A number of the dynamic effects expected from the explosion have been studied. Among them are the formation of the cavity and the subsequent growth of the chimney, the extent of both radial and vertical fractures, ground motion as a function of distance, and the expected concentration of radionuclides in the gas to be withdrawn from the chimney.

The final radius of the spherical cavity and the extent of material failure from the detonation center were predicted by a numerical technique that describes material response to a propagating stress field. This calculation makes use of the relevant equations of motion and depends on the proper equation of state of the material and on a unique description of its failure. For Gasbuggy, samples from the Lewis shale, the Pictured Cliffs sandstone, and the Fruitland coal were measured for static and dynamic compressibilities and failure characteristics. Assuming a yield of about 26 kt, calculations predict a cavity radius of 78 ft and laterial cracking in the Pictured Cliffs formation to 393 ft from the explosive. The same calculations show that the Fruitland coal tongue 334 ft above the explosion center may stop the cracks. The calculations cannot predict actual chimney height. Comparisons with past events show chimney heights to be within about 15% of calculated fracture limits. If the Fruitland coal can prevent the collapse of the over-burden material, then the Gasbuggy chimney may not extend above 330 ft from the energy source. Otherwise it could extend into the Fruitland.

Another study concerns the expected concentrations of radioactive gases from the Gasbuggy chimney. Preparations have been made to sample and analyze this gas. Calculations of expected concentrations have also been used to guide the post-shot sampling and drilling program and to prepare for any possible hazards in withdrawing from the chimney. In addition, a series of computations was performed to evaluate how the rapid flaring of chimney gas and its replacement by formation gas might affect the concentrations of various radionuclides. The results

suggest that rapid flaring could substantially decrease radionuclide concentrations in any gas produced afterwards.

131

131

While several radionuclides have been considered, major emphasis was placed on 1131, Kr3, and tritium, which are the ones most likely to cause problems. For the expected concentrations of I and Kr 85, it was assumed that all of these volatile species produced will end up in the chimney gas. For tritium, the equilibrium distribution of hydrogen between water and the permanent gases expected in the cavity was determined from a thermodynamic calculation. This distribution, as well as the ratio of H2 to CH4, was computed for 1700°C, where about 30% of the hydrogen is in the form of water. The results for 1131, Kr 85, and tritium (in the forms of HT and CH,T) are shown in Figure 12. The amount of tritium in the form of HTO is not included in this figure, since it would not be in the gas phase and hence could be easily removed from the product stream. 1131, with its 8-day half-life, decays rapidly, while both Kr 8 and tritium, whose half-lives are 10.6 and 12.6 years, respectively, show no such decay. Figure 12 also shows the decrease in concentration that is expected to follow the rapid flaring of three chimney volumes of gas.

85

The concentrations of tritium and Kr85 in the combustion products after burning of the gas were also considered. Burners generally dilute each volume of gas by about 200 volumes of air to keep CO and CO, to permissible levels, and a similar reduction (ignoring any pipe-line dilution) would take place in any radionuclides in the gas.

Other studies have yielded expected values of ground motion from the Gasbuggy explosion. These studies were made not only from the consideration of public safety but also as a guide for setting up the surface seismic instrumentation. Ground-motion predictions were based on extensive data from underground detonations at the Nevada Test Site in environments similar to the sandstones and shales at Gasbuggy. In general, surface motion on soil or

Radioactive concentration - C/ft3 of chimney gas at NTP

poorly consolidated materials will be greater than on hard rock. Expected accelerations and velocities as functions of distance are shown in Figures 13 and 14. These curves were derived from data which show considerable scatter. In general, one standard deviation corresponds to a departure from these curves of a factor of 2.

10-1

The measurement program during and immediately following the Gasbuggy explosion will concern itself with subsurface measurements of the outgoing shock wave, surface motion measurements, and a number of safety-related measurements such as precautionary monitoring for radioactivity. The subsurface and surface measurements are being made to establish the validity of the predictive calculations and to aid in the interpretation of the results. Shock instrumentation will be installed in holes GB-1, GB-D, and GB-E. Measurements in GB-1 and GB-E are primarily designed to study the outgoing shock wave, the generation of fractures, and the collapse of the

chimney. Quantities to be measured include peak shock pressure, times of shock arrivals, fracture radii, and the history of chimney collapse. GB-D will be instrumented with acceleration and velocity gages, primarily to study the generation of the seismic wave. The nature and location of the subsurface instrumentation is summarized in Fig. 15. Surface motions are planned to be measured at 47 locations, from a point 100 ft from the collar of GB-E to the rims of the San Juan Basin. Twelve of these stations are within five miles of surface ground zero, while six will measure the response of the Navajo Dam and the El Vado Dam.

The nuclear explosive will be placed in a cylindrical canister 17 in. in diameter and 13 ft long. Part of the space in this canister is occupied by a cooling module designed to keep the temperature inside the canister below 100°F. The canister and cables are designed and tested to withstand an external pressure of over 2000 psi. When all electrical and mechanical systems have been checked, the top of the explosive canister will be attached to a 7-in. drill casing and lowered to the bottom of GB-E by a drill rig. The cables from the explosive and the shock instrumentation will be attached to this casing as it is lowered. After the lowering operation is complete, cement will be pumped into the hole to a height of 1250 ft

above the canister. The balance of the hole, except for the top 50 ft, will be filled with sand. A liquid polymer that sets to a consistency of vulcanized rubber will then be pumped into the remaining space. When everything is in readiness for the det onation, an area out to a radius of 2-1/2 miles from GB-E will be cleared of personnel, and the final electrical connections will be made. An automatic 15min countdown sequence will then be started from the timing and firing trailer at the Control Point. During this sequence a series of electrical signals will energize the instruments and start the recording systems in the recording trailers. The final signal will detonate the explosive.