Completing the CTBT's Verification Regime: Progress and Challenges
Trevor Findlay and Oliver Meier
INTRODUCTION
Four years after the Comprehensive Nuclear Test Ban Treaty (CTBT) was opened for signature, the treaty's verification system is being progressively established. After a slow start, which was mainly due to the novelty of setting up a verification system based on a global network of monitoring stations, implementation is now making good progress. The Preparatory Commission (PrepCom), which marked its third anniversary in April 2000, and the Provisional Technical Secretariat (PTS), both of which are in charge of setting up the International Monitoring System (IMS) and its associated components, have identified the major hurdles which stand in the way of completing the system and have begun to allocate resources accordingly. The Provisional CTBTO is evolving quickly into a fully-fledged international organisation. As of 25 September 2000 the PTS employed 242 staff from 70 signatory states-making it more than two-thirds complete. Although there are still political and technical hurdles to be overcome, such as the development of on-site inspection arrangements, the PTS could complete its work before the treaty enters into force. This will, however, require the undiminished political, technical and financial support of all states signatories.
THE INTERNATIONAL MONITORING SYSTEM (IMS)
Background
The IMS will consist of 321 monitoring stations and 16 radionuclide laboratories located in some 90 countries. Some of these already exist, while others will have to be constructed. Two hundred and one of these stations belong to the primary network, which will be providing data to the IDC which has been collected on a continuous, round-the-clock basis. In many cases IMS stations use existing infrastructure, upgraded and certified for use by the IMS. Sixty-two per cent of the network of 120 auxiliary seismic stations, which will supply data only on request, essentially meet technical specifications already.
Four types of stations are to be established?seismological, infrasound, hydroacoustic and radionuclide. The seismic network will form the core of the verification system. Seismic waves generated by earthquakes, explosions or other phenomena will be detected using 50 primary and 120 auxiliary seismic stations, distributed worldwide. In addition, 11 underwater hydroacoustic stations are being set up. Sixty land-based infrasound stations will use sonar to detect atmospheric tests, while 80 radionuclide stations will measure radioactive particles in the atmosphere from atmospheric nuclear tests or underground tests that vent. Sixteen radionuclide laboratories will analyse filters from the stations, as well as samples taken by inspectors.
The four different technologies operated by the IMS are complementary and are able to detect tests in different environments. Seismic monitoring is best at detecting underground tests - although it might also be able to discern atmospheric tests conducted at low altitudes. Hydroacoustic technology primarily monitors the oceans and infrasound is most efficient at detecting atmospheric tests - although it may also detect some underwater and shallow underground events. Seismic and acoustic detection technologies under specific circumstances might not provide enough conclusive data to reveal whether a large conventional explosion or small nuclear test has taken place. Radionuclide stations could be the most powerful tool in clarifying the nature of an event by detecting radioactive particles.
The four different technologies operated by the IMS are complementary and are able to detect tests in different environments. Seismic monitoring is best at detecting underground tests?although it might also be able to discern atmospheric tests conducted at low altitudes. Hydroacoustic technology primarily monitors the oceans and infrasound is most efficient at detecting atmospheric tests?although it may also detect some underwater and shallow underground events. Seismic and acoustic detection technologies under specific circumstances might not provide enough conclusive data to reveal whether a large conventional explosion or small nuclear test has taken place. Radionuclide stations could be the most powerful tool in clarifying the nature of an event by detecting radioactive particles.
States parties may also contribute data to the IMS from so-called Co-operating National Facilities (CNFs). CNFs are national stations which can be called on by the CTBTO to clarify suspicious events. Such stations are operated by treaty parties but have undergone the same certification procedures, including the authentication of communication links, as IMS station.
Progress in station establishment and certification
After a slow start, the setting up of the required IMS stations is progressing steadily. In the early days of the PTS, many legal and bureaucratic hurdles had to be overcome before construction and/or certification of a station could begin. The PTS first had to establish the legal procedures for setting up stations and establish links with National Authorities and scientific co-operating partners in IMS participating countries.
Another impediment is the complex certification process. In order to certify a station, the PTS has to be assured that technical specifications are substantially met and data from the stations can be authenticated. Finally, a proper link to the global communications infrastructure (GCI) has to be established. Setting up a new IMS station – from the planning stage to certification – takes at least two years.
