Geostatistical Analysis of Spatial Variability of Mineral Abundance and Kd in Frenchman Flat, NTS, Alluvium

Carle, S F ; Zavarin, M ; Pawloski, G A

Lawrence Livermore National Laboratory

关键词: Lithology;    Geology;    Source Terms;    Clinoptilolite;    Petrography;   

DOI  :  10.2172/15004032
RP-ID  :  UCRL-ID-150200
RP-ID  :  W-7405-ENG-48
RP-ID  :  15004032

美国|英语

来源: UNT Digital Library

【 摘 要 】

LLNL hydrologic source term modeling at the Cambric site (Pawloski et al., 2000) showed that retardation of radionuclide transport is sensitive to the distribution and amount of radionuclide sorbing minerals. While all mineralogic information available near the Cambric site was used in these early simulations (11 mineral abundance analyses from UE-5n and 9 from RNM-l), these older data sets were qualitative in nature, with detection limits too high to accurately measure many of the important radionuclide sorbing minerals (e.g. iron oxide). Also, the sparse nature of the mineral abundance data permitted only a hypothetical description of the spatial distribution of radionuclide sorbing minerals. Yet, the modeling results predicted that the spatial distribution of sorbing minerals would strongly affect radionuclide transport. Clearly, additional data are needed to improve understanding of mineral abundances and their spatial distributions if model predictions in Frenchman Flat are to be defensible. This report evaluates new high-resolution quantitative X-Ray Diffraction (XRD) data on mineral distributions and their abundances from core samples recently collected from drill hole ER-5-4. The total of 94 samples from ER-5-4 were collected at various spacings to enable evaluation of spatial variability at a variety of spatial scales as small as 0.3 meters and up to hundreds of meters. Additional XRD analyses obtained from drillholes UE-Sn, ER-5-3, and U-11g-1 are used to augment evaluation of vertical spatial variability and permit some evaluation of lateral spatial variability. A total of 163 samples are evaluated. The overall goal of this study is to understand and characterize the spatial variation of sorbing minerals in Frenchman Flat alluvium using geostatistical techniques, with consideration for the potential impact on reactive transport of radionuclides. To achieve this goal requires an effort to ensure that plausible geostatistical models are used to characterize the spatial variation of minerals. The models must also be plausible from a geological perspective. Spatial variation of mineral abundance is caused not only by random processes but also by ordered geologic processes such as alluvial fan progradation, provenance, and in-situ alteration. Much of the spatial variation of sorbing mineral abundances is attributable to different ''chemofacies'' within which zonal mean abundances are relatively uniform. The chemofacies are, in large part, associated with the alluvial layers interpreted by Warren et al. (2002).