecessor agency of the Department of Energy.
Congressional Research Service
20
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
most recent U.S. nuclear test, in 1992, did so.84 In addition, small amounts of noble gases seeped
into the atmosphere and were detected, onsite only, days to years after five tests from 1984 to
1989,85 and radioactive material was released intentionally after quite a few post-1970 tests, such
as by drilling back into the cavity to collect samples for analysis.86 The 1989 OTA report provided
another view of the effectiveness of post-Baneberry containment: “If the same person had been
standing at the boundary of the Nevada Test Site in the area of maximum concentration of
radioactivity for every test since Baneberry (1970), the person’s total exposure would be
equivalent to 32 extra minutes of normal background exposure (or the equivalent of 1/1000 of a
single chest x-ray).”87 Some U.S. nuclear tests that were not reported as releasing radioactive
material might have done so, but the amount released may have been below the detection
threshold for instruments available at the time.
While there is no publicly-available information on whether North Korea attempted to contain its
second test, and if so what methods it used, containment could have resulted from one of the
following factors, or a combination of several:
Lessons learned from the first test: As noted, a nuclear test provides data for many disciplines
involved with the test. North Koreans involved in containment would have learned lessons from
the first test applicable to the second test, including how such factors as stemming methods, depth
of burial, and type of rock affect containment.
Lessons learned from the experience of other nations: These lessons deal with such factors as
depth of burial, type of stemming, and geologic considerations, as discussed previously.
Use of a higher-yield nuclear device: It can be harder to contain lower-yield underground nuclear
explosions than higher-yield ones. The latter produce more energy, which pushes outward against
the surrounding rock, which then rebounds toward its original position. OTA states, “the
rebounded rock locks around the cavity forming a stress field that is stronger than the pressure
inside the cavity. The stress ‘containment cage’ closes any fractures that may have begun and
prevents new fractures from forming.”88 A nuclear explosion melts rock, forming a glass-like
substance. These effects can seal leak paths, especially fractures in the rock through which noble
gases or particulates might escape. Sealing is more likely for a higher-yield test. While both North
Korean tests were of low yield, the second reportedly had several times the yield of the first.
Good luck: North Korea may have, by chance, selected a test site with solid rock having no
fissures, or with another geology favorable to containment, and may have used enough material to
seal the test shaft or tunnel solidly enough to contain radioactive material. As noted, even in the
period before 1963, when there was no particular effort made to contain U.S. underground
nuclear tests, quite a few of them did not have measured releases of radioactivity.
84 U.S. Department of Energy. Nevada Operations Office, United States Nuclear Tests, July 1945 through September
1992, DOE/NV—209-REV 15, December 2000, pp. 30-89, http://www.nv.doe.gov/library/publications/historical/
DOENV_209_REV15.pdf.
85 U.S. Department of Energy, Radiological Effluents Released from U.S. Continental Tests, 1961 through 1992, pp.
210, 211, 222, 223, 231.
86 U.S. Department of Energy, Radiological Effluents Released from U.S. Continental Tests, 1961 Through 1992.
87 Office of Technology Assessment, The Containment of Underground Nuclear Explosions, pp. 4-5; original text bold.
88 Ibid., p. 34.
Congressional Research Service
21
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Venting below the detection threshold: The standard for IMS radioxenon equipment is the ability
to detect one atom of xenon-133 decaying per second per thousand cubic meters of air (the
minimum detectable concentration).89 In practice, the equipment is more sensitive than
that.90 While this threshold has become lower over the years, it is greater than zero. It is thus
possible that the test vented, but that the quantity of material released was below the detection
threshold.
Nonnuclear explosion: The May 25 event may not have been a nuclear test, which would explain
the lack of radioactive effluents.
While evidently not the case for the 2009 test, as Figure 2 implies, attention to atmospheric
conditions could impede detection of radionuclides and thus contribute to the appearance of
containment. Waiting to conduct a test until wind currents were blowing away from IMS or
national radionuclide stations could prevent these stations from collecting radionuclides.
Given the learning curve, potential failure modes of containment, and the sensitivity of detection
equipment, it would be a significant achievement if North Korea had, by design, been able to hold
venting of its second test to below the current detection threshold. At the same time, one test that
apparently did not release radioactive effluents is too small a sample size from which North
Korea, the United States, or other nations could draw firm conclusions as to North Korea’s
containment capability. That nation’s ability to contain any future tests thus bears close watching.
Potential Value of Containment for North Korea
The ability to contain radioactive material from the 2009 test offers several potential benefits for
North Korea. First, careful attention to containment should reduce the likelihood of a major
venting of fallout similar to Baneberry. Venting would arguably not be in North Korea’s interests.
Fallout reaching China could harm North Korea’s relationship with its major ally, perhaps leading
China to increase pressure on North Korea to halt nuclear testing or even its nuclear weapons
program. Fallout reaching Russia could have a similar effect. Fallout on Japan or South Korea
would likely antagonize them. Fallout on North Korea could contaminate land. Avoiding fallout is
reason enough for North Korea to try to improve its containment capabilities.
