1. INTRODUCTION
To evaluate the postclosure performance of a potential repository at Yucca Mountain, a Total System Performance Assessment (TSPA) will be conducted. Nine documents called Process Model Reports (PMRs), of which this document is one, have been developed to support the TSPA for Site Recommendation (TSPA-SR). TSPA is an ongoing iterative activity at the Yucca Mountain Site Characterization Project (YMP). The nine PMRs that support TSPA-SR discuss the following topics:1.1 Objectives
All PMRs have the shared objective of documenting the necessary and sufficient technical information that the YMP will rely upon to make its site suitability evaluation and potential licensing argument. Specific reports cover designated technical topics and are "stand alone" reports. The purpose, objectives, and scope of this Disruptive Events PMR are contained in the associated technical product development plan (CRWMS M&O 2000d) and are described below. Objectives for this Disruptive Events PMR include summarizing the results of the supporting analyses and the approach to and results of FEPs screening for disruptive events; providing historical information on disruptive events analyses; and discussing how information contained in the report, or the associated AMRs, addresses issues raised by the NRC and other oversight groups (see Section 3.3). The report provides the overview framework for why the AMRs for disruptive events were initiated and where and how the results were used, including their uses in the TSPA-SR. This Disruptive Events PMR contains discussion of the treatment of disruptive events in previous TSPAs to support traceability of the history of this analysis. The report documents the exchange of information between different organizations that ensures consistency of approach between the analyses within this Disruptive Events PMR and those performed for similar events by other organizations, especially those analyzing preclosure EBSs and WPs. The report enhances defensibility, traceability, and transparency of the supporting analyses and calculations by placing them in context with each other and other PMR analyses. An objective of the report is to clarify the bases of project comments on specific NRC KTIs and acceptance criteria. Also documented is consideration of alternative conceptual models proposed by the NRC and other oversight groups and by non-project researchers who developed new information for consideration since completion of the two expert elicitation projects: the PSHA and PVHA.1.2 Scope
This document summarizes information from the following activities and provides roadmapping information linking the analyses to each other and to key issues or Project requirements identified below. The scope includes the following:1.3 Quality Assurance for Disruptive Events Analyses and the Disruptive Events PROCESS MODEL REPORT
Pursuant to evaluations performed in accordance with QAP-2--0, Conduct of Activities, it was determined that activities supporting development of this Disruptive Events PMR and its documentation were quality affecting activities subject to the QA requirements of the Quality Assurance Requirements and Description (DOE 2000). The Disruptive Events PMR was prepared according to the associated technical development plan (CRWMS M&O 2000d). This Disruptive Events PMR complies with DOE Interim Guidance (Dyer 1999). This Disruptive Events PMR was prepared in accordance with AP-3.11Q, Technical Reports, and reviewed in accordance with AP-2.14Q, Review of Technical Products. The QA procedures under which the supporting AMRs and one calculation were prepared are described in the AMRs and calculation and their respective planning documents. The primary procedure under which the AMRs were prepared is AP-3.10Q, Analyses and Models, and the procedure for the calculation was AP-3.12Q, Calculations. Data used in those AMRs were qualified in accordance with AP-SIII.2Q, Qualification of Unqualified Data and the Documentation of Rationale for Accepted Data. Information used in this report has been managed, and the quality status of it tracked, in accordance with AP-3.15Q, Managing Technical Product Inputs. The key software codes used in the analyses that supported this report are listed below; they were managed in accordance with AP-SI.1Q, Software Management.1.4 Relationship of Disruptive Events PROCESS MODEL REPORT to Work Under Other Process Model Reports and Key Project Documents
As stated in Section 1.1, this Disruptive Events PMR is one of several upper level documents (the PMRs) that summarize the analyses, models, and calculations that contribute to the TSPA-SR. The relationship between this Disruptive Events PMR, other PMRs from which data were received, the TSPA-SR, SR, and License Application (LA) is shown in Figure 1-1. This Disruptive Events PMR directly supports the development of descriptive material needed for SR and LA and also supports the development of TSPA calculations, which are needed to evaluate the postclosure performance of the potential repository. This report describes how various site characterization activities are used in the disruptive events analysis (Section 2.1). These documents included the Site Characterization Plan (DOE 1988) and the Yucca Mountain Site Description (CRWMS M&O 1998a, c, d, e, f, g, h). Two YMP expert elicitations that produced hazard analyses for volcanism, ground motion, and fault displacement also contain evaluations of the geologic framework and FEPs that are characteristic of the site (CRWMS M&O 1996; Wong and Stepp 1998). Chapter 2 also describes the role of previous TSPAs in shaping the type of disruptive events analysis that was performed for TSPA-SR. The TSPA documents are listed in Section 2.1. As shown in Figure 1-1, TSPA-SR analysis required SSC consequence information to support disruptive events analyses that was, in large part, provided by data and analyses from other PMRs. The AMR Miscellaneous Waste Form FEPs Screening Arguments (CRWMS M&O 2000o), which supported the Waste Form Degradation Processs Model Report (CRWMS M&O 2000t), contains an analysis that provides waste particle size information to support volcanic eruption analysis in the disruptive events AMR Igneous Consequence Modeling for TSPA-SR (CRWMS M&O 2000l). The Waste Package Degradation Process Model Report (CRWMS M&O 2000u) provided information, through the calculation Waste Package Behavior in Magma (CRWMS M&O 1999b), on the behavior of a waste package (WP) in the thermal environment caused by magma in an emplacement drift. TSPA-SR will require inputs from other PMRs for analyses downstream of the Disruptive Events PMR analyses to support the final output for TSPA-SR. These inputs include the Biosphere PMR supporting analyses Disruptive Event Biosphere Dose Conversion Factors Analysis (CRWMS M&O 2000s) and Evaluate Soil/Radionuclide Removal by Erosion and Leaching (CRWMS M&O 2000m). The drip shield damage abstraction AMR EBS Radionuclide Transport Abstraction (CRWMS M&O 2000r) provided by the PA group will feed into TSPA-SR downstream of the disruptive events AMRs to support disruptive events analyses. Repository design information was provided by the Enhanced Design Alternative (EDA) II (CRWMS M&O 1999a) and TSPA-VA, Volume 2 (DOE 1998b).2. Previous Disruptive Events Work and TSPA Approach for Site Recommendation
This Disruptive Events PMR summarizes the results of analyses and one calculation that will support TSPA-SR. TSPA is a risk assessment that quantitatively estimates how the potential Yucca Mountain repository system will perform in the future under the influence of specific FEPs, incorporating uncertainty in the models and data (DOE 1998a, p. A-41). The purpose of TSPA is to2.1 Previous YUCCA MOUNTAIN SITE CHARACTERIZATION PROJECT Geologic Work Related to Disruptive Events
The analysis of disruptive events was based on the geologic framework developed from the intensive investigations conducted to characterize the geologic setting of the Yucca Mountain region. Site characterization studies have led to the development of the geologic framework described in the following subsections. It is through these studies that the geologic FEPs of importance to volcanism, ground motion, and fault displacement have been described. The site descriptions and AMRs contain the conceptual models of the processes related to volcanic and seismic hazards.2.1.1 Yucca Mountain Geologic Framework
