Strategic Defense Initiative:
Estimates of Brilliant Pebbles' Effectiveness Are Based on Many Unproven Assumptions
NSIAD-92-91: Published: Mar 27, 1992. Publicly Released: Mar 27, 1992.
- Full Report:
Pursuant to a congressional request, GAO reviewed the Strategic Defense Initiative Organization's (SDIO) assessment of the effectiveness of Brilliant Pebbles, a proposed space-based weapon for the Global Protection Against Limited Strikes (GPALS) Strategic Defense System.
GAO found that: (1) as of December 1991, SDIO had not evaluated through computer simulations Brilliant Pebbles' performance against a variety of interceptors deployed against certain hypothetical ballistic missile attacks; (2) to assess effectiveness, SDIO has identified over 40 hypothetical attack scenarios or threats against the United States and its allies, which include short, intermediate, and long-range ballistic missile attacks originating from all over the world and submarine-launched attacks against the United States; (3) Brilliant Pebbles could not intercept missiles with ranges of less than 400 to 600 kilometers or those with altitudes less than 80 to 100 kilometers which means that Brilliant Pebbles could not attack some missiles currently owned by third world countries; (4) the number of weapons deployed and the angle at which the rings of weapons cross the equator have a direct effect on the performance of the Brilliant Pebbles constellation against a given threat; (5) SDIO has performed simulations in which the number of weapons in orbit varies from a few hundred to over a thousand, which indicates that the effectiveness of a hypothetical Brilliant Pebbles constellation improves as the number of weapons in orbit increases; (6) the simulations SDIO uses to estimate the effectiveness of Brilliant Pebbles are relatively immature and use too many unproven assumptions about the performance and operation of the constellation; and (7) as Brilliant Pebbles development progresses, SDIO plans to replace early assumptions about Brilliant Pebbles' performance with data obtained through testing and continued simulation.