Preparing for the next major southern California earthquake: utilizing HAZUS with soils maps and ShakeMaps to predict regional bridge damage and closures

PDF

Download

Open document

Flip pages

Summarize

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Preparing for the Next Major Southern California Earthquake:
Utilizing HAZUS with Soils Maps and ShakeMaps to Predict Regional Bridge Damage and Closures

by

Drew Cover

A Thesis Presented to the
Faculty of the USC Graduate School
University of Southern California
In Partial Fulfillment of the
Requirements for the Degree
Master of Science
(Geographic Information Science and Technology)

May 2018

Copyright © 2017 by Drew Ryan Cover

To my Father, Mother, and Sister.
As well as my extended family from Greater Los Angeles

iv

Table of Contents
List of Figures ................................................................................................................................ vi
List of Tables ................................................................................................................................ vii
Acknowledgements ...................................................................................................................... viii
List of Abbreviations ..................................................................................................................... ix
Abstract ........................................................................................................................................... x
Chapter 1 Introduction .................................................................................................................... 1
1.1. Motivation ...........................................................................................................................3
1.2. Research Goals....................................................................................................................6
Chapter 2 Background and Related Works ..................................................................................... 7
2.1. Southern San Andreas Fault and Earthquake Risk .............................................................8
2.2. Regional Geology and NEHRP Soils Classification ..........................................................9
2.3. HAZUS-MH Implementation Research ...........................................................................11
2.4. Bridge Damage States, Damage Probabilities, and Bridge Closure .................................13
Chapter 3 Data and Methodology ................................................................................................. 15
3.1. Datasets Used ....................................................................................................................16
3.1.1. HAZUS Census tract and default inventory data .....................................................17
3.1.2. Shear Wave Velocity Point Data .............................................................................19
3.1.3. STATSGO2 Soils Maps ...........................................................................................20
3.1.4. USGS Geological Unit Map for California .............................................................21
3.1.5. Custom NEHRP classified Soils Maps ....................................................................22
3.1.6. USGS Scenario ShakeMaps .....................................................................................27
3.1.7. FHWA National Bridge Inventory...........................................................................30
3.2. HAZUS Earthquake Modeling using Soils Maps and ShakeMaps ..................................31
3.2.1. Setting up HAZUS and loading User Supplied Datasets .........................................32
v

3.2.2. Setting Hazard Event Scenarios and Running Analysis ..........................................33
3.2.3. Exporting HAZUS Output Data...............................................................................34
3.3. Bridge Closure Analysis ...................................................................................................34
3.3.1. Bridge Damage States to California Bridge Closures..............................................35
Chapter 4 Results .......................................................................................................................... 37
4.1. Final Results Summary .....................................................................................................37
4.2. Bridge Damage Probabilities to Exceed At Least Moderate Damage ..............................38
4.2.1. M7.4 Scenario ShakeMap ........................................................................................39
4.2.2. M7.4 NEHRP STATSGO Data Map .......................................................................40
4.2.3. M7.4 NEHRP GeoUnit Data Map ...........................................................................41
4.2.4. M7.4 No User-Supplied Data ..................................................................................42
4.2.5. M8.0 Scenario ShakeMap ........................................................................................43
4.2.6. M8.0 NEHRP STATSGO Data Map .......................................................................44
4.2.7. M8.0 NEHRP GeoUnit Data Map ...........................................................................45
4.2.8. M8.0 No User-Supplied Data ..................................................................................46
4.3. Bridge Restrictions and Closures Analysis Results ..........................................................47
4.3.1. Regional Bridge Closures ........................................................................................48
4.4. Summary ...........................................................................................................................56
Chapter 5 Conclusion .................................................................................................................... 58
5.1. Implications.......................................................................................................................59
5.2. Limitations ........................................................................................................................61
5.3. Future Research ................................................................................................................62
References ..................................................................................................................................... 63
APPENDIX A TABLES ............................................................................................................... 67