On 28 July 2000, the first three IMS stations (primary seismic facilities in Canada, Norway and the United States) were certified. The PTS estimates that the number of IMS station that have a site survey completed, are installed, or are certified, will increase dramatically over the next couple of years. As the figures below show, this is true for all four monitoring technologies. Current plans are that by the end of 2001, ninety-six per cent of site surveys will be completed, 56 per cent of stations will be installed and sending data to the IDC, and 41 per cent of the stations will have been certified.1
Calibration of IMS stations has been slow. Four conventional explosions have been conducted to calibrate IMS stations, three in Kazakhstan, in cooperation with the United States, and one in the Red Sea conducted by Israel. An explosion in Kazakhstan in October 1999 was also used for an on-site inspection exercise by the PTS. The most recent explosion in Kazakhstan destroyed the last tunnel at the former Soviet nuclear test site Semipalatinsk, which is now officially dismantled.
Legal frameworks
Certification of stations must be covered by an agreed legal framework, 'facility agreements or arrangements', between the PrepCom and host states. These must be approved by the PrepCom if they differ substantially from the model agreement provided by the Secretariat. The Executive Secretary of the PTS has urged those governments that have not yet negotiated facility agreements to do so.2 So far, fourteen have been concluded. Two hundred and eighty IMS facilities in 60 countries were covered by some kind of legal arrangement by the end of September 2000.3 Site surveys for 125 of the 201 IMS stations in the primary network are complete.4
For about 30 stations, and one radionuclide laboratory, new sites had to be found when site surveys revealed that the co-ordinates given in Annex 2 of the treaty were unsuitable. The reasons included excessive background noise or because the locations were at sea or in other unsuitable areas.5
International Data Centre (IDC) and Global Communications Infrastructure (GCI)
Integrating data on a large scale from many different sources poses a completely new monitoring and verification challenge to an international organisation, but is also likely to result in great synergies. The IDC, which is being progressively commissioned in Vienna, will receive and process data from all IMS monitoring facilities via a dedicated Global Communications Infrastructure (GCI).
The GCI will use very small aperture terminals (VSATs) to ensure the swift and secure transport of up to 11.4 gigabytes of data between facilities, the IDC and states parties. By March 2000, VSATs had been installed at 25 IMS facilities. Fifty are expected to be installed by the end of 2000.6 Three communications 'hubs', which receive data from IMS stations in a particular region and forward it to the IDC, are located in Germany (European hub), Italy (Atlantic and Indian Ocean hubs) and California (Pacific Ocean hub). They are all complete and transmitting data.
Data from seismic and acoustic stations will be received by the IDC in near real time and be available within a few hours to states parties which wish to receive it. Data processing will be largely automated. The second of four software releases for this purpose was installed at the IDC in late 1999. On 20 February 2000 the IDC assumed responsibility for collecting and disseminating data from the IMS stations in operation. Of the 201 stations in the primary network, 32 are sending data to the IDC. This number is expected to rise sharply over the next few years, with more than 50 per cent of stations reporting to the IDC by the end of 2001.
Unlike the four types of 'wave form' data, radionuclide data will be available after a delay of several days because samples have to be physically collected and analysed. New technologies currently being developed may automate data transmission from stations in remote areas, but this will not shorten the time necessary for analysis.
It is the IDC's responsibility to screen out events which are clearly of natural origin, as well as those which are clearly non-nuclear and man-made, such as large conventional explosions. A large percentage of all earthquakes, for instance, occur at depths at which it is impossible to conduct clandestine nuclear tests. By applying screening criteria to the vast amount of data delivered to the IMS, the number of potentially suspicious events can be dramatically reduced. The IDC will issue Standard Event Bulletins, which indicate the degree to which each detected event meets specific screening criteria.7 States without significant national technical and analytical means will naturally look to the IDC for more precise information if suspicions are aroused concerning a particular event. The IDC is expected to assist any state party in the technical analysis of IMS data as well as data provided by other states parties.8
Forty-four signatory states are currently receiving IMS data and 'products' on a trial basis. On 21 February 2000 the IDC started to distribute Reviewed Event Bulletins (REBs) to member states. They have been issued daily for the past few months, even though this was not planned for the current Phase 4 of the establishment of the IDC. Due to a shortage of staff and the constraints of a 5-day working week, REBs are currently being distributed with several days' delay.
Some member states have urged the PTS to ensure that the IDC distributes products on a more timely basis. In order to do this under current operational limitations, PrepCom Working Group B (responsible for verification) suggested the IDC do this for 5 daily REBs per week. This will be done by expanding the IDC's work schedule to a 6-day working week. The immediate goal is to demonstrate that the IDC is capable of distributing data in near-real time, around the clock, without implementing this on a permanent basis prior to entry into force. It is planned that during Phase 5 of the IDC's installation (commencing in early 2001), the timely production of REBs will be increased from 5 to 7 per week.
