“What's really strikes me about ACA is the potential to shape the next generation of leaders on arms control and nuclear policy. This is something I witnessed firsthand as someone who was introduced to the field through ACA.”
– Alicia Sanders-Zakre
June 2, 2022
Building Up the Regime for Verifying the CTBT
Share this

Ambassador Tibor Tóth

Over the last decade, considerable progress has been made in building up the unique verification regime of the Comprehensive Nuclear-Test-Ban Treaty (CTBT) to monitor the globe for nuclear explosions.

The CTBT’s global alarm system has grown not only quantitatively through the increase in number of monitoring facilities, but also qualitatively. As a result of scientific and technological progress in monitoring technologies, including in the automatic processing of data, the system, although still incomplete, is already more powerful than expected by the treaty’s negotiators. Therefore, there is a very high probability today that a militarily significant nuclear test anywhere on the planet will be detected by the system. This capability will further increase as more and more monitoring stations join the global network of the Preparatory Commission for the Comprehensive Nuclear-Test-Ban Treaty Organization (CTBTO), the organization mandated to establish the verification regime so that it is operational when the treaty enters into force.

The Three Pillars of Verification

The CTBT verification regime consists of three main elements. The first is the International Monitoring System (IMS) with its 337 monitoring facilities—321 monitoring stations and 16 radionuclide laboratories—around the globe. The IMS comprises four verification technologies that work in synergy to enable the detection, location, and identification of potential nuclear explosions. Seismic stations “feel” the ground for shock waves generated by explosions, infrasound and hydroacoustic stations “listen” for corresponding sound waves, and radionuclide stations “sniff” the atmosphere for traces of radioactive particles and gases, which indicate whether a given explosion is nuclear.

IMS facilities, located in 89 countries worldwide from Iceland to Tahiti, provide complete global coverage. Many of the monitoring stations are deliberately constructed in remote areas to ensure global coverage or reduce disturbing background signals from settlements or traffic. As of August 2009, 248 monitoring facilities, or more than 73 percent of the planned IMS network, had been certified as meeting CTBTO specifications.

Data collected by the IMS are transferred in real time via six geostationary satellites and secure terrestrial communication lines of the Global Communications Infrastructure to the International Data Centre (IDC) in Vienna, the second pillar of the verification regime. The data are analyzed to detect, locate, and identify natural and man-made events, including potential nuclear events. From the IDC, data and analysis, both automated and human, are forwarded to the CTBTO member states, in near real time for the automated data.

The third and final element of the global alarm system is the on-site inspection (OSI) regime, the in-the-field, eyes-on-the-ground component. It provides clarity on an event recorded by the IMS and analyzed by the IDC. Although an OSI can be invoked only after entry into force of the CTBT, OSI procedures have already been established and tested in the field.

Seismic Monitoring

Seismic monitoring stations are usually the first to detect underground nuclear explosions, which make up three-quarters of the nearly 2,050 nuclear tests conducted until the CTBT’s opening for signature in September 1996. (Six underground nuclear tests were conducted afterward by India, Pakistan, and the Democratic People’s Republic of Korea [DPRK].)[1] Seismic stations are equipped with sensors that measure elastic waves—similar to sound waves, but in a solid environment—generated by seismic events such as earthquakes or explosions. Their measurements help to identify the location, magnitude, and nature (e.g., earthquake or explosion) of a seismic event.

IMS seismic stations are grouped into primary and auxiliary networks. The primary seismic network transmits its data continuously to the IDC in real time. The auxiliary seismic stations also operate continuously but send data on request only. Their data are used to corroborate detection of an event by the primary seismic stations and to determine its characteristics with greater confidence.

By August 2009, 80 percent of the network of 50 primary seismic stations and 75 percent of the planned 120 auxiliary stations had been certified. Another 27 seismic stations have already been installed or are under construction, including a primary seismic station in Turkmenistan.

