Silencing the Bomb Page 21

  ISRAELI AND SOUTH AFRICAN NUCLEAR COOPERATION

  Reed and Stillman claim, “After the Yom Kippur War of 1973, Israel and South Africa established closer nuclear ties, and Israel was assured of a uranium supply well into the future.” In 1997 South African officials announced that Israel had helped it develop nuclear weapons.

  On September 22, 1979, a U.S. satellite detected a “double flash” typical of a small nuclear explosion over the southern oceans. U.S. hydrophones on Ascension Island in the equatorial Atlantic, designed to be used for listening for underwater sound, picked up signals that were consistent with a nuclear explosion in the Indian Ocean on or near South Africa’s Prince Edward Islands. Seymour Hersh, an American investigative journalist, stated in 1991 that a flotilla of South African and Israeli military ships had been tracked by the U.S. National Security Agency to a site near the Islands. Some media reports, apparently based only on the satellite data, placed the event incorrectly in the South Atlantic.

  Many people in the U.S. weapons labs and the Defense Department concluded then, and continue to think, that it was a small nuclear explosion, perhaps of a neutron bomb that Israel thought it needed for close combat with tanks and Arab forces as occurred during the 1973 war. The double flash could not have occurred at a less opportune time for the Carter administration as the president was about to submit the second Strategic Arms Limitation Treaty (SALT II) to the Senate and to run for reelection on his success with nonproliferation. At the time, American foreign policy was in jeopardy in Iran following the overthrow of the shah and the capture of American hostages.

  Carter convened a panel of scientists, including Lamont’s William Donn, to examine classified data related to a possible small nuclear explosion. Their mandate, however, was to examine only technical data. Hersh says, “Our capturing it [the double flash] fortuitously was an embarrassment, a big political problem, and there were a lot of people who wanted to obscure the event.” The panel, which was chaired by Jack Ruina of MIT, concluded that the flashes probably were not caused by a nuclear explosion but perhaps by a micro-meteorite hitting the satellite. Others attributed it to a small nuclear explosion close to the ground by Israel in cooperation with South Africa. Nevertheless, the global monitoring system is certainly capable of detecting and identifying such an event today. That is one reason the International Monitoring System includes more than just seismic monitoring.

  Israel signed but has not ratified the Comprehensive Nuclear Test Ban Treaty (CTBT), nor has it signed the Nonproliferation Treaty. Israel does furnish data to the International Monitoring Center in Vienna, including information from a seismic array.

  SOUTH AFRICAN NUCLEAR PROGRAM

  South Africa developed first-generation nuclear weapons during the apartheid era. In the 1970s it became concerned about armed forces that opposed its regime in the former Portuguese colonies of Angola and Mozambique, in Southern Rhodesia (now Zimbabwe), and in SouthWest Africa (now Namibia). South Africa was further distressed by the introduction of Cuban troops into Angola and by warfare in SouthWest Africa. These forces, as well as embargoes against it, led South Africa to pursue a nuclear weapons program.

  Soviet satellites detected South Africa’s development of a nuclear test site in the Kalahari Desert in July 1977. They and the United States put pressure on South Africa not to test a crude, first-generation uranium weapon, which it did not do.

  Late in 1993 President F. W. de Klerk disclosed details about South Africa’s abandoning its nuclear weapons program in 1989 and dismantling six gun-type uranium 235 nuclear weapons, similar to the one dropped on Hiroshima in 1945. In is unknown whether those crude weapons would have worked, but many people assumed that they would. The government of de Klerk did not want those weapons to fall into the hands of a new government under Nelson Mandela. The International Atomic Energy Agency declared that the South African shafts in the Kalahari Desert had not been used for nuclear testing and, in 1993, that they had been rendered useless for nuclear tests.

  South Africa is the only country that has destroyed its inventory of nuclear weapons. Sweden actively planned to acquire nuclear weapons in 1945 but then halted those programs in 1968. On June 1, 1996, Ukraine became “nuclear free” after returning the last of its 1900 Soviet-era strategic nuclear weapons to the Russian Federation. All three countries went on to sign and ratify the Nonproliferation Treaty and the CTBT.