Second, if particulates containing uranium or plutonium vented and could be collected at a
distance, other nations could analyze them in an attempt to gain data on weapon characteristics,
helping to track problems and progress of North Korea’s nuclear weapons program. This is
another reason for North Korea to focus on containment of its underground explosions.
Third, absence of radionuclides from a nuclear test, as a result of containment, could make it
harder to muster the 30 votes in the 51-member CTBTO Executive Council needed to authorize
an OSI by providing scientific cover to nations that wanted to deny a request for an OSI on
political grounds. This approach could be more significant for a nation with more allies than
North Korea has. On the other hand, a lack of radioactive noble gases combined with a nuclear
89 Mika Nikkinen, Matthias Zahringer, and Robert Werzi, “The Radionuclide Processing System of the CTBTO,”
poster presented at International Scientific Studies 2009, http://www.ctbto.org/fileadmin/user_upload/ISS_2009/Poster/
RN-30%20%28PTS%29%20-%20Mika_Nikkinen%20etal.pdf.
90 “Responses to Jonathan Medalia,” August 2009, p. 1.
Congressional Research Service
22
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
explosion-like seismic signal and other technical evidence would provide a compelling technical
case for requesting an OSI. Of course, the surer way for North Korea to avert OSIs would be for
that nation not to ratify the CTBT, keeping it from entering into force.
Fourth, and more speculatively, successful containment could enable other nations to conduct
nuclear tests in North Korea. This does not appear to have happened, but Iran is a possible
candidate. The two have a record of conventional arms trade and missile cooperation.91 Events in
2009, such as the discovery of a covert facility for uranium enrichment, increased suspicions that
Iran is pursuing nuclear weapons. It is not unprecedented for one nation to “host” another’s
nuclear tests: the United Kingdom conducted 24 tests jointly with the United States at the Nevada
Test Site between 1962 and 1991.92
Iranian testing in North Korea would aid the latter by providing data for weapons development
and giving the impression that its nuclear weapons program was proceeding rapidly. Such testing
would aid Iran by helping it develop nuclear weapons while potentially avoiding consequences of
a test in Iran, such as an attack on its nuclear facilities. In particular, an extremely low yield test
(e.g., 0.5 kilotons) conducted in Iran might be interpreted as a failure, inviting attack before the
weapons program developed further, while a larger test (e.g., 20 kilotons) in Iran might deter
attack. Conducting one or two tests in North Korea might avert the former contingency. This
arrangement would demand high confidence in North Korea’s containment ability so as to deny
radioactive samples by which the test could be attributed to its partner. Analysis of these samples
might reveal if the bomb fuel was uranium or plutonium, details of the bomb design, and perhaps
which reactor produced the fuel. While particulate samples convey more data than do gases, the
ability to contain gases would imply a strong ability to contain particulates.
Several factors argue against this scenario. Iran might use nuclear tests in Iran to demonstrate its
nuclear capability as a deterrent, to gain leverage in the Middle East, and to show its people that
other nations could not dictate its nuclear policies. Iran might believe it could deter a U.S. strike
on its nuclear facilities by its nuclear threat or the prospect of retaliation against U.S. forces in the
area. Iran might discount the threat of an Israeli strike if it felt that Israel could only inflict a
temporary setback to its nuclear program. North Korea might halt nuclear tests if it thought it
could make major gains in the Six-Party Talks. North Korea and Iran might not have high
confidence in North Korea’s ability to contain nuclear tests. There are questions about the
reliability of media reports on Iranian-North Korean cooperation in the nuclear area.
Issues for Congress
The 2009 North Korean test raises at least two issues for Congress. What does the test imply for
U.S. ability to verify compliance with the CTBT? And what unilateral and multilateral steps
might Congress mandate or encourage to improve monitoring and verification capability?
91 See, for example, Jim Wolf, “North Korea, Iran Joined on Missile Work: U.S. General,” Reuters, June 11, 2009,
http://www.reuters.com/article/newsOne/idUSTRE55A4E720090611. See also CRS Report RL30613, North Korea:
Terrorism List Removal, by Larry A. Niksch.
92 U.S. Department of Energy, United States Nuclear Tests, July 1945 through September 1992, pp. xvi, 18-89.
Congressional Research Service
23
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Implications for the CTBT
Supporters and opponents of the CTBT will likely draw opposing conclusions on what the
absence of detectable radionuclides from the 2009 test indicates. Here are points they might raise
in a future debate.
Opponents might argue that without detection of radionuclides there is no proof that North Korea
conducted a nuclear test. For example, the May 2009 event might have been a large conventional
explosion conducted to inflate the appearance of progress in North Korea’s weapons program.