vi

List of Figures
Figure 1 Southern California Study Region Map ........................................................................... 2
Figure 2 San Andreas Fault Map with Historical Earthquakes and Rupture Sections ................... 4
Figure 3 San Andreas Fault and Earthquake Probability Near Gorman, CA ................................. 5
Figure 4 UCERF3 earthquake probabilities for CA by Fault ......................................................... 9
Figure 5 Flowchart of Overall Methodology ................................................................................ 15
Figure 6 HAZUS Default Highway Bridge Data .......................................................................... 19
Figure 7 Study Region Shear Wave Velocity (Vs) Points ............................................................ 20
Figure 8 STATSGO2 General Soil Types for California ............................................................. 21
Figure 9 USGS Geological Units for California by Dominant Lithology .................................... 22
Figure 10 STATSGO NEHRP Soil Class by Type ....................................................................... 26
Figure 11 USGS CA Geological Units by NEHRP Class Type ................................................... 27
Figure 12 M8.0 Scenario Shakemap Peak Ground Acceleration in G-Force ............................... 29
Figure 13 M7.4 Scenario Shakemap Peak Ground Acceleration in G-Force ............................... 30
Figure 14 2015 National Bridge Inventory National Highway System Bridges .......................... 31
Figure 15 M7.4 Shakemap At Least Moderate Bridge Damage Probabilities ............................. 39
Figure 16 M7.4 STATSGO At Least Moderate Bridge Damage Probabilities ............................ 40
Figure 17 M7.4 Geologic Units At Least Moderate Bridge Damage Probabilities ...................... 41
Figure 18 M7.4 Default Data At Least Moderate Bridge Damage Probabilities.......................... 42
Figure 19 M8.0 Shakemap At Least Moderate Bridge Damage Probabilities ............................. 43
Figure 20 M8.0 STATSGO At Least Moderate Bridge Damage Probabilities ............................ 44
Figure 21 M8.0 Geo Unit At Least Moderate Bridge Damage Probabilities ............................... 45
Figure 22 M8.0 Default Data At Least Moderate Bridge Damage Probabilities.......................... 46
Figure 23 M7.4 Shakemap Restrictions and Closures .................................................................. 48
Figure 24 M7.4 STATSGO Restrictions and Closures ................................................................. 49
Figure 25 M7.4 Geological Unit Restrictions and Closures ......................................................... 50
Figure 26 M7.4 Default Data Restrictions and Closures .............................................................. 51
Figure 27 M8.0 Shakemap Restrictions and Closures .................................................................. 52
Figure 28 M8.0 STATSGO Restrictions and Closures ................................................................. 53
Figure 29 M8.0 GeoUnit Restrictions and Closures ..................................................................... 54
Figure 30 M8.0 Default Data Restrictions and Closures .............................................................. 55

vii

List of Tables
Table 1 Bridge Damage States with corresponding BDI values and Closure Status.................... 14
Table 2 Inventory Data, Soil Maps, Shakemaps, Vs Point Data, and Bridge Point Data ............. 17
Table 3 NEHRP Classes by Vs Values ......................................................................................... 23
Table 4 STATSGO Class Averages .............................................................................................. 24
Table 5 Geo Units Class Averages ............................................................................................... 25
Table 6 Economic Losses and >15% Probability of At Least Moderate Damage ....................... 47
Table 7 HAZUS Scenario Runs and Number of Bridges Restricted or Closed............................ 56
Table 8 HAZUS Highway Bridge Types ..................................................................................... 67
Table 9 M8.0 ShakeMap Bridge Closures ................................................................................... 70
Table 10 M8.0 Default Data Bridge Closures ............................................................................. 75
Table 11 M8.0 STATSGO Bridge Closures ................................................................................ 76
Table 12 M8.0 GeoUnit Bridge Closures .................................................................................... 76
Table 13 M7.4 ShakeMap Bridge Closures .................................................................................. 76
Table 14 M7.4 STATSGO Bridge Closures ................................................................................. 77
Table 15 7.4 GeoUnit Bridge Closures ......................................................................................... 77
Table 16 M7.4 Default Data Bridge Closures .............................................................................. 77

viii

Acknowledgements
I am grateful for the ongoing support and guidance from Dr. Darren Ruddell and my other USC
faculty who gave me assistance when I needed it. I would like to thank the members of my
thesis committee, Dr. Jennifer Swift and Dr. Laura Loyola for providing technical assistance and
conceptual direction for my thesis. I would also like to thank Dr. Su Jin Lee for introducing me
to HAZUS and GIS based spatial modeling. I am forever grateful to my family and friends for
their unwavering support, which I have leaned on and has accompanied me through the entire
process.

ix

List of Abbreviations
HAZUS HAZards United States
UCERF3 3
rd
Uniform California Earthquake Rupture Forecast
M or MM Magnitude or Moment Magnitude
NEHRP National Earthquake Hazards Reduction Program
STATSGO STATe Soil GeOgraphic database
GeoUnit USGS Geological Units for California
Vs Shear Wave Velocity at 30M under surface
PGA Peak Ground Acceleration
PGV Peak Ground Velocity
PSA0.3 Peak Ground Acceleration at 0.3 seconds
PSA1.0 Peak Ground Acceleration at 1.0 seconds
FHWA Federal HighWay Authority
FEMA Federal Emergency Management Agency
SCEC Southern California Earthquake Center
USGS United States Geological Service
USDA United States Department of Agriculture
CGS California Geological Survey
CDMS HAZUS Comprehensive Database Management System
.mdb Microsoft Access Database File
.kml Kernel Markup Language – Google Earth File
GIS Geographic information system

