Calibrated Seismic Verification of the Threshold Test Ban Treaty
Picard, Richard, Bryson, Maurice, Journal of the American Statistical Association
Since 1976 nuclear weapons tests conducted by the United States and Soviet Union have been limited by the Threshold Test Ban Treaty (TTBT) to a maximum yield of 150 kilotons (kt). This agreement aimed to restrain development of first-strike weapons. Verification was to be provided by "national technical means," based primarily on interpretation of seismic data (U.S. Dept. of State 1986). Monitoring procedures requiring on-site presence would have been difficult to negotiate at the time (Greb and Heckrotte 1983) and were not pursued.
The TTBT's original protocol provided for the United States and Soviet Union to exchange descriptions of their test areas' geologies as well as other data, including yields of "two nuclear weapon tests for calibration purposes from each geophysically distinct testing area where underground nuclear weapon tests have been and are to be conducted" (Dahlman and Israelson 1977, p. 367). In principle, such an information exchange would facilitate calibration of algorithms used to interpret seismic magnitudes; in practice, there is understandably some reluctance to accept unvalidated data. Largely because of verification concerns, the TTBT was not promptly ratified by the U.S., and the exchange did not take place.
An alternative approach to verification favored by the Reagan administration advocated using on-site instrumentation known as CORRTEX for all nuclear tests designed to exceed a specified yield level. Using this instrumentation, yield is inferred hydrodynamically through direct measurement of the explosion's shock front. Widely described (e.g., Reagan 1987) as having a 30% estimation error, CORRTEX became accepted as a de facto U.S. standard for effective TTBT verification. As a matter of official U.S. policy, seismic monitoring came to have less importance.
This policy appeared to change in 1989. Following a meeting of U.S. Secretary of State Baker and Soviet Foreign Minister Shevardnadze came the statement;
The Secretary and the Foreign Minister ... reached agreement to incorporate hydrodynamic and seismic monitoring, as well as on-site inspection, into the verification protocol for the 1974 Threshold Test Ban Treaty, as well as the levels above which these measurements would occur. In order to obtain a statistically significant number of data points to improve the national technical means of each side, each side will guarantee the other side the right to make on-site hydrodynamic yield measurements of at least two tests per year during the first five years following ratification of this treaty. After five years, each side shall guarantee one such hydrodynamic measurement a year thereafter ... (U.S. Arms Control and Disarmament Agency 1989).
Not only would on-site measurements verify TTBT compliance, but the resulting data base could aid in calibrating a seismic monitoring system.
A new protocol, completed in early 1990, called for hydrodynamic measurements of tests designed to exceed 50 kt and for the aforementioned two-test-per-year minimum. These provisions, and others permitting in-country seismic monitoring, allowed subsequent ratification of the TTBT.
More recently, an attempted military coup in the Soviet Union sought to change the course of world events, but failed. Not long thereafter, the "Soviet Union" ceased to exist and its TTBT responsibilities were assumed by the Russian republic, which sent scientists to the Nevada Test Site to exercise treaty rights and temporarily suspended its own nuclear testing. Still other events with implications for arms control and verification may have occurred between the time of this writing and the publication of this article.
The future role of seismic monitoring vis-a-vis on-site measurement is difficult to fully characterize. For example, it is conceivable that the developing concern with nuclear proliferation could expand the targets of a monitoring system beyond the United States and Russia. …