Technical Basis Document No. 14: Low Probability Seismic Events Revision 1 

June 2004

1. INTRODUCTION 

1.1 OVERVIEW 
This technical basis document provides a summary of the conceptual understanding of 
low-probability seismic events. Low-probability seismic events are unlikely but possible 
earthquake fault displacements and vibratory ground motions that could affect the Yucca 
Mountain site. This document and the associated references form an outline of the ongoing 
development of the postclosure safety analysis that will be included in the license application 
(LA). The conceptual understanding summarized here also provides a framework for addressing 
open Key Technical Issue (KTI) agreements between the U.S. Nuclear Regulatory Commission 
(NRC) and the U.S. Department of Energy (DOE) that are related to low-probability seismic 
events and postclosure repository performance. Because the characteristics of low-probability 
seismic events are also an input to preclosure safety analyses, the conceptual framework 
presented in this report additionally supports the resolution of KTIs related to preclosure seismic 
design. 
This technical basis document is one in a series that is being prepared for each component of the 
repository system that is relevant to postclosure performance. This document focuses on the 
prediction of postclosure seismic hazards and the potential effects of these hazards on 
performance of the emplacement drifts and the engineered barrier system (EBS). The 
postclosure seismic hazards are vibratory ground motion and fault displacement. The major 
elements of the EBS are the waste package, drip shield, emplacement pallet, and the fuel rod 
cladding. 
Seismic hazards are also a consideration in demonstrating preclosure performance. While a 
complete discussion of the seismic aspects of preclosure performance is beyond the scope of this 
report, the development of seismic inputs for preclosure analyses is largely identical to the 
development of seismic inputs for postclosure analyses. Thus, this report describes the 
development of seismic inputs for both time frames. It does not, however, describe how the 
preclosure seismic inputs are used in demonstrating compliant preclosure performance. 
The relationship of low-probability seismic events to the other components of postclosure 
repository performance is illustrated in Figure 1-1. As higher levels of ground motion are 
considered, their annual probabilities of exceedance become lower. At some point, seismic loads 
could contribute to mechanical degradation of the waste-emplacement drifts and generate 
rockfall that impinges on drip shields. The residual tensile stresses induced in drip shields by 
rockfall could lead to accelerated drip shield corrosion, if those stresses exceed a particular 
threshold. Low-probability vibratory ground motion could cause impacts between adjacent 
waste packages, between the waste package and its emplacement pallet, between the waste 
package and the drip shield, between the drip shield and emplacement pallet, and between the 
drip shield and invert. Residual tensile stresses induced by such impacts could lead to 
accelerated drip shield and waste package stress corrosion cracking if those stresses exceed 
material-specific thresholds. Seismically induced impacts also could damage the cladding of the 
spent nuclear fuel waste form inside the waste packages and degrade the cladding’s performance 
as a barrier to radionuclide migration. Seismically induced rockfall would change the shape of 
waste-emplacement drifts, which could affect the seepage rate of water into the drifts. The 
June 2004 No. 14: Seismic Events Revision 1 
presence of seismically induced rockfall also would change thermal properties within the drifts. 
Volcanic events are commonly preceded and accompanied by numerous earthquakes that 
indicate progressive rock failure as magma rises to the ground surface, and such volcanic-related 
earthquakes are considered in the evaluation of seismic hazard. Seismic effects on the other 
natural systems and processes illustrated in Figure 1-1 also have been considered, but no other 
potentially significant seismic effects have been identified. 
1.2 PURPOSE AND SCOPE 
This document provides an overview of the current understanding of low-probability seismic 
events and their possible effects on the components of the repository that are potentially 
important to long-term (postclosure) repository performance. The focus is on the potential 
damage to EBS components from severe, low-probability events because less severe but more 
likely seismic events do not have the potential to disrupt the drip shield, the waste package, the 
emplacement pallet, and the cladding. 
This document also discusses less severe, higher probability (i.e., more frequent) ground motion 
that is considered in preclosure seismic design and safety analyses. The document does not, 
however, describe preclosure seismic strategy or preclosure seismic design analyses. These 
aspects of preclosure application are described in other Yucca Mountain Project documents, such 
as Preclosure Seismic Design Methodology for a Geological Repository at Yucca Mountain 
(BSC 2004a) and Preliminary Seismic Analysis for Preclosure Safety Analysis (BSC 2003a). 
They are beyond the scope of this report. Similarly, hydrologic changes caused by seismic 
activity are detailed in Technical Basis Document No. 3: Water Seeping into Drifts. 
June 2004 1-2 No. 14: Seismic Events Figure 1-1. Components of the Postclosure Technical Basis for the License Application 
The body of this report documents the conceptual framework for responses to open KTI 
agreements between the DOE and the NRC that are related to low-probability seismic events and 
repository postclosure performance. Detailed responses to these KTI agreements are provided in 
Appendices A through D, as shown in Table 1-1. 
Table 1-1. Key Technical Issue Agreements Related to Low-Probability Seismic Events Addressed in 
This Report 
KTI Agreement/AIN Appendix 
A 
CLST 3.10, 
TSPAI 3.06 
RDTME 2.01, 
RDTME 2.02, 
RDTME 3.03, and 
SDS 2.02 B 
C SDS 2.01 AIN-1 
D SDS 2.04 AIN-1 
Short Description 
Rockfall and vibratory loading effects on the mechanical failure of cladding, 
and methodology used to implement the effects of seismic effects on cladding 
Documentation of preclosure seismic design inputs, preclosure seismic 
design methodology, and features, events, and processes related to 
potentially disruptive igneous or seismic activity 
Documentation of expert elicitation process 
Documentation of seismic fragility curves and seismic risk analysis 
For the postclosure period, 10 CFR 63.114 requires that DOE conduct a performance assessment 
that consider only events that have at least 1 chance in 10,000 of occurring over 10,000 years. 
No. 14: Seismic Events 
Revision 1 
June 2004 1-3 Seismic events that have this probability of occurrence have been analyzed and are discussed in 
this technical basis document. In addition, 10 CFR 63.102(j) provides that, for the postclosure 
period, the event classes analyzed in the performance assessment should consist of all possible 
specific initiating events that are caused by a common natural process (e.g., the event class for 
seismicity includes the range of credible earthquakes for the Yucca Mountain site). Additional 
studies are currently being conducted to ascertain constraints on credible maximum ground 
motion, and results of those studies are not discussed in this document. The additional studies do 
not affect the technical bases or conclusions described in this document. The results of those 
studies will be included in the license application, as appropriate. 
1.3 SCREENING OF SEISMIC-RELATED FEATURES, EVENTS, AND PROCESSES 
The identification of seismic effects that could be significant to repository postclosure 
performance was accomplished as part of a comprehensive effort to identify and address the 
features, events, and processes (FEPs) that could affect repository performance. Each identified 
FEP is screened in or out of the total system performance assessment (TSPA), depending on the 
consequences and probability that have been assessed for that FEP. As documented in Features, 
Events, and Processes: Disruptive Events (BSC 2004b), the seismic-related FEPs are: 
• Tectonic activity—large scale 
• Fault displacement damages EBS components 
• Seismic ground motion damages EBS components 
• Seismic-induced rockfall damages EBS components 
• Seismic-induced drift collapse damages EBS components 
• Seismic-induced drift collapse alters in-drift thermal-hydrology 
• Seismicity associated with igneous activity 
• Hydrologic response to seismic activity 
• Seismic activity changes porosity and permeability of rock 
• Seismic activity changes porosity and permeability of faults 
• Seismic activity changes porosity and permeability of fractures 
• Seismic activity alters perched water zones. 
Each of these seismic-related FEPs and its screening status are discussed briefly. More detailed 
discussion and supporting references are provided in Features, Events, and Processes: 
Disruptive Events (BSC 2004b). 
Tectonic Activity—Large Scale–Large-scale tectonic activity includes regional uplift, 
subsidence, folding, mountain building, and other processes related to plate tectonics. These 
tectonic processes could alter the physical and thermal-hydrologic properties of the earth’s crust 
in the Yucca Mountain region. These changes, in turn, would have the potential to alter the 
groundwater flux through the repository and the amount of water contacting EBS components 
and, thereby, alter the performance characteristics of the repository and its engineered 
components. Yucca Mountain lies within the Walker Lane domain, an area of ongoing tectonic 
deformation (see Section 2.1.1). However, the rate of tectonic deformation since at least the 
beginning of the Quaternary Period, 1.8 million years ago, is too slow to significantly affect 
Yucca Mountain during the regulatory compliance period of 10,000 years. Therefore, this FEP 
1-4 June 2004 No. 14: Seismic Events 
Revision 1 Revision 1 
has been screened out on the basis of low consequence (BSC 2004b, Section 6.2.1.1) and is not 
further addressed in this report. 
Fault Displacement Damages EBS Components–This FEP describes a geologic fault that 
intersects waste-emplacement drifts within the repository and that undergoes movement during 
an earthquake. The EBS components experience related movement or displacement such that 
performance is degraded by component shearing due to vertical fault displacement or by drip 
shield separation due to tilting of components. This FEP has been screened in (BSC 2004b, 
Section 6.2.1.2) and is addressed in Section 5.5. 
Seismic Ground Motion Damages EBS Components–In this FEP, earthquake vibratory 
ground motion shakes the EBS components (drip shield, waste package, pallet, and invert). The 
vibration could directly damage the drip shields, waste packages, and waste package internals or 
cause impacts between EBS components that could lead to damage. Such damage could degrade 
the performance of the EBS components as barriers to radionuclide migration. This FEP has 
been screened in (BSC 2004b, Section 6.2.1.3) and is addressed in Section 5.4. 
Seismic-Induced Rockfall Damages EBS Components–Seismic ground motion has the 
potential, because of the additional imposed stresses, to cause blocks of rock to be shaken loose 
from the walls of emplacement drifts. The mechanical impact of rockfall could damage the drip 
shields. Drip shield damage, in turn, could lead to more seepage water contacting the waste 
packages. The waste packages would not be directly impacted by rockfall because of the 
protection afforded by the drip shields. This FEP has been screened in (BSC 2004b, 
Section 6.2.1.4) and is addressed in Section 5.3. 
Seismic-Induced Drift Collapse Damages EBS Components–In this FEP, stresses induced in 
the walls of the emplacement drifts by seismic ground motion lead to collapse of all or part of the 
drift. The resulting rubble rests partially upon the drip shields and imposes a static load. 
However, engineering calculations summarized in Section 5.3.2 of this report indicate that the 
resulting static loads would be insufficient to cause structural failure of the drip shields. The 
waste packages are not directly affected by drift collapse because of the protection afforded by 
the drip shields. The consequences of the dead weight of rockfall on drip shield corrosion will 
be discussed in Technical Basis Document No. 6: Waste Package and Drip Shield Corrosion. 
This FEP is screened out of TSPA on the basis of low consequences (BSC 2004b, 
Section 6.2.1.5). 
and is addressed in Section 4.7 of Technical Basis Document No. 3: Water Seeping into Drifts. 
Seismic-Induced Drift Collapse Alters In-Drift Thermal-Hydrology–Downwardly 
percolating groundwater encountering a waste-emplacement drift will be partly diverted around 
the drift opening because of the capillary barrier effect, which refers to the tendency of water to 
be held in the pores of unsaturated rock rather than drip into a large opening. As a result, the 
water seepage flux is expected to be smaller than the local percolation flux. This barrier effect is 
an attribute of the unsaturated zone at Yucca Mountain. Seismic activity could lead to drift 
collapse and resulting rockfall rubble throughout part or all of an emplacement drift. The change 
in drift profile and presence of rubble would reduce the effectiveness of the capillary barrier and 
lead to increased water seepage. This FEP has been screened in (BSC 2004b, Section 6.2.1.6) 
Analyses supporting that technical basis document indicate an approximately 2% to 5% increase 
June 2004 1-5 No. 14: Seismic Events Revision 1 
in seepage for worst-case changes to the drift profile. The presence of rubble also would affect 
thermal properties within the EBS. The presence of rubble around the drip shield will also cause 
changes in the temperature and relative humidity of EBS components. The magnitude of these 
changes has been estimated with thermal-hydraulic calculations using the multiscale model (BSC 
2004c). These calculations predict changes in temperature and relative humidity for eight 
different waste package emplacement configurations, using bounding (high or low) values for 
the thermal conductivity of the rubble surrounding the drip shield. The results from these 
calculations (DTN: LL040310323122.044) are included in the appropriate models for TSPA. 
Seismicity Associated with Igneous Activity–Volcanic eruptions are commonly preceded and 
accompanied by numerous earthquakes that indicate progressive rock failure as magma rises to 
the earth’s surface. Seismic events associated with igneous activity, like seismic events 
associated with tectonic activity, have the potential to disrupt the integrity of components of the 
EBS and thereby affect repository performance. This FEP has been screened in (BSC 2004b, 
Section 6.2.1.7) and is addressed in the probabilistic seismic hazard analysis (PSHA) described 
in this report (Section 3.2.1). 
Hydrologic Response to Seismic Activity–Seismic activity associated with fault movement 
could create new or enhanced flow pathways or connections between rock units, or it may 
change the rock stress and, therefore, fluid pressure within the rock. These effects have the 
potential to change hydrologic system attributes, such as the surface water and groundwater flow 
directions and water level. The effects on groundwater flow and radionuclide transport by 
seismic-induced changes to fracture systems have been investigated in sensitivity studies. The 
results indicate that changes in fracture aperture confined to fault zones have minimal effect on 
transport behavior in the unsaturated zone. Increased fracture aperture applied over the entire 
unsaturated zone domain results in effects that are no more significant than other uncertainties 
related to infiltration (in particular, uncertainties related to the range of future climatic 
conditions). A number of other studies on the effect of seismic activity on the water table level 
indicate that changes caused by seismic activity would be, at most, a few tens of meters and 
would not reach the repository level. These changes would be transient and would not 
significantly affect groundwater flow or radionuclide transport. Because hydrologic response to 
seismic activity does not provide a mechanism to significantly affect performance, this FEP has 
been screened out on the basis of low consequence (BSC 2004b, Section 6.2.1.8) and is not 
further addressed in this report. 