x

Abstract
The San Andreas Faultline is the largest fault in California. On average, this fault has produced
a major earthquake every 150 years, the last to strike the southern section was the magnitude 7.9
Fort Tejon earthquake of 1857. Today the Greater LA Area is one of the largest urban
agglomerations in the world and the second largest metropolitan region in the U.S. The area is
well known for its urban sprawl and expansive highway system. The weak points of any
highway system are the bridges and overpasses. Modeling the effects of an earthquake on this
infrastructure will help inform emergency planning and speed economic recovery.
Experts have predicted a major earthquake, from magnitude 7.0 to 8.0, will strike the
fault within the next 30 years. The goal of this project was to examine enhanced HAZUS hazard
datasets to assess potential earthquake damage to highway bridges and how this may correspond
to bridge closures. NEHRP soils maps were created by joining shear wave velocity data to
STATSGO and Geological Unit data, then classifying each soil unit by the NEHRP class shear
wave velocity range. USGS Scenario ShakeMaps at M7.4 and M8.0 were selected near the area
of fault section with the highest probability of a great earthquake. Eight scenarios were modeled
with this data. Results of this work show that user-supplied datasets for ground motion generally
reduce HAZUS bridge damage outputs, that all earthquake scenarios will significantly damage
southern California bridge infrastructure, and how relative damage state outputs translate into
bridge restrictions and closures.

1

Chapter 1 Introduction
The San Andreas Fault line runs north and south along most of California and has a history of
producing major earthquakes approaching Magnitude (M) 8.0. The last major earthquake on the
San Andreas was the 1906 M7.8 San Francisco Earthquake. Before the San Francisco
earthquake the fault produced a M7.9 earthquake in 1857 near Fort Tejon, and before that, the
southernmost section of the fault near the Salton Sea produced a M7.7 earthquake in 1680 (Jones
2011). Studies suggest that the southern section of the San Andreas Fault will experience a
major earthquake sometime in the next 30 years (Fialko 2006; Bird 2009; Jones 2011; Field
2013). Southern California planners have been working on a comprehensive freeway and
express system since the 1940’s, today it is an interconnected system serving the entire southern
California region of 22 million people (Figure 1). Bridges are the weak link in any highway
system, and the region has over 9,000 road and highway bridges (Faigin 2015). Even though the
system is highly redundant and California has high seismic engineering standards, a major
earthquake will cause significant damage to bridges and immediately impair emergency response
while leading to days or weeks of economic impacts (Moehle 2003). With over 10.1 million
people, Los Angeles County is the most populous county within the Southern San Andreas Fault
zone, and much of its infrastructure will be affected by an earthquake in the southern San
Andreas Fault (US Census 2016; Porter 2011).
2

Figure 1 Southern California Study Region Map
The Federal Emergency Management Agency maintains free natural hazards modeling
software called HAZUS, which runs on top of ArcMap10.4 and allows users to estimate physical
damage, economic and social losses resulting from earthquakes (FEMA 2017). This thesis
investigated the effects of a southern San Andreas major earthquake recurrence, with a focus on
examining how different datasets determining ground motion will affect damage to bridges. The
research examined how to run a custom HAZUS analysis utilizing soils maps and ShakeMaps for
enhanced ground motion modeling and how HAZUS damage state probability outputs may be
used to identify bridge closures. The project implemented two HAZUS earthquake scenarios,
each on the southern San Andreas, with different magnitude values representing a range of
3

‘major’ earthquake magnitudes. To evaluate model output sensitivity to updated datasets, each
earthquake scenario (with event data taken from the selected Scenario ShakeMaps) was run once
with default HAZUS data.
1.1. Motivation
Scholars predict another major to great earthquake along the southern San Andreas Fault
within the next 30 years (Fialko 2006; Bird 2009; Field 2013). The San Andreas Fault line
averages a major earthquake every 150 years and the southern section of the fault is far past this
average recurrence frequency (Figure 2). An earthquake is classified as ‘major’ when its
magnitude is M7.0-7.9 and ‘great’ at magnitude 8.0 or more. The magnitude scale is logarithmic
and not linear, meaning that for every whole number increase in magnitude, the amplitude of
ground motion goes up ten times. (Michigan Tech 2017). Modeling a major or great earthquake
on infrastructure in a Geographic Information System will help us better understand
vulnerabilities and plan appropriate emergency measures to prevent loss of life and economic
interruption.
4

Figure 2 San Andreas Fault Map with Historical Earthquakes and Rupture Sections
Source: whatcausesearthquakes.com

A major earthquake recurrence on the southern San Andreas would affect all southern
California and cause significant destruction near the epicenter and along the ruptured section.
The section of Fault line with the highest likelihood to rupture at M8.0 or greater lies north of
Los Angeles, near the town of Gorman (Figure 3). Gorman is a small town in between Los
Angeles and Bakersfield that sits atop the Tejon Pass in the San Gabriel Mountains. Gorman
does not have a large population, but the I-5 freeway travels directly past it and is the main
North-South freeway artery for all southern California.
5