This section provides a summary level discussion, based on past YMP work, of the regional setting, stratigraphy, and structural features that form the geologic framework of Yucca Mountain. Section 2.1.2 focuses on past geologic studies related to Yucca Mountain region volcanism, and Section 2.1.3 focuses on past geologic studies related to Yucca Mountain region seismicity and structural deformation. These three sections summarize the geologic picture for the Yucca Mountain region that has been developed and provides a foundation for disruptive events AMR analyses for TSPA-SR. A comprehensive description of the site geology is presented in the Yucca Mountain Site Description (CRWMS M&O 1998a, c, d, e, f, g, h) and is being updated (CRWMS M&O 1999i). The following discussion is based on the updated document unless otherwise noted. The Yucca Mountain site is located on the western boundary of the Nevada Test Site (NTS), where scientists have conducted geologic investigations since the 1950s. Studies related to nuclear waste disposal have focused on Yucca Mountain since the late 1970s and have included careful mapping of the rocks at the surface and the subsurface in more than 10 km (6 mi) of tunnels and drilling and logging of numerous wells and boreholes (CRWMS M&O 1998e, Section 3.1.3). The characterization of the geology of Yucca Mountain is nearing completion, and it provides the framework for understanding the natural processes important to assessment of disruptive events and the safety of the potential repository. Yucca Mountain is located in the Basin and Range tectonic province of the western United States, within the region known as the Great Basin (CRWMS M&O 1998e, Section 3.4.1.1). The Great Basin encompasses nearly all of Nevada and parts of Utah, Idaho, Oregon, and California. The Basin and Range draws its name from its characteristic, generally north-south aligned mountain ranges. These ranges are separated by basins containing thick deposits of sediment (mostly sand and gravel) derived from erosion of the adjacent ranges over millions of years. The tectonic structure of the Basin and Range has developed over a period of more than 30 million years. In southern Nevada, including Yucca Mountain, the pattern of mountains and valleys has been formed in the past 15 million years from the movement of faults on one or both sides of the ranges (Fridrich 1999). The highest rates of modern tectonic activity in the southwestern Great Basin (i.e., active faulting and volcanism) occurs to the south, west, and northwest of Yucca Mountain in a regional context (CRWMS M&O 1998e, Section 3.2.1). Among the most active areas are the Furnace Creek-Death Valley fault zone, the Sierra Nevada front (i.e., the Owens Valley and Mammoth Lakes area), and the area north of the Garlock fault in the Mojave Desert (CRWMS M&O 2000c, Figure 1). This domain includes modern basins and ranges with great structural relief, such as the Death Valley basin and the Panamint Range. Modern faulting and volcanic activity are caused by the continuation of the same tectonic extension that resulted in the formation of the entire Basin and Range. The crust on the western edge of the Great Basin (the Sierra Nevada) is gradually moving to the west relative to the eastern edge of the basin (the Wasatch Front in Utah).2.1.1.1 Yucca Mountain Regional Stratigraphy
The geologic system at Yucca Mountian forms a fundamental framework for understanding the performance of the site as a potential geologic repository for high-level nuclear waste (HLW). The exposed stratigraphic sequence at Yucca Mountain is dominated by mid-Tertiary volcanic rocks, consisting mostly of pyroclastic flow and fallout tephra deposits with minor lava flows and reworked materials (CRWMS M&O 1998e, Section 3.5.1). Rocks and sedimentary deposits exposed in the region surrounding Yucca Mountain range from Precambrian, or more than 570 million years old, to surficial Holocene deposits, or less than about 10,000 years old. However, with the exception of two limited areas, Calico Hills and Bare Mountain, surface outcrops in the potential repository site area range from Miocene to Recent (Day et al. 1998). Understanding the distribution of rock types is important because it enables geologists to understand the geologic history of the area, which is fundamental to analyses of geologic hazards such as seismic and volcanic risk. Rock types below and around Yucca Mountain influence the regional flow of groundwater and directly control the migration of any potential releases from the repository system. The stratigraphic sequence of volcanic rocks at Yucca Mountain is the result of two stages of regional volcanism, an early silicic and a later basaltic stage. Between about 15 and 7.5 million years ago, during the Miocene Epoch of the Cenozoic Era, a series of large-scale silicic volcanic eruptions resulted in the formation of the southwestern Nevada volcanic field (CRWMS M&O 1998e, Section 3.9), which consists of six major volcanic centers, or "calderas," in which Yucca Mountain is located. The Timber Mountain Caldera Complex, one of six major calderas in the southwestern Nevada volcanic field, includes the Claim Canyon Caldera located north of Yucca Mountain. The silicic caldera forming eruptions occurred during a period of intense tectonic activity associated with active faulting caused by rapid extension of the earth’s crust. The Claim Canyon Caldera was the probable eruptive source of the approximately 13-million-year-old rock units that now form the mountain ridges at the potential repository site. These eruptions, along with all of the silicic activity from the southwest Nevada volcanic field, ended over seven million years ago. Based on geology of similar systems in the Great Basin, it appears that the silicic volcanic cycle is complete and will not recur. Basaltic volcanism in the region began approximately 11 million years ago and has continued into the Quaternary period. The basaltic volcanic events were much smaller in magnitude and less explosive than those of the silicic episode. Two episodes of basaltic volcanism have occurred. An older episode of basaltic volcanism occurred between 9 and 7.2 million years ago, while a second one occurred between 4.7 and 0.075 million years ago. The more recent events consisted of small volume volcanoes, in the form of cinder cones with lava flows and volcanic ash, that erupted to the west and south of Yucca Mountain. Four cinder cones formed between about 1.17 and 0.77 million years ago in Crater Flat, west of Yucca Mountain. The latest volcanic episode, about 80,000 years ago, created the Lathrop Wells Cone, about 16 km (10 mi) south of the potential repository site. Additional detail on the Miocene to Quaternary volcanic history of the Yucca Mountain region is provided in CRWMS M&O 1998e (Section 3.9.3). Surficial deposits in the Yucca Mountain region provide a record of the evolution of surface processes and climate conditions over the past several hundred thousand years (CRWMS M&O 1998e, Section 3.4.3). Most surficial deposits are composed of sands and gravels, known as alluvium if they are deposited by flowing streams or as colluvium if they originate from hill slopes as flows of debris. Eolian deposits (wind-blown deposits, such as sand dunes) are generally a minor component of the surficial deposits in the region. The ages of surficial deposits range from less than 1,000 years to more than 760,000 years, but most deposits exposed at the surface were deposited during the last 100,000 years. Determining the ages and distributions of these deposits is important to understanding the age and movement of faults in the area.2.1.1.2 Yucca Mountain Site Stratigraphy
Yucca Mountain consists of successive layers of volcanic rocks that generally thin from north to south. These rocks are described in detail in the Yucca Mountain Site Description (CRWMS M&O 1998e, Section 3.5.3; stratigraphic unit ages are shown in Figure 3.5-1). Three volcanic tuff layers are present between the surface and the elevation of the potential repository: the Tiva Canyon welded tuff at the surface, the Topopah Spring welded tuff at the level of the potential repository, and an intervening nonwelded tuff. As a result of faulting over the last 13 million years, these layers are all tilted to the east about 10 degrees. Figure 2-1 shows these tilted volcanic tuffs. Most of the surface of Yucca Mountain above the potential repository location is composed of the Tiva Canyon Tuff of the Paintbrush Group. This unit is a large-volume, regionally extensive ash-flow tuff with a thickness that ranges from 50 to 175 m (165 to 575 ft). A layer of nonwelded tuff underlies the Tiva Canyon Tuff near the site of the potential repository. The nonwelded layer includes two separate ash flows, the Yucca Mountain Tuff and the Pah Canyon Tuff. In the vicinity of the potential repository the total thickness of the nonwelded units ranges from 30 to 50 m (100 to 165 ft). The lowermost unit in the Paintbrush Group is the Topopah Spring Tuff, which forms the host rock for the potential repository (CRWMS M&O 1998e, Section 3.5.3.7). The Topopah Spring Tuff was formed by an eruption about 12.8 million years ago and has a maximum thickness of about 380 m (1,250 ft) near Yucca Mountain. Based on surface mapping and studies of boreholes and underground exposures, the Topopah Spring Tuff has been subdivided into several lateral layers according to chemical composition, mineral content, the size and abundance of pumice and rock fragments, and other variations in texture and appearance. An important characteristic of the layers is the presence and abundance of lithophysae, which are bubble-like holes in the rock caused by volcanic gases that were trapped in the rock matrix as the ash-flow tuff cooled. The nature, size, and abundance of lithophysae in tuff may affect its thermal, mechanical, and hydrologic properties. The lower and middle portions of the Topopah Spring Tuff have been divided into four layers according to the amount of lithophysae they contain. Because these layers are tilted, and the drifts in the potential repository would be near-horizontal, the potential repository horizon crosses the lithophysal zones. Like the Tiva Canyon Tuff, the Topopah Spring Tuff is fractured throughout, and these fractures provide the main pathway for groundwater to flow through the rock unit. Beneath the Paintbrush Group, the Calico Hills Formation is a series of mostly nonwelded rhyolite tuffs and lavas that erupted approximately 12.9 million years ago. The formation thins southward across the potential repository site, from a total thickness of as much as 460 m (1,500 ft) to only about 15 m (50 ft) (CRWMS M&O 1998e, Section 3.5.3.6). The water table below the potential repository is located within the Calico Hills Formation. The geologic units below the water table contain volcanic rocks composed mainly of welded and nonwelded ash-flow tuffs of the Crater Flat Group and older undifferentiated Miocene volcanics. The volcanic rocks are underlain by Paleozoic limestones and dolomites. Although the older volcanic rocks and the Paleozoic rocks lie deep beneath the surface near Yucca Mountain, they are found at much shallower depths (and even at the surface) to the south, where they are an important component of the hydrologic flow system.2.1.1.3 Yucca Mountain Faulting and Local Structural Geology