Seismic Activity Changes Porosity and Permeability of Rock–Earthquake fault displacement 
and vibratory ground motion have the potential to change rock stresses and create strains that 
affect flow properties. Specifically, seismic activity could cause strains that alter the 
permeability in the rock matrix. The net effects of rock matrix permeability changes are a 
temporary rise or decline of the water table level and changes in groundwater flow and 
radionuclide transport in the unsaturated zone. However, as discussed for the preceding FEP, 
these effects are not significant to repository performance and, therefore, this FEP has been 
screened out of TSPA on the basis of low consequence (BSC 2004b, Section 6.2.1.9). It is not 
further addressed in this report. 
Seismic Activity Changes Porosity and Permeability of Faults–Earthquake fault displacement 
and vibratory ground motion could cause movement along rock fractures and change stresses and 
June 2004 1-6 No. 14: Seismic Events Revision 1 
strains that alter the permeability along faults. The redistribution of strain could reactivate 
preexisting faults or generate new faults. These effects could alter or short-circuit the flow paths 
and flow distributions close to the repository and create new pathways through the repository, 
leading to decreased radionuclide transport times. However, the effects of changes to fracture 
systems due to geologic effects have been investigated in sensitivity studies. The results indicate 
that changes in fracture aperture confined to fault zones show virtually no effect on transport 
behavior in the unsaturated zone. In addition, an early, corroborative analysis of the effect of a 
fault on flow in the saturated zone suggests that there would be negligible impact on 
performance, even for fault hydraulic conductivities varying over five orders of magnitude. This 
FEP has been screened out on the basis of low consequence (BSC 2004b, Section 6.2.1.10) and 
is not further addressed in this report. 
Seismic Activity Changes Porosity and Permeability of Fractures–This FEP is similar to the 
preceding FEP, but it involves fractures rather than faults. Fractures are breaks in rock that, 
unlike faults, show little or no displacement parallel to the fracture surface. Earthquake fault 
displacement or vibratory ground motion could change stresses and strains that alter the 
permeability along fractures. Seismic activity could reactivate existing fractures or generate new 
fractures. Generation of new fractures and reactivation of existing fractures could change 
groundwater flow and radionuclide transport paths, alter or short-circuit the flow paths and flow 
distributions close to the repository, and create new pathways through the repository. These 
effects could decrease radionuclide transport times. The rock at Yucca Mountain is highly 
fractured, and existing fractures are likely to accommodate seismically induced stresses. 
Therefore, generation of new fractures as a result of seismic activity is likely to be negligible. 
And, as already noted, the effect on groundwater flow and radionuclide transport of changes to 
fracture systems due to geologic effects, such as fault displacement and seismic ground motion, 
have been investigated in sensitivity studies. The results indicate that increased fracture aperture 
applied over the entire unsaturated zone domain results in effects that are no more significant 
than other uncertainties related to infiltration. Therefore, this FEP is screened out on the basis of 
low consequence (BSC 2004b, Section 6.2.1.11) and is not further addressed in this report. 
Seismic Activity Alters Perched Water Zones–Strain caused by stress changes from seismic 
events could lower rock permeabilities, effectively sealing a volume of rock against groundwater 
flow and leading to the formation and persistence of perched-water zones (volumes of 
water-saturated rock above the general water table). The generation of a perched-water zone 
above the repository could potentially affect the flow of water to waste-emplacement drifts. This 
potential effect is addressed indirectly in the model for water seepage into drifts, which shows 
that rock heterogeneity will lead to focusing of water flow (see Technical Basis Document No. 3: 
Water Seeping into Drifts, Section 2.2). Below the repository, the potential to release perched 
water because of stress changes and fracture openings has the potential to result in a pulse of 
radionuclides, if the perched water contains radionuclides and is allowed to drain. However, 
considering the amount of water in the fracture domain below the repository and the 
radionuclides that could be contained in this water, such a pulse would cause a minimal change 
in the radionuclide mass flux at the water table. Therefore, this FEP has been screened out on 
the basis of low consequences (BSC 2004b, Section 6.2.1.12) and is not further addressed in this 
report. 
June 2004 1-7 No. 14: Seismic Events Revision 1 
1.4 ORGANIZATION AND APPROACH 
This report describes the information that supports the assessment of repository system 
performance, considering seismic disruption, as illustrated in Figure 1-2. 
Section 2 summarizes the data supporting the Yucca Mountain seismic hazard analysis. This 
information base includes: 
• Regional and local tectonic interpretations 
• Regional and local seismicity compilations (location, frequency, and magnitude of 
earthquakes) coming primarily from earthquake catalogs that document earthquakes 
recorded by seismic monitoring networks and records of earthquake felt effects 
• Regional and local geologic maps of earthquake faults 
• Paleoseismic reconstructions of the timing and magnitude of earthquakes on individual 
faults 
• Geodetic measurements of crustal strain rates in the Yucca Mountain region 
• Studies of how seismic waves propagate through the earth’s crust and attenuate with 
distance, based on the analysis of ground motion recordings. 
Section 3 summarizes the PSHA that was conducted for the Yucca Mountain site. The PSHA 
provides a structured framework for documenting the range of current scientific thinking about 
the factors that control seismic hazard and the associated uncertainties. In the Yucca Mountain 
PSHA study, seven seismologists and earthquake engineers who are experts in ground motion 
estimation evaluated available worldwide and regional strong ground motion recordings and 
other relevant data about seismic wave propagation in the Yucca Mountain region. Through an 
expert-elicitation process, these experts provided interpretations of ground motion levels and 
uncertainties for a range of magnitude, distance, and faulting style combinations. Their 
interpretations consisted of median values and associated standard deviations (variability due to 
the inherent randomness of earthquake processes). In developing their interpretations, the 
experts considered alternative ground motion models that are consistent with the available 
information. In evaluating those models, the experts gave more weight to those that they 
determined were closer to physical reality for Yucca Mountain. For both the median and 
standard deviation, the experts also provided assessments of uncertainty resulting from the 
inability of the available data to discriminate among the possible models. 
A similar process was followed in the PSHA study for the identification of seismic sources. 
Seismic sources are defined as areas or volumes of the earth’s crust that can be characterized by 
a specified relationship between earthquake recurrence rates and earthquake magnitude and by a 
probability distribution for maximum earthquake magnitude. In this case, six expert teams, 
rather than individual experts, provided the interpretations. Each expert team consisted of three 
earth scientists who collectively had sufficient expertise in seismology, geology, geophysics, and 
tectonics to be able to interpret seismic source zones based on the available information. 
June 2004 1-8 No. 14: Seismic Events Revision 1 
Following an expert elicitation process, each team developed alternative seismic-source 
interpretations and weights for those interpretations. Uncertainty in current scientific 
understanding of seismic sources is expressed by the range of interpretations and the assigned 
weights. Variability in earthquake location, timing, and magnitude due to the random nature of 
earthquake occurrences is expressed by random variables that are specified in each 
interpretation. 
The six expert teams also developed interpretations of earthquake fault displacement. The teams 
settled on two basic, alternative approaches. The first, called the “earthquake approach,” 
parallels the PSHA formulation for ground shaking hazard and relates the occurrence of 
displacement on a geologic feature to the magnitude and distance of earthquakes in the site 
region. The feature in question may be a fault, a minor shear, a fracture, or unbroken ground. 
The second, called the “displacement approach,” utilizes the geologic history of fault 
displacement observed at the site of interest to quantify the hazard, without invoking a specific 
mechanism for its cause. 
The final ground motion products of the PSHA study are plots of the annual rate of exceeding a 
particular measure of vibratory ground motion (e.g., peak acceleration) versus ground motion 
level. These probability distributions are referred to as “hazard curves.” There is one hazard 
curve for every combination of alternative seismic-source interpretations (from each expert 
team) and every alternative ground motion interpretation (from each ground motion expert). The 
weight assigned to each hazard curve is the combined weights of the associated source 
interpretation and ground motion interpretation. The hazard curves express the variability in 
ground motion due to the inherent randomness of earthquake processes, and the range of hazard 
curves expresses scientific uncertainty due to limits on available information. The variability 
and uncertainty are both carried forward into the TSPA and into the development and sampling 
of multiple ground motion time histories for postclosure rockfall and structural response 
calculations. Variability and uncertainty are also incorporated into the development of time 
histories for use in preclosure seismic design and safety analysis. For any annual rate of 
exceedance, the mean hazard curve can be deaggregated to identify the magnitudes and distances 
of those earthquakes contributing to the hazard. 
The final fault-displacement products of the PSHA study are plots of the annual rate of 
exceeding particular fault-displacement levels. The fault-displacement hazard curves are defined 
for 17 different Yucca Mountain site locations and conditions. The site locations include 
locations on the known major faults, and the conditions specify a range of generic conditions 
(e.g., a small fault with a 2-m offset located somewhere between the Solitario Canyon Fault and 
Ghost Dance Fault). The range of locations and conditions is intended to cover the locations of 
potential interest at the Yucca Mountain site. 
June 2004 1-9 No. 14: Seismic Events Revision 1 
Figure 1-2. Components of the Assessment of Seismic System Performance 
June 2004 1-10 No. 14: Seismic Events Revision 1 
Section 4 describes the site-specific ground motion model and results. The ground motion 
hazard curves that resulted from the PSHA study apply to a hypothetical rock outcrop that has 
the same seismic-wave propagation properties as the rock at the repository horizon inside Yucca 
Mountain. The hazard curves at this reference rock location have to be modified to reflect 
site-specific rock properties for application to the actual locations of subsurface and surface 
facilities. The seismic site response model accounts for the effects of the approximately 300 m 
of rock overburden for the waste emplacement drifts and the effects of near-surface rock and soil 
at surface locations. The seismic response model also accounts for uncertainties in 
measurements of the site-specific soil and rock properties. 
The site response model generates measures or descriptors of the level of vibratory ground 
motion for different annual exceedance rates at specified locations. For example, one product of 
the site response model is the peak ground velocity that has an annual exceedance rate of 
1.0 × 10-6 at the repository horizon. The locus of values for different exceedance rates defines 
the peak ground velocity hazard curve at the specified location. For engineering analyses for 
which ground motion time histories are required, the site response model is used to generate 
suites of ground motion time histories representative of ground motion that would be expected 
from earthquakes in the contributing magnitude and distance ranges for a specified annual 
exceedance rate. Suites of time histories with considerable variability between the time histories 
are provided to model the natural variability in earthquake ground motion that results from the 
inherent randomness of earthquake processes. 
The PSHA ground motion and fault-displacement hazard curves and the outputs of the site 
response model are equally applicable to the preclosure period of repository operations 
(approximately 100 years) and to the postclosure regulatory compliance period (10,000 years). 
This report focuses on their use in assessments of postclosure seismic system performance. For 
descriptions of their use in preclosure seismic design and safety analyses, see Development of 
Earthquake Ground Motion Input for Preclosure Seismic Design and Postclosure Performance 
Assessment of a Geologic Repository at Yucca Mountain, NV (BSC 2004d), Preclosure Seismic 
Design Methodology for a Geological Repository at Yucca Mountain (BSC 2004a), and 
Preliminary Seismic Analysis for Preclosure Safety Analysis (BSC 2003a). 
Section 5 summarizes the calculated effects of seismic loading on the waste emplacement drifts 
and other EBS components. As noted in the discussion of seismic-related FEPs (Section 1.3), 
seismically induced stresses in the walls of the emplacement drifts can cause blocks of rock to be 
shaken loose and lead to accelerated drift degradation and rockfall. The mechanical impact of 
rockfall on the drip shields is not expected to cause immediate mechanical rupture but, rather, to 
leave residual stresses in the titanium plates of the drip shields. Similarly, vibratory ground 
motion is not expected to lead to immediate tensile failure but could, at very low probability 
levels, cause significant separation between adjacent drip shields. Severe, low-probability 
ground motion also could cause impacts between adjacent waste packages, between the waste 
package and its emplacement pallet, and between the waste package and the drip shield. These 
impacts are not calculated to cause an immediate puncture or tear in the barrier but would impose 
residual stresses in the Alloy 22 (UNS N06022) outer shell of the waste package. Given 
calculated residual stress levels in the drip shields and waste packages and a set of failure criteria 
that specify the residual tensile stress levels at which accelerated corrosion is expected to occur, 
the EBS seismic effects analysis then calculates the percent of the surface areas of the drip 
June 2004 1-11 No. 14: Seismic Events Revision 1 
shields and waste packages that fail due to accelerated corrosion. In TSPA, immediate drip 
shield and waste package failure is assumed once their residual stress thresholds are exceeded, 
resulting in a network of through-wall stress corrosion cracks in the failed areas that exceed the 
stress threshold. On the drip shield, this network of cracks is expected to plug from evaporationinduced 
precipitation of calcite and other minerals in the groundwater, thereby preventing 
advective flow through the cracks. On the waste package, the crack network is assumed to 
always allow diffusive transport of radionuclides through the effective area of the network. The 
capacity of the crack network to support advective flow through the waste package is currently 
under investigation. Analogous to the consequence analysis for seismic loading from vibratory 
ground motion, the EBS seismic effects analysis also calculates failed waste package surface 
areas from extremely low probability fault-displacement events. 
