办事指南

Technology: Testing time for a total test ban

点击量:   时间:2019-02-28 03:14:05

By DANIEL CLERY Next week representatives from more than 115 countries will meet at the United Nations in New York to discuss extending the partial ban on the testing of nuclear weapons to a total ban. The US and Britain oppose amending the current ban because they believe that testing weapons is essential to maintaining a credible nuclear deterrent, and because they do not think a comprehensive ban could be adequately verified. But there is mounting scientific evidence that verification is achievable. The Partial Test Ban Treaty limits its signatories to testing weapons underground but contains a commitment to work towards a total ban. Many of the non-nuclear signatories wish to amend the treaty to a total ban but three nuclear members, the US, USSR and Britain, have a veto on changes to the treaty. So it is important for the advocates of a total ban to show that verification is possible (see ‘Iraq: nuclear secrets or rumours?’, this issue). A considerable amount of research has been carried out over the last few years into the feasibility of verification of a total ban. It would be virtually impossible to identify every conceivable nuclear explosion but identification may be possible down to yields that are not militarily useful or where disguising the blast would be prohibitively expensive. Detonations could be disguised behind large conventional explosions or earthquakes, or carried out in large underground cavities so that the shock wave is ‘decoupled’ from the Earth’s surface and is fainter and harder to detect. Detecting such faint seismic signals is difficult if the monitoring stations are thousands of kilometres away from the test site because so-called teleseismic waves, which travel deep through the Earth’s interior, become scattered. The solution is to put monitoring stations nearer the test site. One such station was set up in the Soviet Union by the British Seismic Verification Research Project at Garm near the Afghan border, 1,400 kilometres from the Soviet test site at Semipalatinsk. The experiment, carried out during 1989, used a single, commercially available seismometer. Everything they detected was identified says Roger Clark, a geophysicist from the University of leeds, including 12 announced nuclear explosions and one unannounced test at the Soviet’s arctic test site at Novaya Zemlya, confirmed by a Norwegian monitoring station. These results were achieved despite the fact that it was only a single station and the site was far from ideal, says Clark. Their results give a definite stamp on current capability of in-country monitoring, states Clark. ‘Now we have the data,’ he says, identification of explosions down to the kilotonne range ‘will be routinely possible in the near future.’ The beginnings of a network of seismic stations has been set up in the USSR by a consortium of US universities and colleges involved in geophysics known as IRIS. In 1984, IRIS began to set up a global network of seismological stations for geophysical research. In 1988, IRIS installed six stations in the USSR with the support of the US congress which was interested in studying possibilities for seismic verification. Unlike the British experiment, IRIS stations have three-component seismometers which can measure shock waves in any direction. IRIS has also developed an electronic data logging system which converts seismic data into digital code. They hope soon to be able to collect data from the stations remotely through telephone lines. The consortium’s research have made two significant findings relevant to verification says Paul Richards, a professor of geological sciences at Columbia University in New York. First that high-frequency seismic waves in the range of 30 hertz, which are important in the identification of explosions, propagate efficiently in the geological structure of the USSR. Also that within a continent, waves that travel deep within the Earth suffer less from scatter. This would enable verification monitoring to obtain a better estimate of an explosion’s yield. But verification need not depend on seismic monitoring alone. Recent research carried out by the Verification Technology Information Centre (Vertic) in London, believes it would be possible to verify a total ban by relying on more than one form of detection. Violations detected by one system could be checked by a combination of other techniques. Vertic’s proposal includes four elements: a worldwide network of seismic monitoring stations, direct observation of possible test sites using remote-sensing satellites, a network of sensors to monitor radioactive material in the air that may have leaked from underground tests, and an agreement for one-site inspections both on a routine basis and special inspections when violations are suspected. It also advocates setting up a central agency, similar to the International Atomic Energy Authority, to coordinate the collection and analysis of data. Vertic predicts that if current off-the-shelf seismic technology were used, a total of 1347 stations would be required – with 130 in the USA and 273 in the USSR – at a cost of nearly $340 million. The pressure group Parliamentarian Global Action, in New York, has been working with the six non-nuclear signatories who proposed amending the test ban treaty to draw up a draft protocol for a verification regime. Its scheme is cheaper than the one proposed by Vertic and has fewer stations because it subjects the nuclear power to greater scrutiny than other nations. Any nation that has conducted more than one nuclear test would be required to have a seismic network capable of detecting tests down to the equivalent of five tonnes of TNT. Any other country – including India which has conducted one test, reputedly for ‘peaceful’ purposes – need only have verification for tests greater than 500 tonnes. PGA argues that a country seeking to evade the verification regime and conduct its first nuclear test will not have the technology to successfully decouple their test. Advanced nuclear nations on the other hand could do this so a more sensitive network is needed. This regime would require a maximum of 127 off-the-shelf seismometers for non-nuclear countries along with 59 more sensitive purpose-built sensors for the nuclear powers and 33 for international territory. It estimates this would cost a total of $150 million for the equipment alone. The regime also requires atmospheric monitoring, portable seismometers to investigate suspected violations in more detail,