The distribution and properties of faults and fractures in the volcanic bedrock are important elements of the structural geology of the potential repository at Yucca Mountain. The potential main repository emplacement area is bounded on the west by the Solitario Canyon fault and on the east by the Ghost Dance fault. No faults with significant displacement (more than a few meters) occur within the area defined for emplacement (Wong and Stepp 1998). Detailed studies of the faults within the emplacement area indicate that they are not active faults; thus they are considered to have an extremely low probability of being active in the future (CRWMS M&O 2000c, Section 6.3.2). The structural geology of Yucca Mountain is dominated by block-bounding faults spaced 1 to 4 km (0.6 to 2.5 mi.) apart. These faults include (from west to east) the Windy Wash, Fatigue Wash, Solitario Canyon, Bow Ridge, and Paintbrush Canyon faults (see Figure 2-2). The faults generally are steeply dipping, north-south striking normal faults, and typically exhibit some left-lateral displacement. Displacement between the block-bounding faults occurs along multiple smaller faults, which may intersect block-bounding faults at oblique angles. The Ghost Dance and Sundance faults are examples of smaller "intrablock" faults near the potential repository.2.1.1.4 Yucca Mountain Fracture Characteristics
The distribution and characteristics of fractures at Yucca Mountain are important, because in many of the hydrogeologic units at the site (particularly the welded tuffs) fractures are the dominant pathways for groundwater flow in both the UZ and SZ. The fracture systems play a major role in the performance of the potential repository. The following discussion was summarized primarily from Book 1, Section 3 of the Yucca Mountain Site Description (CRWMS M&O 1998e, Section 3.6.3). Fractures at Yucca Mountain are generally of three types: early cooling joints, later tectonic joints caused by faulting and rock stress, and joints caused by erosional unloading. At Yucca Mountain, cooling and tectonic joints have similar orientations but can be distinguished from each other because cooling joints are smoother. Cooling joints form two orthogonal (at 90° angles to each other) sets of steeply dipping fractures and, locally, a set of subhorizontal fractures. Four steeply dipping sets and one subhorizontal set of tectonic joints have been identified. In general, joint orientation is significant in disruptive events analyses; the relationship between the orientation of emplacement drifts and joint and fault orientations has an effect on rock fall. Joint orientation, which affects the size and number of key blocks, controls the rock sizes, shapes, and numbers that can fall (CRWMS M&O 2000f). Fracture density, connectivity, and hydraulic conductivity are highest in the densely welded tuffs and lowest in the nonwelded units. The Tiva Canyon and Topopah Spring welded units are characterized by well-connected fracture networks, whereas the Paintbrush nonwelded units and the Calico Hills tuffs generally do not exhibit connected fractures. Additionally, the non-lithophysal welded units tend to have fractures with longer trace lengths, while the units with higher lithophysal content tend to have fractures with shorter trace lengths. It is reasoned that the presence of lithophysae inhibits the propagation of fractures (Sweetkind et al. 1997, pp. 61-66). In all units, fracture density varies both vertically and laterally because of the variation in tuff properties. Fractures related to faults may affect the hydraulic properties around fault zones and provide fast flow paths through hydrologic units that are otherwise not prone to fracture flow. Even nonwelded units, such as the Pah Canyon and Calico Hills tuffs, may allow groundwater flow in fractured zones adjacent to faults. The extent of rock property modification due to faulting (i.e., fracture zones related to faulting) generally correlates with the amount of movement on the fault (i.e., faults with larger displacements have larger fractured zones).2.1.2 Yucca Mountain Site Characterization Project Previous Work: Volcanism
In this section discussion of previous YMP geologic study and site characterization focuses on summarizing the evaluations of conceptual models and data for volcanism from the PVHA. A discussion of the contribution of past PA analysis of volcanism through the TSPA-VA is also contained in this section.2.1.2.1 Volcanism Studies for Site Characterization
Assessment of the volcanic hazard at Yucca Mountain evaluated late Tertiary and Quaternary igneous activity. Volcanism studies have been ongoing for the past two decades as part of the site characterization to determine the ages and character of past volcanic episodes in the Yucca Mountain region and to understand the tectonic setting with which volcanic activity is associated. These investigations included:2.1.2.2 Probabilistic Volcanic Hazard Analysis
Founded upon this extensive base of data, analyses and interpretation, a Probabilistic Volcanic Hazard Analysis (PVHA) (CRWMS M&O 1996) was conducted to determine the probability of igneous activity intersecting the potential repository. To ensure appropriate quantification of scientific uncertainty in the hazard analysis, the DOE identified ten experts to evaluate data, volcanic processes, and features. The product of the PVHA was a quantitative assessment of the probability of a basaltic dike intersecting the potential repository and the uncertainty associated with the assessment. The result of this expert elicitation is volcanic hazard that reflects a diversity and range of alternative scientific interpretations. The major procedural steps in the PVHA were: (1) selecting the expert panel members, (2) identifying the technical issues, (3) eliciting the experts’ evaluations in a series of workshops, and (4) performing probabalistic calculations. From more than 70 nominees, 10 individuals were selected to evaluate volcanic processes and models and develop input interpretations. The panel was carefully balanced with respect to technical expertise (physical volcanology, geochemistry, and geophysics) and institutional/organizational affiliation (CRWMS M&O 1996, Table 1-2). At the core of the PVHA elicitation were four workshops. The primary objective of the workshops was to ensure the experts’ understanding of the issues, volcanic processes, alternative volcanic models, volcanic features, and the data. The first three workshops focused on the data, volcanic processes and models, and interpretations relevant to the PVHA. The workshops included presentations of data and interpretations by technical specialists from Los Alamos National Laboratory, the U.S. Geological Survey (USGS), the University of Nevada, Las Vegas, the Center for Nuclear Waste Regulatory Analysis, and some PVHA project experts. During the fourth workshop, the experts reviewed the preliminary evaluations developed by the panel members, after which the individual evaluations were revised based on feedback received. Two field trips held during the course of the PVHA provided the opportunity for the panel members to observe volcanic features and relationships pertaining to eruptive style and the distribution and timing of volcanic activity in the Yucca Mountain region. The experts developed temporal and spatial models of volcanic activity for hazard calculation. Temporal models describe the frequency of occurrence of volcanic activity and include homogeneous and nonhomogeneous (time varying) models. Many of the experts used homogeneous Poisson models to define the temporal occurrence of volcanic events. Homogeneous Poisson models assume a uniform rate of volcanism based on the number of volcanic events that occurred during various periods in the past. Nonhomogeneous models were used by some experts to describe