The EBS seismic effects analysis input to the TSPA is in the form of seismic damage 
abstractions. Based on the results of the EBS seismic effects analysis, the abstractions provide 
probability distributions for waste package and drip shield failed surface areas as functions of 
peak ground velocity. As described in Section 6, in each run or realization of the TSPA a peak 
ground velocity value is sampled randomly from the probability distribution defined by the peak 
ground velocity hazard curve for the repository horizon. This hazard curve is determined from 
the PSHA results and the ground motion site-response modeling. Also, in each realization, a 
distribution on upper-bound peak ground velocity is sampled. This distribution represents 
uncertainty in physical constraints on the levels of peak ground velocity that can be attained at 
Yucca Mountain. In each realization, if the sampled peak ground velocity exceeds the sampled 
upper-bound peak ground velocity, its value is reset to the value of the upper bound. Given this 
peak ground velocity, the TSPA then samples from the corresponding probability distributions 
for waste package and drip shield failed surface areas. These sampled seismic damage states, 
along with sampled values from probability distributions for other model inputs, are then used by 
the TSPA to calculate a dose to the reasonably maximally exposed individual, as defined by 
NRC regulations at 10 CFR 63.312. The TSPA is rerun many times to generate a statistically 
valid estimate of the mean annual dose. The mean annual dose is compared to the individual 
dose standard at 10 CFR 63.311 and is the primary measure of the effect of seismic events on 
postclosure performance of the repository. Additional measures of performance are projected 
radiation levels from specific radionuclides in a representative volume of groundwater, as 
defined by 10 CFR 63.332, which must meet groundwater protection standards at 
10 CFR 63.331. 
Section 7 provides conclusions regarding low-probability seismic events, and references are 
listed in Section 8. 
1.5 SUMMARY OF CONCLUSIONS 
This report documents the case for the following six key conclusions about low-probability 
seismic events: 
1. Much is known about seismic hazards at Yucca Mountain. The geology and 
seismology of the site have been studied intensively for more than 20 years. 
June 2004 1-12 No. 14: Seismic Events Revision 1 
2. Potentially significant seismic effects on repository performance have been considered 
and addressed, as appropriate. 
3. The seismic hazard analysis that has been carried out for Yucca Mountain accounts for 
both the variability in vibratory ground motion and fault displacement due to the 
inherent randomness of earthquake processes and the uncertainty due to limitations in 
scientific knowledge. 
4. Large ground motion predicted by the PSHA at annual exceedance probabilities of 
1.0 × 10-6 and below overestimates the severity of low-probability ground motion at 
Yucca Mountain. This overestimation of ground motion is being addressed through 
determination of constraints on maximum ground motion imposed by the 
stress-release characteristics of seismic sources and by limits on strain that can be 
propagated by seismic waves through rocks at Yucca Mountain. The constraint on 
maximum ground motion will be incorporated in the abstraction of seismic 
consequences that feeds TSPA. 
5. Seismic and tectonic effects on the natural systems at Yucca Mountain will not 
significantly affect repository performance. 
6. The EBS components are robust under seismic loads and will provide substantial 
protection of the waste form from seepage water, even under severe seismic loading. 
1.6 NOTE REGARDING THE STATUS OF SUPPORTING TECHNICAL 
INFORMATION 
This document was prepared using the most current information available at the time of its 
development. This document and its appendices providing KTI agreement responses were 
prepared using preliminary or draft information and reflect the status of the Yucca Mountain 
Project’s scientific and design bases at the time of submittal. In some cases, this involved the 
use of draft analysis and model reports and other draft references whose contents may change 
with time. Information that evolves through subsequent revisions of the analysis and model 
reports and other references will be reflected in the LA as the approved analyses of record at the 
time of LA submittal. Consequently, the Yucca Mountain Project will not routinely update 
either this technical basis document or its KTI agreement appendices to reflect changes in the 
supporting references prior to submittal of the LA. 
June 2004 1-13 No. 14: Seismic Events INTENTIONALLY LEFT BLANK 
1-14 No. 14: Seismic Events 
Revision 1 
June 2004 Revision 1 
2. DATA SUPPORTING SEISMIC HAZARD ANALYSIS 
The assessment of seismic hazards at Yucca Mountain focuses on characterizing the potential 
vibratory ground motion and fault displacement that will be associated with future earthquake 
activity near the site. The PSHA was conducted based on the evaluation of a large set of data 
pertaining to earthquake sources, fault displacement, and ground motion propagation in the 
Yucca Mountain region (see Section 3). Tectonic models that have been proposed for the Yucca 
Mountain area and information from analog sites in the Basin and Range Province provide the 
basis to characterize the patterns and amounts of fault displacement. The data set also contains 
information providing the history of prehistoric earthquakes on nearby Quaternary faults (faults 
active during the last 1.8 million years). The historical earthquake record and information on the 
attenuation of ground motion are also important components of this data set. 
This section summarizes the data sets and interpretations that support the PSHA for Yucca 
Mountain. The section begins by describing the seismotectonic framework of the Yucca 
Mountain region, followed by descriptions of historical seismicity and prehistoric earthquakes, 
followed by a discussion of data and analyses carried out to understand seismic ground motion 
and its attenuation in the region surrounding Yucca Mountain. The discussion is intended to 
summarize conclusions regarding these topics. A more detailed description is provided in the 
Yucca Mountain Site Description (Simmons 2004) and reference is made therein to associated 
documentation. 
Site characterization activities that have resulted in data for understanding vibratory ground 
motion and fault displacement hazards at Yucca Mountain have included: 
• Compilation of a historical catalog of earthquakes to support analyses of earthquake 
recurrence rate and magnitude distribution 
• Establishment of a network of seismometers and strong-motion instruments to monitor 
and characterize contemporary seismicity 
• Reconnaissance geologic surveys of known and suspected Quaternary faults within 
100 km of the site to characterize their extent and rates of activity 
• Geophysical surveys to identify and characterize the orientation of subsurface faults 
• Geologic mapping of faults and geologic units in the local Yucca Mountain area 
• Paleoseismic studies of known and suspected Quaternary faults in the immediate vicinity 
of Yucca Mountain to provide information on past earthquakes, including their number, 
size, extent, and timing 
• Geodetic monitoring to understand rates of regional deformation 
2-1 June 2004 No. 14: Seismic Events Revision 1 
• Analysis of ground motion data from local earthquakes to evaluate the local attenuation 
of seismic waves 
• Analysis of ground motion data from extensional tectonic regimes to provide 
information on the regional rate of attenuation. 
The data sets that have been developed as a result of these site characterization activities over the 
past 20 years provide a strong basis for the PSHA interpretations (Section 3). 
2.1 SEISMOTECTONIC FRAMEWORK 
The assessment of earthquake hazards is a function of the seismotectonic framework of the 
region and vicinity of Yucca Mountain. The seismotectonic framework is characterized by the 
geologic history and structures that are present in the region, such as faults; the nature of tectonic 
processes and stresses that are currently operating; the seismicity that has been observed during 
the historical period of observation; and the location and rate of activity on regional and local 
faults. Understanding these processes and their rates of occurrence provides a fundamental basis 
for assessing seismic hazards. This section summarizes the understanding of the seismotectonic 
framework and the associated geologic and seismologic data sets that have been developed to 
improve that understanding. This information is given in the discussions of seismotectonic 
framework and characterization of faulting by Whitney (1996) and in the Yucca Mountain Site 
Description (Simmons 2004, Sections 3 and 4). 
A complex history of faulting and volcanism has occurred at Yucca Mountain over the past 
several million years. Because the regional tectonic setting of Yucca Mountain (i.e., the Great 
Basin) presently is undergoing deformation, faulting can be expected to continue to occur near 
Yucca Mountain during the present tectonic regime. Tectonic models can represent the geologic 
structure of a portion of the earth’s crust and illustrate how the structures have evolved through 
time. A tectonic model incorporates the geometry of and the spatial relations among structures, 
such as faults, and the mechanisms by which structures interact and respond to regional stress. 
When tectonic models seek to demonstrate the relations among processes of deformation and to 
include the pattern of historical earthquakes, they can provide insights into the seismotectonics of 
a region. 
A number of alternative tectonic models for Yucca Mountain have been proposed to explain 
observed geologic structures and geophysical data in light of the history of volcanism and fault 
movement, uplift and subsidence, and lateral extension (Simmons 2004, Section 4.1.2). These 
models vary from detachment fault models involving extension to lateral-shear pull-apart 
models. The focus of the assessment of seismic hazards, as expressed in a “seismotectonic 
framework,” is the assemblage of contemporary geologic structures and crustal stresses that will 
give rise to future seismicity. Resolving which tectonic model is “correct” is not necessary for 
conducting a seismic hazard analysis. The models are based on the data and observations 
regarding the location of structures, the orientation of tectonic stresses, rates of deformation, and 
observed seismicity. The assessment of seismic sources for the PSHA began first with a careful 
consideration of these data sets and the various tectonic models that have been derived from 
them (see Section 3). These data and observations are summarized below. 
2-2 June 2004 No. 14: Seismic Events Revision 1 
2.1.1 Regional Tectonic Setting 
“Tectonic setting” refers to the geological architecture or structural configuration of the different 
rock masses in the area. The overall tectonic setting, the Great Basin, generally consists of 
fault-bounded basins and mountain ranges (including Yucca Mountain) complicated by volcanic 
activity that has occurred within the past 15 million years. Typically, faults in this setting 
include normal (extensional displacement perpendicular to the fault trace) and strike-slip (lateral 
displacement parallel to the fault trace) faults that reflect the extensional deformation caused by 
plate tectonic interactions at the western margin of the North American continent during the 
middle and late Cenozoic Era (65 Ma to present). The Great Basin is segmented into tectonic 
domains, structurally bounded blocks of the earth’s crust characterized by deformations that 
distinguish them from adjacent domains. Three regional tectonic domains characterize Yucca 
Mountain and its surrounding environs: the Walker Lane domain, which includes the site; the 
Basin and Range domain to the northeast; and the Inyo-Mono domain to the southwest 
(Figures 2-1 and 2-2). 
Yucca Mountain lies within the Walker Lane domain, an approximately 100-km-wide structural 
belt along the western side of the Basin and Range domain (Figure 2-2). The domain extends 
northwestward from the vicinity of Las Vegas, subparallel to the Nevada-California border, into 
northern California. The domain is characterized as an assemblage of crustal blocks separated 
by discontinuous northwest-striking right-lateral faults (with rock movement to the right) and 
northeast-striking left-lateral faults (Stewart 1988, p. 686; Carr 1990, p. 284). Because of its 
structural heterogeneity, the Walker Lane is recognized as a tectonic terrane distinct from the 
Basin and Range at a regional scale. 
The geologic setting of Yucca Mountain is characterized structurally by two distinctly different 
tectonic deformation styles: an earlier compressional orogenic, or mountain-building, style of 
regional folding and overthrust faulting and a later extensional “basin forming” style of regional 
normal and strike-slip faulting. The compressional style records orogenic events that occurred 
primarily during Paleozoic (544 to 248 Ma) deposition, followed by a peak event that occurred in 
the Mesozoic (248 to 66 Ma) and terminated marine deposition. Studies of the extensional 
tectonics in the central Basin and Range (Snow and Wernicke 2000, p. 659) conclude that from 
250 to 300 km of extension have occurred by a west-northwest motion of the Sierra Nevada 
block away from the Colorado Plateau, at rates initially as great as 2 cm/yr and at 1.5 to 1 cm/yr 
during the last 5 Ma. Research suggests that most of the current extension, as indicated by strain 
measurements and seismicity, is concentrated along the eastern and western margins of the Basin 
and Range (Thatcher et al. 1999, Figure 1; Martinez et al. 1998, p. 569). 
The main extensional features within the tectonic setting of Yucca Mountain were established by 
about 15 Ma, namely a basin and range structural pattern defined chiefly by north-south-oriented 
basins or troughs and fault zones associated with the Walker Lane, including the Rock Valley 
fault zone. The culminating tectonic event in the geologic evolution of the region and, 
coincidentally, the initiating event for the structural formation of Yucca Mountain was the 
creation of the southwestern Nevada volcanic field. This field was produced by a succession of 
at least five voluminous and numerous smaller eruptions that occurred over 7.5 Ma, from about 
15 to 7.5 Ma. The greatest of these eruptions created the volcanic sequence (the Paintbrush and 
Timber Mountain groups) of which Yucca Mountain is a part. From about 11 to 7 Ma, the style 
June 2004 2-3 No. 14: Seismic Events Revision 1 
of tectonic deformation in the Yucca Mountain region became more clearly one of narrow basin 
subsidence, possibly accompanied by adjacent range uplift. 
Source: Simmons 2004, Figure 2-1. 
Figure 2-1. Approximate Locations of the Physiographic Subdivisions of the West-Central Southern 
Great Basin 
June 2004 2-4 No. 14: Seismic Events Revision 1 
Source: Simmons 2004, Figure 2-3. 
NOTE: FC = Furnace Creek Fault, G = Garlock Fault, OV = Owens Valley Fault, RV = Rock Valley Fault. 
2.1.2 Site Geology 
Volcanic rocks called tuffs dominate the exposed bedrock formations in the Yucca Mountain site 
area (Figure 2-3). The deposits consist mostly of pyroclastic flows (high-density mixture of hot, 
dry rock fragments and hot gases) and fallout tephra (fragments that traveled through the air), 
with minor lava flows (molten rock). The volcanic rocks were deposited mainly between 11 and 
14 million years ago, during the Miocene Epoch of the Tertiary Period. The stratigraphy of these 
rocks forms the basic framework for the modeling and analyses of rock properties, mineral 
Figure 2-2. Regional Tectonic Domains for Yucca Mountain and Surrounding Environs, plus Zones of 
Historical Seismic Activity 
June 2004 2-5 No. 14: Seismic Events Revision 1 
distributions, faulting and fracturing, hydrologic flow, and radionuclide transport. Descriptions 
of the stratigraphic units are given in the Yucca Mountain Site Description (Simmons 2004, 
Section 3.3.4, Table 3-1). 
For seismic hazard analysis, the knowledge of the site geologic units and their physical 
properties is important principally in characterizing the site conditions for ground motion (see 
Section 4) and for consideration of rockfall during seismic shaking (BSC 2003b). The dynamic 
response of the rock in the emplacement drifts during seismic shaking is expected to vary 
according to whether the drift lies within the lithophysal or nonlithophysal members of the 
volcanic rock (see Section 5). Lithophysae are spherical-to-oblate cavities in the volcanic rock 
formed during their deposition. Presence or absence of lithophysae is one mechanism for 
subdividing formations and members of formations at Yucca Mountain. This is particularly true 
within the Topopah Spring Tuff, where the repository emplacement drifts will lie, and the 
overlying Tiva Canyon Tuff. As discussed in Drift Degradation Analysis (BSC 2003b), the 
mechanical response of the drifts and associated rockfall varies with whether the drift lies within 
lithophysal or nonlithophysal units. Likewise, the effects on the EBS from seismically induced 
rockfall will vary as a function of this location (see Section 5.3). 