Figure 3 San Andreas Fault and Earthquake Probability Near Gorman, CA
(UCERF3 .kml data from SCEC 2015)
HAZUS is a natural hazards model using sophisticated loss-estimation methods; it was
developed by the Federal Emergency Management Agency and National Institute of Building
Sciences and released in 2003 (Yeats 2013). HAZUS is used to model natural disasters
including earthquakes, runs as a custom application inside ArcGIS10.4, and its software and data
are freely available to download from the FEMA website.
HAZUS models ground shaking on highway bridges and provide outputs in terms of
relative damage level probabilities (0-1 probability for slight, moderate, extensive, complete
damage), total structure loss (with $ repair cost), or percent reduced functionality (%
functionality at day 1, 3, 7, 14, 30, and 90) (Kircher 2006). Additional analysis of this data
predicted how highway bridge damage outputs could be translated into bridge closures over the
road network. A bridge with complete damage or total loss cannot be crossed, but what
6

functionality or traffic capacity will a moderately or extensively damaged bridge provide the
road network? While default HAZUS analysis uses only one soils class to determine ground
motion, this analysis can be enhanced by adding classified soils maps or ShakeMaps to provide
better data for ground motion modeling (FEMA 2015). A HAZUS analysis utilizing local user-
supplied data typically provide lower damage estimates (Neighbors. 2013). A major earthquake
will affect the entire region, and its unknown to what spatial extent the damages will occur. This
project will implement a HAZUS earthquake analysis for all 10 southern California counties
using soils maps or ShakeMaps, and employ methods to convert the damage state outputs for
highway bridges into thresholds for bridge closures and restrictions.
1.2. Research Goals
The research questions:
• To what degree will a major earthquake on the southern San Andreas Fault damage
bridges in southern California?
• How will user-supplied datasets for ground motion effect bridge damage outputs
compared to default HAZUS data?
• How might HAZUS bridge damage state outputs correspond to bridge closures and
restrictions?
7

Chapter 2 Background and Related Works
This study area for this project is determined by the 10-county definition of Southern California.
This more extensive definition includes the counties of Kern and San Luis Obispo. This
definition was decided on due to the spatial distribution of the San Andreas Fault line and the
location for the highest probability epicenter located near Gorman, on the edge of Los Angeles
and Kern Counties. Also, the predicted lengths of fault rupture and energy released vary greatly
between the range of a M7.0 to M8.0 earthquake event predicted in the next thirty years, and the
study area needed to account for a low to mid end event magnitude and a maximum event
magnitude. There is software other than HAZUS for predicting earthquake losses, such as the
open source software SELENA-RiSE (NORSAR/ICG 2010). HAZUS was selected due to its
integration with ArcGIS and its comprehensive modeling system and outputs for building
damage, economic/social losses and damage to lifeline systems. SELENA-RiSE borrows many
of the earthquake hazard algorithms used in HAZUS but currently only outputs losses due to
building damage (NORSAR/ICG 2010). NEHRP classified soils maps for input to HAZUS or
SELENA define site-specific ground amplifications and must be created by the user based on
local knowledge. Studies for classifying NEHRP maps based on soils and geological units are
discussed in this chapter. Also discussed in this chapter are studies implementing HAZUS
evaluating model sensitivity to input parameters, a HAZUS scenario compared to real-world
losses, and a HAZUS scenario for a hypothetical earthquake on the southern San Andreas meant
to inform emergency preparedness and disaster mitigation. Finally, the outputs for bridge
damage state probabilities are discussed as well as studies that relate these outputs to a Bridge
Damage Index used by engineers and corresponding bridge closures which a local transportation
agency may impose (Park 2001; Shiraki 2007; Richardson 2015).
8

2.1. Southern San Andreas Fault and Earthquake Risk
The San Andreas is the dominant fault line across the state of California. In southern
California, the San Andreas runs from NE San Luis Obispo County along the San Gabriel, San
Bernardino, and San Jacinto mountain ranges, then terminates near the Salton Sea in Imperial
County. Earthquake Risk on the Southern San Andreas has been assessed by many researchers
and organizations. The first researcher to predict a 30-year timeframe for a major earthquake on
the San Andreas was Yuri Fialko from the Scripps Institution of Oceanography at UCSD. In
March 2006 he presented a seminal article in Nature about slip deficit and strain accumulation
on the southern section of the San Andreas Fault. Using synthetic aperture radar data for high-
resolution measurements, he found slip deficit on the southern San Andreas to be in the order of
7-10 meters, which is comparable to the maximum co-seismic offset ever documented on the
fault (Fialko 2006). Other researchers followed with studies predicting California earthquake
likelihood, but the largest is the Working Group on California Earthquake Probabilities, a group
of researchers from several Federal and State agencies and many universities who began
publishing the Uniform California Earthquake Rupture Forecast reports (Field 2013). Field et al.
(2013) produced the UCERF3 Report, detailing rupture probabilities for significant faults
throughout the state of California (Figure 4). UCERF3 is a probability model utilizing
supercomputers to predict earthquake rates at specific magnitudes along major fault lines. Some
of the fault rupture probabilities have decreased compared to UCERF2, the newer model shows a
20% chance of a M6.7 or greater earthquake occurring on the Southern San Andreas Fault within
the next 30 years. The UCERF3 report provides the main motivation, background, and basis for
this studies’ earthquake scenario.
9