The increase manifested itself in the nuclear test announced by the DPRK on May 25, when a total of 61 seismic stations registered the event, compared to 22 for the DPRK’s first announced nuclear test on October 9, 2006. Although the greater magnitude of the 2009 event compared to the one in 2006 contributed to this increase, so did the fact that additional seismic stations were operating, 18 of which registered the 2009 event.

A 2002 report by the U.S. National Academy of Sciences on technical issues related to the CTBT stated that “underground nuclear explosions can be reliably detected and can be identified as explosions, using IMS data down to a yield of 0.1 kilotons (100 tons) in hard rock if conducted anywhere in Europe, Asia, North Africa and North America.”[2]

National Technical Means

The CTBT explicitly recognizes a role for national technical means of verification. Data collected through national technical means may be used to request an OSI. This feature is particularly relevant for seismic monitoring. The primary seismic network alone can already detect seismic events of a magnitude as low as four on the Richter scale (the rough equivalent of a one-kiloton test) anywhere in the world and below that level for many regions. The thousands of additional seismometers deployed throughout the world will detect events of considerably smaller magnitude. It is worth noting that the detection capability could increase even further when seismic data are combined with, for example, radionuclide/noble gas data and on-site inspections.

Hydroacoustic Monitoring

The IMS hydroacoustic network comprises 11 hydrophone and T-phase (seismic) stations. Only a few stations are required because water is a highly efficient medium for the transmission of sound. These stations monitor the world’s oceans for evidence of nuclear tests by measuring changes in water pressure caused by sound waves emanating from underwater explosions. One layer in the water “traps” sound waves and carries them for thousands of kilometers. This layer, where the speed of sound is slowest, is called the Sound Fixing and Ranging Channel, which is typically at a depth of about 1,000 meters.

The ocean is a very inhospitable working environment, and the station equipment must be of exceptional quality, able to withstand the huge pressure, destructive waves, strong currents, and rugged undersea volcanic terrain, in addition to freezing temperatures and the saline corrosiveness of ocean waters. The underwater equipment is designed for longevity; stations must last for as long as 20 years. The length of cables between the hydrophones and the shore can exceed 100 kilometers. The hydrophone stations are the costliest and most technically demanding ones of the entire IMS.

Ten of the 11 hydroacoustic stations have already been certified, the latest being the station at Wake Island in June 2007. The network’s capabilities have been greatly enhanced. An explosion triggered as part of a marine seismic survey off the coast of Japan was detected approximately 16,000 kilometers away by IMS hydrophone sensors at Juan Fernandez Island. Its yield was the equivalent of a mere 20 kilograms of TNT. By comparison, one kiloton, the unit for measuring the yield of nuclear weapons, is 1,000,000 kilograms.

Infrasound Monitoring

Infrasonic sensors measure micropressure changes in the atmosphere that are generated by the propagation of infrasonic waves. These very low frequency waves can also be created by atmospheric nuclear explosions. The IMS infrasound network is unique because it had to be built from scratch rather than building on an existing network, such as the seismic technologies. It will comprise 60 stations when complete. A total of 41 infrasound stations in 25 countries has been certified to date, meaning that almost 70 percent of the network’s stations already comply with the CTBT’s stringent requirements. Another five stations are currently under construction.

An infrasound station must be located at a site with the lowest possible background noise level in order to monitor the atmosphere effectively for infrasound waves from explosions. For infrasound stations, a main concern is direct exposure to winds. For that reason, such stations are often situated in forests. An example is the station at Freyung, Germany, in the Bavarian Forest; that station recorded the explosion at an oil depot near London in December 2005.

Radionuclide Monitoring

Collection and analysis of radioactive particles or gas from a nuclear explosion can provide conclusive evidence that an explosion has been nuclear. At entry into force of the CTBT, the total radionuclide network will consist of 80 stations, of which 57 have already been certified.