2.1.3 Contemporary Deformation 
Large earthquakes on range-front faults during the past 100 years, such as the Dixie Valley and 
Fairview Peak earthquakes, indicate that Basin and Range extension is still underway. Epicenter 
distribution patterns and geodetic strain data indicate that strain presently is concentrated 
primarily north of Yucca Mountain, in a zone along latitude 37°N, in the eastern California shear 
zone, and in the central Nevada seismic zone (Bennett, Davis et al. 1999, p. 373, Figure 1) 
(Figure 2-2). High geodetic extension rates characterize these active areas (Bennett, Wernicke, 
et al. 1998, p. 566; Savage et al. 1995, p. 20266). Northwest motion of the Sierra Nevada block 
is accomplished by a combination of east–west extension on north-striking normal faults and by 
right-lateral motion on northwest-striking strike-slip faults of the Walker Lane and eastern 
California shear zone (Dixon et al. 1995, p. 762). 
June 2004 2-6 No. 14: Seismic Events Revision 1 
Source: Potter et al. 2002. 
NOTE: Faults are shown with solid lines, although large segments of some are concealed or inferred beneath 
Quaternary deposits. ECRB = Enhanced Characterization of the Repository Block. 
Figure 2-3. Map of Yucca Mountain Site Area Showing Distribution of Principal Stratigraphic Units, 
Major Faults, and Locations of Geographic Features 
June 2004 2-7 No. 14: Seismic Events Revision 1 
Strain surveys show that the direction of extension in the Great Basin is toward the northwest, 
the direction of the least compressive stress. The northern Basin and Range appears to be 
moving at 4.9 ± 1.3 mm/yr west-southwest with respect to the continental interior and the 
southern Great Basin by means of crustal extension (Savage et al. 1995, p. 20265). The 
relatively high strain rate of the northern Basin and Range is at least partly accommodated by the 
central Nevada seismic zone (Figure 2-2). Trilateration (measuring changes in distances 
between three or more points) data show that at least 2.7 mm/yr of extension is taken up by this 
zone (Savage et al. 1995, p. 20265). The southern Basin and Range has an extension rate of 
3 mm/yr or less (Sauber 1989, p. 123). The zone of seismicity along latitude 37ºN 
accommodates the differential extension between the northern and southern Basin and Range 
provinces; the slip rate in this zone is estimated at about 3.2 mm/yr (Savage et al. 1995, 
p. 20267). The high strain rate zones, such as the central Nevada seismic zone, may represent 
concentrations of active deformation among relatively stable crustal blocks. 
The kinematic boundary condition for Basin and Range deformation (the relative motions of the 
Pacific and North American plates) has been nearly constant for at least the past 3.4 Ma (Harbert 
and Cox 1989, p. 3061). During this time, tectonic activity has gradually shifted westward, from 
the Death Valley–Furnace Creek Fault to the Owens Valley Fault. The Walker Lane likely 
accommodates significant right-lateral shear. The central Nevada seismic zone trends obliquely 
across the older Walker Lane. Therefore, it would seem that the westward migration of 
tectonism in the Inyo-Mono domain and the historical surface-rupturing earthquake activity 
along the central Nevada seismic zone represent a concentration of crustal strain of regional 
extent and significant longevity. This strain zone appears to be shifting westward and perhaps 
northward away from Yucca Mountain. 
2.1.4 Regional and Local Faults 
The structural geology of Yucca Mountain and vicinity is dominated by a series of north-striking 
normal faults (Figure 2-4) along which Tertiary volcanic rocks were tilted eastward and 
displaced hundreds of meters. Movement occurred primarily during a period of extensional 
deformation in middle to late Miocene time but has continued at a low level into the Quaternary 
Period (the last 1.8 million years). These through-going faults divided the site area into several 
blocks, each of which is further deformed by minor faults. Block-bounding faults within the 
Yucca Mountain site area are spaced 1 to 6 km apart, and include, from east to west, the 
Paintbrush Canyon, Bow Ridge, Solitario Canyon–Iron Ridge, Fatigue Wash, and Windy Wash 
faults (Figures 2-4 and 2-5). Fault scarps commonly dip from 50° to 80° to the west. 
Displacements are mainly dip-slip (direction of slip down the fault plane), down to –the west, 
with subordinate strike-slip or oblique-slip components of movement exhibited along some 
faults. Numerous intrablock faults occur within the individual structural blocks, representing 
local adjustments in response to stress created, for the most part, by the large displacements that 
took place along block boundaries (Simmons 2004, Section 3.5). 
June 2004 2-8 No. 14: Seismic Events Revision 1 
Source: Potter et al. 2002. 
NOTE: All faults are shown with solid lines, although many segments are concealed or inferred. Symbols and 
abbreviations: bar and bell, downthrown side of fault; arrows, relative direction of strike-slip movement; 
ESF = Exploratory Studies Facility; CD = ECRB Cross-Drift. 
Site area defined by black border. Blue line is the approximate line of cross section given in Figure 2-5. 
Figure 2-4. Distribution of Faults in the Yucca Mountain Site Area and Adjacent Areas to the South and 
West 
June 2004 2-9 No. 14: Seismic Events June 2004 
Source: Day et al. 1998, cross section B-B'. 
NOTE: Approximate line of section is also shown on Figure 2-4. 
Figure 2-5. East–West Structure Section across Yucca Mountain Site Area 
2-10 No. 14: Seismic Events 
Revision 1 Revision 1 
From the standpoint of seismic hazard analysis, Quaternary faults are important potential sources 
of earthquakes. The assessment of the age of faults is made using a combination of geomorphic 
analysis, mapping of Quaternary deposits, local mapping of faulting in existing exposures, and 
exploratory trenching to examine the relationships between faulting and Quaternary deposits. 
Figure 2-6 shows the faults near Yucca Mountain that are assessed to have been active during the 
Quaternary Period. 
Several of the block-bounding faults at Yucca Mountain show evidence of Quaternary 
displacements that influenced depositional patterns of surficial materials on hillslopes and on 
adjacent valley or basin floors and in places produced visible scarps in surficial deposits along 
some fault traces. However, low rates of offset and long recurrence intervals between successive 
faulting events on faults in the site area during Quaternary time have resulted in subtle 
landforms. 
Information on the stratigraphic relations among surficial deposits was augmented considerably 
by the large-scale mapping and detailed descriptions of vertical sequences that were freshly 
exposed in trenches excavated, or enhanced stream or roadcut exposures, across those faults 
suspected of being active during Quaternary time (Whitney and Taylor 1996). Although studies 
concluded that there is no evidence of Quaternary displacement, trenches were also excavated 
across the Ghost Dance Fault because of its location relative to the area of waste emplacement. 
Within the site area, 52 trenches were emplaced across such faults to determine the extent of 
Quaternary tectonic activity with respect to (1) the number, amount of displacement, and age of 
individual surface-rupturing events causing earthquakes; (2) fault slip rates; and (3) recurrence 
intervals between successive events on a given fault. Figure 2-6 shows the locations of those 
trenches in the site area that exhibited evidence of Quaternary age displacement or, in the case of 
the Ghost Dance Fault, fractures with no displacement that might be associated with Quaternary 
activity. Trenches were also excavated across the Bare Mountain Fault and Rock Valley Fault to 
determine activity on the more significant regional faults. These assessments, which combine 
the mapping of Quaternary deposits at the surface with the characteristics of faulting observed in 
subsurface trenches, are called paleoseismic investigations. The number of trenches and the 
level of detailed investigation of faults in the local Yucca Mountain area provide a database for 
interpretations of the paleoseismic behavior of the local faults. Paleoseismic investigations and 
associated data are described in the Yucca Mountain Site Description (Simmons 2004, Section 
4.3.2) and in more detail by Whitney and Taylor (1996). These paleoseismic data serve as a 
fundamental resource for the seismic source characterization activity in the PSHA (see 
Section 3.2). An example log of a mapped enhanced exposure cut across the Paintbrush Canyon 
Fault is shown in Figure 2-7. 
June 2004 2-11 No. 14: Seismic Events Source: Fault map is from Simonds et al. 1995; Pezzopane, Whitney et al. 1996. 
NOTE: Only trenches that exhibited evidence of Quaternary age displacements are shown. In addition, trenches 
across the Ghost Dance Fault that exhibited fractures in the Quaternary deposits, but not displacement, are 
shown. 
Figure 2-6. Local Faults and Paleoseismic Study Sites at Yucca Mountain 
2-12 No. 14: Seismic Events 
Revision 1 
June 2004 Revision 1 
Figure 2-7. Example Log from Paleoseismic Studies of Late Quaternary Faults near Yucca Mountain: 
Paintbrush Canyon Fault at Busted Butte 
Source: Menges and Whitney 1996, Figure 4.4-1. 
Geophysical surveys have also contributed to the understanding of local and regional faults in the 
Yucca Mountain vicinity (Oliver et al. 1995; Majer et al. 1996). Geophysical techniques 
employed at Yucca Mountain include seismic reflection (e.g., Brocher et al. 1998) and gravity 
and magnetics (e.g., Ponce and Langenheim 1994; Ponce 1996). These investigations provided 
information on the subsurface geometry of faults and on the location of buried faults with no 
surface expression. 
2.2 HISTORICAL SEISMICITY OF THE YUCCA MOUNTAIN REGION 
Seismic hazard evaluations rely on a description of the temporal and spatial distribution of 
earthquakes, their magnitudes, and how they relate to the seismotectonic processes of the region. 
The temporal and spatial occurrence of earthquakes for a given region is evaluated from 
two sources: historical (instrumental records and reported effects) and prehistoric (paleoseismic 
data). 
Seismic monitoring of the southern Great Basin began in the early 1900s with isolated stations 
installed and operated by the University of California, Berkeley, and the University of Nevada, 
2-13 June 2004 No. 14: Seismic Events Revision 1 
Reno. The history of seismic monitoring in the Southern Great Basin is found in the Yucca 
Mountain Site Description (Simmons 2004, Section 4.3.1.1) and Seismicity in the Vicinity of 
Yucca Mountain, Nevada, for the Period October 1, 1995, to September 30, 1996 (von Seggern 
and Smith 1997). In the late 1950s and early 1960s, a global network (the Worldwide 
Standardized Seismograph Network) came into existence, thereby providing the capability to 
record earthquakes larger than about magnitude 4 in the southern Great Basin. Later, networks 
of stations were installed to monitor and study specific areas such as the Nevada Test Site and 
portions of the western United States. 
Seismic monitoring specifically to characterize the Yucca Mountain region began when a 
47-station network was installed in 1978 and 1979. The network extended to a distance of 
160 km (99 mi) from the site (Figure 2-8a) and employed the analog technology of the day. In 
1981, six seismograph stations in the immediate vicinity of Yucca Mountain were added to the 
network. In 1995, seismic monitoring near Yucca Mountain was reconfigured with the 
establishment of a network of three-component seismograph systems with digitally recorded and 
telemetered data. The Southern Great Basin Digital Seismic Network, which is operated by the 
University of Nevada, Reno, spans a radius of 50 km (31 mi) from Yucca Mountain and, in 
2001, consisted of 30 stations (Figure 2-8b). The higher quality data obtained from the digital 
network of three-component seismograph systems have enabled the detection of smaller 
magnitude earthquakes and have resulted in improved accuracy of earthquake locations. 
In addition to the seismograph network, 27 strong motion stations consisting of three-component 
accelerometers also monitor Yucca Mountain. These stations are designed to record larger 
ground motion from nearby moderate to large earthquakes. Seventeen of the three-component 
strong-motion stations are located at the surface, one is sited in Alcove 5 of the Exploratory 
Studies Facility (ESF), and nine are located in three boreholes in the surface facilities area near 
the North Portal of the ESF. For the boreholes, in each borehole one station is located at the 
surface, one at mid-depth, and one at the bottom. 
2.2.1 Historical Seismicity Catalog for the Yucca Mountain Region 
In preparation for the PSHA, a catalog of earthquakes was compiled for the region within 
300 km of Yucca Mountain using both historical and instrumental records (CRWMS M&O 
1998, Appendix G). The resulting combined catalog contains 271,223 earthquakes occurring 
from 1868 to 1996. Earthquakes with magnitudes down to approximately M 1 were recorded in 
the more recent periods as the sensitivity of local and regional seismic networks was 
substantially increased. Figure 2-9 shows events in the catalog with magnitude greater 
than moment magnitude (Mw) 3.5. The accuracy of information in the historical catalog is 
affected by several variables (e.g., accuracy of historical accounts, detection capability, and 
instrumental precision), especially the variability in seismic network coverage as a function of 
time. The spatial distribution of seismicity in the 300 km catalog depends both on the density of 
inhabitants and the density of seismographic network coverage in a particular region over time 
and is somewhat an artifact of the more thoroughly represented aftershock sequences of the 
modern period. For example, a significant portion of the catalog in terms of number of events is 
derived from a recent series of moderate-sized earthquakes, aftershock sequences, and volcanic 
swarm activity in the Mammoth Lakes, California, area. Although these events figure 
prominently in terms of the number of earthquakes in the catalog, they represent an insignificant 
2-14 June 2004 No. 14: Seismic Events Revision 1 
portion of the total seismic energy release along the western Basin and Range Province in the 
past two decades. 
Since the Yucca Mountain catalog was compiled from several source catalogs, each using a 
variety of different magnitude scales that also changed with time, a uniform magnitude scale was 
required to compute the earthquake recurrence for the region. In addition, it was necessary to 
assign magnitudes to historical earthquakes that occurred prior to calibrated seismographic 
instrumentation. Nevada Test Site explosions and their induced earthquake aftershocks and 
reservoir-induced seismicity events at Lake Mead were identified in the earthquake catalog. 