  MONITORING THE NEVADA TEST SITE

  Nettles and I also studied seismic events from 1992 to 2008 within 62 miles (100 km) of the Nevada Test Site (NTS). Focal mechanism solutions of the very long period type, measurements of Ms-mb, and data on high-frequency waves at regional stations indicate that all seismic events near NTS of magnitude greater than 3.3 could be identified as earthquakes. Unlike earthquakes near the Chinese and Pakistani test sites, shocks on and near NTS are quite shallow and hard to identify using just the seismic waves pP and sP (figure 3.1). Nevertheless, the three other methods sufficed for positive identification of them as earthquakes even though NTS is a region characterized by poor propagation of seismic P waves.

  IRAN AND THE MIDDLE EAST

  The 2012 National Academies report The Comprehensive Nuclear Test Ban Treaty examined the monitoring of the CTBT for Iran and other parts of the Middle East. Understandably, those countries are of major concern to many in terms of their possibly acquiring and testing nuclear weapons. Iran signed but has not ratified the CTBT. About a decade ago, it allowed the International Monitoring Service (IMS) to operate seismic stations on its territory. After the IMS certified them, however, data were no longer transmitted abroad even though the stations still exist.

  Major concern about Iranian nuclear intentions led me to devote particular attention recently to the geology and earthquakes of Iran. If Iran acquires nuclear weapons, Egypt, Saudi Arabia, Turkey, and some Gulf states may also decide to do so. Iran still has some centrifuges running that produce enriched uranium, claiming it enriches materials solely for the generation of nuclear power. Its reactors and the amounts of enriched uranium are subject to the agreement reached in 2015 by Iran, the five main nuclear powers, Germany, and the European Union.

  Since earthquakes occur often in Iran, distinguishing their seismic signals from those of underground nuclear explosions is of concern to many nations. Hence, I now describe the tectonic settings of Iranian earthquakes pertinent to their identification.

  Iran is largely a region of continental convergence caught between the Arabian and Eurasian plates. Similar deformation extends into northeastern Iraq, the Caucasus, eastern Turkey, and Turkmenistan. Compression causes the continental crust of Iran to be squeezed outward into its surrounding countries and the southern Caspian Sea. In contrast, the oceanic plate beneath the Arabian Sea is being subducted along the Makran plate boundary of southeastern Iran and southern Pakistan. Subduction is inhibited today elsewhere in Iran.

  The Arabian plate underthrusts the southwestern side of the Zagros Mountains along the Persian Gulf in southwestern Iran and northeastern Iraq. The crystalline rocks of the crust beneath the Zagros are part of the Arabian plate. Oceanic crust was subducted along the northeastern side of the Zagros about 80 million years ago. The suture zone and the Main Reverse Fault that remain from that former subduction are largely devoid of earthquakes. A second period of crustal shortening occurred in the Zagros during the last few million years. GPS measurements indicate that active crustal shortening is concentrated along the frontal, southwestern part of the Zagros belt.

  Iran has a long history of damaging and deadly earthquakes. In 1997 M. Berberian stated that earthquakes had killed 126,000 Iranians during the previous hundred years. In their 1982 book A History of Persian Earthquakes, N. N. Ambraseys and C. P. Melville list many shocks and the destruction of cities going back thousands of years. The Tabas earthquake (magnitude Mw 7.4) of September 1978 in east-central Iran and the Rudbar-Tarom earthquake (magnitude Mw 7.7) of June 1990 in northwestern Iran were the most catastrophic ear
thquakes in Iran during the twentieth century, killing more than 20,000 and 40,000 people, respectively. Financial losses from the June 1990 earthquake were about $7.2 billion, about 10 percent of Iran’s gross national product. About 30,000 deaths occurred in the moderate-size Bam earthquake of December 2003 (magnitude 6.6) in southeastern Iran. Poor construction and the shallow depths of those earthquakes contributed to the large loss of life and high damage.

  Large earthquakes (figure 14.3) do not occur randomly throughout Iran but are concentrated in the Alborz Mountains in the north, in northwestern Iran, along faults surrounding the otherwise nearly nonseismic Lut block between 27 and 34 degrees north and 57.5 to 60 degrees east in eastern Iran, and in the Kopet Dag Mountains along Iran’s northeastern border with Turkmenistan. Since 1918 most of the earthquakes along the Zagros Mountains have not exceeded magnitude 6.5, but many shocks of small to moderate size have occurred there. Central Iran, which is located between these zones of higher activity, has few large or moderate-size earthquakes.