One-half of the stations will be equipped with noble-gas technology, a relatively new technology that was in an experimental stage at the time of the treaty’s negotiations in the mid-1990s. As with the infrasound network, the CTBTO is the first to build up a global noble-gas monitoring capability.

Radioactive noble gases are particularly relevant for detecting underground nuclear explosions. Noble gases, such as xenon, are unreactive and can leak out, or vent, through fissures in the rock after an underground test. Once in the atmosphere, the plumes of xenon isotopes can be blown for thousands of miles and detected by CTBT noble-gas systems. This constitutes an additional means of detecting a potential treaty violator, in addition to telltale seismic, infrasound, or hydroacoustic signals from the explosion. In October 2006, for example, following the DPRK’s first nuclear test, the IMS detected traces of xenon-133 although the noble-gas network was only 25 percent installed. The traces were picked up by the CTBTO’s station in Yellowknife, Canada, some 7,500 kilometers away from the DPRK’s test site.

The DPRK’s Recent Nuclear Test

Although no noble gas was measured from the DPRK’s second announced nuclear test, it is important to note that the seismic evidence and the precision of the data would have been enough to provide a firm basis for a decision by the CTBTO’s future executive council to dispatch an OSI. The CTBT foresees a size limit of 1,000 square kilometers for the inspected area, and the seismic data determined the location of the DPRK’s test well within this limit. The data would have provided a clear lead to the inspection team.

Seismic data alone cannot identify whether an explosion is nuclear or conventional, but it can give an indication of the explosion’s yield. With regard to yield estimations of the recent event in the DPRK, verification technology experts such as Professor Paul Richards of the Lamont-Doherty Earth Observatory at Columbia University considered a “nuclear bluff” scenario as highly implausible. Several thousand metric tons of conventional explosives would have needed to be fired instantaneously, and preparations would have required a massive logistical undertaking that would have been virtually impossible under the prevailing circumstances and would not have escaped detection.[3]

The findings from the DPRK’s tests in 2006 and in 2009 show how the different elements of the verification regime—waveform and radionuclide technology and on-site inspections—are designed to complement each other to ensure that any clandestine nuclear test will be caught. In fact, at the time of the DPRK’s second nuclear test, the noble-gas network had been considerably enhanced, with new systems installed in China, Japan, Mongolia, and Russia. The system’s detection capability was excellent and would have detected the venting of even one part per thousand (0.1 percent) of the newly formed noble gases.

The fact that apparently no more than 0.1 percent of the noble gases escaped is most likely due to containment. In one such scenario, an explosion may melt the surrounding rock, which, in certain circumstances, may solidify into an effective barrier against leakage. Experience from U.S. and Soviet nuclear testing, however, shows that such containment is extremely difficult to cause deliberately because it depends on a multitude of factors that vary strongly depending on the circumstances and are hard to predict, especially for countries with nascent nuclear weapons programs.

Data Processing and Analysis

The CTBTO is completing an upgrade of its computer systems. The newest computers at the IDC are now more than 50 times faster at processing and analyzing IMS monitoring data than the original ones installed in 1997. This additional capacity enables the IDC to keep up efficiently with the further increase in IMS data, which doubles every few years, as well as add new capabilities.

The CTBT verification regime was negotiated at a time during which the Internet grew substantially and high-bandwidth, low-cost communication became globally available. New software to publish, index, and retrieve information through the Web was developed for a wide consumer market, which has enabled the IDC to provide levels of service to member states far beyond what was originally envisioned for the organization. One example is the ability to provide e-learning to station operators.

Data distribution using the CTBTO’s own communication network, the Global Communications Infrastructure, and the power of modern Web tools has helped the National Data Centres fulfill their monitoring obligation. Sophisticated software developed and maintained by the IDC is available to the member states to assist in their role as treaty monitors.

Better Understanding the Data

With up-to-date computational capabilities, the IDC is upgrading nearly all aspects of its monitoring capabilities. As the seismic network grows, more and smaller seismic events are detected, which increases the workload for the analyst staff. New developments have focused on improving automatic processing to reduce the number of false alarms while maintaining a low threshold for detection.

