Barometric gas transport along faults and its application to nuclear test-ban monitoring

Underground nuclear explosions produce a unique but evanescent set of radionuclide gases that potentially can be used in the context of an on-site, test-ban monitoring program to differentiate them from other detected events such as earthquakes or mining activity. In Part I of this report we describe an experiment to evaluate the upward transport of gases from an underground explosion using two gas tracers with very different diffusivities that were released in a 400- m-deep, chemical explosive detonation. The less diffusive (more massive) tracer was detected on a nearby geologic fault 50 days following the detonation while the more diffusive tracer was-- detected 375 days after release. Computer simulations indicate that the arrival time and the chromatographic behavior of transport are characteristic of barometrically induced flow in a fractured, porous matrix regime. For a hypothetical 1-kiloton fission explosion subject to the same weather and gas transport conditions of the chemical explosion, simulations predict the delectability of argon-37 after 80 days in spite of depletion by radioactive decay. Largely because of the earlier arrival of xenon-133, owing to its lower binary gas diffusivity, the exceedingly short lived isotope should also be detectable-arriving about 30 days earlier than argon. in Part II we consider that our prediction of the delectability of argon and xenon is based upon the small volume (0.00001 M3) sampling technique of the NPE tracer-gas sampling study while actual sampling for radionuclides would involve drawing much larger volume (possibly 0.1- 1 M3) gas samples from the near-surface