Figure 4 UCERF3 earthquake probabilities for CA by Fault
(.kml data from SCEC)
2.2. Regional Geology and NEHRP Soils Classification
The history of earthquakes and unique geology indicates that the the study region is
seismically vulnerable and provides justification for strict earthquake engineering standards.
Olsen et al. (2005) utilized the TerraShake model to show that a chain of sedimentary basins
between San Bernardino and Los Angeles act as an effective waveguide that channels Long
waves (the most destructive earthquake waves) along the southern edge of the San Bernardino
and San Gabriel Mountains. They found that this effect can produce unusually high long-period
ground motion over the region and produce intense shaking from variations in the waveguide
10

cross section. The Olsen study sheds light on a significant portion of my study areas’ lithology,
it lends to the importance of using soils maps to make this HAZUS analysis more accurate and
provides some assumptions on the general qualities of the Greater Los Angeles areas soil basins.
The National Earthquake Hazards Program (NEHRP) provides the soils classification
scheme used in HAZUS modeling software, these soils maps should be custom classified based
on local field data from borehole observations. NEHRP soils classes are defined by the shear
wave velocity (Vs) values of the soil at 30 meters.
Recently, the use of NEHRP classified soils has been validated for use in HAZUS
analyses. Medves (2009) made use of enhanced NEHRP classified soils maps in a HAZUS
analysis for his master’s thesis at the College of Charleston. He incorporated SSURGO and
STATSGO polygon soil data into his analysis by joining shear wave velocity point data from
boreholes to soil mapping units, then assigned NEHRP soil classes according to the soil units
shear wave velocity values. Soils units with no Vs data were assigned the average values of
similar units, then classified for NEHRP values. Thitimakorn et al. (2016) derived NEHRP
classified soils maps for Lamphun City, Thailand using shear wave velocity point values derived
from multi-channel analysis of surface wave (MASW) data. The Vs values were joined to
geological unit polygon data for the area surrounding Lamphun City. NEHRP classifications
were assigned to geological units according to the average Vs value. The authors found the
alluvium and terrace geologic units surrounding the city to be classified as NEHRP type D and
C. These previous analyses provided information and methods for determining NEHRP
classifications for soils maps used in this HAZUS analysis.
11

2.3. HAZUS-MH Implementation Research
HAZUS has been available to the public for almost fifteen years. There are several
studies that address model parameters, ShakeMaps, soil maps, updated datasets, and model
sensitivity. A major resource for HAZUS users is the extensive HAZUS User and Technical
Manuals that outline how to create hazard scenarios and explain modeling parameters,
methodologies, and details for inputs and outputs. FEMA began the HAZUS initiative in 1997
under an agreement with the National Institute of Building Sciences and released HAZUS-MH in
2003 to assist US municipalities or businesses with hazard loss estimation. HAZUS-MH 4.0
uses ArcMap10.4 to map building inventories, soil conditions, faults, critical infrastructure, and
lifelines to estimate economic loss, physical damage, and social impacts (Yeats 13). The quality
of HAZUS modeling can also be enhanced with the addition of soils maps, ShakeMaps, and the
AEBM; which provides analysis for individual buildings as opposed to the general building
stock data aggregated at the census block level (FEMA 2015a). Porter et al. (2011) used
HAZUS in the Shake Out Scenario to evaluate a hypothetical 7.8 magnitude earthquake striking
the Southern San Andreas Fault. Physics-based modeling was used to create shaking intensity
and peak ground motion maps. A custom HAZUS analysis utilized 18 special studies to analyze
the effects of the hypothetical earthquake on regional infrastructure. The earthquake scenario
studied caused 1800 deaths, 53,000 casualties requiring emergency room care, 1600 fires
destroying 200 mil ft. of building stock, and a total of 191 billion in economic losses. The study
also found widespread damage to highway bridges and major interstates, taking up to seven
months to repair. Emergency response activities are also depicted showing activities over time.
Using HAZUS to model the effects of the next major earthquake on bridge infrastructure data
12