To evaluate the work of the IDC, Working Group B has commissioned an external expert evaluation. Led by the UK's Ian Kenyon, formerly of the Organization for the Prohibition of Chemical Weapons (OPCW), six international experts will spend two to three weeks in Vienna to evaluate whether the IDC is fulfilling its mandate and what possible improvements could be made. Topics to be considered by the Evaluation Team include: the implementation of PrepCom guidelines by the IDC; the overall state of the IDC and the GCI; as well as the interaction between the IDC and other parts of the PTS, states signatories and the broader scientific community; and possible improvements in the scientific methods and software used by the IDC. The experts will report to the PrepCom in November 2000. If successful, external consultants might be used to evaluate other parts of the PTS.
THE CONTRIBUTION OF NON-IMS CAPABILITIES
Member states can also submit data from national technical means (NTM) of verification to the Executive Council in the course of consultation and clarification procedures or to support a request for an on-site inspection. It is likely that states will use a very broad definition of NTM, enabling them to table almost any information that they deem relevant and politically acceptable. Thus, a state could, for example, table information from its own signals intelligence, satellite imagery obtained from its own satellites or those operated commercially, or data collected by scientists anywhere in the world.
Data from scientific networks might carry greater weight because they will have been collected transparently and been peer reviewed. In any case, whatever information states might table in the Executive Council, suspicious events are likely to be detected, discussed, analysed and evaluated outside the Council. As has happened in past, for example when India and Pakistan conducted their nuclear tests in May 1998 and when the US government accused Russia of having conducted clandestine nuclear tests in August 1997, data from non-IMS stations will be distributed and discussed widely within the international scientific community.
While the CTBT does not explicitly ban nuclear test preparations, commercial satellite pictures can provide an important addition to the IMS because they can detect such preparations. Non-governmental organisations are already monitoring nuclear test sites using such data.9
The question is therefore not whether non-IMS capabilities can strengthen monitoring of compliance with the nuclear test ban but how this can be best achieved. Attempts to introduce new monitoring techniques would complicate the task of the PrepCom as it seeks to fully establish the verification system already provided for in the treaty. On the other hand, the treaty does commit states parties to 'cooperate with the Organization and with other States Parties in the improvement of the verification regime and in the examination of the verification potential of additional monitoring technologies…with a view to developing, when appropriate, specific measures to enhance the efficient and effective verification of this Treaty'.10 It specifically mentions electromagnetic pulse detection and satellite monitoring in this context.
ISSUES TO BE RESOLVED
On-Site Inspections Manual
In parallel to setting up the IMS, the PrepCom is also laying the groundwork for on-site inspections (OSIs). OSIs may be ordered by the Executive Council to clarify suspicious events?on the basis of signals detected by the IMS and /or information from NTMs submitted by a state party. The CTBTO will not, unlike the OPCW, have a standing OSI inspectorate, but will draw on a pool of trained inspectors nominated by member states. This pool needs to be geographically representative and large enough to provide a team of up to 40 inspectors within six days. Inspectors will require a diverse range of skills and the ability to work in harsh climates or terrain. By October 2000 one hundred and fifty participants from 69 states signatories had successfully undertaken three introductory courses conducted by the PTS.
OSI teams will be permitted to spend up to 130 days on an inspected state's territory and will therefore require significant in-country support. Substantial amounts of portable equipment will also be needed, including geophysical and radionuclide equipment, drilling equipment, communications equipment and the means to conduct overflights.
The development of an Operations Manual (OpsMan) for on-site inspections is proving one of the most difficult areas of the PrepCom's work, largely because too many fundamental issues were left unresolved by the treaty negotiators. While OSI provisions received insufficient attention during the early days, when establishing the IMS was the first priority, in November 1999 the PrepCom took steps to speed up the development of OSI procedures. The budget for developing an OSI capacity was doubled and a group of Friends of the OSI Programme Coordinator was established, open to participation by all signatories, to draft a text for an OSI manual.