Nuclear explosions and related aftershocks, which appear as the prominent clusters of epicenters 
to the north and east of Yucca Mountain, were excluded from the catalog for purposes of 
calculating seismicity recurrence rates for the seismic hazard analysis. 
The larger earthquakes documented in the historical catalog occurred within the Central Nevada 
Seismic Zone and the Eastern California Shear Zone, 100 to 300 km to the northwest, west, 
southwest, and south of Yucca Mountain (Figures 2-2 and 2-9). These zones of activity include 
the 1872 Owens Valley, California, earthquake (Mw 7.8), the 1932 Cedar Mountain, Nevada, 
earthquake (Mw 7.1), the 1954 Fairview Peak and Dixie Valley, Nevada, earthquakes (Mw 7.1 
and Mw 6.8), and the 1992 Landers, California, earthquake (Mw 7.3). The 1999 Hector Mine, 
California, earthquake (Mw 7.1) also occurred in the Eastern California Shear Zone, but occurred 
in 1999, after the time period shown on Figure 2-9. 
W 5.5 are located within 100 km of 
W 6.1 earthquake that occurred 
W 5.6 earthquake that occurred near Little 
Three events during the historical period of greater than M 
Yucca Mountain (Figure 2-10). The largest event is the 1916 M 
in Death Valley. These events also include the 1992 M 
Skull Mountain, about 15 km southeast of Yucca Mountain, and a MW 5.7 event in 1910 about 
85 km to the northwest (Figure 2-10). 
Various analyses have shown that earthquakes in the southern Great Basin occur predominantly 
between depths of about 2 and 15 km (e.g., Rogers et al. 1987, pp. 55 to 56; Harmsen and Bufe 
1992, pp. 42 to 43). Focal mechanisms (analyses to determine the orientation and direction of 
faulting) of recent earthquakes within the southern Great Basin indicate that right-lateral slip on 
north-trending faults is the predominant mode of stress release near the site (Pezzopane, Bufe 
et al. 1996, p. 7-3). Normal and oblique-slip faulting is also observed. Focal mechanisms for the 
period 1971 through 1992 indicate roughly equal proportions of strike-slip and normal faulting 
(Pezzopane, Bufe et al. 1996, p. 7-38, Table 7-3). 
2.2.2 Seismicity in the Vicinity of Yucca Mountain 
While the southern portion of the Nevada Test Site, southeast of Yucca Mountain, is one of the 
more seismically active regions in the southern Great Basin, the area immediately around Yucca 
Mountain (within 10 km) itself is rather aseismic (Figure 2-10). Studies, including an 
experiment in high-resolution monitoring of seismicity at the site, have shown that the Yucca 
Mountain zone of quiescence is a real feature of the contemporary seismicity and not an artifact 
of network design or detection capability (Gomberg 1991a, pp. 16411 to 16412; Gomberg 
1991b, pp. 16397 to 16398; Brune et al. 1992, p. 164). Modeling of the strain field in southern 
2-15 June 2004 No. 14: Seismic Events Revision 1 
Nevada suggests that this area is not accumulating significant strain and that Yucca Mountain is 
an isolated block within the structural framework of the southern Great Basin. 
Source: Simmons 2004, Figure 4-17. 
NOTE: YMP = Yucca Mountain Project. 
Figure 2-8. Seismograph Stations Operating in the Southern Great Basin in 1980 (a) and 2001 (b) 
June 2004 2-16 No. 14: Seismic Events Revision 1 
Source: DOE 2002, Figure 4-162. 
NOTE: Shown are earthquakes from 1868 to 1996. Part (a) shows earthquakes with magnitude greater than or 
equal to 3.5; part (b) shows the subset of earthquakes with magnitude greater than or equal to 6. This 
seismicity catalog was compiled from a variety of sources. Coverage of older seismicity is sparse because 
of the absence or limited availability of seismographic coverage in the late 1800s and early 1900s. The 
cluster of earthquakes near the southern boundary of the Nevada Test Site represents the 1992 Little Skull 
Mountain earthquake and its numerous aftershocks. Many of the events in the northern part of the Nevada 
Test Site occurred in response to underground nuclear weapons tests. Significant earthquakes are labeled 
with year of occurrence. 
Figure 2-9. Historical Earthquake Epicenters within 300 km of Yucca Mountain 
June 2004 2-17 No. 14: Seismic Events Source: Simmons 2004, Figure 4-19. 
NOTE: Shown are earthquakes from 1904 to 1998. Earthquakes associated with the 1999 Scotty’s Junction and 
1999 Frenchman Flat sequences are also shown. Significant earthquakes or earthquake sequences are 
shown with years of occurrence. 
Figure 2-10. Historical Earthquake Epicenters within 100 km of Yucca Mountain 
2-18 No. 14: Seismic Events 
Revision 1 
June 2004 Revision 1 
L 
From the initiation of Southern Great Basin Digital Seismic Network operation in October 1995 
through September 2002, microearthquakes that have occurred in the immediate vicinity (within 
10 km) of Yucca Mountain are shown in Figure 2-11. The events ranged in magnitude from 
about -1 to 1 M 1. These microearthquakes occurred throughout the Yucca Mountain block and 
had focal depths between about 3 and 13 km. Focal mechanisms were determined for some of 
the events and are consistent with normal to normal-oblique slip on faults with northwestern to 
northeastern orientations. 
While the immediate Yucca Mountain area has been quiescent during the historical period, 
paleoseismic evidence indicates active Quaternary faults exist near the site. Paleoseismic events 
exhibit very long times between events, thus little or no microseismicity may occur on the faults 
during the historical period of observation. Many faults in the Great Basin with paleoseismic 
evidence for prehistoric surface-rupture earthquakes have little or no associated historical 
seismicity. 
Seismicity to the east and southeast of Yucca Mountain is spatially associated with the Rock 
Valley, Mine Mountain, and Cane Springs fault zones (RV, MM, and CS, respectively, on 
Figure 2-12). This activity forms a wide, northeast-trending zone that includes the 1973 Ranger 
Mountain sequence, the 1992 Little Skull Mountain sequence, the 1993 Rock Valley sequence, 
the 1999 Frenchman Flat sequence, and other earthquake clusters (Figure 2-10). The main 
shocks from the Little Skull Mountain and Frenchman Flat sequences, near the ends of the 
seismicity zone, exhibited normal faulting on northeasterly striking planes. The Rock Valley 
sequence in the middle of the zone exhibited strike-slip faulting. Some seismicity in the Yucca 
Mountain area is also spatially associated with the southern boundary of the Timber Mountain 
caldera. 
The larger earthquakes or earthquake sequences recorded by the Southern Great Basin Digital 
Seismic Network (or its predecessor, the Southern Great Basin Seismic Network) are the 
following (Figure 2-10): 
• 1992 Little Skull Mountain sequence, including the June 29, 1992, mainshock with 
MW 5.6 
• 1993 earthquakes in the Rock Valley fault zone. Following the Little Skull Mountain 
earthquake in 1992, several earthquakes of up to MW 3.7 occurred. 
• 1999 Frenchman Flat earthquake sequence on January 27, 1999 (MW 4.7), preceded by 
an extended foreshock sequence that included a ML 4.2 earthquake. 
• 1999 Scotty’s Junction earthquake sequence on August 1, 1999 (ML 5.7, with two 
aftershocks ML 4.8 and 4.9). 
1 
used to indicate them, are as follows: M 
Historically, earthquake magnitude has been determined using a variety of techniques, resulting in a number of 
different earthquake magnitude scales. The various magnitude scales mentioned in this report, and the symbols 
L = local magnitude, MW = moment magnitude. 
June 2004 2-19 No. 14: Seismic Events Revision 1 
Figure 2-11. Seismicity at Yucca Mountain from October 1, 1995, to September 30, 2002 
Source: Simmons 2004, Figure 4-22. 
2.3 PREHISTORIC EARTHQUAKES 
2.3.1 Paleoseismic Studies 
The identification and documentation of earthquakes occurring prior to historical times, termed 
paleoseismology, is possible by studying the geologic record of past events. Larger events that 
rupture to the surface often leave geological evidence in the form of offset strata and 
characteristic earthquake-related deposits. A program of geologic fault studies at Yucca 
Mountain reveals that the recurrence times of large-magnitude earthquakes are on the order of 
tens of thousands of years, much longer than the approximately 130-year historical earthquake 
record of the Yucca Mountain region. Thus, the prehistoric earthquake history of the Yucca 
Mountain site spans at least the past several hundred thousand years and is particularly important 
for the PSHA because it extends the record for larger magnitude events. 
Paleoseismic studies are the basis for identifying the occurrence of large-magnitude 
surface-rupturing earthquakes and evaluating their size, age, and occurrence rate. Regional 
investigations were conducted to identify faults in the Yucca Mountain area that have evidence 
of Quaternary displacements within 100 km of the site (Figure 2-12). 
June 2004 2-20 No. 14: Seismic Events Revision 1 
Displaced or deformed sedimentary deposits record late Quaternary faulting along local faults in 
the Yucca Mountain area. These faults include, from west to east, the Northern Crater Flat, 
Southern Crater Flat, Windy Wash, Fatigue Wash, Solitario Canyon, Iron Ridge, Stagecoach 
Road, Bow Ridge, and Paintbrush Canyon faults (Figures 2-6 and 2-12). Paleoseismic data for 
these faults provide the basis for interpretations made as part of the PSHA. These data and 
interpretations are summarized in the next section. 
Source: DOE 2002, Figure 4-164. 
NOTE: (a) Faults within 100 km of Yucca Mountain. (b) Detail of (a), showing faults near Yucca Mountain; these 
faults are the Abandoned Wash (AW), Black Cone (BC), Boomerang Point (BP), Crater Flat (CRF), Drill 
Hole Wash (DHW), Dune Wash (DW), East Busted Butte (EB), East Lathrop Wells (ELW) Fatigue Wash 
(FW), Ghost Dance (GD), Iron Ridge (IR), Midway Valley (MWV), Pagany Wash (PW), Paintbrush Canyon 
(PC), Sever Wash (SW), Solitario Canyon (SC), Stagecoach Road (SCR), Sundance (SD), and Windy Wash 
(WW) faults. For detailed identification of each fault labeled with initials in (a) but not identified in this note, 
see the Yucca Mountain Site Description (Simmons 2004, Figure 4-23). 
2.3.2 Paleoseismic Interpretations 
Geologic studies of the local faults in the Yucca Mountain area resulted in data and 
interpretations of key paleoseismic characteristics that define the size and rate of occurrence of 
prehistoric earthquakes on these faults. These characteristics include fault slip rate, 
displacements associated with individual paleoseismic events, lengths of rupture, and recurrence 
intervals between individual paleoseismic events. Paleoseismic investigations of these faults are 
Figure 2-12. Known or Suspected Quaternary Faults and Other Notable Faults in the Yucca Mountain 
Region 
June 2004 2-21 No. 14: Seismic Events Revision 1 
described in the Yucca Mountain Site Description (Simmons 2004, Section 4.3.2) and in more 
detail by Whitney and Taylor (1996). 
Fault Slip Rate–Fault slip rate is the time-averaged rate of displacement on a fault in 
millimeters per year. Slip rate is an important paleoseismic parameter; it is a standardized 
measure of activity on the fault, which can be used for comparing the activity of many different 
faults. It can also provide direct input to calculations of earthquake recurrence rate for seismic 
hazard analyses. Fault slip rates are calculated by dividing the amount of cumulative observed 
net slip by the age of a specific faulted deposit or horizon. Uncertainties in the age of the unit 
and in the amount of offset are incorporated into uncertainty estimates of the slip rate. 
Estimates of slip rates for faults at Yucca Mountain vary from 0.001 to 0.07 mm/yr. These rates 
are low when compared to those for other faults in the Basin and Range Province. For example, 
in dePolo’s scheme to classify the activity of Basin and Range Province faults (dePolo 1994, p. 
49), the Yucca Mountain faults fall into the moderately-low-to-low activity class. Also, slip 
rates for faults at Yucca Mountain are equal to or lower than values in a compilation of slip rates 
developed by McCalpin (1995, Table 2) for the Basin and Range Province. 
Per-Event Displacement–Per-event displacements are an important paleoseismic parameter for 
estimating the maximum magnitude of prehistoric earthquakes. The most precise per-event 
displacements were estimated directly from trench log data. Multiple measurements of per-event 
displacement are available for local faults at Yucca Mountain. Fewer data on per-event 
displacements are available for regional faults, and they are based primarily on the projected 
measurement of geomorphic surfaces across topographic scarps or the surface trace of the faults. 
Estimated per-event displacements vary from near 0 to 257 cm. For example, the range of 
estimates is 0 to 130 cm for the Solitario Canyon Fault, 1 to 80 cm for the Bow Ridge Fault, and 
0 to 257 cm for the Paintbrush Canyon Fault. Fracture events with no detectable offsets are 
sometimes identified. Displacements per event are larger (80 to less than 362 cm) for the Rock 
Valley Fault and Bare Mountain Fault relative to the block-bounding faults at Yucca Mountain. 
Available estimates of single-event displacements for regional faults are in the general range of 
those for site faults, with the exception of displacements of 240 to 470 cm on the Death Valley– 
Furnace Creek fault system. 
Surface-Rupture Length–Coseismic surface-rupture length is an important parameter used to 
define the maximum magnitudes of prehistoric earthquakes. Determination of rupture lengths 
for prehistoric earthquakes is estimated using surficial mapping data, the locations of trenches, 
and displacement-event timing data. The length of a rupture is determined by the lateral extent a 
given event can be identified along one or more faults. In some cases, this identification is 
accomplished via event correlations between sites using similarities in ages and other supportive 
geologic criteria. Surface-rupture lengths may be equivalent to the total fault length or may be 
restricted to a specific portion, or segment, of the fault. 
Rupture lengths of individual site faults and regional faults in the Yucca Mountain area range 
from 1 to about 25 km. Fault lengths in the region surrounding the site range from several 
kilometers to more than 300 km for the Death Valley–Furnace Creek–Fish Lake fault system. 