will assist in regional planning, and emergency response, as well as help mitigate economic
losses.
ShakeMap datasets and understanding model outputs for damage states are important to
this study. Kircher et al. (2006) focus on building-related methods of HAZUS loss estimation as
well as the Advanced Engineering Building Module and the use of ShakeMaps for ground
motion data to assess potential damage immediately following an earthquake. The study created
a comparison of losses between actual 1994 Northridge Earthquake data and losses generated by
HAZUS with ShakeMaps ground motion data collected from the same earthquake. Damage state
probabilities and fragility curves are also discussed in detail, these are datasets which HAZUS
outputs to assess building and infrastructure damage. This paper provides insights into
ShakeMap use in HAZUS and evaluates damage state probability outputs compared to ground
truth data. Another sensitivity analysis for HAZUS earthquake model outputs was examined by
Neighbors et al. (2013) using a case study in King County, Washington. The research addresses
hazard input parameters for earthquake scenarios and how sensitive ground motion and
economic loss estimates are to earthquake variables. Primarily concerned with monetary
building damages, their results show economic loss scenarios in the study area are more sensitive
to changes in earthquake hazard source parameters than changes in site conditions from user-
supplied data sets and that changes in source parameters for earthquake intensity influence large
variability in building damage. They found that user-supplied datasets can produce lower
damage results than default HAZUS data This research provides assumptions on output
relationships between scenario variables, and output differences between default data and user-
supplied data.
13

Extracting HAZUS results data and utilizing appropriate highway bridge damage state
outputs addressed by Curtis (2016), who implemented a HAZUS transportation analysis in the
Dallas/Fort Worth metro area to obtain shelter needs, shelter points, and bridge damage state data
for an emergency services location-allocation analysis as her master’s thesis at USC. Levels of
bridge damage states were used to identify bridge closures used as impediment points for the
Location-Allocation analysis.
2.4. Bridge Damage States, Damage Probabilities, and Bridge Closure
Once the enhanced datasets that will used were defined, earthquake scenarios established,
analysis run and data extracted, damage states and damage state probabilities that are acceptable
for use in a bridge closure analysis also need to be identified. Additionally, damage state levels
that qualify for bridge closures in California must be delineated. The California Geological
Survey (CGS) released a report in June 2009 detailing HAZUS loss estimation for California
Scenario Earthquakes. The CGS reported building and infrastructure to be in moderate damage
states when a damage state probability is at least 50% (CGS 2009). Richardson et al. (2015)
used the Southern California Planning Model (SCPM) and the Early Post Earthquake Damage
Assessment Tool (EPEDAT) to model a hypothetical M7.1 earthquake on the Elysian Park Fault
near downtown Los Angeles. While the authors found bridge damage states to be standardized
among earthquake models, corresponding bridge functionality for the highway network was
highly subjective. They recognize different acceptable risk levels at moderate and severe
damage states for bridge closures and suggest different traffic restriction mitigations for
maintaining some traffic capacity in a moderate damage state. The importance of this study for
my research lies in the methods the authors used to convert relative qualitative damage
descriptions (‘moderate damage,’ ‘severe damage’) into a corresponding range of engineering
14

Bridge Damage Index (BDI) values and then thresholds for bridge closure or traffic restriction
based on a transportation agencies level of acceptable risk. Shiraki et al. (2007) examine
transportation network delay from earthquake damage in Los Angeles and Orange County using
simulations with bridge fragility curves to evaluate bridge damage states in terms of bridge
damage index. Bridge fragility curves indicating damage as ‘at least minor/moderate/major’ and
‘collapse’ were related to BDI values. This research reinforces Richardson et. al. (2015) and
relates quantitative bridge damage states to relative states, such as ‘Moderate’ or ‘Extensive.
This research is summarized in Table 1 below. Park et al. (2001) provided background on the
Bridge Damage Index Method and what index values correspond to in terms of structure
damage. The damage index method uses the change in modal strain energy of the pre-damage
and post-damage structure to detect, locate, and size damage in a structure. The authors found a
strong correlation between predicted BDI and observed damages as well as a significant
influence from environmental conditions during wet and dry seasons.

Table 1 Bridge Damage States with corresponding BDI values and Closure Status

Bridge Damage
State
Damage
State
Acceptance
Probability
Bridge
Damage
Index
Value
Richardson
(2015) CA
Closure Status
No Damage > 0.50 0 Open
At Least Slight
Damage
> 0.50 0.1 Open
At Least
Moderate
Damage
> 0.50 0.3 Restricted
At Least
Extensive
Damage
> 0.50 0.75 Closed

15

Chapter 3 Data and Methodology
This chapter outlines the data used and the methodology implemented for this study. This
project consisted of a HAZUS earthquake analysis for a 10-county definition of southern
California and uses the outputs for highway bridge damage to determine bridge closures and
restrictions immediately following the event, as illustrated in Figure 5. Datasets used for this
project are discussed first, then the methodology developed in the HAZUS model runs is
discussed, followed by the methodology for determining bridge closures and restrictions.