The treaty’s supporters might respond that the ability of the IMS seismic component to pick out
signals characteristic of an explosion originating in the area of the suspected test from the many
seismic signals occurring each day shows the capability of the IMS. As another example, the IMS
detected the 2009 event seismically, and identified it as an explosion, before North Korea
announced it.93 More generally, the 2009 test shows that an attempt to evade detection would also
have to contain radionuclides and suppress other signatures, a more difficult task than suppressing
only one signature. Suspicious seismic signals and an absence of radionuclides, it is argued,
would surely lead to calls for an OSI.
Opponents might counter that OSIs could not happen unless the treaty entered into force, and that
North Korea is unlikely to ratify the treaty, thereby preventing entry into force, as long as it has
any interest in future nuclear tests. Even if North Korea ratified the treaty, it could bar inspections
of its territory, and if it allowed them, inspectors might not find proof that a test occurred. While
the case could be referred to the United Nations, CTBT opponents would see only a slim
likelihood of that body taking effective action.94
Supporters recognize that OSIs could not be conducted under the treaty without entry into force,
and see that as a benefit of entry into force. They believe that OSIs have a good chance of finding
a “smoking gun,” and that the U.N. would adopt stringent sanctions on North Korea in response
to nuclear tests conducted after the treaty had entered into force. They see a refusal by a state
party to permit inspections as prima facie evidence of a violation.
One generally-accepted means of evading detection of nuclear tests, especially low-yield tests, is
“decoupling,” testing in a large underground cavity to muffle the seismic signal. Opponents could
argue that a decoupled test conducted in a manner that prevented release of radionuclides, such as
deep under a mountain, might go undetected by radionuclide sensors as well as seismographs, and
that the other two IMS technologies, infrasound and hydroacoustic, would not be expected to
detect a test of this sort, so all IMS technologies might be circumvented simultaneously.
Opponents might argue further that the ability of the IMS to detect the 2006 and 2009 tests does
not show that that system can detect clandestine tests because neither test was conducted
evasively. (North Korea announced both of them.) In this view, any cheater would use evasive
methods, so the IMS has merit only insofar as it can detect evasive tests.95
93 Personal communication, Annika Thunborg, Comprehensive Nuclear-Test-Ban Treaty Organization Preparatory
Commission, December 8, 2009.
94 See Senator Jon Kyl, “Why We Need to Test Nuclear Weapons,” Wall Street Journal, October 21, 2009, p. 23.
95 For discussions of decoupling and other evasion scenarios, see David Hafemeister, “CTBT Evasion Scenarios:
Possible or Probable?,” CTBTO Spectrum, issue 13, September 2009, pp. 22-25, http://www.ctbto.org/fileadmin/
user_upload/pdf/Spectrum/09_2009/Spectrum13_hafemeister_p22-25.pdf, and Robert Barker, “CTBT Monitoring
(continued...)
Congressional Research Service
24
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Supporters question the feasibility of decoupling, or otherwise hiding, a test of more than 1 or 2
kilotons. They note that the IMS has detected seismic signals down to a small fraction of a
kiloton; that it is difficult to hide the rock that must be removed to create an underground cavity;
and that, despite precautions, an evader cannot count on near-perfect containment. The merits of
various evasion techniques have been debated for decades.96
Entry into force requires ratification by North Korea, among others, yet that nation’s ratification
may be difficult to obtain. To circumvent the problem, some CTBT supporters have suggested
bringing the treaty into force provisionally.97 This apparently would mean that states that had
ratified the treaty would behave among themselves as if the treaty had entered into force,
permitting OSIs among these states and formal operation of the structures of the CTBTO. But
there are problems with provisional entry into force. First, a state not party to provisional entry
into force might conduct a large conventional explosion designed to simulate a nuclear explosion
so as to give the impression of progress on its nuclear program. Since the IMS is designed to
detect nuclear tests only, it would not detect signatures that would identify a test as nonnuclear.
Second, since a state could conduct a test in a host state that was not party to provisional entry
into force, it would be important to attribute the test to learn if a state party to provisional entry
into force had conducted the test; that would be difficult to do if the test were well contained.
“Regular” entry into force would address both concerns. OSIs could reveal if the test was nuclear
or conventional, and attribution would not matter for the treaty’s verification regime because any
nuclear test would violate the CTBT.
Improving Monitoring and Verification Capability
Key problems for analyses of the 2009 North Korean event were determining whether it was a
nuclear explosion and learning more about it. In examining budgets and programs, Congress may
wish to consider various means of improving U.S. and international ability to monitor nuclear
testing by North Korea and other nations. The preceding sections of this report lead to several
possible means to do this. They fall into several categories: (1) conduct research to better
characterize nuclear explosions and containment, (2) deploy more monitoring equipment, (3)
(...continued)
Limitations & Verification Implications: Cheating Scenarios,” presentation to the National Academies’ CTBT Review
Committee, September 9, 2009. The latter document is available through the committee’s website,
http://www8.nationalacademies.org/cp/projectview.aspx?key=49131; follow the link to the Public Access Records
Office at the bottom of the page, and use that link to file a request.