Except for very long, likely segmented faults, the total measured length of a fault represents its 
June 2004 2-22 No. 14: Seismic Events Revision 1 
maximum rupture potential. This interpretation is considered reasonable, especially for the 
relatively short faults at Yucca Mountain. For consideration in the PSHA, a number of rupture 
scenarios have been developed. These scenarios include permissible combinations of faults and 
fault segments that could, based on the available paleoseismic data, rupture synchronously 
during individual earthquakes. 
Recurrence Interval–“Recurrence interval” is defined as the time interval between successive 
surface-rupture earthquakes. It is an important temporal measure of fault behavior in seismic 
hazard analysis. Recurrence intervals are calculated for individual faults at relevant sites using 
paleoseismic data of the number and timing of coseismic surface-rupture events. Individual 
recurrence intervals are estimated in cases for which there is adequate age control to isolate the 
timing between specific pairs of events. Uncertainties in both the dating of units and the number 
of possible events are incorporated into the reported ranges of recurrence intervals. 
At least one and commonly two or more recurrence interval estimates are made along individual 
faults at Yucca Mountain. Average recurrence intervals of individual faults range from 5 to 
270 ka. The long recurrence intervals reflect relatively small numbers of observed displacements 
in middle Pleistocene deposits and are consistent with the estimated low fault slip rates. 
2.4 VIBRATORY GROUND MOTION INFORMATION 
In a seismic hazard analysis, assessments are made of the size and recurrence rate of earthquakes 
that seismic sources might generate, as well as the propagation of earthquake energy from the 
source to the Yucca Mountain site. The level of ground motion that will be experienced at the 
site for any given earthquake location and magnitude must be assessed and is described by 
attenuation relations. To the extent possible, the vibratory ground motion adopted for the 
seismic design of the repository and analysis of its postclosure performance should incorporate 
the effects of the seismic sources, propagation path, and local site geology specific to the Yucca 
Mountain region and site. Ideally, recorded ground motion from large earthquakes in the Yucca 
Mountain region would be used directly to develop attenuation relations for application at Yucca 
Mountain. However, because large earthquakes occur infrequently in the region, such data are 
few and are insufficient to constrain empirical models. This situation exists even when the entire 
Basin and Range Province, extending from southern Oregon and Idaho to northern Mexico and 
western Texas, is considered. The few data recorded from larger earthquakes and the 
geophysical and seismological properties derived for the region, nevertheless, provide some 
information for estimating ground motion at the repository site. This information forms part of 
the basis for the expert interpretations and assessments of uncertainty that were part of the PSHA 
for Yucca Mountain. 
Characterizing ground motion at Yucca Mountain using existing attenuation relations involves 
resolving whether (and to what extent) those relations for the western United States are 
applicable to the Basin and Range Province, in general, and to Yucca Mountain, in particular. 
The seismological questions include whether differences in the factors that influence ground 
motion in the Yucca Mountain region and in the western United States would lead to significant 
differences in ground motion estimates for the two regions. These factors include seismic source 
properties, regional crustal properties, and shallow geologic site properties at the repository. 
2-23 June 2004 No. 14: Seismic Events Revision 1 
Generally, comparisons must be made between Yucca Mountain factors and average factors 
inherent in the strong-motion database for the western United States. 
To address these issues, six ground motion studies were carried out as part of site 
characterization activities and are discussed in the following sections. The first study was an 
empirical analysis of worldwide ground motion data from extensional regimes. The second 
study comprised numerical modeling of selected scenario earthquakes near Yucca Mountain in 
which ground motions were estimated using seismological models of the source, path, and site 
effects. The numerical modeling allowed the region-specific crustal structure and site-specific 
rock properties to be incorporated in the ground motion estimates. The third study used weak 
motion recordings to characterize the near-surface seismic wave attenuation at Yucca Mountain. 
The fourth study examined earthquake stress drops in extensional regimes to compare them to 
those for earthquakes used to develop western United States ground motion attenuation relations. 
Stress drop is a factor in determining the level of high-frequency ground motion. The fifth study 
investigated the possible constraint that precariously balanced rocks can provide on the levels of 
ground motion that have occurred in the past. The sixth study is the ground motion 
characterization performed as part of the PSHA project and is the most comprehensive of the six. 
This study was a formal elicitation of a panel of ground motion experts, which incorporated 
results from the other five studies and resulted in ground motion attenuation relations specific to 
Yucca Mountain. The ground motion characterization performed as part of the PSHA is 
discussed in Section 3.3. 
2.4.1 Attenuation of Strong Motion in Extensional Regimes 
The Basin and Range Province in which Yucca Mountain is situated is an extensional tectonic 
regime. At the time that the PSHA for Yucca Mountain was getting underway, a ground motion 
attenuation relation had not been developed specifically for extensional regimes. Thus, to 
develop a ground motion attenuation relation appropriate for the site region, data from 
extensional regimes were compiled and analyzed (Spudich, Fletcher et al. 1996). Because the 
number of events in the Basin and Range Province is limited, the database includes ground 
motions recorded in extensional regimes worldwide. 
Several representative attenuation relations based on western United States data were compared 
to the extensional data by Spudich, Fletcher et al. (1996). The mean residual, or bias, was 
computed for each attenuation relation and indicates whether that relation systematically 
underpredicts or overpredicts the extensional strong-motion data. In general, the computed 
residuals indicate that the standard western United States attenuation relations overpredict 
ground motion from extensional regimes by about 15% to 35% on average (Spudich, Fletcher 
et al. 1996, Table 9; Abrahamson and Becker 1996, Figure 10.5-3). 
Spudich, Fletcher et al. (1997) developed an attenuation relation (SEA96) to estimate ground 
motion in extensional regimes. Comparisons of median predictions from this relation with those 
from several western United States attenuation relations illustrate their differences. In general, at 
short to moderate periods, the predictions of the SEA96 relation are less than or lie at the lower 
limit of the values predicted by other western United States relations. At long periods, the 
SEA96 relation is similar to the western United States relations. Notably, however, the SEA96 
2-24 June 2004 No. 14: Seismic Events Revision 1 
relation has a much larger standard deviation at long periods than is usual for the western United 
States relations. 
Spudich, Joyner et al. (1999) updated the attenuation relation for extensional tectonic regimes 
using a larger data set and corrected minor errors in the data set used for the earlier analysis. At 
short distances (5 to 30 km), ground motion predicted by the new relation (SEA99) are up to 
20% higher than those predicted by the SEA96 relation, while at longer periods (1.0 to 2.0 s) and 
larger distances (40 to 100 km), they are about 20% lower (Spudich, Joyner et al. 1999, p. 1156). 
When compared to ground motion determined from the relation of Boore et al. (1994), results 
average about 20% lower, except for short distances at periods around 1.0 s. For this 
combination, the ground motion from SEA99 exceed those determined from the Boore et al. 
(1994) relation by up to 10%. The standard deviation of SEA99 is considerably lower than that 
of SEA96, probably because of correction of errors in the data set used for the regression. 
2.4.2 Ground Motion from Yucca Mountain Scenario Earthquakes 
Due to the lack of near-fault strong-motion data from earthquakes in the Basin and Range 
Province (including the Yucca Mountain region), a study was carried out to estimate vibratory 
ground motion for several earthquake scenarios potentially affecting Yucca Mountain (Schneider 
et al. 1996). The six participants in the study used established numerical modeling methods to 
simulate ground motion that was appropriate to the specific conditions at Yucca Mountain. As 
part of the modeling exercise, both median ground motion and its variability were estimated for 
each earthquake scenario. 
Six earthquake scenarios were evaluated based on two criteria: (1) the postulated sources are 
likely to have generated significant earthquakes in the past, and (2) they are considered likely to 
produce ground motion that would impact seismic hazard estimates at Yucca Mountain. The six 
scenarios include four normal faulting events (Mw 6.3 to 6.6) at source-to-site distances of 1 to 
15.5 km and two strike-slip faulting events (Mw 6.7 and 7.0) at source-to-site distances of 25 and 
50 km, respectively. 
Six modeling approaches were included in the study. The methods vary significantly in their 
treatment of wave propagation, site response, and overall level of complexity, but all the 
methods accommodate the essential aspects of seismic energy being generated from a finite 
source and propagated along a path to a site at the earth’s surface. Differences in resulting 
predictions capture an important component of the uncertainty in ground motion in these 
scenario earthquakes that can be applied to the variability of other simulations. 
The study included a validation phase in which the models were calibrated using ground motion 
recordings from the nearby Little Skull Mountain earthquake (MW 5.6). The six participants 
incorporated various Yucca Mountain source, path, and site parameters into their models to fit 
observed ground motion from five sites that recorded the Little Skull Mountain earthquake. 
Most of the six methods had also been previously calibrated against recordings from other 
earthquakes. 
The models produced ground motion estimates that were comparatively unbiased for periods of 
less than 1 s (Abrahamson and Becker 1996, Figure 10.7-5), indicating that they are applicable to 
2-25 June 2004 No. 14: Seismic Events Revision 1 
estimating ground motion in the Basin and Range Province. However, the bias for periods 
greater than 1 s indicates that the numerical simulation models do not work well for the 1992 
Little Skull Mountain earthquake at long periods. 
Using the Little Skull Mountain–calibrated models, the six teams computed motions for the six 
faulting scenarios. Five of the teams whose models were numerical simulations (i.e., all except 
an empirical underground nuclear explosion model) ran multiple realizations of the source 
process and computed a mean spectrum for each scenario. The computed median spectral 
accelerations for the scenario events based on the team’s models are shown on Figure 2-13. 
Ground motion computed for the normal faulting scenario events at close distances (Bow Ridge, 
Solitario Canyon, and Paintbrush Canyon faults) are large (Schneider et al. 1996, Figure 10-10): 
34 Hz spectral accelerations range from 0.5 to 1.0 g at distances of 1 to 3 km. 
The model simulations were compared with several western United States empirical attenuation 
relations (Sadigh et al. 1993; Boore et al. 1994). The simulated median ground motion for the 
four normal faulting scenario events exceed the western United States predictions by about 60% 
at distances less than 5 km (3 mi) and by about 20% at 15 km (9 mi) (Abrahamson and 
Becker 1996, p. 10-39). The differences are largest at high frequencies, attributable primarily to 
low site attenuation in the shallow rock at Yucca Mountain and to larger crustal amplification for 
the Basin and Range Province. At long periods, the difference is attributed to the larger crustal 
amplification and directivity effects (stronger ground motion in the direction of rupture 
propagation than in other directions from the earthquake source). 
June 2004 2-26 No. 14: Seismic Events Revision 1 
Source: Abrahamson and Becker 1996, Figure 10.8-15. 
Figure 2-13. Median Spectral Acceleration of Scenario Earthquakes in the Yucca Mountain Region 
For the more distant strike-slip faulting earthquake scenarios, the simulated median ground 
motion results are greater than the western United States attenuation predictions by about 30% at 
a distance of 25 km at high frequencies (Abrahamson and Becker 1996, p. 10-40). This increase 
is similarly attributed to low site attenuation and larger crustal amplification. At 50 km, the 
simulated longer period ground motion results are consistent with western United States 
empirical attenuation predictions because the effect of local site attenuation is not as significant. 
The simulated higher ground motion at high frequencies is consistent with records from the 1992 
Little Skull Mountain earthquake. The high-frequency ground motion from this event is 
significantly larger than those predicted by western United States empirical attenuation relations. 
June 2004 2-27 No. 14: Seismic Events Revision 1 
The variability of the simulated motions is also greater than that computed for western United 
States empirical attenuation relations. The standard error is about 0.15 natural log units larger 
than that found for empirical attenuation relations. 
2.4.3 Site Attenuation 
Recordings of regional earthquakes at Yucca Mountain were used by Su et al. (1996) to evaluate 
the near-surface attenuation. Results indicated that site attenuation at Yucca Mountain is lower 
than for typical California soft rock. Therefore, at low levels of shaking, damping from the tuff 
is less than that for California soft-rock conditions. This leads to larger high-frequency ground 
motion on the tuff as compared to that on California soft rock, assuming that all other parameters 
are the same. The results of Su et al. (1996) were used to provide a site attenuation value for the 
Yucca Mountain PSHA. 
2.4.4 Earthquake Stress Drop 
Stress drop is the difference in stress across the fault before and after an earthquake. Stress drop 
affects high-frequency ground motion. If stress drops in the Yucca Mountain region are greater 
than the typical value for western United States earthquakes, then larger high-frequency motions 
would be expected at Yucca Mountain relative to motions determined from western United 
States empirical attenuation relations. To provide data to assess this potential effect, an 
evaluation of stress drops for earthquakes in extensional regimes was performed in support of the 
ground motion characterization effort in the PSHA project (CRWMS M&O 1998, p. 5-11). 
The analysis used a data set composed of earthquake records from extensional tectonic regimes 
(Spudich, Fletcher et al. 1997), including both normal and strike-slip events. These data were 
supplemented with data from the 1995 Dinar, Turkey, earthquake, which were not available to 
Spudich, Fletcher et al. (1997). The final data set comprised 210 horizontal components from 
140 sites in 24 earthquakes, a magnitude range of Mw 5.1 to 6.9, and distances from 0 to 102 km. 
The data were fit to a standard earthquake source model. A two-step inversion process was 
adopted to decouple the inversions for site attenuation and for stress drop. Stress drops 
computed for each earthquake were weighted to yield a median value for each mechanism. 
The median stress drop for normal-faulting earthquakes was about 4.5 MPa, and the value for 
strike-slip earthquakes (in extensional regimes) was about 5.5 MPa. In comparison, stress drops 
for western United States earthquakes are about 7 to 10 MPa (e.g., Atkinson 1995, p. 1341). 
These differences in stress drop contribute to lower high-frequency motions in extensional 
regimes compared to transpressional regimes, such as coastal California. This information was 
considered in the ground motion characterization component of the PSHA for Yucca Mountain. 