Hydroacoustic processing has exceeded initial expectations, with the hydrophone arrays being capable of measuring very precise azimuths (the direction of the source) for recorded sound waves. For infrasound monitoring, the IDC is developing a new processing and analysis system to interpret infrasound data more efficiently. Advances in computer graphics enable the analysts to visualize the data in greater detail and much more efficiently.

For the radionuclide technologies, the processing of the particulate measurements has been re-engineered to improve automatic treatment of CTBT-relevant isotopes. Along with new analysis software, the result is more objective and consistent. In the area of noble-gas monitoring, entirely new data processing and analysis methods had to be introduced.

A giant step forward for the CTBT monitoring capabilities over the past decade, enabled directly by the increase in computer capability, is in the field of atmospheric transport. Reliably tracking the movement of air volumes is crucial for estimating the location of sources of radionuclide particulates and noble gases detected at IMS stations. For this tracking, the CTBTO applies a technique called Atmospheric Transport Modeling (ATM) to reconstruct the three-dimensional travel path of a radionuclide particle or noble-gas atom. To improve the accuracy of the ATM, the CTBTO cooperates with the World Meteorological Organization (WMO) to exchange results from other WMO data centers also performing the ATM. Merging the independent results potentially reduces the uncertainties involved in the backtracking procedure.

To be able to precisely identify radionuclides or radioactive noble gases as coming from a nuclear explosion, it is important to increase the knowledge of the “background,” the natural or man-made concentration of such substances. To further its knowledge in the field of noble-gas background, the CTBTO performed a series of measuring campaigns using mobile noble-gas stations in areas not previously covered by regular stations. This project, funded by the European Union, helped to improve characterization of the background xenon signals from civilian sources, such as isotope production facilities.

In the next step, referred to as data fusion, radionuclide observations and the results of ATM are compared to the data harvested through the seismic, hydroacoustic, and infrasound technologies. The synergy between the two types of monitoring allows analysts to link a seismo-acoustic event to a radionuclide or noble-gas release.

On-Site Inspections

Although two elements of the CTBT verification regime, the IMS and the IDC, are already functioning on a provisional basis and have successfully proved themselves with the DPRK’s tests, the situation is different for the third element, the OSI. The CTBTO is limited here in what it can practice or demonstrate because real OSIs will only be possible after the treaty’s entry into force.

Regardless, the organization is training intensely; the CTBTO has regularly been conducting field exercises to test the different scientific methods, at times through several exercises per year. A quantum leap was made last September when different aspects of an OSI were tested together in the Integrated Field Exercise in 2008 (IFE08), conducted at the former Soviet nuclear test site at Semipalatinsk, Kazakhstan. This involved a whole range of challenges, from transporting more than 50 metric tons of equipment and around 200 participants to an extremely remote area, to negotiating with an unwilling inspected country (the fictitious state of Arcania—not Kazakhstan, which was an extremely helpful host country for the exercise), to performing various scientific measurements.

The IFE08 provided invaluable lessons for the further development of OSI capabilities. It helped to identify where further buildup of personnel and equipment is necessary to meet the CTBT’s ambitious requirement of being able to launch a full-fledged OSI at short notice anywhere on the globe. Yet, this exercise also demonstrated that the CTBTO has the capacity to conduct an OSI. This final element of the CTBT’s verification regime promises to act as a strong and reliable deterrent to potential violators of the ban on nuclear explosions once the treaty enters into force.


The verification regime is constantly being improved. The global system to monitor the world for nuclear explosions is ready to verify the CTBT once it enters into force. Given the considerable progress that has been made in the past decade and the experience gained, there is a very high probability today that member states would be able to discover any nuclear test using data generated by the CTBT verification regime and other assets available to individual states.