96 John Holdren (chair) et al., Technical Issues Related to Ratification of the Comprehensive Test Ban Treaty,
Committee on Technical Issues Related to Ratification of the Comprehensive Test Ban Treaty, National Academy of
Sciences, National Academy Press, Washington, 2002, pp. 46-48. See also Harold Karan Jacobson and Eric Stein,
Diplomats, Scientists, and Politicians: The United States and the Nuclear Test Ban Negotiations, University of
Michigan Press, Ann Arbor, 1966, pp. 153-154.
97 United Nations. General Assembly. “Letter dated 29 June 2006 from the Permanent Representative of Sweden to the
United Nations addressed to the Secretary-General,” Annex, “Weapons of Terror: Freeing the World of Nuclear,
Biological, and Chemical Arms,” document A/60/934, July 10, 2006, p. 13, http://www.wmdcommission.org/files/
english.pdf; and Martin Matishak, “Nuclear Test Ban Could Become Reality Without North Korea, Experts Say,”
Global Security Newswire, June 4, 2009, http://www.globalsecuritynewswire.org/gsn/nw_20090602_5876.php. Jessica
Tuchman Mathews said, “If only North Korea and Iran remain [as states that must ratify the CTBT for it to enter into
force], the more than 160 nations that have joined the treaty will not allow them to block it. An amendment will be
drawn that allows provisional entry into force without them.” Jessica Tuchman Mathews, “This Time, Ban the Test,”
International Herald Tribune, October 21, 2009, http://carnegieendowment.org/publications/index.cfm?fa=view&id=
24021.
Congressional Research Service
25
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
improve the performance of monitoring systems, and (4) look for new signatures to help
determine if a test is nuclear. Note that government agencies that conduct programs to improve
monitoring and verification capability develop strategic plans for their work and update them
annually; the options presented here would need to be prioritized against existing programs.98
Conduct Research to Better Characterize Nuclear Explosions and Containment
Conduct Basic Research on Containment
Radioactive gases, and especially radioactive noble gases, are an important sign of a nuclear
explosion. Yet Raymond Jeanloz, Professor of Earth and Planetary Science, University of
California at Berkeley, said: “The science underlying the containment of gases in the Earth’s crust
is poorly understood. The U.S. nuclear test program focused on containment of particulates. The
program did not try to gain a full understanding of what determines how or when gases are
contained, but instead developed practical solutions to containment.”99 A better understanding of
the science of how the Earth contains gases, especially in the case of nuclear tests, should help
evaluate North Korean containment efforts.
Evaluate the Adequacy of Monitoring of North Korean Containment
Given the potential significance of North Korean efforts to contain radioactive material from
nuclear tests, it may be of value to have the Intelligence Community analyze the first two North
Korean nuclear tests to see what containment methods were used, and in what ways, if any, North
Korea modified those methods for its second test. Similarly, it may be of value for that
community to pay particular attention to containment methods in monitoring North Korean
preparations for any future nuclear test. The Intelligence Community could report its findings on
a classified basis to the congressional committees of jurisdiction.
Conduct Research to Improve Atmospheric Transport Modeling
Improving atmospheric transport modeling could improve the accuracy with which it could track
radionuclides back to their location of origin, or predict the path of an air mass carrying
radioactive materials.
98 As an example of a strategic plan, see U.S. Department of Energy. National Nuclear Security Administration.
“Nuclear Explosion Monitoring Research and Engineering Program: Strategic Plan,” https://na22.nnsa.doe.gov/ndd/
strategicplan.
99 Personal communication, June 23, 2009. Jeanloz continues, “Interestingly, the question of containment of gases in
the Earth’s crust is also important for energy and environmental issues such as carbon sequestration. While the time
scale is much longer for sequestration (centuries to millenia) than for nuclear explosion monitoring (hours to days), the
issue of gas containment in the crust is pretty much the same, and study of long-term sequestration would benefit from
a better understanding of short-term containment, such as for CTBT monitoring.” Jonathan Katz, Professor of Physics
at Washington University in St. Louis, states, “the issues of noble gas seepage and carbon dioxide (CO2) sequestration
are not quite the same. Unlike noble gases, CO2 will react with some rocks and ground water, and will liquefy under
pressure at room temperature. Both effects make a difference for diffusion.” Personal communication, August 4, 2009.
Congressional Research Service
26
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Provide Fellowships
Breakthroughs in analysis could enable seismologists to extract more information from seismic
data and lead to improved concepts for future seismic networks.100 However, such advances, as
well as maintaining long-term monitoring capability more generally, will require training of
graduate students in nuclear explosion monitoring disciplines. One expert stated,
The past decade has seen sharp fluctuations in funding of programs in this area by two key
sources, the Air Force Research Laboratory and the National Nuclear Security
Administration, and funding has been far below the level recommended by a 1997 National
Research Council report. Low and erratic funding has disrupted graduate student training. As
a result, it is becoming hard to sustain adequate numbers of experts in nuclear explosion
monitoring, as evidenced by recent difficulties in replacing seismologists who retired.101
To address this issue, the United States could support, at a steady level, fellowships and programs
adequate to produce enough experts in nuclear explosion monitoring to meet national needs.