volcanic clustering in time or to describe the possible waning or waxing of volcanic activity in the region. In order to capture the uncertainty in the location of future volcanic events in the Yucca Mountain region, the PVHA experts used different spatial models. Three types of models were used. Volcanic source zones represent regions in which the future occurrence of volcanoes is spatially homogeneous. These source zones are defined using several criteria and observations: the spatial distribution of observed basaltic volcanic centers (especially post-5 million year old centers), structurally controlled regions, regions defined based on geochemical affinities, and tectonic provinces. Parametric models represent the spatial distribution of future volcanic events that follow a given distribution (field shape), such as a bivariate Gaussian distribution. Nonparametric estimation techniques define the spatial distribution of future events by "smoothing" the locations of known events using a smoothing function. The PVHA experts included alternative source zone configurations and smoothing parameters in their models to reflect the diversity and range of scientific interpretations. Formal elicitation followed the third workshop. The process consisted of a two-day individual interview with each expert. To provide consistency the same interview team was used for all elicitations. Following the elicitation interview each expert was provided with a written summary of their elicitation that was prepared by the interview team. The expert reviewed and clarified the summary and had the opportunity to revise any assessments. To promote a full understanding of each individual’s evaluations, the assessments were presented and discussed at the fourth workshop. Following this workshop each expert had a final opportunity to revise the evaluations (CRWMS M&O 1996, Appendix E) before the results of the PVHA were finalized. The product of the PVHA was a quantitative assessment of the probability of a basaltic dike intersecting the potential repository (CRWMS M&O 1996, Figure 4-32). Specifically, the hazard is a probability distribution of the annual frequency of intersection of a basaltic dike with the repository footprint. A probability distribution of the annual frequency of intersection of the repository footprint by a dike that typically spanned approximately 2 orders of magnitude was computed for each of the ten experts (CRWMS M&O 1996, Figure 4-31). From these individual probability distributions an aggregate probability distribution was computed that reflected the uncertainty across the entire expert panel (CRWMS M&O 1996, Figure 4-32). The distributions of individual experts were combined using equal weights. The mean value of the aggregate probability distribution was 1.5 x 10-8 dike intersections per year with a 90 percent confidence interval of 5.4 x 10-10 to 4.9 x 10-8 (CRWMS M&O 1996, p. 4-10). These values have been updated for the current repository footprint, Enhanced Design Alternative II (EDA II) Design B (CRWMS M&O 1999a), as discussed in Sections 3.1.1 and 3.1.4 of this Disruptive Events PMR. The composite distribution for intersection frequency spanned about three orders of magnitude. The range in the mean frequencies of intersection for the individual experts’ interpretations spanned about one order of magnitude (CRWMS M&O 1996, Figure 4-32). The variance for frequency of intersection defined by the composite distribution was disaggregated to identify the contributions from each of the sources of uncertainty, including variability between the experts’ interpretations (CRWMS M&O 1996, Figure 4-33). Generally, the uncertainty in characterizing a hazard arose from uncertainty in an individual expert’s evaluations of volcanic processes and model interpretations of the hazard, rather than differences in interpretations between the experts (CRWMS M&O 1996, p. 4-10, Figure 4-33). The probability distribution arrived at by the PVHA accounted for undetected events (buried volcanic vents or intrusive activity that never reached the surface). The undetected event frequency ranged from 1 to 5 times that of observed events, with most estimates in the range of 1.1 to 1.5 (CRWMS M&O 1996, Figure 3-62). The PVHA results indicated that the uncertainty in estimating the event rate was the largest component of intraexpert uncertainty (CRWMS M&O 1996, p. 4-10, Figure 4-33). The next largest uncertainty was in the appropriate spatial model. Other important spatial uncertainties included the spatial smoothing distance, Gaussian field parameters, zonation models, and event lengths. The temporal issues of importance included the time period of interest, event counts at a particular center, and the frequency of hidden events (CRWMS M&O 1996, Figure 4-33).2.1.2.3 TSPA-VA Analysis of Volcanism
The PVHA, which focused on the volcanic hazard at the site, provided significant input to assessment of volcanic risk for the TSPA-VA analysis (DOE 1998a, Section 4.4). Details of the analysis of volcanic disruptive events scenarios were described in the Total System Performance Assessment-Viability Assessment (TSPA-VA) Analyses Technical Basis Document in Chapter 10, Disruptive Events (CRWMS M&O 1998b, Section 10.4). The disruptive events analyses for volcanism in TSPA-VA were constructed based on FEPs scenarios developed from the immediately preceding TSPAs (see list of previous TSPAs in Section 2.0). PAs previous to TSPA-VA used generalized event trees for constructing disruptive scenarios that lead to understanding the processes that could contribute to increased radionuclide releases from disruptive events. In addition to analyzing FEPs that were determined to be important from previous TSPAs, disruptive events volcanism analyses for TSPA-VA were prepared with the view of addressing the two subissues and acceptance criteria of the NRC’s IRSR for Igneous Activity, Rev. 1 (NRC 1998e, Section 5). The volcanism analysis for TSPA-VA used probability information and descriptions of the nature of volcanic processes and events from the PVHA expert elicitation (CRWMS M&O 1996). Three igneous activity effects scenarios were analyzed for TSPA-VA: (1) direct release, (2) enhanced source term, and (3) indirect effects. Two of these scenarios are taken forward for analysis in TSPA-SR (1 and 2); the third was screened out from further consideration. Screening arguments for excluding indirect effects (i.e., hydrologic response) from further analysis are contained in the AMR Disruptive Events FEPs (CRWMS 2000h, FEP 1.2.10.02.00). The disruptive events FEPs AMR is summarized in Section 3.1.6 of this PMR, and a brief description of the FEP 1.2.10.02.00 screening argument is presented. The event tree method was used in TSPA-VA to determine potential consequences of igneous activity from whether a rising basaltic dike intersected emplacement drifts to the possibility of formation of a surface cinder cone and a contaminated ash plume. The TSPA-VA consequence scenario analysis process as understood at the time of TSPA-VA is captured in Figures 2-3 through 2-7 [2-3, 2-4, 2-5, 2-6, 2-7]. The event tree in Figure 2-3 depicts alternative consequences and decision points of a basaltic dike intersecting the potential repository and possibly contacting WPs. This represents the intrusive phase of volcanism that is common to both eruptive and intrusive events. The TSPA-VA analysis looked at consequences of both magma and ash particles contacting WPs, and the event tree for pyroclasts contacting WPs and waste is represented in Figure 2-4. WP breach was assumed to be by contact from pyroclasts (not by melting). Figure 2-5 is the event tree for waste entrainment in a volcanic ash cloud during an eruptive event. An event tree for release of waste from WPs engulfed in magma, but not entrained during an eruptive event, is depicted in Figure 2-6. This scenario analysis was called enhanced source term and analyzed release of waste into groundwater that entered WPs encased in cooled basalt years after the packages were compromised. The last TSPA-VA event tree, Figure 2-7, depicts the indirect effects scenario that was excluded from further analysis after TSPA-VA. This scenario represents a dike emplaced in the SZ having a significant effect on groundwater flow. TSPA-VA sensitivity studies (and subsequent disruptive events FEPs screening arguments) provided the basis for concluding that there would be no significant effect on dose from this scenario. The direct release scenario (renamed as the volcanic eruption release for TSPA-SR) for TSPA-VA was one in which a volcanic eruption dispersed contaminated ash on the ground 20 km from the potential repository site. The processes included in the direct release scenario for the TSPA-VA are depicted in Figures 2-3, 2-4, and 2-5. The enhanced source term scenario (renamed igneous intrusion groundwater release for TSPA-SR) was liquid magma intersecting the repository drifts and engulfing WPs, compromising their integrity and leaving the contents exposed in the basaltic rock that formed from the cooled magma. The contents were then assumed to be available for transport in encroaching groundwater using the UZ and SZ flow TSPA models. Figures 2-3 and 2-4 were combined to support analysis of this scenario for TSPA-VA. The indirect igneous effects scenario was for the possible effects on groundwater flow in the SZ from dike emplacement assuming two possibilities, that the dike was either more or less permeable than the country rock it intruded. Figure 2-7 supported analysis of this scenario for TSPA-VA and, as previously mentioned, this scenario was not analyzed for TSPA-SR. Further details of the assumptions and methods used in the TSPA-VA analysis can be found in Section 3.1.5, where the disruptive events AMR Igneous Consequence Modeling for TSPA-SR (CRWMS M&O 2000l) is discussed.2.1.3 YMP Previous Work: Seismicity and Structural Deformation