  FIGURE 14.3

  Locations and magnitudes of large earthquakes (Mw>5.4) in Iran and surrounding regions from 1900 to 2009.

  Source: R. Engdahl, personal communication, 2013.

  Other large earthquakes have occurred in adjacent countries. The largest, of magnitude 8, occurred in 1945 off the coast of Pakistan, not Iran, along the Makran subduction zone (figure 14.3). Uplifted terraces along the coast in the Iranian part of Makran are indicative of past great shocks.

  Monitoring possible nuclear testing by Iran involves the identification of seismic signals of possible explosions as distinct from those of the many moderate-to-small earthquakes that occur every year. Accurate determination of depths is key to identifying many seismic events as earthquakes.

  In 2004 M. Tatar and colleagues obtained the most accurate depths of seismic events in the Zagros Mountains using local portable stations. The earthquakes they studied occurred within the upper 5 to 10 miles (8 to 16 km) of the crust; none were deeper than 20 miles (32 km). They report that these earthquakes are likely located in the upper part of crystalline crust below the very thick sedimentary layers of the Zagros. The thick Hormuz salt, which is about 430 to 600 million years old, is found at the base of those sediments and the top of the crystalline basement.

  This is fortunate for verification because most of the earthquakes they examined beneath the Zagros were much deeper than any nuclear explosions that could be detonated there. Hence, accurate determinations of depths using data from local seismic stations are exceedingly valuable for the identification of events as either explosions or earthquakes.

  Many sources of data on Iran’s earthquakes and geology are accessible to those interested in seismic verification. The great loss of life and destruction from past shocks led the Iranian government many decades ago to work on reducing earthquake losses through the operation of seismic networks, engineering of buildings to better withstand strong shaking, geologic studies, and mapping of active faults. Many papers on these topics, as well as on Iran’s petroleum resources and their geology, have been published over the past century.

  In 2009 Michael Pasyanos and colleagues made an extensive study of thousands of paths that seismic waves travel from earthquakes in the Middle East to tens of regional seismic stations. They also used data from stations and earthquakes within Iran. In 2013 Mark Fisk of Alliant Techsystems and Scott Phillips of the Los Alamos Lab studied hundred of thousands of paths traversed by four different types of seismic waves in Asia, Europe, and the Middle East. The two studies show in detail how well or how poorly seismic waves are detected at various frequencies for many seismic paths and how well events can be identified as being either explosions or earthquakes. Both studies provide a rationale for choosing data from seismic stations best suited to examining and identifying future seismic events.

  IRAQ

  The United States invaded Iraq in 2003. Government claims that Iraq possessed an active program to develop nuclear weapons turned out to be false. Another false accusation was that Iraq had obtained uranium from Niger in central Africa. France, however, gets uranium from Niger, a former colony, and carefully controls Niger’s export. The United States is said to have destroyed Iraq’s uranium (yellow cake) earlier, during the first Gulf War. Another false claim was that Iraq had conducted one or more decoupled nuclear explosions beneath a large lake in 1989. Iraq subsequently signed and ratified the CTBT.

  DEALING WITH PROBLEM SEISMIC EVENTS GLOBALLY: A SUMMARY

  The numbers and sizes of various problem or anomalous seismic events decreased dramatically from 1960 to mid-2009 (figures 14.4 and 14.5). It should be remembered that these are a tiny fraction of the earthquakes and chemical explosions that are reported every year. Most problem seismic events are small, occurring near the lower end of seismic detectability at the time.

  FIGURE 14.4

  Unidentified nuclear explosions, alleged explosions, unidentified seismic events, and identified nuclear explosions by India, Pakistan, and North Korea from 1997 to mid-2009. Seismic magnitude is at the left side and yield in kilotons (kt) at the right. Downward-pointing arrows indicate that events were equal in size or smaller.

  FIGURE 14.5

  Events not identified and problem (anomalous) seismic events identified as earthquakes through special studies: Eastern Kazakhstan, E Kaz; Kara Sea, Kara; Kola Peninsula, Kola; Nevada Test Site, NTS; Novaya Zemlya, NZ. CCD 72 are seismic events from a U.S. document tabled in 1972 at the UN’s Conference of the Committee on Disarmament.