Similarly, Congress, in P.L. 111-140, Nuclear Forensics and Attribution Act, found, “The number
of radiochemistry programs and radiochemists in United States National Laboratories and
universities has dramatically declined over the past several decades.”102 In response, this act
would establish a National Nuclear Forensics Expertise Development Program.
Deploy More Monitoring Equipment
Add Radionuclide Stations and Radioxenon Equipment
Since venting or seepage of radioxenons is more likely than venting of particulates, many agree
that it would be desirable for the IMS to have radioxenon equipment at all 80 radionuclide
stations instead of the 40 currently planned. The treaty (Protocol, paragraph 10) provides for this
expansion once it enters into force: “At its first regular meeting, the Conference [of the States
Parties] shall consider and decide on a plan for implementing noble gas monitoring capability
throughout the network.” The CTBTO PrepCom indicates that adding radioxenon equipment to
the remaining 40 particulate stations would not be technically difficult, but is more a matter of
political will and financial resources.103 The PrepCom also states, “there has been a strong interest
in building up and strengthening the noble gas capability since the 2006 DPRK declared test
within the CTBTO PrepCom.”104 The European Union has made voluntary contributions for this
purpose,105 and the United States has made technical contributions to this effort.106 It may also be
100 Email from Raymond Jeanloz, Professor of Geophysics, University of California, Berkeley, April 20, 2009.
101 Information provided by Thorne Lay, Professor of Earth and Planetary Sciences, University of California, Santa
Cruz, personal correspondence, April 20, 2009. The 1997 report is Thorne Lay et al., Research Required to Support
Comprehensive Nuclear Test Ban Treaty Monitoring, National Research Council, National Academies Press, 1997.
102 H.R. 730 passed the Senate with an amendment by unanimous consent on December 23, 2009. On January 21, 2010,
the House agreed to the Senate amendment, 397-10. The President signed the bill into law February 16. 2010.
103 Personal communications, July 30, 2009, and December 8, 2009.
104 Personal communication, September 15, 2009.
105 In 2007 and 2008, the European Union provided a total of €3.986 million, part of which was to be used for noble gas
monitoring. See “Council Joint Action 2007/468/CFSP of 28 June 2007,” Official Journal of the European Union, July
6, 2007, pp. L 176/31-L 176/38; and “Council Joint Action 2008/588/CFSP of 15 July 2008,” Official Journal of the
European Union, July 17, 2008, pp. L 189/28-L 189/35.
106 Personal communication, NNSA staff, November 20, 2009.
Congressional Research Service
27
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
desirable to equip all 80 stations with radioxenon equipment, or to increase the number of
radionuclide stations with radioxenon equipment beyond 80, before the treaty enters into force.
Additional stations in Japan and South Korea, and in areas of China and Russia close to North
Korea, would be of particular value for monitoring any testing by that nation.
Procure Mobile Radionuclide Collection Equipment for Rapid Deployment
Even completing the IMS and having radioxenon equipment at all 80 radionuclide stations would
leave gaps in coverage, as Figure 2 shows for North Korea. These gaps pose a problem for
monitoring and verification. For example, a test might release radioactive material that wind
currents blow away from IMS stations, or the wind might loft such material high above these
stations, which are at ground level. Mobile detection systems ready to deploy immediately after
detecting a nuclear test would help address this problem. These systems could include ships and
radionuclide collection aircraft to deploy on or over international waters, and land-mobile
systems to deploy in nations near the suspected test. These systems might or might not be part of
the IMS depending on how they were handled pursuant to Article IV (verification), paragraphs
23-25 of the treaty, “Changes to the International Monitoring System.” Even if these systems
were not included in the IMS, states could still share the resulting data with the International Data
Center pursuant to Article IV, paragraphs 27 and 28, “Cooperating National Facilities.”
Mobile systems offer many advantages. The ability to collect over broad ocean areas would close
some gaps in the IMS. Since mobile systems could collect data close to and soon after a suspected
detonation, they might be able to collect particulates before they dropped back to Earth.
Particulates can provide high confidence that the material originated from a nuclear test; they can
also provide data on certain weapon characteristics. Gathering radioxenons quickly is of
particular value for analyzing the ratio of xenon-135 to xenon-133, which can also provide high
confidence that a test was nuclear. The rapid decay of xenon-135 (half-life, 9.14 hours), plus the
decay of iodine-133 and -135 into xenon-133 and -135, respectively, precludes such analysis after
a short time. Close-in, rapid collection should result in higher concentration of radionuclides,
facilitating analysis, because they would have less time to dilute in the atmosphere. Hafemeister
states that airborne sensors or ground sensors closer to a test can enhance the concentration of
radionuclides by a factor of more than a million.107 Close-in collection should also result in more
confident determination of which nation conducted a test by greatly reducing the number of
countries from which the radionuclides could have originated.