Civil and Scientific Applications of the CTBT Monitoring System

In addition to its principal use of verifying a nuclear test ban, the monitoring network established by the Preparatory Commission for the Comprehensive Nuclear-Test-Ban Treaty Organization (CTBTO) has a number of important civil and scientific applications.

Seismic: When CTBTO experts analyze the events located by means of seismic data from the International Monitoring System (IMS), most often they are looking at an earthquake. More than 250,000 earthquakes of an approximate magnitude of four and higher on the Richter scale have been detected by the IMS in the past 10 years, compared to only two nuclear tests, by the Democratic People’s Republic of Korea. (In both cases, the CTBTO recorded the data, detected and located the event, and made the location available to member states immediately.) Nearly all seismic events can be discarded as irrelevant for the purposes of test ban monitoring, but the real-time data have proven to be very fast and reliable and therefore very useful to the tsunami warning centers in high-risk areas. After the devastating Sumatra tsunami on December 26, 2004, CTBTO member states decided that real-time primary and auxiliary seismic data should be provided to tsunami warning centers in the Indian and Pacific Ocean areas.

Hydroacoustic: Hydroacoustic technology is of great use in a range of civil and scientific fields. Like seismic monitoring data, hydroacoustic data are provided to tsunami warning centers based on a decision of CTBTO member states. It is up to the member states to explore other potential uses of this monitoring technology, such as climate change studies and research on whale populations and their migration patterns. IMS hydroacoustic data also could help increase shipping safety by providing data on underwater volcanic explosions or on the breakup of ice shelves and the creation of large icebergs.

Infrasound: CTBTO member states could explore applying CTBT infrasound technology to civil aviation safety. Large ash plumes caused by volcanic eruptions can cause jet engines to malfunction or fail completely. There are 600 volcanoes active today, which have already caused four near-disasters in commercial aviation since 1982. CTBT infrasound technology could assist in the detection of volcanic eruptions by registering the very low frequency sound waves they emit. Infrasound technology is also important for the monitoring and scientific study of the upper atmosphere and for collecting data on hurricanes and tropical storms.

Radionuclide: The radionuclide particulate and radio-xenon networks are unprecedented in terms of detection sensitivity and global coverage. The data gathered from these networks could thus be used, should CTBTO member states decide to do so, for scientific studies on worldwide background radioactivity levels. The 80 air monitoring stations collect standardized daily filter samples, which constitute a large and permanently growing worldwide archive for historical studies of pollutants and microorganisms. (The latter are sensitive indicators for climate change.) Moreover, the data allow for meteorological studies of the dispersion and long-range transport of airborne pollutants. Finally, certain radionuclides can be used as tracers to help to understand the stratosphere-troposphere exchange and to verify and validate Global Circulation Models, which describe the dynamics of the atmosphere.




Ambassador Tibor Tóth has been executive secretary of the Preparatory Commission for the Comprehensive Nuclear-Test-Ban Treaty Organization since August 2005. He has been actively involved with arms control issues since the early 1980s, including as Hungarian deputy state secretary of defense, ambassador at large for nonproliferation, and permanent representative of Hungary to the Conference on Disarmament in Geneva.


1. There are various ways of counting nuclear tests. For this article, the method used by the Stockholm International Peace Research Institute has been used. Under that method, for example, India’s detonations on May 11 and May 13, 1998, are counted as one test for each date. See Vitaly Fedchenko and Ragnhild Ferm Hellgren, “Nuclear Explosions, 1945–2006,” SIPRI Yearbook 2007 (New York: Oxford University Press, June 2007), pp. 555-557.

2. National Academy of Sciences, Technical Issues Related to the Comprehensive Nuclear Test Ban Treaty (Washington, D.C.: National Academy Press, 2002), p. 57, www.nap.edu/catalog/10471.html.

3. See CTBTO Preparatory Commission, “Experts Sure About Nature of the DPRK Event,” June 12, 2009, www.ctbto.org/press-centre/highlights/2009/experts-sure-about-nature-of-the-dprk-event/.