  Source: Sykes and Nettles, unpublished poster, 2009.

  Figure 14.4 shows unidentified nuclear explosions, alleged nuclear explosions, and seismic events that were not positively identified initially, as well as recorded nuclear explosions by India, Pakistan, and North Korea from 1970 to mid-2009. Nuclear explosions that were not identified soon after they occurred are denoted by solid triangles. They decreased in size from about 8 kilotons in 1965 to about 0.3 kilotons in 1985. We found no unidentified nuclear explosions or any unidentified events larger than magnitude 2.5 (or 0.005 kilotons, 5 tons, if they were nuclear) after 1989.

  Solid squares indicate nuclear explosions by India, Pakistan, and North Korea. They are the only countries that tested after the Comprehensive Nuclear Test Ban Treaty was open for signature in September 1996. The three did not sign the treaty. India and Pakistan last tested in 1998.

  Seismic waves were not found, despite diligent searches, for one alleged nuclear explosion on September 19, 1989, in Iraq. That claim, which was publicized by dissidents, is almost certainly false. The seismic event in North Korea on May 22, 1989, which concerned some people in the United States, could not be identified at the time. There is no indication, however, that North Korea tested as early as 1989. Seismic waves were not detected for two suspect Russian tests on September 8 and 23, 1999, at Novaya Zemlya, as reported by Bill Gertz in the Washington Times. Noise measurements indicate that those events were not larger than magnitude 2.0 and 2.5. Both may have been permitted subcritical tests—that is, ones with no release of nuclear energy.

  Figure 14.5 shows problem or “anomalous” seismic events that were later identified as earthquakes through special studies. Their magnitudes decreased from 5.6 in the mid-1960s to 2.7 by 2009. If they had been nuclear explosions, which they were not, their detectability in terms of yield improved over time from about 30 kilotons to 0.003 kilotons (3 tons). Nettles and I identified the problem event of 2003 near China’s Lop Nor test site as an earthquake by five different techniques.

  Nettles and I also identified (not shown) several large mine collapses in various countries of magnitude 2.8 to 5. They are typically richer in long-period seismic waves than earthquakes, making their identification as distinct from underground nuclear explosions even easier. Mechanism solutions for some of those events also indicate that they were collapses and not explosions. Several very large chemical explosions were detected and identified as well.

  In summa
ry, the detection and identification of problem seismic events of several kinds—earthquakes, nuclear explosions, alleged nuclear explosions, mine collapses, and chemical explosions—have improved dramatically over the past five decades. A residuum of less than one problem event per year in countries of concern to the United States can be reduced further with special studies.

  15

  SENATE REJECTION OF THE CTBT IN 1999

  President Clinton signed the Comprehensive Nuclear Test Ban Treaty (CTBT) along with a number of other world leaders at the United Nations in September 1996. When it came up for a vote in 1999, the U.S. Senate failed to ratify the treaty. Nevertheless, it still remains on the Senate’s agenda. Ratifying a treaty requires the Senate to pass it by a two-thirds vote and the president to then sign it. Those countries that have signed but not ratified it are nonetheless still bound by the Vienna law of treaties to obey the terms of the CTBT. None of the signers has violated the terms of the CTBT by conducting nuclear tests since September 1996. This was a considerable accomplishment.

  ACTIONS PRIOR TO SENATE DEBATE

  Several developments in 1997 and 1998 preceded the Senate’s vote on the treaty. On September 22, 1997, President Clinton submitted the CTBT to the Senate for what is termed its advice and consent. He also included the following six safeguards with his submission:

  1. A Science-Based Stockpile Stewardship program to ensure a high level of confidence in the safety and reliability of U.S. nuclear weapons in the active stockpile, including the conduct of a broad range of effective and continuing experimental programs.

  2. The maintenance of modern nuclear laboratory facilities and programs in theoretical and exploratory nuclear technology that will attract, retain, and ensure the continued application of our human scientific resources to those programs on which continued progress in nuclear technology depends.

  3. The maintenance of the basic capability to resume nuclear test activities prohibited by the CTBT should the United States cease to be bound to adhere to this treaty.