Comprehensive geologic and geophysical studies have been conducted to assess the seismic hazard at Yucca Mountain. The previous studies included: (1) ongoing site characterization activities that establish the site geologic framework (see CRWMS M&O 1998e), (2) the seismic topical reports (YMP 1997a, b; CRWMS M&O 1999h) that establish the methodology to be followed in assessing seismic hazard and preclosure design inputs, and (3) the PSHA that establishes the seismic hazard (Wong and Stepp 1998). Scientific investigations and evaluation conducted over the past twenty years provide the basis for assessment of seismic hazards at Yucca Mountain (CRWMS M&O 2000c, Section 6.1.2). Building upon earlier investigations of the (NTS) region, studies of the Yucca Mountain site have included:2.1.3.1 Seismic Topical Reports
Two seismic topical reports have been prepared, and a third is in preparation, that together document the basis for seismic design of the potential repository. The seismic hazard results were developed principally for preclosure analyses; however, they also provide the basis for the postclosure PA analyses that are the focus of this PMR. Two of these reports have been presented to the NRC for its review and comment. The third report is currently being prepared for completion after TSPA-SR. A probabilistic seismic hazard analysis (the PSHA) was conducted based on the methodology developed in Topical Report 1. Topical Report 2 and Topical Report 3 document the preclosure seismic design methodology and development of the seismic design basis inputs, respectively. Seismic Topical Report 1, Methodology to Assess Fault Displacement and Vibratory Ground Motion Hazards at Yucca Mountain (YMP 1997a), contains a description of the DOE methodology for probabilistic assessment of vibratory ground motion and fault displacement hazards. The methodology involves a series of workshops structured so that multiple experts can interact to evaluate hypotheses and models using the geological, geophysical, and seismological data sets from the Yucca Mountain area. The methodology requires that the experts specifically evaluate all hypotheses and models that have credible support in the data. The product of the methodology is multiple interpretations by the experts of seismic sources, source properties, and evaluations of ground motion, all of which include specific expressions of uncertainty. Comprehensive and consistent consideration of data and documentation of all interpretations is required by the methodology. This topical report guided the process followed for the PSHA expert elicitation. Seismic Topical Report 2, Preclosure Seismic Design Methodology for a Geologic Repository at Yucca Mountain (YMP 1997b), contains a description of the design methodology and criteria that the DOE intends to implement to provide reasonable assurance that vibratory ground motion and fault displacements will not compromise the preclosure function of repository systems important to safety. The report establishes hazard probability levels that are appropriate for determining design basis vibratory ground motions and design basis fault displacements. Acceptance criteria for both surface and underground facilities are provided for vibratory ground motion and fault displacement design. The report also provides criteria for fault avoidance and seismic design considerations for WPs. Seismic Topical Report 3, Preclosure Seismic Design Bases for a Geologic Repository at Yucca Mountain (described in its development plan CRWMS M&O 1999h), the last of the methodology reports, will contain a description of the development of seismic design basis inputs for appropriate frequencies of occurrence as defined in Topical Report 2. The results of the probabilistic seismic hazard analyses (PSHA) will be summarized, including the characterizations of seismic sources, fault displacement, and ground motion attenuation developed by the two panels of experts (described in the discussion of the PSHA that follows). Design basis earthquake ground motions will be defined for three specific sites that represent the range of locations and conditions where repository facilities would be located.2.1.3.2 Probabilistic Seismic Hazard Analysis
A PSHA that assessed both ground motion and fault displacement hazards was conducted for the potential repository at Yucca Mountain. The PSHA combines seismic source zones and their associated earthquake recurrence with appropriate attenuation relationships to produce "hazard curves" in terms of level of ground motion and an associated probability of that ground motion being exceeded annually. The study, Probabilistic Seismic Hazard Analyses for Fault Displacement and Vibratory Ground Motion at Yucca Mountain, Nevada (Wong and Stepp 1998), was a four-year multidisciplinary project that was based on expert elicitation. The disruptive events AMR Characterize Framework for Seismicity and Structural Deformation at Yucca Mountain, Nevada (CRWMS M&O 2000c) contains more detail and roadmapping to sections of the PSHA and is summarized in Section 3.2.1.2 of this PMR. Many scientists and engineers participated in and contributed to the PSHA (Wong and Stepp 1998, Appendices A, C, and D). These individuals were associated with universities or government agencies or were experts from industry. Six teams of three experts each, who together formed a composite expertise in the seismicity, tectonics, and geology of the Yucca Mountain region, made seismic source characterizations. Seven individual experts made the ground motion assessments. Many other researchers participated in workshops and field trips devoted to the discussion of available data and possible interpretations of these data. The experts’ interpretations specifically incorporated uncertainties related to the data and to resolving different hypotheses and models. The uncertainties that were factored into the analyses reflected the range of views of the many individuals that contributed to the hazard assessment. The experience level and diversity of PSHA participants in a wide variety of tectonic environments supported an appropriate representation of uncertainty through the composite distribution of views represented by diverse participants from the scientific community. The objectives of the PSHA analyses were to support assessments of the potential repository’s long-term performance and seismic design criteria development for facility design (Wong and Stepp 1998, Section 1.1). Quantitative hazard results were developed in the form of annual exceedence probabilities for various levels of fault displacement at selected locations and vibratory ground motion at a hypothetical rock outcrop at the ground surface. Both the preclosure and postclosure performance periods of the repository were addressed in the PSHA study. Three primary activities of the study were:2.1.3.2.1 PSHA Summary: Seismic Source Location and Geometry
A seismic source is defined as a region of the earth’s crust that has relatively uniform seismicity characteristics, is distinct from those of neighboring sources, and can be used in approximating the locations of future earthquakes. It is a construct developed for seismic hazard analysis as a means of approximating the locations of earthquake occurrences. Seismic sources can be categorized into two basic source types: fault sources and areal sources (Wong and Stepp 1998, Section 4.1.2). Fault sources are represented as lines or planes and represent the occurrence of earthquakes along a known or suspected fault trace. Areal sources represent areas of distributed seismicity that are not apparently associated with specific, known faults. Areal sources can be divided into three types: a source whose boundary encloses a concentrated zone of seismicity, a source defined by regional seismotectonic characteristics, and a regional background source (typically applying to a larger region than is defined by the other area sources). The boundaries of areal sources delineate areas that have relatively uniform seismic potential in terms of earthquake occurrence and maximum earthquake magnitude. The basic characteristics that must be defined for all source types are the same (i.e., location, maximum magnitude, and recurrence); however, the particular parameters and data sets that are used to define these characteristics may be quite different. The seismic source study includes analysis of seismic moment, which is a measure of the strength of the earthquake. Magnitude, also a measure of earthquake size, is determined by taking the