2.4.5 Implications for Vibratory Ground Motion from Studies of Precarious Rocks at 
Yucca Mountain 
The existence of precariously balanced rocks in the Yucca Mountain region may place some 
constraints on the level of vibratory ground motion experienced at the site over the past several 
tens of thousands of years. Precariously balanced rocks provide evidence that past levels of 
strong vibratory ground motion have been insufficient to topple them. In areas where strong 
ground motion is known to have occurred historically, precarious rocks are not observed. For 
June 2004 2-28 No. 14: Seismic Events Revision 1 
example, based on reconnaissance field surveys in southern California, Brune (1996, p. 43) 
concluded that no precarious rocks are found within 15 km of zones of high-energy release of 
historical large earthquakes. Laboratory physical modeling, numerical modeling, and field tests 
provide confidence that rough estimates of the accelerations required to topple precarious rocks 
can be made without extensive controlled testing (Brune 2000, p. 1107; Anooshehpoor and 
Brune 2000, pp. 2 to 4). Brune and Whitney (2000, p. 18) noted that numerous precarious rocks 
exist along Solitario Canyon and argued that accelerations at Yucca Mountain have not exceeded 
about 0.3 g at the surface during the past 75,000 to 80,000 years. Vibratory ground motion at the 
depth of waste emplacement would be less than those at the earth’s surface. 
Precarious rocks have also been used to test ground motion attenuation relations. Brune (2000, 
p. 1107) notes that, in contrast to observations near strike-slip faults, precarious and 
semiprecarious rocks are found nearly to the fault trace on the footwall side of normal faults in 
Nevada and California. Comparison of estimated toppling accelerations with accelerations 
predicted by a ground motion attenuation relation based largely on data for strike-slip 
earthquakes suggests that the attenuation relation may be conservative. The attenuation relation 
overestimates accelerations on the footwall of normal faults at near distances (Brune 2000, 
Figure 2). This result is consistent with results from dynamic foam rubber models of strike-slip 
and normal faulting earthquakes (Brune and Anooshehpoor 1999). That is, the models indicate 
that ground motion near the fault trace is less for normal faulting earthquakes than for strike-slip 
earthquakes. The implication of this observation is that seismic hazard estimates using ground 
motion attenuation curves based largely on data from strike-slip earthquakes may result in 
conservative values of hazard for sites such as Yucca Mountain, where normal faulting 
dominates. 
June 2004 2-29 No. 14: Seismic Events INTENTIONALLY LEFT BLANK 
2-30 No. 14: Seismic Events 
Revision 1 
June 2004 Revision 1 
3. PROBABILISTIC SEISMIC HAZARD ANALYSIS 
To assess the seismic hazards of vibratory ground motion and fault displacement at Yucca 
Mountain, a PSHA was performed. The PSHA provides quantitative hazard results to support an 
assessment of the repository’s long-term performance and to form the basis for developing 
seismic design criteria for the LA. The PSHA methodology that the DOE has used for Yucca 
Mountain is consistent with that documented in Methodology to Assess Fault Displacement and 
Vibratory Ground Motion Hazards at Yucca Mountain (YMP 1997). This methodology also is 
consistent with state-of-the-practice guidance provided by the Senior Seismic Hazard Analysis 
Committee (Budnitz et al. 1997). Key attributes of the PSHA methodology for Yucca Mountain 
are (1) utilization of a geologic and seismologic database developed over a 20-year period in the 
Yucca Mountain region; (2) explicit consideration and quantification of uncertainties regarding 
alternative seismic-source, ground motion, and fault-displacement interpretations; and (3) use of 
a formal, structured expert elicitation process to capture the informed scientific community’s 
views of key inputs to the PSHA. 
The PSHA methodology for vibratory ground motion has become standard practice for deriving 
vibratory ground motion hazards for design purposes. Less commonly, probabilistic fault 
displacement analyses are conducted to provide quantitative assessments of the location and 
amount of differential ground displacement that might occur. Both analyses provide hazard 
curves, which express the probability of exceeding various amounts of ground motion (or fault 
displacement). The probability is usually expressed as a frequency of exceedance per year. The 
resulting seismic hazard curves represent the integration over all identified earthquake sources 
and magnitudes of the probability of future earthquake occurrence and, given an occurrence, its 
effect at a site of interest. 
The basic elements of a PSHA for vibratory ground motion are: 
a) Identification of seismic sources that contribute to the vibratory ground motion hazard at 
Yucca Mountain and characterization of their geometry. This establishes the distribution 
of source-to-site distances for each source. 
b) Characterization of seismic sources by the recurrence rate of earthquakes of various 
magnitudes and the maximum magnitude. This establishes the distribution of magnitudes 
and their rate of occurrence for each source. 
c) Ground motion attenuation relations that provide for the estimation of a specified ground 
motion parameter as a function of magnitude, source-to-site distance, local site 
conditions, and, in some cases, seismic source characteristics. This establishes the 
distribution of ground motion that will be experienced at the site when a given magnitude 
earthquake occurs on a given source. 
d) Integration of the seismic source characterization and ground motion attenuation 
evaluations, including associated uncertainties, into a seismic hazard curve and associated 
uncertainty distribution. 
3-1 June 2004 No. 14: Seismic Events Revision 1 
Probabilistic fault displacement hazard analysis follows a similar path: 
a) Identification of fault sources of fault displacement (principal faults) 
b) Characterization of the frequency, size, and locations of displacements on principal faults 
c) Characterization of the amounts and locations of subsidiary displacements as a function 
of distance from principal faults and magnitudes 
d) Integration of source characterization and distance distribution, including associated 
uncertainties, into a fault displacement hazard curve and associated uncertainty 
distribution. 
PSHAs incorporate both randomness and uncertainty. For discrete variables, the randomness, 
also termed aleatory variability, is parameterized by the probability of each possible value. For 
continuous variables, the randomness is parameterized by the probability density function. An 
example of randomness is the amplitude of ground motion that would occur at a particular 
location from repeated earthquakes having exactly the same magnitude at exactly the same 
distance (say, magnitude 6 at 25 km distance). Variations in ground motion amplitude are 
expected due to unknowable complexities in earthquake-to-earthquake source properties and in 
the propagation path. 
Uncertainty, also termed epistemic uncertainty, is the scientific uncertainty in the model of the 
process. The uncertainty is due to limited data and knowledge and is characterized by alternative 
models. For discrete random variables, the uncertainty is modeled by alternative probability 
distributions. For continuous random variables, the uncertainty is modeled by alternative 
probability density functions. Examples of uncertainty are alternative ground motion attenuation 
relations that express the amplitude of ground motion at a particular site as a function of distance 
to the source and earthquake magnitude. Unlike randomness, uncertainty is potentially reducible 
with additional knowledge and data. 
Given the input evaluations, the hazard calculation method integrates over all values of the 
variables and estimates the mean probability of exceedance of any ground-shaking amplitude at 
the site. A plot of these results is the seismic hazard curve. The hazard curve quantifies the 
variability of the earthquake occurrence and ground-shaking attenuation. In addition to the 
variability of the seismic hazard, however, is uncertainty about the seismotectonic environment 
of a site. Significant advances in development of methodology to quantify uncertainty in seismic 
hazard have been made in the past 20 years (EPRI 1986, Volume 1; Budnitz et al. 1997). These 
advances involve the development of alternative interpretations of the seismotectonic 
environment of a site by multiple experts and the structured characterization of uncertainty. 
Evaluations by multiple experts are made within a structured expert elicitation process designed 
to minimize uncertainty due to uneven or incomplete knowledge and understanding (Budnitz 
et al. 1997). The weighted alternative interpretations are expressed by use of logic trees. Each 
pathway through the logic tree represents a weighted interpretation of the seismotectonic 
environment of the site for which a seismic hazard curve is computed. The result of computing 
the hazard for all pathways is a distribution of hazard curves representing the full variability and 
uncertainty in the hazard at a site. 
June 2004 3-2 No. 14: Seismic Events Revision 1 
3.1 PROBABILISTIC SEISMIC HAZARD ANALYSIS PROCESS 
By necessity, evaluations of seismic source characteristics, earthquake ground motion, and fault 
displacement involve interpretations of data. These interpretations have associated uncertainties 
related to the ability of data to fully resolve various hypotheses and models. In the PSHA, a 
formal expert elicitation process was used to develop inputs that specifically included estimates 
of variability and uncertainty in interpretations (CRWMS M&O 1998). The expert elicitation 
process followed was generally consistent with the Branch Technical Position on the Use of 
Expert Elicitation in the High-Level Radioactive Waste Program (Kotra et al. 1996) and the 
Recommendations for Probabilistic Seismic Hazard Analysis: Guidance on the Uncertainty and 
Use of Experts (Budnitz et al. 1997). 
Six teams of three earth science experts characterized seismic sources in the Yucca Mountain site 
region (within a distance of about 100 km) and fault displacement potential at 
nine demonstration points. Seven ground motion experts characterized ground motion 
attenuation in the site region. Interpretations for hazard assessment were coordinated and 
facilitated through a series of workshops. Each workshop was designed to accomplish a specific 
step in the elicitation process and to ensure that the relevant data were being appropriately 
considered and integrated. An important goal in the elicitation process was to reduce variability 
in interpretations due to a lack of common understanding of the available data and probabilistic 
models used in the analysis. The final PSHA results were presented as mean, median, and 
fractile hazard curves representing the total variability and uncertainty in input interpretations. 
The essential elements of the seismic source characterization, ground motion attenuation 
evaluation, fault displacement characterization, and PSHA results are summarized in 
Sections 3.2, 3.3, 3.4 and 3.5, respectively. 
3.2.1 Source Geometries 
As used in PSHA, 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. Two main types of seismic sources were 
characterized by the seismic source expert teams: fault sources and areal source zones. Fault 
sources are used to represent the occurrence of earthquakes along a known or suspected fault. 
Uncertainty in the definition of fault sources is expressed by considering alternative rupture 
lengths, alternative fault dips, and possible linkages with other faults. In addition, an evaluation 
3.2 SEISMIC SOURCE CHARACTERIZATION 
Seismic sources were interpreted by the PSHA expert teams. For each source zone, the expert 
teams characterized the zone’s geometry, maximum magnitude, rate of earthquake occurrence, 
and the magnitude distribution of those earthquakes. Summaries of the source-zone 
characterization performed by the six expert teams can be found in Characterize Framework for 
Seismicity and Structural Deformation at Yucca Mountain, Nevada (CRWMS M&O 2000, Table 
5); more complete descriptions are found in Probabilistic Seismic Hazard Analyses for Fault 
Displacement and Vibratory Ground Motion at Yucca Mountain, Nevada (CRWMS M&O 1998, 
Appendix E). 
3-3 June 2004 No. 14: Seismic Events Revision 1 
is made of the probability that a particular fault is active and capable of generating 
moderate-to-large earthquakes. 
Areal source zones represent regions of distributed seismicity that are not associated with 
specific known faults. The events are considered to be occurring on unidentified faults or 
structures whose areal extents are best characterized by zones. Uncertainty in defining areal 
zones typically was expressed by considering alternative zonations of the region surrounding the 
Yucca Mountain site. For each source zone, the expert teams characterized the zone’s maximum 
magnitude, its rate of earthquake occurrence, and the magnitude distribution of those 
earthquakes. Uncertainties in all of these parameters are also quantified. 
Two types of fault sources were considered in the PSHA: 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 larger. Local faults were defined as being located within about 15 km 
of Yucca Mountain. The specific faults that required detailed characterization were determined 
based on factors including fault length and location relative to Yucca Mountain, displacement of 
Quaternary deposits, direct relation with seismicity, structural relation to other Quaternary faults, 
orientation within the contemporary tectonic stress regime, and considerations of alternative 
tectonic models. 
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. For the characterization 
of local faults, varying behavioral and structural models were employed by the expert teams to 
represent the range of possible rupture patterns and fault interactions. Interpretations regarding 
the potential simultaneous rupture of multiple faults were included in all of the expert 
assessments. 
Areal source zones were defined by all teams to account for background earthquakes that occur 
on buried faults or other faults not explicitly included in their characterizations. Seismicity 
related to volcanic processes (specifically to basaltic volcanoes and dike injection) was 
considered by all six expert teams but explicitly characterized as distinct source zones by only 
two expert teams. Volcanic-related earthquakes were not interpreted as separate seismic sources 
by the other four teams because the low magnitude and frequency of volcanic-related seismicity 
was accounted for by earthquakes in the areal zones. 
3.2.2 Maximum Earthquake Magnitude 
Maximum earthquake moment magnitude (Mmax) was determined for each source by each expert 
team to represent the largest earthquake that the source is capable of generating. Two basic 
approaches were used to assess maximum magnitude. The primary approach, which was used 
for faults, was based on estimates of the maximum dimensions of fault rupture. Multiple sources 
of uncertainty were considered in estimating physical dimensions of maximum rupture on faults, 
including uncertainty in rupture length, rupture area, and displacement per event. The second 
approach considered historical data on the seismicity of the region. This approach was used 
June 2004 3-4 No. 14: Seismic Events Revision 1 
primarily for areal source zones. For each of the sources included in the PSHA, the uncertainty 
in Mmax is expressed as a probability distribution. 
3.2.3 Earthquake Recurrence 
Earthquake recurrence relations express the rate or annual frequency of different magnitude 
earthquakes occurring on a seismic source. Methods for developing these relations are usually 
different for fault sources than for areal sources. For fault sources, the expert teams used 
approaches based on estimates of fault slip rates and the average slip per event and on seismic 
moment rates. For areal sources, earthquake recurrence relationships were determined from the 
catalog of historical and instrumental earthquakes within 300 km of Yucca Mountain. 
3.2.4 Example of Seismic Source Characterization Assessments 
The six expert teams considered a variety of alternative models and parameters in their 
characterization of seismic sources. This section identifies the assessments that were made to 
illustrate the range of interpretations, as well as the uncertainties that were quantified. Logic 
trees provided a mechanism for describing and quantifying the uncertainties. Key assessments 
for the PSHA are given as nodes of the logic tree and alternative models or parameter values are 
given on the branches at each node. Weights are assigned to the alternative branches based on 
expert judgment regarding the relative credibility of the alternative models and parameters. 
