DETECTION OF PAST UNDERGROUND NUCLEAR EVENTS BY GEOCHEMICAL SIGNATURES IN SOILS

This paper describes a research program to determine if a geochemical signature is retained in surface and sub-surface soils over a well-con tained underground nuclear explosion. Samples of soil at surface and a bout 10-15 cm below were collected on orthogonal traverses up to about 1 km from ground zero (GZ, surface projection of blast point) at thre e locations in central (Ville) and southern Yucca Flat (Laredo and Pre sidio) on the Nevada Test Site (NTS). They were analyzed for total ele ment content by instrumental neutron activation (INAA) and, using sele ctive leaches, for partial element content, namely that amount bound t o amorphous Fe oxyhydroxide and that amount bound to Mn oxide. Forty-t wo elements were determined in these leaches of both surface and sub-s urface soils, by inductively coupled plasma mass spectrometry (ICP-MS) , ICP emission spectrometry (ICP-ES) and hydride generation quartz tub e atomic absorption spectrometry (HG-QTAAS). These partial leaches wer e chosen on the premise that the extreme temperature and pressure cond itions created by an underground blast would volatilize elements in th e surrounding rock and transport them along with fission products to t he surface where they would be trapped by these phases in the soil. Th e INAA data generally show consistent concentrations of major, minor a nd trace elements within each location. In contrast, the selective lea ch data show distinct geochemical anomalies in numerous elements. Iodi ne, As, Mo, and U demonstrate particularly consistent positive anomali es at and around GZ at all three locations by both leaches. The amplit udes of these anomalies can exceed two orders of magnitude over backgr ound, particularly for I. This positive response is shown by more elem ents associated with Mn oxide than associated with Fe oxyhydroxide at all three locations; furthermore these patterns are more coherent and intense for some elements. The Mn oxide phase itself is more consisten tly distributed than is the Fe oxyhydroxide phase at the sites within each location. In most cases, significant depletion of Fe as amorphous Fe oxyhydroxide coincides with the multiple positive anomalies in thi s phase. The number and group of elements displaying distinct patterns is dependent upon the location as well as on the leach. The concentra tion ranges of elements extracted from the Fe oxyhydroxide or Mn oxide phase are similar for sites at Laredo and Presidio but dissimilar fro m those at Ville, probably a reflection of the differing geological se ttings. Whereas some anomalies are spatially quite compact, extending only about 50 m from GZ (e.g. Rb, Cs, Th), others such as I and Sb ext end up to 300 m. The geochemical behaviour of the surface soil samples , although slightly noisier, is similar to that of the sub-surface sui te. These results indicate that analysis of specific geochemical phase s in soils over a suspected underground nuclear test has an important role in the on-site inspection component of the verification of the Co mprehensive Test Ban Treaty. (C) 1997 Elsevier Science B.V.