common logarithm (base 10) of the largest ground motion recorded by a seismograph and applying a correction for the distance to the earthquake. Several scales have been defined, but the most commonly used are: (1) local magnitude (ML), commonly referred to as "Richter Magnitude," (2) surface-wave magnitude (Ms), (3) body-wave magnitude (mb), and (4) moment magnitude (Mw). Scales 1 to 3 have limited range and applicability and do not satisfactorily measure the size of the largest earthquakes. The moment magnitude scale, based on the concept of seismic moment, is uniformly applicable to all sizes of earthquakes but is more difficult to compute than the other types. The seismic source expert teams considered two types of fault sources: regional faults and local faults. Regional faults were defined by most teams as Quaternary faults within 100 km of Yucca Mountain, but outside the local vicinity of the site, that were judged to be capable of generating earthquakes of Mw 5 and greater. Local faults were defined as being located within about 15 km of Yucca Mountain. Paleoseismic data from numerous references (see Wong and Stepp 1998, Appendix B) were used by all the teams to identify and characterize fault sources, some of which were regional. Faults were considered, but not judged relevant to the hazard analysis, if they had short lengths or no significant Quaternary displacement (Wong and Stepp 1998, p. 4-49).2.1.3.2.2 PSHA Summary: Regional Faults, Local Faults, Areal Source Zones and Volcanic Sources
The number of regional faults considered by the expert teams ranged from 11 to as many as 36. This reflected, in part, the judgments of the teams regarding the activity of various faults as well as the decision by some teams to also include potentially active faults. Some teams also considered areal source zones as adequately representing regional faults. All the teams modeled the regional faults as simple, planar faults to maximum seismogenic depth with generalized dips depending on the style of faulting (preferred values of 90° for strike-slip faults and 60° or 65° for normal-slip faults). Alternative fault lengths for most of the faults were included by all the teams to express uncertainty in their mapped lengths. Of the regional faults, the most significant were the Furnace Creek and Death Valley faults, despite their relatively great distances from the Yucca Mountain site (> 50 km), because of their high slip rates (2.5 to 8 mm/yr.) and potential to generate maximum magnitude (Mmax) earthquakes of about Mw 7.5. Figure 2-8 shows the known or suspected Quaternary faults and potentially significant local faults within 100 km of Yucca Mountain; local faults in the immediate vicinity of Yucca Mountain are shown in Figure 2-9. Varying behavioral and structural models were employed by the expert teams to represent the full range of possible rupture patterns and fault interactions in the characterization of local faults. Most teams preferred a planar fault model. Some of the faults could have been interconnected, with linkages along strike or coalescence down-dip. Some type of simultaneous rupture of multiple faults was included in all models. In general, preferred models for multiple fault rupture included two to four coalescing fault systems. Several teams used detachment models to constrain the extent and geometry of local fault sources. A seismogenic detachment fault was considered, but not strongly favored, as a source of large earthquakes by the teams. The possibility that right-lateral shear is accommodated in the Yucca Mountain region by a buried strike-slip fault was considered by all expert teams. Most teams included some variation of a regional buried strike-slip fault source, though with low probability. Areal source zones were defined by the expert teams to account for background earthquakes that could occur on potential buried faults or faults not explicitly included in their model. Some teams included alternative areal zone models in their characterization within a 100 km radius of the Yucca Mountain site. The teams also defined areal zones that extended beyond 100 km from the Yucca Mountain site to completely express uncertainty in the seismic source interpretations. Several teams defined a site area, or zone, solely for assigning a lower Mmax to the area where more detailed investigations had been conducted and the inventory of fault sources was more complete. Seismicity related to volcanic processes, specifically to basaltic volcanoes and dike-injection, was considered by all teams, but explicitly modeled as distinct source zones by only two expert teams (Wong and Stepp 1998, Table 4-1). Volcanic-related earthquakes were not modeled as a separate seismic source by the other four teams because the low magnitude and frequency of volcanic-related seismicity was assumed to be accounted for by earthquakes in the areal zones.2.1.3.2.3 PSHA Summary: Maximum Earthquake Magnitudes
Mmax earthquakes were defined for each seismic source by each team to represent the largest earthquake that the source was capable of generating, regardless of how frequently it occurred (Wong and Stepp 1998, p. 4-49). As discussed in Section 2.1.3.2.2, numerous seismic sources were characterized, and each of these different sources has been assigned a maximum magnitude. The maximum earthquakes from all sources were incorporated in the vibratory ground motion hazard assessment (described in Section 2.1.3.2.5). There are two basic approaches to assessing maximum magnitudes for seismic sources: constraints provided by estimates of maximum dimensions of fault rupture and constraints provided by historical seismicity. As is common in most parts of the world, the historical seismicity record is too short to have observed and recorded with certainty the maximum earthquakes on seismic sources in the Yucca Mountain region. Hence, estimates of fault rupture dimensions are the principal means of estimating maximum magnitudes. Uncertainties in estimating the physical dimensions of the maximum rupture on the faults were explicitly incorporated into the analysis. The approach used to evaluate the Mmax for faults was to estimate the maximum dimensions of rupture and then compare those dimensions in empirical relationships between rupture dimensions and earthquake magnitude. The types of empirical relationships available were: magnitude versus rupture length, magnitude versus rupture area, magnitude versus maximum surface displacement, and magnitude versus average surface displacement. For areal sources the Mmax for the zone was based primarily on consideration of the historical seismicity record. The Mmax could also have been selected as representing the largest earthquake determined to occur on any of the faults within the areal zone. If an areal zone was used to model the occurrence of earthquakes on unknown faults, the Mmax for the zone was determined by the largest fault mapped within the zone or the largest earthquake that was not associated with surface faulting. This ensured that any unknown or unidentified faults were accounted for.2.1.3.2.4 PSHA Summary: Earthquake Recurrence
Earthquake recurrence relationships express the rate or annual frequency of earthquakes occurring for a single seismic source. Seismic sources generate a range of earthquake magnitudes up to the maximum magnitude. A magnitude-distribution model defines the relative number of earthquakes having particular magnitudes. Methods for developing recurrence relationships are usually different for fault sources than for areal sources. Recurrence rates for fault sources are usually estimated from geologic data, while for areal sources historical seismicity data are used. Two approaches were used to estimate the earthquake recurrence relationships for fault sources (Wong and Stepp 1998, Section 4.3.1.2). The first involved estimating the frequency of large-magnitude, surface-rupturing earthquakes on the fault either by dating of paleoearthquakes or by dividing an estimate of the fault slip rate by an estimate of the average slip per event. The second approach was to translate the estimated fault slip rate into a seismic moment rate and then partition the moment into earthquakes of various magnitudes according to the magnitude-distribution model used. For areal sources, earthquake recurrence relationships were determined from the historical seismicity. The earthquake catalog for the region within a 300-km radius of the Yucca Mountain site was compiled from all available regional and national earthquake catalogues. All known NTS blasts were identified and removed. The catalog was analyzed to identify and remove dependent events (earthquakes that were aftershocks or foreshocks of larger earthquakes). Figure 2-10 compares the combined distribution for earthquake recurrence from all seismic sources and the mean results for the six expert team characterizations. There is generally less than an order of magnitude range in uncertainty in the estimation of regional seismicity rates. At smaller magnitudes, the range reflects the differences in how the teams characterized the regional source zones. At larger magnitudes, the assessments from the individual teams lie within the uncertainty in the occurrence rates of earthquakes based on the historical record. Because the ground motion hazard, at least for high spectral frequency ground motions, is influenced largely by nearby seismic sources, the larger uncertainty in recurrence rates for the local sources has a significant effect on the uncertainty in the ground motion hazard.2.1.3.2.5 PSHA Summary: Vibratory Ground Motion Hazard