Key elements of the seismic source characterization by the expert teams were the following: 
• Alternative tectonic models 
– Planar fault models 
– Shear models (buried strike-slip faults or fault systems) 
– Detachment models 
– Synchronous volcanic-tectonic events 
• Thickness of earthquake-generating crust 
• Areal seismic source zones 
– Source geometry 
– Recurrence model 
– Seismicity catalog for recurrence calculation 
– Spatial smoothing model and parameters 
– Maximum magnitude 
• Regional fault sources 
– Probability of activity, slip type, and geometry 
– Maximum magnitude 
– Recurrence approach and parameters 
3-5 June 2004 No. 14: Seismic Events Revision 1 
• Local fault sources 
– Probability of activity 
– Synchroneity of ruptures 
– Coalescence of faults at depth 
– Downdip geometries 
– Rupture dimensions (length, per-event displacement, area) 
– Maximum magnitude 
– Recurrence approach and parameters 
• Other potential sources 
– Buried regional right-lateral shear zone 
– Earthquake-generating detachment 
– Volcanic source zone 
– Fault interpreted from gravity anomalies 
– Cross-basin fault 
– Highway 95 or Carrara Fault. 
For illustration, Table 3-1 presents ranges of key parameter values assessed by the experts for the 
PSHA (CRWMS M&O 1998). The table summarizes the probability that the fault is active and 
the range of maximum magnitudes, slip rates, and recurrence intervals assessed by the experts. 
3.3 GROUND MOTION CHARACTERIZATION 
The goal of the ground motion evaluation for the Yucca Mountain PSHA was to formulate 
attenuation interpretations describing vibratory ground motion at the repository. Seven experts 
evaluated various proponent models. The experts provided point estimates of ground motion for 
a suite of prescribed faulting cases, and these point estimates were subsequently regressed to 
attenuation equations. The ground motion constituted response spectral values (horizontal and 
vertical components) for specified spectral periods. The point estimates constituted an estimate 
of the median ground motion, its randomness (aleatory variability), and the uncertainty in each 
(epistemic uncertainty). Each faulting case corresponded to a particular magnitude earthquake, 
fault geometry, and source-site distance. The cases were designed to sample the 
magnitude-distance-faulting space at Yucca Mountain in sufficient detail to provide a robust 
regression. 
The ground motion estimates and, thus, the resulting attenuation relations were developed for a 
reference rock outcrop (Figure 3-1). This reference rock outcrop was defined to have 
geotechnical conditions identical to those at the depth of the repository. The reference rock 
outcrop was used because site-specific data on the velocities and dynamic properties of the upper 
300 m of rock and soil were limited at the time of the PSHA. For design analyses and analyses 
supporting performance assessment, the effect on ground motion of the material between the 
waste emplacement level and the earth’s surface (site response) will be taken into account (see 
Section 4). 
3-6 June 2004 No. 14: Seismic Events Table 3-1. Summary of Fault Parameters from Probabilistic Seismic Hazard Analysis Seismic Source 
Characterization 
Bare Mountain 
Black Cone 
Bow Ridge 
Dune Wash 
Fatigue Wash 
Iron Ridge 
Midway Valley 
Solitario Canyon 
Windy Wash 
Fault 
Crater Flat Fault system 
Central Crater Flat 
Southern Crater Flat 
Northern Crater Flat 
East Busted Butte 
East Lathrop Wells Cone 
Fatigue Wash–Windy Wash 
Ghost Dance Fault zone 
Iron Ridge–Solitario Canyon 
Paintbrush Canyon 
Paintbrush Canyon–Stagecoach 
Road 
Paintbrush–Stagecoach–Bow Ridge 
Stagecoach Road 
South Windy Wash 
North Windy Wash 
Source: CRWMS M&O 1998. 
NOTE: Parameter ranges developed from all teams reporting (i.e., one to six teams); all parameter ranges were 
provided as probability distributions. 
Recurrence 
Interval (ka) 
20 to 200 
- 
40 to 350 
-
- 
40 to 180 
120 to 160 
-
-
- 
50 to 250 
-
-
-
-
- 
20 to 270 
15 to 120 
10 to 75 
35 to 180 
5 to 75 
35 to 100 
20 to 60 
- 
Proponent Models–The experts computed their ground motion point estimates by considering 
existing proponent models. The proponent models fell into several classes: empirical attenuation 
relations, hybrid-empirical, point source numerical simulations, finite-fault numerical 
simulations, and empirical blast models (CRWMS M&O 2000, Table 8). 
Because no empirical attenuation models exist for the Yucca Mountain region or the Basin and 
Range Province, the empirical models used in this study resulted from regression analyses of 
strong-motion records primarily from California earthquakes. These empirical relations required 
adjustments so they would better fit conditions in the Yucca Mountain region. The hybrid 
No. 14: Seismic Events 
Maximum 
Magnitude 
5.8 to 7.5 
5.0 to 7.0 
5.2 to 7.0 
5.3 to 7.0 
5.3 to 7.0 
5.4 to 7.0 
5.5 to 7.0 
4.9 to 7.2 
4.5 to 7.2 
4.6 to 6.9 
5.5 to 7.3 
5.6 to 7.2 
4.5 to 7.0 
5.1 to 7.0 
5.5 to 7.2 
4.9 to 7.1 
5.9 to 7.4 
5.6 to 7.3 
5.5 to 7.6 
5.6 to 7.4 
5.3 to 7.1 
6.6 to 7.5 
5.7 to 7.1 
5.6 to 7.2 
Probability 
Fault Is Active 
1.0 
0.8 
0.4 to 1.0 
1.0 
0.6 
1.0 
1.0 
0.1 
0.4 
1.0 
1.0 
1.0 
0.05 to 0.1 
0.1 to 1.0 
1.0 
0.1 
1.0 
1.0 
1.0 
1.0 
1.0 
1.0 
1.0 
1.0 
3-7 
Revision 1 
Slip Rate 
(mm/yr) 
0.005 to 0.25 
0.001 to 0.005 
0.002 to 0.007 
0.001 to 0.003 
0.001 to 0.005 
0.002 to 0.02 
0.001 to 0.005 
0.0001 to 
0.001 
0.0005 to 
0.003 
0.005 to 0.003 
0.002 to 0.02 
0.005 to 0.024 
0.0001 to 
0.002 
0.001 to 0.005 
0.005 to 0.024 
0.0001 to 
0.001 
0.002 to 0.03 
0.009 to 0.05 
0.005 to 0.02 
0.002 to 0.04 
0.01 to 0.07 
0.01 to 0.027 
0.01 to 0.04 
0.001 to 0.005 
June 2004 Revision 1 
empirical model is derived from these relations and implicitly includes conversion factors that 
must be separately applied to the empirical relations. The empirical relation for extensional 
regimes developed by Spudich et al. (1997) (Section 2.4.1) was one of the relations considered 
by the experts. 
conversion factors are presented in Probabilistic Seismic Hazard Analyses for Fault 
Figure 3-1. Relation of Reference Rock Outcrop to Sites for which Seismic Inputs Are Developed 
The blast models are based on empirical records from underground nuclear explosions at the 
Nevada Test Site (Schneider et al. 1996, pp. 3-15 to 3-17). Three blast models were assessed, 
each with a different approach to account for differences in earthquake sources and 
explosion sources. 
The numerical simulations were tailored to Yucca Mountain conditions and required no 
adjustments. The point source models are the simplest numerical models and also the best 
understood. The finite-fault numerical simulations were derived from the six models evaluated 
in the scenario earthquake modeling study previously described (Section 2.4.2). The experts 
chose three model approaches for their analyses: 
• Stochastic method with ù2 subevents 
• Composite fractal source method 
• Broadband Green’s function method. 
Conversion Factors–Depending on the nature of the data sets upon which they were based, the 
empirical relations typically represented source, path, and site conditions different from those 
encountered at Yucca Mountain. Suites of conversion factors were consequently computed as 
part of the study. They were developed using the results of numerical finite-fault simulations, 
stochastic point source simulations, and empirical attenuation relations. Summaries of the 
June 2004 3-8 No. 14: Seismic Events Revision 1 
Displacement and Vibratory Ground Motion at Yucca Mountain, Nevada (CRWMS M&O 1998, 
Section 5). The factors included corrections for the following: 
• Source–Western United States compressional and strike-slip seismic sources to Yucca 
Mountain extensional seismic sources 
• Crust–Western United States crust to Yucca Mountain crust 
• Site–Yucca Mountain surface conditions to reference rock outcrop. 
Additionally, many of the proponent models did not include the full range of ground motion 
parameters required. For example, not all the empirical models included vertical ground motion. 
Thus, a variety of scaling factors was also developed and applied in the same manner as the 
conversion models to develop the required parameters. 
Attenuation Relations–Each expert developed a set of point estimates for the defined cases 
covering different faulting styles, event magnitudes, source geometries, and source-site 
distances. The estimates comprised median ground motion, its variability, and the uncertainty in 
both. The estimates were determined directly from the models, the conversion factors, the 
adjustment factors described above, and other judgments by the experts. These estimates were 
then parameterized in the form of attenuation relations. 
The seven sets of attenuation relations predict median ground motion that differs by less than a 
factor of 1.5. The experts’ horizontal aleatory (randomness) estimates, the epistemic 
uncertainties in their median estimates (with one exception), and the epistemic uncertainties in 
their aleatory estimates all vary by less than about 0.1 natural log unit (CRWMS M&O 1998, 
p. 6-4). 
3.4 FAULT DISPLACEMENT CHARACTERIZATION 
Several approaches to characterize the fault displacement potential were developed by the 
seismic source characterization expert teams. These approaches were based primarily on 
empirical observations of the pattern of faulting from historical ruptures throughout the Basin 
and Range and at the site. 
The potential for fault displacement was categorized as either principal or distributed faulting. 
Principal faulting is the faulting along the main plane (or planes) of crustal weakness responsible 
for the primary release of seismic energy during the earthquake. Where the principal fault 
rupture extends to the surface, it may be represented by displacement along a single narrow trace 
or over a zone that is a few to many meters wide. Distributed faulting is rupture that occurs on 
other faults near the principal rupture in response to the principal displacement. It is expected 
that distributed faulting will be discontinuous in nature and occur over a zone that may extend 
outward several tens of meters to many kilometers from the principal rupture. A fault that can 
produce principal rupture may also undergo distributed faulting in response to principal rupture 
on other faults. 
3-9 June 2004 No. 14: Seismic Events Revision 1 
Both principal and distributed faulting are important to the assessment of the fault displacement 
hazard at the Yucca Mountain site. Nine locations at or near Yucca Mountain, with multiple 
assumed conditions at some locations, were identified to demonstrate the fault displacement 
methodology (Figure 3-2). These locations were chosen to represent the range of potential 
faulting conditions throughout the Yucca Mountain site. Some of these locations lie on faults 
that may experience both principal faulting and distributed faulting. The other points are sites of 
only potential distributed faulting. 
The approaches developed by the seismic source expert teams for characterizing the frequency of 
displacement events can be divided into two categories: the displacement approach and the 
earthquake approach. The displacement approach provides an estimate of the frequency of 
displacement events directly from the geologic history of displacement, as interpreted from 
observed feature-specific or point-specific data. The earthquake approach involves relating the 
frequency of slip events to the frequency of earthquakes on the various seismic sources defined 
by the seismic source characterization interpretations for the ground motion assessment. Both 
approaches are used for assessing the fault displacement hazard for principal faulting and 
distributed faulting. 
June 2004 3-10 No. 14: Seismic Events Source: CRWMS M&O 1998, Figure 4-9. 
NOTE: See Table 3-2 for descriptions of demonstration points. 
Figure 3-2. Location of Nine Points for Demonstration of Fault Displacement Hazard Assessment 
3-11 No. 14: Seismic Events 
Revision 1 
June 2004 Table 3-2. Mean Displacement Hazard at Nine Demonstration Sites 
Mean Displacement (cm) 
Location 
Bow Ridge Fault 
Site 
1 
10-4 
<0.1 
<0.1 Solitario Canyon Fault 2 
<0.1 Drill Hole Wash Fault 3 
<0.1 Ghost Dance Fault 4 
<0.1 Sundance Fault 5 
<0.1 Unnamed fault west of Dune Wash 6 
7 About 100 m east of the Solitario 
Canyon Fault with the condition: 
7a <0.1 Small fault with 2 m cumulative 
displacement 
<0.1 7b Shear with 0.10 m cumulative 
displacement 
<0.1 Fracture 7c 
<0.1 Intact rock 7d 
8 Between Solitario Canyon and Ghost 
Dance Fault with the condition: 
8a <0.1 Small fault with 2 m cumulative 
displacement 
<0.1 8b Shear with 0.10 m cumulative 
displacement 
<0.1 Fracture 8c 
<0.1 Intact rock 8d 
5 × 10-5 
Annual Exceedance Probability 
10-5 10-6 10-7 
73 7.8 <0.1 
180 32 <0.1 
15 <0.1 <0.1 
13 <0.1 <0.1 
5.5 <0.1 <0.1 
13 <0.1 <0.1 
2.1 <0.1 <0.1 
1.0 <0.1 <0.1 
0.11 <0.1 <0.1 
– <0.1 <0.1 
2.0 <0.1 <0.1 
0.89 <0.1 <0.1 
0.10 <0.1 <0.1 
– <0.1 <0.1 
11 0.11 <0.1 
220 
490 
75 
56 
39 
71 
18 
5.6 
0.53 
– 
18 
5.5 
0.52 
– 
67 <0.1 Midway Valley 9 
3.5 SEISMIC HAZARD RESULTS 
3.5.1 Vibratory Ground Motion Hazard 
Vibratory ground motion hazard was computed at a defined reference rock outcrop having the 
properties of tuff at a waste emplacement depth of about 300 m below the ground surface at 
Yucca Mounta