The anomalous groundwater mound and resulting flow regime associated with the Aardvark underground nuclear explosion have been examined. The initial anomalous groundwater mound was estimated to be about 200 m high and 250 m in radius with a central depression. The major mechanism for mound development was probably compaction of nearly saturated rock surrounding the explosion. Results of the calculations indicate that water first flowed into the collapsed zone, then radially outward with early velocities approximately 200 times preshot flow velocities. After about 400 days the magnitude of the anomalous flow had declined to or below the magnitude of the preshot groundwater flow rates. Ideally ‘traced tagged’ water particles initially near the edge of the collapsed zone are estimated to have moved outward about 6.2 m during the first 100 days after the explosion. It is shown that the anomalous mound does not significantly increase the migration of radionuclides from this explosion environment.

Til* rtport n« impend u «. account at OmnnuiA iptiaaoml wort. Haltbcr tb-t caltad •tabu, ear lb* Cooauwloa. nor ur ptraoB acting on bthalT of tfaa ComoUaaJen:A. lUktaroirvarrantr ornpfa*muilon,npreai^orlmiiI-dI>.ibrupKi toihaaccti' racj. compliLnni, or aacfalsua of (ha taAirmailMi caeuiaad 1B ibla report, or tint ttw aw of u; Ltbrutlm, apparauu, method, or prmra diKlntud In Ihl* report ma? oat Lnfrlnc* pHralalf gntd rlckU; or B. Aamnca tuj llibltf lltl with raipKl u Uw oat of, or lor dtouxra moiling from the OM of »j lafamalion, ipparatiia, nutted, or proem dlaclaaad U tfcla report.A* OMd In in* sbOT*, "paraoa nrtlaa: oo btnalf of tbc CouusualoB" loclafea «aj •mpfcijaa or contractor of Ika Coounlialon, or MSplayw of inch comncior, ID DM eaieni thai nek. onplajM or coatraetor of tit* Conotlailon, or tajAoft* of mch contractor praparu, dUttaUaam, or prorltSta KOM to. an? iDfbrouuloD parmai to kin •motojoMit or eoBlract wltk tka Cao.DL.nlcm. or

TU« rtptn m pnnroj M M KCCiat at Oo*f raaunt tfeumt^i "ork, Htllhtr UP uait*d •at**, w ito CcaalMkm. BC »J |»TM» lelh* <* htkilf ol tt» Caanltnoav A. Hik«»u]rw»ruijorNBiMfBliUaB,(xp T ttudor lmpUc«>«itb niptctWL'iaicmr«r. eoaai*W**M, or UH(U1M»I ol uw InJoriMttnt. coatalMd Is thji r*port, or uV.i a* ui DI H

niikw, or proctii 4j*dM:d In lh!i report mirnvt tnlrlafa prlnmjoMMd rtfblti or B, AHOMI uy lubllllU* wtu FMPMI (OIWUH ot, or for dtn*rMO«£tlB|o

The OGRE code was originally developed to evaluate the effects of nuclear stimu lation of oil and gas reservoirs. As a result, the mathematical model follows closely some of the two-dimensional models used by the petroleum industry. The nomenclature also follows that used by the industry but, since the processes being analyzed are the same as those which take place in the viscous flow of any fluid through porous media, the model can be used for a more general class of problems. Analysis is carried out by the solution of finite difference approximations to the differential equations describing the flow of one or two compressible fluids in the verti cal plane. The medium may be homogeneous or heterogeneous and, in the heterogeneous case, several standard geometries or a random distribution of permeability and porosity may be used. Like many working codes, OGRE is in a continuing process of modit'cation, and thus the presentation here represents its current form. increment less than or equal to At is chosen and then increased periodically as the problem progresses. If, of course, the second phase appears during the progress of the problem, the time increments must also satisfy Eq, (88). An additional constraint on the time increment exists whenever a problem includes a heterogeneous permeability distribution. In this case. and * n < At n < c T j,k **j,k J ' k V °j,k •JJL* 1 3,k V 2 j.k 3 j,k 4 j.k/ (89) (90) where B" _ **Ja l, k ' %k ^g •5 Z j.k ou. .k 8° e j.k J." < ^.-u n j.k 'j.k °i,k rf-^^.kK^k'^-J 2A°*£ °j.k For any problem, then, the time increment is the minimum value of at determined from all these criteria.

One promising application of engineering uses for nuclear explosives is the creation of large underground rubble chimneys into which liquid wastes can be injected. To assess the feasibility of this application, a predictive capability has been developed by matching laboratory model studies with mathematical analogs. The effects of varying permeability, porosity, input flow rates, and chimney dimensions were examined. Variation between laboratory and mathematical model results was less than 10%. This paper describes the laboratory model studies of two hydrologic environments, the development of the mathematical models, and the numerical solutions obtained with the mathematical models. Also summarized are two examples of potential applications of nuclear explosives to the field of waste disposal. The results indicate that, for many selected hydrologic conditions, it is feasible to use nuclear rubble chimneys in permeable but unsaturated formations for the disposal of liquid wastes. (