This process faces several difficulties. First, there is no agreed understanding of the scope and the purpose of the manual. Israel, which is wary of intrusive OSIs for reasons unrelated to the CTBT, favours a minutely detailed manual which explains the purpose, methodology and parameters of the activities to be undertaken by inspectors. Others, including the United States, prefer a manual that outlines general responsibilities of the inspectors, but leaves room for flexibility and is within the spirit of the treaty's OSI provisions. Specific questions which need to be resolved include: should an inspection team only be allowed to look for evidence relating to an ambiguous event, or should it be allowed to 'look around' the inspection area? What kind of data should it be allowed to collect? What kind of managed access provisions are necessary in case the team needs to access or inspect a facility? What are the rules for inspecting so-called restricted access areas?11
Decisions about equipment to be used by inspectors are also affected by disagreement about the nature of OSIs. Thus, it has been argued that inspection equipment must not reveal information irrelevant to the inspection's purpose. This has made it more difficult to use off-the-shelf technology for OSIs, for example to analyse samples for radionuclide traces.
Second, the development of the manual has until very recently depended on national contributions, since the PTS was not allowed to propose language. Fortunately, this is no longer the case and the PTS has contributed several working papers to the OpsMan draft.
Third, the current drafting method, in which national contributions are simply compiled into a rolling text that now amounts to 1,000 pages, is too slow and ineffective. Working Group B has initiated a new process intended to speed up the drafting. Based on contributions received so far, the Chairman of the Friends of the OSI Programme Coordinator will compile a draft Rolling Text for the OpsMan with the assistance of others. Discussions on the basis of this Rolling Text will be initiated at a special meeting of Working Group B in February 2001. More time in the intersessional periods has been allocated to the drafting process and it is hoped that more delegations will be drawn into the drafting process at the February meeting.
Finally, there is a danger of linkage between completion of the OSI manual and entry into force of the treaty. At least one state whose ratification is required for entry into force has indicated that it might not ratify before substantial parts of the OSI arrangements have been agreed. It would be deplorable if disagreement over the least important element of the CTBT's verification system for day-to-day monitoring of compliance were to delay entry into force.
Clearly, any attempt to renegotiate the CTBT through the back door of negotiations on the OSI provisions is a cause for concern. A flexible mandate for inspectors, within clearly defined boundaries, will increase the chances that clandestine tests are identified. Experience in other verification regimes also shows that many of the fears articulated during treaty negotiations and the preparatory process prior to entry into force turn out to be exaggerated when regimes begin functioning. This is even true where intrusive inspections are conducted on a regular basis and affect private companies, as is the case with the 1993 Chemical Weapons Convention.
Confidentiality Rules
Because the IDC will soon start to distribute data and products to member states on a large scale – including daily 'Fused Event Bulletins', ad hoc event bulletins and analyses of data – the implementation of the 'confidentiality' provisions of the treaty has become a controversial issue in the PrepCom. The treaty itself provides only that it is the duty of the Technical Secretariat to 'make available all data, both raw and processed, and any reporting products, to all States Parties' (Article IV paragraph 14.e). It is unclear whether this excludes the possibility of making information available to others.
Scientific and humanitarian relief organisations are especially interested in receiving data from IMS stations. Data from the seismic network is of interest to seismologists in improving their ability to predict earthquakes and other natural phenomena. Hydroacoustic stations could give early warning of tsunamis, while infrasound stations could warn of volcanic eruptions.
China and other states have argued that for security reasons access should be restricted to governments. Some Western states and others favour a more open policy, arguing that IMS data has little national security relevance. It will in any case be difficult to prevent leakage of the data, since data centres in all CTBT parties will have direct access to it. In order to evaluate confidentiality rules, the PTS is planning a phased release of certain types of data to a limited number of non-state recipients. Thus, humanitarian organisations could promptly receive IMS data for disaster relief operations, while others would have only delayed access. The proposed test of a delayed release of certain types of IMS data beyond states parties' National Data Centres has not begun because of the continued resistance of at least one state party.
Costs
The 2000 PrepCom budget is $US 79.9 million, compared with $US 74.7 million in 1999 and $US 58.4 million in 1998. The collection rate for assessed contributions to the budget was approximately 96 per cent for the 1999 budget and 92 per cent for 2000. This is a good record compared with most international organisations, but needs to be maintained.
Even though the PrepCom's two Working Groups and the PrepCom itself still define the parameters of the work of the PTS, it is gradually becoming more independent. While the PTS has generally accepted responsibility for paying for the operating and maintenance costs of primary IMS stations after certification, a question remains as to whether it should wholly or partially pay the operating costs for auxiliary seismic stations after their certification. These stations are 'dual use' and mostly serve scientific purposes. They need to be certified by PTS, but will only report to the IDC on request (for example to clarify a suspicious event).