There could be obstacles to airborne or seaborne collection systems operating on or over
international waters. For example, according to press reports, North Korea fired several surface-
to-air missiles around the time of the 2009 nuclear test that “appeared to be aimed at keeping U.S.
and Japanese surveillance planes away from the nuclear test site.”108
107 David Hafemeister, “Input to the NAS CTBT Study,” presentation to the CTBT Review Committee of the National
Academy of Sciences, September 9, 2009, p. 24.
108 Jean Lee, “Defying World Powers, N. Korea Conducts N Test,” Associated Press, May 25, 2009,
http://www.breitbart.com/article.php?id=D98DCSF00&show_article=1.
Congressional Research Service
28
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Improve the Capability of Monitoring Systems
Increase the Sensitivity of Detection Equipment
While the capability to detect radioactive noble gases is very good, it could be improved. So
doing would increase the probability of detection, both remote and onsite, by enabling a detector
to pick up a signal from a radioisotope at a lower concentration or for a longer time. Figure 5
illustrates the point for onsite detection. It shows a detection limit or threshold for argon-37 and
xenon-133 (horizontal lines). It shows the signal from these two isotopes diminishing over time
(diagonal lines), with the xenon signal diminishing faster than the argon signal because the
former has a shorter half-life, 5.3 days vs. 34.8 days. The “window,” or the period in which an
isotope can be detected (vertical lines), opens when the gas reaches the surface, with xenon and
argon reaching the surface about 50 and 80 days, respectively, after a detonation as a result of
barometric pumping, and closes when the amount of either gas falls below the detection
threshold.109 Thus if the detection threshold can be lowered, the window closes later. While the
graph shows the signal as starting out at its peak, in practice the signal would begin at zero and at
some point would rise rapidly. As a result, a more sensitive detector might also “open” the
window slightly earlier.
109 Charles Carrigan, “Using OSI Field Studies and Tests to Define Noble Gas Sampling and Analysis Requirements,”
presentation at INGE [International Noble Gas Experiment] 2009, Daejeon, Korea, November 9-14, 2009, Lawrence
Livermore National Laboratory document LLNL-PRES-41961, p. 11. This prediction assumes good test containment
and the barometric and geologic conditions present for a specific (nonnuclear) test conducted at the Nevada Test Site.
This test, the Non-Proliferation Experiment (NPE), was conducted on September 22, 1993, and used some 1,400 tons
of chemical explosive, along with small quantities of gases intended to simulate certain radioactive gases. Other
conditions could produce different results. For further information on NPE, see U.S. Department of Energy. Symposium
on the Non-Proliferation Experiment: Results and Implications for Test Ban Treaties, CONF-9404100, April 19-21,
1994, https://na22.nnsa.doe.gov/cgi-bin/prod/nemre/index.cgi?Page=Symposium+1994.
Congressional Research Service
29
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Figure 5. Detection “Window” for Argon-37 and Xenon-133
Lowering the Detection Limit Keeps “Window” Open Longer
100
Ar
A -37 Detec
e
t
tec i
t on
Li
on m
Li it
1
Ar
Ar
A -37 Signal
i
"W
"
indow"
ndow
. m
u
/c
0.01
0.
ls
re
Xe
X -
e 133
Detec
e
t
tec i
t o
i n Lim
n Li it
que
e
Xe
e
B
0.0001
"W
"
indow"
ndow
Xe-
e 133 Si
133 S gnal
-6
10
50
100
150
200
250
300
350
400
450
45
Days Po
a
st-De
-D to
t n
o at
a io
i n
Source: Charles Carrigan, “Using OSI [On-Site Inspection] Field Studies and Tests to Define Noble Gas
Sampling and Analysis Requirements,” Paper presented at International Noble Gas Experiment-2009 Conference,
Daejeon, Korea, November 9-14, 2009, Lawrence Livermore National Laboratory, LLNL-PRES-41961.
Notes: 1 Becquerel = 1 radioactive disintegration per second. Calculations are derived from a model based on
data obtained from the Non-Proliferation Experiment (NPE), an underground blast conducted in 1993 at the
Nevada Test Site that used nonnuclear explosives to simulate a 1-kiloton nuclear explosion. It used helium-3 gas
to simulate argon-37 and sulfur hexafluoride gas to simulate xenon-133.
Study Numbers, Types, and Basing for Aircraft That Collect Nuclear Debris
The Air Force Technical Applications Center (AFTAC) operates two WC-135 “Constant
Phoenix” aircraft, which are designed to collect particulates and gases from a nuclear explosion.