The level of ground shaking, expressed as the amplitude of ground motions, is a function of three main elements: the seismic source, the source-to-site path, and the site conditions. The source conditions include the magnitude of the earthquake, style of faulting, and geometry of the coseismic fault rupture. The second element is the travel path of seismic waves from the source of the earthquake to a particular site. The length of this path is important, because the amplitude of ground motions will decrease, or attenuate, with distance. The third element is the local site condition, or the effect of the uppermost several hundred meters of rock and soil and the surface topography. All three of these elements that control ground motions were explicitly addressed in the Yucca Mountain seismic hazard analysis. When the ground motion analysis is combined with the seismic source characteristics, a probabilistic representation of vibratory ground motion hazard is produced. The seven ground motion experts estimated median ground motion and uncertainties for a matrix of earthquake magnitudes, source-to-site distances and faulting styles (normal- and strike-slip), and for a suite of spectral frequencies (Wong and Stepp 1998, Section 5). Data representative of ground motions at a depth of 300 m below ground level were needed to estimate the hazard at the potential repository; however, there was, and is, very little empirical data for strong ground motions measured at depth. Most geophysical data are taken with the receiver for the signal located at the surface (i.e., the air-rock interface). At the level of the potential repository the returning signal would be traveling in rock, under confining pressure without free air space above. The PSHA experts used both empirical data, taken in "outcrop" conditions at a geologic surface with free air space, and theoretically based calculations that would allow them to make the empirical data more applicable to the case that was being evaluated (i.e., the potential repository level). They performed ground motion calculations as if there were free air space conditions at the point of interest, but the used rock properties that exist at 300 m including bulk density and shear wave velocity (Figure 2-11). The PSHA experts used this technique because there is not a straight forward extrapolation from ground motion data taken at ground level to the equivalent data taken at depth. In the PSHA the point for which all hazard calculations were to be applicable, the potential repository level, is referred to as "Point A". Other points representing different rock conditions were considered, but ground motions were calculated for Point A only and the hazard curves produced apply to Point A. The probabilistic hazard for vibratory ground motion was calculated based on equally weighted inputs from the six seismic source expert teams and the seven ground motion experts (Wong and Stepp 1998, Section 7.3). The probabilistic hazard was calculated for horizontal and vertical peak acceleration; spectral accelerations at frequencies of 0.3, 0.5, 1, 2, 5, 10, and 20 Hz; and peak velocity. It was expressed in terms of hazard curves (see Figure 2-12). The hazard was also expressed in terms of uniform hazard spectra (see Figure 2-13). Disaggregation of the mean hazard or magnitude, distance, and ground motion variability for an annual exceedence probability of 10-4 shows that at 5 to 10 Hz (or other high frequencies) ground motions are dominated by earthquakes of smaller than Mw 6.5 occurring at distances of less than 15 km. Dominant events for low-frequency ground motions, such as at 1 to 2 Hz, display a bimodal distribution, including large nearby events and Mw 7 and larger earthquakes beyond distances of 50 km (see Figure 2-14). The latter contribution is due mainly to the relatively higher activity rates for the Death Valley and Furnace Creek faults.2.1.3.2.6 PSHA Summary: Fault Displacement Characterization
Fault displacement hazard is the hazard related to differential slip that occurs at the surface along a seismogenic fault or along secondary faults triggered by the seismogenic rupture. Several alternative approaches to characterizing fault displacement hazard assessment were developed by the experts (Wong and Stepp 1998, Section 4.3.2). The approaches were based primarily on empirical observations of faulting characteristics at Yucca Mountain and in the Basin and Range province during past earthquakes. The method for assessing probabilistic fault displacement hazard was similar to that for vibratory ground motion hazard. The hazard was represented probabilistically by a displacement hazard curve that is analogous to ground motion hazard curves. Thus the hazard curve was a plot of the frequency of exceeding a fault displacement value. Fault displacement hazard was evaluated at nine locations within the Yucca Mountain site area (CRWMS M&O 2000c, Figure 3). These locations were selected to span the range of known faulting conditions and ranged from block-bounding faults to small fractures and unfaulted rock. All the teams considered the points on the Bow Ridge and Solitario Canyon faults as subject to principal faulting hazard. A few teams also considered some potential for principal faulting hazard at two locations on two intrablock faults. The teams varied widely in their assessments of the probability that distributed faulting could occur in future earthquakes at points that are located off of the block-bounding faults. These assessments were based on fault orientation, cumulative slip, and structural relationship. Four teams considered that the probability of displacement at a point in intact rock due to the occurrence of a future earthquake is essentially zero (i.e., the probability that a new fault will form is essentially zero). With the exception of the block-bounding Bow Ridge and Solitario Canyon faults, the mean displacements are 0.1 cm or less at 10-5 annual exceedence probability. At 10-5 probability, the mean displacements are 8 and 32 cm, respectively, for these two faults. Sites not located on a block-bounding fault—such as sites on the intrablock faults, other small faults, shear fractures, and intact rock—are estimated to have displacements significantly less than 0.1 cm for annual frequencies as low as 10-5 (Wong and Stepp 1998, Table 8-1).2.1.3.3 TSPA-VA Analysis of Seismicity
Prior to the TSPA-VA, analysis of seismic hazard had not been systematically included in TSPAs, although some calculations had been made (Gauthier et al. 1996). Disruptive events seismic hazard analyses for TSPA-VA examined the subissues and acceptance criteria of the NRC IRSR for Structural Deformation and Seismicity (NRC 1999a). However, because of the limited scope of seismic activity analysis, the TSPA-VA did not contribute much toward addressing the subissues of the IRSR (CRWMS M&O 1998b, p.10-57). Potential effects of seismic activity that were identified by the TSPA-VA from previous work included: (1) vibratory ground motion and fault displacement from earthquakes, (2) changes in site hydrologic properties including changes in water table elevation and changes in groundwater flow patterns, and (3) indirect effects such as alteration of groundwater flow paths caused by faulting or dike emplacement in the SZ (DOE 1998a, p. 4-88). The indirect effects scenario for faulting was excluded (screened out) from TSPA-VA analysis by the same sensitivity study that supported screening out indirect effects of volcanism. Section 2.1.2.3 contains a discussion of the indirect effects of volcanism from a dike emplaced in the SZ. The only seismic effect analyzed in TSPA-VA was that for rockfall on a WP caused by vibratory ground motion initiated by an earthquake. Changes in site hydrologic properties were not analyzed by TSPA-VA, except for the aspects of changes in groundwater flow patterns included in the sensitivity analysis for indirect effects. The rockfall scenario was one in which rocks, jarred free of the emplacement drift roof by vibratory ground motion, fell on WPs (DOE 1998a, p. 4-90). Thermal-mechanical stresses from drift excavation and the heat generated by the waste were also considered as a source of rock quality weakening that could contribute to rockfall (DOE 1998a, p. 10-57). The drift’s concrete liners were assumed to have failed within a few hundred years (DOE 1998a, p. 4-90). The result of rockfall was conceptualized either as a split in the WP that allowed immediate access of air and water or as dents in the package that provided locations for accelerated corrosion and premature failure of the WP. Damage to WP walls was a function of time since closure because of thinning by corrosion (DOE 1998a, p. 4-91). The results of TSPA-VA seismic activity modeling showed that, if the outer barrier (corrosion allowance material) was not corroded, a rock larger than allowed by any observed combination of fractures measured in the Exploratory Studies Facility (ESF) was needed to damage the WP (DOE 1998a, p. 4-92). Results showed that, when the outer barrier and half of the inner barrier were corroded, a rock of the dimensions allowed by fractures observed in the ESF could damage the WP; however, this scenario would require more than 100,000 years of wet corrosion conditions. Calculations showed almost no effect on repository performance for the first 1,000,000 years, and over a 10,000-year period "the probability of rockfall causing a WP to split open was essentially zero" (DOE 1998a). For TSPA-SR some TSPA-VA scenarios are being re-examined. Water table rise is the subject of FEP 1.3.07.02.00 in the Project FEPs database, and the screening argument for it is contained in Features, Events and Processes in SZ Flow and Transport (CRWMS M&O 2000q). Rockfall is re-examined for analyses where there is no backfill in the potential repository design.2.1.4 Features, Events, and Processes Analysis for Disruptive Events