Figure 5 Flowchart of Overall Methodology
16

3.1. Datasets Used
All data for this project were freely available online and downloaded from various
organization or federal government agency websites (SCEC 2015; Soil Survey Staff 2016;
Ludington 2005; Young 2015; USGS 2017; FEMA 2017). As mentioned in Chapter 1, HAZUS
comes with a preloaded default data inventory mostly based on 2010 census data aggregated at
the tract level. Statewide hazard and structural inventory datasets were downloaded from the
FEMA website and loaded into the Comprehensive Database Management System (CDMS),
which can be defined to create a highly detailed study region for a specific analysis. The
statewide database inventory in the CDMS is in Microsoft Access (.mdb) format and can be
customized with updated or user-defined datasets uploaded to the statewide inventory through
the CDMS or input as Data Maps in the HAZUS hazard scenario setup menu.
Creation of NEHRP classified soils maps through a spatial join between Shear Wave
Velocity (Vs) point values to STATSGO, and California Geological Units maps involved careful
data pre-processing steps which introduced possible errors as well as assumptions. A detailed
explanation of the methods and underlying goals of the spatial join can be found in Medves
(2009) and Thitimakorn et al. (2016). I this study, a new field for NEHRP classes was added and
filled by the NEHRP class Vs value corresponding to that soil units average Vs value.
Scenario ShakeMaps are hypothetical earthquake scenarios created by USGS
seismologists which reflect the characteristics of a specific fault and display rupture length and
ground motion data (USGS 2016). Two Scenario ShakeMaps at M8.0 and M7.4 were chosen to
represent the likely range of probable values based on UCERF3 forecasts (Field 2013) for a
recurrence of a significant earthquake event on the San Andreas.
17

The 2015 data from the National Bridge Inventory was downloaded from the FHWA
website, with 9 fields matched and formatted for upload to the CDMS (FHWA 2015). There
were 20 different HAZUS bridge classifications identified in this data using the HAZUS CDMS
data dictionary (FEMA 2013). The bridge classification data is 5 years more recent than the
default HAZUS bridge inventory and includes all National Highway System bridges in the study
area. An overview of these datasets can be found in Table 2 below.
Table 2 Inventory Data, Soil Maps, ShakeMaps, Vs Point Data, and Bridge Point Data
Dataset Source
Temporal
Scale
Spatial
Scale
HAZUS
Default
Inventory
FEMA/HAZUS 2010 native data.
Lifelines - Transport systems - Highway
- Bridges.
2010
County or
region
scale
Soils Maps
USDA STATSGO:
USGS Geological Units for California:
STATSGO:
2016
GeoUnit:
2005
Statewide
CA
ShakeMaps USGS Earthquake Scenario Map 2017
Regional
Extent
Shear
Wave
Velocity
Data
USGS Compilation of Vs 30 values in
the US
2015
Regional
Point data
National
Bridge
Inventory
Federal HighWay Authority National
Bridge Inventory (Prepared but not used
in analysis)
2015
Regional
point data

3.1.1. HAZUS Census tract and default inventory data
The HAZUS default data inventory consists of 2010 census data and buildings
aggregated at the tract level as well as point or line data for lifelines and Essential Facilities
(FEMA 2015). Depending on the modules chosen for an individual analysis, HAZUS uses all of
this data to estimate casualties and shelter needs, damage to default structural inventory, and
18

indirect and direct economic impacts. HAZUS does not identify multiple bridges which span
over each other, thus stacked bridge data points will be represented as multiple point objects with
similar coordinates, yet will be designated according to their HAZUS highway bridge class. (See
Appendix A Table 8 for details on HAZUS highway bridge classes).
This project deploys HAZUS, with upgraded ground motion and soils data compared to
HAZUS default soil inputs, to model the earthquake scenarios effects on the HAZUS default
inventory bridge data. The output data of concern is from the Lifelines Transportation Systems
module. Results are provided as point data representing bridges and the discrete probabilistic
damage states of no damage, slight damage, moderate damage, extensive damage, and complete
damage as well as cumulative damage states of ‘at least moderate’ and ‘at least extensive’
(FEMA 2015a). For efficient processing time and to maintain focus on the previously stated
research questions, only the necessary modules for analyzing highway bridge damages were
selected to illustrate the effects of the chosen earthquake scenarios on the default bridge
inventory, and how damage to the bridge data differs between scenarios used in the analysis.
Figure 6 illustrates the 9,516 input bridges in the default HAZUS bridge inventory, represented
as points.
19