Some auxiliary stations are also nominated to back up primary stations in case they cannot report. If entry into force is substantially delayed, states might decide to switch off stations already reporting to the IDC to save costs. One way to resolve this question would be for the PTS to provide financial assistance to states having trouble funding their auxiliary stations.
Cost estimates for non-seismic stations made during negotiations on the CTBT have been consistently too low because expenditures for installation, especially in remote locations, were not taken into account and because little experience with novel monitoring technologies existed.
PrepCom 2000 Budget ($US 79.9 million) · $US 40.2 million for establishing or upgrading IMS stations · $US 12.6 million for the IDC · $US 7.3 million for establishing the global communications infrastructure · $US 2.8 million towards developing an on-site inspection capacity · $US 13 million on administration. Source: CTBTO PrepCom document CTBT/PC-10/1/Annex V '2000 Programme and Budget', Tenth Session, Vienna, 15-19 Nov. 1999.
In its 2001 budget projection the PTS has called for a moderate increase to $US 94.9 million. Some signatories have insisted on a smaller budget and it now seems more likely that the 2001 budget (which will be approved during the November 2000 session of the PrepCom) will be in the order of $US 85 million. While it is positive that there will be an increase in real terms compared to the 2000 budget, the fact that some member states have again insisted on cuts is worrying, for several reasons.
First, a static or shrinking PTS budget is hard to reconcile with the investment required to have the verification system fully functioning at entry into force. The PTS has estimated that the investments necessary for fully establishing the IMS require over $US 100 million per year for at least two years.
Second, budgetary pressures are likely to increase as more IMS stations are certified and the PTS assumes operational and maintenance costs for them. This will make it harder to maintain a rising level of investment in new stations, an important yardstick for assessing progress in completing the verification system.
Third, member states' insistence on budget cuts may signal reduced political support for the treaty and its verification system. It is especially worrisome if those states that are otherwise prominent in promoting the nuclear disarmament agenda insist on budget cuts. In some cases, attempts to micromanage the PTS budget have also led to problems by limiting the time available for discussions on the future of the PTS. In general the PTS should be given more leeway in managing its budget.
THE WAY AHEAD
According to Article IV of the CTBT, the IMS must be able to meet verification requirements at entry into force of the treaty, which will occur six months after all 44 states required to ratify the treaty have done so. From a legal perspective this requires that all three operational manuals-for the IMS, the IDC and OSIs-are ready for adoption by the first conference of states parties. In practical terms, it means that the verification system needs to be workable and have global coverage.
While the exact date of entry into force is unpredictable, the PTS is now planning on the basis that the IMS will need to be completed by 2005 at the latest. However, contingency plans exist should entry into force be achieved earlier. This will depend on the political support of signatories, as well as on their willingness to make the necessary technical and financial contributions.
References and Footnotes
1 Peter Basham, 'The Current Status of the CTBTO Monitoring System', Contribution to the Independent Commission on CTBT Verifiability, http://www.ctbtcommission.org/bashampaper.htm
2 CTBTO PrepCom document CTBT/PC-9/1/Annex III, 2 Sept. 1999, p. 8.
3 'Comprehensive Nuclear-Test-Ban Treaty - Four Years Old' Preparatory Commission of the Comprehensive Nuclear-Test-Ban Treaty Organization, Provisional Technical Secretariat, Press Release, 25 September 2000.
4 Peter Basham, 'The Current Status of the CTBTO Monitoring System', Contribution to the Independent Commission on CTBT Verifiability, http://www.ctbtcommission.org/bashampaper.htm
5 CTBTO PrepCom document CTBT/PC-10/1/Annex II, 24 Nov. 1999.
6 Peter Basham, 'The Current Status of the CTBTO Monitoring System', Contribution to the Independent Commission on CTBT Verifiability, http://www.ctbtcommission.org/bashampaper.htm.
7 Protocol to the CTBT, Part I, Section F, paragraph 18.
8 Protocol to the CTBT, Part I, Section F, paragraph 20.
9 The Institute for Science and International Security, for example, has published recent images of the Pakistani test site in the Kharan Desert on the internet. These can be found at http://www.isis-online.org/publications/southasia/satindex.html
10 CTBT, Article IV.11.
11 On these questions see Mordechai Melamud, 'Background Paper on On-Site Inspections (OSI) Main Elements and Expectations', http://www.ctbtcommission.org/melamudpaper.htm