The following discussion serves two purposes. It is a summary of the FEPs scenario development process currently in use by the DOE and employed for disruptive events FEPs analysis for TSPA-SR. Because it is taken from the disruptive events FEPs AMR Disruptive Events FEPs (CRWMS M&O 2000h, Section 1), it also serves as part of the summary of that AMR in this Disruptive Events PMR. The rest of the summary for the disruptive events FEPs AMR is provided in two other sections of this PMR. The summary of FEPs analysis results for FEPs associated with volcanism is contained in Section 3.1.5, and FEPs associated with tectonics, seismicity and structural deformation are summarized in Section 3.2.4. The following discussion is a summary of the origin and methods of the FEPs scenario development process for TSPA-SR. Under the provisions of the DOE’s Interim Guidance (Dyer 1999),, the DOE must provide a reasonable assurance that the performance objectives for the potential repository can be achieved for a 10,000-year postclosure period. This assurance must be demonstrated in the form of a PA that:2.1.4.1 FEPs Identification and Analysis
The development of a comprehensive list of FEPs relevant to the YMP is an ongoing process based on site-specific information, guidance documents, and proposed regulations. The YMP FEPs Database (CRWMS M&O 2000j) contains 1,786 entries derived from the following sources:2.1.4.2 FEPs Screening Process
The first step in the scenario development process was the identification and analysis of FEPs. The second step in the scenario development process included the screening of each FEP against project criteria. Each FEP was screened against criteria stated in DOE’s Interim Guidance (Dyer 1999) and in the EPA’s proposed rule 40 CFR Part 197 (64 FR 46976). Each FEP is screened against the guidance, assumptions, or specific criteria stated in NRC’s proposed rule 10 CFR 63 (64 FR 8640) and in the EPA’s proposed rule 40 CFR Part 197 (64 FR 46976) (CRWMS M&O 2000h, Section 1.3). The screening criteria are discussed in more detail in the AMR Disruptive Events FEPs, (CRWMS M&O 2000h, Section 1.3); they are summarized here:2.1.4.3 Disruptive Events FEPs
The primary purpose of the Disruptive Events FEPs AMR (CRWMS M&O 2000h) was to identify and document the analysis, screening decision, and TSPA disposition, or screening argument, for the 21 FEPs that were recognized as disruptive events FEPs (see Table 2-1). FEPs addressed in the Disruptive Events FEPs AMR represent natural systems processes that have the potential to significantly affect the potential repository performance. The FEPs are related to geologic processes such as structural deformation, seismicity, and igneous activity. Of the 21 Primary disruptive events FEPs (See Table 2-1), 16 were addressed explicitly and fully in the AMR Disruptive Events FEPs (CRWMS M&O 2000h, Section 6). The remaining five Primary disruptive events FEPs are addressed in the Disruptive Events FEPs AMR document with only short summaries and with references to the other AMRs that provide the explicit and full discussion of the FEP. FEPs addressed in the Disruptive Events FEPs AMR represent natural systems processes that have the potential to significantly affect the potential repository performance. The FEPs are related to geologic processes such as structural deformation, seismicity, and igneous activity. Of the 21 Primary disruptive events FEPs (See Table 2-1), 16 were addressed explicitly and fully in the AMR Disruptive Events FEPs (CRWMS M&O 2000h, Section 6). The remaining five Primary disruptive events FEPs are addressed in the Disruptive Events FEPs AMR document with only short summaries and with references to the other AMRs that provide the explicit and full discussion of the FEP. Table 2-2 provides a summary of the disruptive events FEPs screening decisions and the basis for "Exclude" decisions. A detailed discussion of the "Exclude" decision process is presented in the AMR Disruptive Events FEPs (CRWMS M&O 2000h, Section 6). Shaded FEPs are Primary; others are Secondary. Not all Secondary FEPs are shown in Table 2-2 because many of the Secondary FEPs are either redundant or Secondary FEP descriptions, which are insufficient to allow resolution. FEPs screening provided decisions regarding which analyses will be included in TSPA-SR. Section 2.1.4.2 of this Disruptive Events PMR explains the screening criteria and the significance of the "Include" and "Exclude" screening decisions. The next section of Chapter 2 discusses the overall approach to disruptive events analysis for SR that evolved from previous work and from technical workshops held in early 1999. A discussion is provided regarding how disruptive events analyses work together to produce the current approach. The impact of design on analyses is also discussed at a summary level.2.2 Approach to Disruptive Events Analysis for Site Recommendation
Site characterization work, expert elicitations, TSPAs, and other analyses and calculations by the YMP and other researchers discussed in previous sections of this chapter contributed to developing the bases for the analysis of volcanism and seismicity for TSPA-SR. In addition, a series of Project workshops held in February of 1999 brought together analysts from disciplines that had contributed to disruptive events analysis in three areas: volcanism, seismicity, and criticality. At the workshops the results of TSPA-VA analyses and major unresolved key technical issues were discussed. Potential analytical approaches were discussed and the outcome led to development of work plans that were used as the bases for the technical development plans that support TSPA-SR AMRs. An initial list of FEPs from the YMP FEPs database, sorted into subject areas, was distributed at the workshops for discussion of association to process model topics. A list of the FEPs to be addressed in the Disruptive Events FEPs AMR was selected from this process. In April of 1999 the procedural framework that guides the TSPA-SR was significantly reworked and the AMR and PMR structure was developed. The structure of disruptive events analysis was developed to be based on eight AMRs and one calculation. The feeds from one AMR or calculation to another (or others) and support from AMRs or calculations performed outside of the disruptive events group is illustrated in Figures 2-16 and 2-17. Section 2.2.2 contains a summary level discussion of the relationship between the analyses and calculations shown in Figures 2-16 and 2-17. The tables in Sections 3.1.1 through 3.1.5 [3.1.1, 3.1.2, 3.1.3, 3.1.4, 3.15] and Sections 3.2.1 through 3.2.3 [3.2.1, 3.2.2, 3.2.3] that summarize the inputs and outputs of the AMRs and calculation contain further information to support the figures. Each AMR is written to the outline provided in procedure AP-3.10Q, in which input data are listed in Chapter 4, assumptions are given in Chapter 5, the analysis is provided in Chapter 6, and conclusions are listed in Chapter 7. Conclusions include outputs that are used in other AMRs (as parameters ready for use or to be further reduced), or are used directly in the TSPA analysis. The following discussion will provide other summary level information regarding the overall approach to analysis, including the approach to incorporation of new data and how the analyses responded to design changes over the period of development of the AMRs. The issue of how and whether to incorporate new data into analyses as the data become available was addressed in a letter to the NRC (Brocoum 1997). The following discussion of treatment of new data is taken from that letter. Although the letter was written after a technical exchange on the topic of igneous activity, the new-data policy applies to new data for all topics. At the time of the PVHA and PSHA expert elicitations the experts had access to all the applicable data that had been developed by the YMP and other researchers. It was recognized that new data would continue to be collected that might be relevant to the hazard analysis results; therefore, a policy was established by the DOE to review new data. The letter describing the approach to new data states: