Fault-zone deformation in welded tuffs at Yucca Mountain, Nevada, USA

Abstract

Field and microstructural analyses of faults with millimeters to hundreds of meters of displacement within welded tuffs at Yucca Mountain, Nevada, has led us to identify four different fault zone architectures. We designate these fault zones as Classes A–D. We conclude that Classes A, C, and D are genetically related and that observed differences in their morphology and deformation mechanisms are related to differences in displacement magnitude and changes in the rheology of faulted rocks. We show that Class A, C, and D fault zones were formed by progressive cataclasis, leading to the development of a foliated gouge core and wide damage zone at the highest displacement magnitudes. By inference, displacement variations across individual high displacement faults are expected to produce significant lateral changes in fault zone architecture. The broad range of fault rock textures and architectures found on the Class A, C, and D continuum may produce important corresponding variability in hydrologic properties in ignimbrites.

By contrast, Class B fault zones have unique mineralogies and microtextures relative to other Yucca Mountain fault zones. Most notable is the occurrence of distinctive jigsaw puzzle texture within fault core breccias, which consists of angular wall rock fragments floating in a coarsely crystalline secondary calcite cement matrix. The calcite cement constitutes more than 65% of the Class B fault zones. Comparisons with other occurrences of secondary minerals within open lithophysal cavities, which appear to have formed in unsaturated conditions, indicate that the secondary mineralization history at Yucca Mountain was complex and likely polygenetic. The age and origin of the secondary calcite minerals in the Class B fault zones remain undetermined.

Introduction

Yucca Mountain, Nevada (Fig. 1a), the potential site for a high-level nuclear waste repository, consists of a ridge of welded and nonwelded pyroclastic flows and air-fall tuffs that erupted from the Southwest Nevada Volcanic Field calderas in the Late Miocene, between approximately 11 and 15 Ma (Sawyer et al., 1994). Tuffs exposed at Yucca Mountain include units of the Crater Flat and Paintbrush Groups as well as the Calico Hills Formation. The repository is currently planned to be housed within moderately to densely welded members of the Topopah Spring Tuff, which is one of the main tuff units of the Paintbrush Group.

Yucca Mountain lies within the southern Great Basin of the central Basin and Range Physiographic Province (Fig. 1b). It is part of the Crater Flat Basin, a half-graben bounded on the west by the east-dipping Bare Mountain fault and on the east by a series of mostly west-dipping faults that are antithetic to the Bare Mountain fault (Fig. 1c). Within this structural framework, Yucca Mountain is a tilted fault block that dips approximately 5–10° to the east (Fig. 2a and b). Yucca Mountain is bounded on the west by the Solitario Canyon fault and on the east by the Bow Ridge and Paintbrush Canyon faults (Potter et al., 2004) (Fig. 2b). Most faults within the Yucca Mountain block are steeply dipping normal faults; however, oblique slip, strike-slip, high-angle reverse, and low-angle thrust faults are also present. Growth strata in the Paintbrush Group adjacent to some of the faults indicate faulting initiated before or during tuff deposition (e.g. Day et al., 1998). Faulting activity was most prominent in the Late Miocene between 11 and 12 Ma (Fridrich et al., 1999). Faulting has remained active to the present, as indicated by offset Quaternary deposits across many large faults at Yucca Mountain (e.g. Stepp et al., 2000) and historic seismicity (e.g. Rogers et al., 1991).

The US Code of Federal Regulations at 10 CFR Part 63 requires the potential Yucca Mountain repository to safely contain high-level radioactive waste. The unprecedented long design life of the facility and the technical challenge of permanently disposing high-level radioactive waste in a geologic repository have prompted the US Department of Energy (DOE) to develop a 10.4-km-long network of tunnels at a maximum depth of 300 m within Yucca Mountain. The Exploratory Studies Facility (ESF) is a 7.6-m-diameter, 7.8-km-long tunnel from which the 6.1-m-diameter, 2.6-km-long Enhanced Characterization of the Repository Block or Cross Drift tunnel was bored (CRWMS M&O, 1998). Miocene volcanic tuffs of Yucca Mountain have sufficient strength to allow nearly continuous three-dimensional exposures of rock (CRWMS M&O, 1997).

Our analyses of faults and fault zones in the ESF and Cross Drift were initiated in 1998, mainly to refine models of faulting used in the US Nuclear Regulatory Commission (NRC) repository performance assessment codes. Such models help evaluate uncertainties in the performance estimates from potential direct disruption of the emplaced waste by fault movements. Accurate models of faulting require a detailed description and understanding of faults at the repository host–horizon level (Gray et al., 1998, Gray et al., 1999). Results of our fault studies from the ESF and Cross Drift also bear on other components potentially affecting repository performance. In particular, our study provides important constraints for hydrologic models and interpretations of thermochronology at Yucca Mountain (Gray et al., 2000).

The ages, origins, and significance of secondary mineralization at Yucca Mountain have been the subject of debate for more than a decade. Much of the most recent research has centered on the textures, isotope geochemistry and fluid inclusion microthermometry in mineral occurrences within lithophysae and fractures (e.g. Paces et al., 2001, Whelan et al., 2002, Wilson et al., 2003a, Wilson et al., 2003b). Based on these occurrences, Paces et al. (2001) concludes that the repository horizon has been within the vadose zone since deposition of the tuffs. They point out that most lithophysal cavities are unfilled or only partially filled with secondary minerals and that mineralization is generally restricted to the floors of the cavities. Strontium, oxygen, and carbon isotopic signatures have been interpreted by these researchers to indicate that the source of calcium carbonate was meteoric water from above throughout the postdepositional history of the tuffs (Whelan and Moscati, 1998, Paces et al., 2001, Whelan et al., 2002, Wilson et al., 2003a, Wilson et al., 2003b). A comprehensive fluid inclusion microthermometry study on samples collected in the ESF and Cross Drift yielded homogenization temperatures (equivalent to trapping temperatures in this case) for two-phase primary fluid inclusion assemblages within calcite averaging between 45 and 60 °C and reaching a maximum of 83 °C. In contrast, Dublyansky et al. (2001) conclude that at least some of the secondary minerals found in lithophysal cavities were precipitated from water with elevated temperature derived from below Yucca Mountain. They suggest that warm water invaded the repository horizon several times in the past and potentially in the recent past. Dublyansky et al. (2001) pointed out that two-phase fluid inclusion assemblages, some containing evidence of hydrocarbons, are present in calcite. Our observations of fault-related mineralization must be examined in light of existing conflicting interpretations of the history and significance of secondary mineralization at Yucca Mountain. Although our results do not resolve the ongoing debate, they provide additional information that should be addressed and incorporated in models of secondary mineralization at Yucca Mountain.

Our approach for gaining a better understanding of faulting was to examine, in detail, the fault rocks that comprise the fault zones. Fault-zone deformation in the upper crust produces a wide variety of morphologies indicative of the conditions and history of faulting. Idealized faults consist of two textural zones, a fault core and a damage zone (Caine et al., 1996, Seront et al., 1998). The fault core is a zone of relatively high strain that typically accommodates most of the fault displacement by shear in gouge, cataclasite, breccia, or mylonite. The surrounding damage zone is less deformed, accommodates less displacement, and contains subsidiary structures such as joints, veins, and minor faults. The dimensions of a fault zone are typically expressed in terms of thickness (measured normal to the fault zone boundaries) or width (measured horizontally between fault zone boundaries).

Deformation mechanisms govern the behavior, textural morphology, and ultimately, the gross architecture of fault zones. At Yucca Mountain, the protolith (welded volcanic tuff) has undergone brittle deformation dominated by cataclasis at shallow levels in the upper crust. Changes in deformation mechanisms with time and increased fault displacement, the presence or absence of fluids in the fault zone, mineral transformations, and syndeformational mineralization affect the rheology of the fault zone, causing the active parts of the faults to widen or narrow with increasing displacement. Two end-member possibilities exist (Wojtal and Mitra, 1986), (1) deformation produces fault rocks that inhibit further slip in the fault core (i.e. strain hardening) so additional fault displacement causes the protolith to fracture and the fault zone to widen with time, and (2) the fault rocks become progressively easier to deform (i.e. strain localization or strain softening) so deformation is localized to a progressively narrower portion of the fault zone. In strain softening, the intensely deformed fault core incrementally accommodates more of the deformation, and there is no further increase in fault-zone width. Investigations of faulting at Yucca Mountain, especially studies of fault zones exposed in the ESF, reveal that all of these deformation processes and related features are present, and subsets of fault zones with common architecture may be defined on the basis of this approach.

In this paper, we present results from detailed observations of fault zones exposed within Yucca Mountain. In particular, we present our findings of four different fault zone architectures, which we denote as Classes A–D. Based on these observations, we interpret the genetic relationships between different fault zone types. The ramifications of these findings are then discussed in light of ongoing studies of secondary mineralization, paleohydrology, and thermochronology at Yucca Mountain.