The WC-135 is a component of the U.S. Atomic Energy Detection System.110 AFTAC states,
The Air Force maintains one primary and one backup WC-135 to support airborne nuclear
collections. The aircraft are operated by the 55th Wing, 45th Reconnaissance Squadron at
110 U.S. Air Force. “WC-135 Constant Phoenix,” fact sheet, http://www.af.mil/information/factsheets/factsheet.asp?id=
192.
Congressional Research Service
30
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Offutt Air Force Base, Neb. Both the primary and backup aircraft are equipped with an
AFTAC collection suite that provides the capability to collect the gaseous and particulate
debris that might be released from a nuclear explosion. The Air Force is conducting an
Analysis of Alternatives to determine solutions that can support changing mission
requirements and will provide long term, viable alternatives to the current capability as it
reaches end of life (the WC-135 airframes are almost 50 years old).111
The Air Force elaborated on the preceding statement: “Currently, the WC-135 must be evaluated
for re-skinning the wings in around 2022, and AFMC [Air Force Materiel Command] asserts TF-
33 engine sustainment through 2040. Otherwise, there is not an explicit end-of-life date.” The
analysis of alternatives “will examine the WC-135, plus other manned and unmanned aircraft,
and assess the number of each aircraft appropriate for the mission.”112
One of the two WC-135s is the primary aircraft; the other is a backup. The Air Force stated,
“Either aircraft can fly the mission. Also both aircraft can be flown simultaneously. However, one
of the aircraft is usually in Primary Depot Maintenance (PDM) and therefore would be
unavailable. If both aircraft are out of PDM then they both can support mission.”113 Thus the
USAEDS airborne collection asset at most times is a single aircraft nearly 50 years old based
thousands of miles from North Korea and also from Iran.
Given the prospect that several nations, over a vast geographic area, might conduct nuclear tests,
it would be of value to collect samples as soon as possible to narrow the region where the test
occurred, to minimize the loss of samples with time, and to have a chance of obtaining samples of
xenon isotopes with forensic value. Accordingly, it may be desirable to have more than two
aircraft for this purpose, and to have some forward-based. Forward basing might be less costly for
large, long-range Reaper- or Global Hawk-type drones than for WC-135s and similar aircraft.114
For example, since these drones are remotely operated, personnel controlling the mission and
operating the sensors (as distinct from the ground crew) would not have to be forward-based.
Also at issue is whether to extend the service life of WC-135s or procure new aircraft. Finally, it
may be worth considering whether to add airborne sensors to the IMS, which would have to be
done in accordance with Article IV, paragraphs 11 and 23, of the CTBT.
Congress is aware of the importance of collecting samples promptly. In the Nuclear Forensics and
Attribution Act, it found,
111 Information provided by Air Force Technical Applications Center through Air Force Legislative Liaison Office,
email, October 26, 2009.
112 Information provided by Air Force Legislative Liaison Office, email, October 29, 2009.
113 Information provided by U.S. Air Force Legislative Liaison Office, November 24, 2009.
114 The MQ-9 Reaper is a large remotely-piloted aircraft designed for ground attack or intelligence missions. Several
characteristics may make it suitable for collecting radionuclide samples. Its range is 3,682 miles, its payload is 3,750
pounds, its ceiling is up to 50,000 feet, it has long endurance, and it can be loaded into a container for deployment by
aircraft, e.g., C-130 Hercules, worldwide. U.S. Air Force. “MQ-9 Reaper,” fact sheet, http://www.af.mil/information/
factsheets/factsheet.asp?id=6405. The RQ-4 Global Hawk might also be used to collect radionuclide samples. Its
mission is long-range high-altitude intelligence, surveillance, and reconnaissance. It is larger than the Reaper, has a
range of 10,939 miles, a payload of 2,000 or 3,000 pounds, depending on the model, and a ceiling of 60,000 feet. U.S.
Air Force. “RQ-4 Global Hawk,” fact sheet, http://www.af.mil/information/factsheets/factsheet.asp?id=13225. By way
of contrast, the Predator has a range of 454 miles, a payload of 450 pounds, and a ceiling of 25,000 feet. U.S. Air
Force. “MQ-1 Predator,” fact sheet, http://www.af.mil/information/factsheets/factsheet.asp?fsID=122.
Congressional Research Service
31
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Many of the radioisotopes produced in the detonation of a nuclear device have short half-
lives, so the timely acquisition of samples is of the utmost importance. Over the past several
decades, the ability of the United States to gather atmospheric samples—often the preferred
method of sample acquisition—has diminished. This ability must be restored and modern
technologies that could complement or replace existing techniques should be pursued.
Improve Onsite Inspection Capability
Since the previous administration did not seek entry into force of the CTBT but did favor
improving means of monitoring nuclear testing, it requested only those funds for the CTBTO
Preparatory Commission that directly supported the IMS. It requested these funds in the State
Department’s International Affairs Function 150 budget in the Nonproliferation, Antiterrorism,
Demining, and Related Programs account. The FY2007 budget justification, for example, stated
that the requested funds, $19.8 million, would “pay the U.S. share for the ongoing development
and implementation of the International Monitoring System (IMS), which supplements U.S.
capabilities to detect nuclear explosions. Since the United States does not seek ratification and
entry-into-force of the CTBT, none of the funds will support Preparatory Commission activities
that are not related to the IMS.”115 While the PrepCom budget shows no nation-by-nation link
between funds received and funds spent, this quotation illustrates the attitude toward expenditures
that would be of value only if the CTBT were to enter into force, such as OSIs. Consistent with
this policy, the administration directed U.S. R&D funding away from OSI issues and toward IMS
technologies.
With the Obama Administration favoring the CTBT, and with OSIs a key part of the verification
regime, more might be done to make them more effective. One approach would be to develop
more sensitive detectors. A second would be to integrate geophysical methods for detecting
anomalies hundreds of feet underground with gas sampling techniques to help inspectors locate a
suspected test more precisely. A third would be to conduct field experiments on how noble gases
reach the surface. The only OSI-type gas sampling experiment was performed in conjunction with
the 1993 NPE (see “Increase the Sensitivity of Detection Equipment”). Conducting similar
experiments (perhaps releasing gases in mine shafts to reduce costs) under various geologic,
containment, hydrologic, and barometric conditions would help develop and calibrate computer
models of gas leakage from underground tests, making results of the type found in 1993
applicable to a wider range of conditions under which OSIs might be conducted.
Conduct Further R&D on Satellite Detection
Given the immense value of data provided by satellites, Congress might explore whether
additional R&D might be warranted on satellite-borne means of detecting signatures that a
clandestine nuclear program or test, or preparations for a nuclear test, might produce. While the
IMS does not include satellites, the Provisional Technical Secretariat, which operates the IMS,
might want to conduct such R&D as well, both because it can utilize commercial satellite imagery
and because the IMS might, at some point, have access to its own satellite data. The CTBT
provides (Article IV, paragraphs 11 and 23) for the possibility of adding monitoring technologies
such as satellites to the IMS if agreed pursuant to Article VII (Amendments).
115 U.S. Department of State. Summary and Highlights: International Affairs Function 150, Fiscal Year 2007 Budget
Request, p. 40, at http://www.state.gov/documents/organization/60297.pdf.
Congressional Research Service
32
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Evaluate Classified Projects
The United States is presumably conducting classified R&D in areas related to the subject matter
of this report. The relevant congressional committees may wish to determine what efforts, if any,
are being made along these lines and whether the level of effort in each such area is appropriate.
For example, many evasion scenarios set forth by CTBT opponents were created decades ago,
some in the 1950s. They have been studied and debated ever since, and study of such scenarios
remains a daily concern of the Intelligence Community. The ability to defeat evasive scenarios
would increase confidence in detection capability, while an inability to defeat them would help
guide detection R&D. Either way, efforts to develop and to defeat these scenarios would
challenge scientists working on detection, just as with any other offense-defense competition.
However, few new evasion scenarios or technologies appear in the public record out of concern
that public discussion of them could aid would-be evaders.
Look for New Signatures to Help Determine If a Test Is Nuclear
Examine Costs and Benefits of Long-Range Detection of Argon-37
As noted earlier, it may be feasible and useful to detect argon-37 at a distance. However, moving
from “may be feasible” to an operational system would require characterizing background levels
of this isotope; determining the value that might be gained by detecting this isotope in addition to
detecting xenon-133; studying the worldwide distribution of calcium, especially at likely test
sites; developing automated detection equipment that could be used at remote locations, such as
at IMS radionuclide stations; and determining whether the cost of this effort is worth the benefit.
Study Signatures of a Chemical Explosion
As the case of the 2009 North Korean test shows, it would be useful to determine if an explosion
was nuclear or chemical in order to reveal if a nation had conducted a nuclear test or was bluffing.
Effluents of a chemical explosion would probably not permit making that determination because
they would be hydrocarbons and there is a huge atmospheric background of such materials from
vehicles, industry, forest fires, mining explosions, etc. However, there might be other signatures,
such as in the preparation, seismic waves, or post-event activity. For example, detailed study of
seismic waves might reveal slight differences between those generated by nuclear or chemical
explosions. This was a goal of the 1993 NPE, though the apparent inability to prove conclusively
that the 2009 North Korean event was or was not nuclear indicates that more work along these
lines may be warranted. At the same time, it would be important to guard against the prospect that
a nation could create signatures of a chemical explosion as cover for a nuclear test. The ability to
monitor signatures of nonnuclear explosions is especially important in situations like the 2009
event prior to CTBT entry into force. After entry into force, should it occur, onsite inspections
would be available to help resolve such situations.
Congressional Research Service
33
North Korea’s 2009 Nuclear Test: Containment, Monitoring, Implications
Author Contact Information
Jonathan Medalia
Specialist in Nuclear Weapons Policy
jmedalia@crs.loc.gov, 7-7632
Congressional Research Service
34