Figure 6 HAZUS Default Highway Bridge Data
3.1.2. Shear Wave Velocity Point Data
NEHRP soils classes are defined by different ranges of shear wave velocities (Vs) in the
upper 30M of soil (FEMA 2015b; Medves 2009). In seismic hazard analysis, the Vs at a given
location is of interest because it gives an indication of whether the expected shaking in response
to specified earthquake event may be high or low. Vs point data was obtained from the USGS
Earthquake Hazards Program online data map (Yong 2015). Values from the southern California
area were selected and exported in .csv file format. Some values lie to the north outside the
study region, these are included so these locations could contribute to overall spatial continuity
through spatial joins for soil and geological mapping units which lie in the study area and extend
20

over the northern boundary. The .csv data contains 777 Vs points with latitude and longitude
values, Vs values, total depth values, and the survey method used (Figure 7). The latitude and
longitude point values were displayed using ArcMap and exported as a shapefile layer for the
spatial join to soils and geological mapping unit data.

Figure 7 Study Region Shear Wave Velocity (Vs) Points
3.1.3. STATSGO2 Soils Maps
Statewide STATSGO2 soils maps were downloaded from the USDA Web Soil Survey
data warehouse (Soil Survey Staff 2017). STATSGO2, the US General Soil Map, was
developed by the National Cooperative Soil Survey. It is a broad-based inventory of soils and
non-soils areas shown at a scale designed for planning and management over state and multi-
21

state areas. The dataset was created by generalizing more detailed soil maps (USDA 2017). For
this reason, it was selected for this study region. Soil Taxonomy class names for the dataset are
joined to the spatial data within ArcMap using the USDA Soil Data Viewer extension for
ArcMap (Figure 8).

Figure 8 STATSGO2 General Soil Types for California
3.1.4. USGS Geological Unit Map for California
A Geological Units map for the state of California was downloaded from the USGS
Mineral Resources online spatial data warehouse. The data was originally organized by Jennings
et al. (1977) and updated by Ludington et al. in 2005. Intended to be combined into regional
maps to depict age and lithology of map units, this data is aggregated for the state of California
22

and is intended for geographic scales smaller than 1: 500,000. The field ‘RockType1’ is of main
interest for this study area and describes the dominant lithology type in each mapping unit. The
dominant lithology for southern California is Alluvium (Figure 9).

Figure 9 USGS Geological Units for California by Dominant Lithology
3.1.5. Custom NEHRP classified Soils Maps
Mapping unit polygons represent distinct soil or lithology classes. Vs data points were
joined by spatial location to the mapping units they overlay. NEHRP classes of the mapping
units were then determined by the mapping units’ average Vs values and queried by a range of
Vs values in NEHRP classes; these layers included a NEHRP Class field added and filled by
class type. NEHRP classes range from Class A (highest Vs values and lowest amplification) to
23

Class E (lowest Vs values and highest amplification). Table 3 below displays the Vs range for
each class and general lithology characteristics typical of that class. Average Vs point value data
joined to STATSGO soil units or USGS GeoUnits serves as the ground truth Vs value for any
individual mapping unit.
Table 3 NEHRP Classes by Vs Values (Definitions from USGS Earthquake Hazards Program)
NEHRP
Soil Type
Shear
Wave
Velocity
(Vs)
Generalized Lithology Definition
A
Vs > 1500
m/sec
Includes unweathered intrusive igneous rock. Soil types A and B do not
contribute greatly to shaking amplification.
B
1500
m/sec > Vs
> 750
m/sec
Includes volcanics, most Mesozoic bedrock, and some Franciscan bedrock.
C
750 m/sec
> Vs > 350
m/sec
Includes some Quaternary sands, sandstones and mudstones, some Upper
Tertiary sandstones, mudstones and limestone, some Lower Tertiary
mudstones and sandstones, and Franciscan melange and serpentinite.
D
350 m/sec
> Vs > 200
m/sec
Includes some Quaternary muds, sands, gravels, silts, and mud. Significant
amplification of shaking by these soils is generally expected.
E
200 m/sec
> Vs
Includes water-saturated mud and artificial fill. The strongest amplification
of shaking due is expected for this soil type.

Mapping units with no data joined to them in either STATSGO or GeoUnit maps were
assigned the average Vs value of all similar STATSGO taxonomy classes or GeoUnit rocktype1
classifications and given the appropriate NEHRP class category according to the average class
value (Table 4 and Table 5). Soil taxonomy and rocktype1 classes with no Vs data were
assigned NEHRP classes based on general lithology from the Geologic Map Unit Classification
6.1 guide (USGS Staff 2002) or soil class characteristics from the USDA Soil Taxonomy guide
(USDA 1999) and similar rock domains while using the descriptions in Figure 10 as a guide.
24

Table 4 STATSGO Class Averages
Grouped Taxonomy
Classes
Vs
Avg
NEHRP
Class
#
Objects
# Vs
Points
No Tax Name 584.2 C 407 374
Clayey-Skeletal 447.3 C 9 3
Coarse-Loamy 394.2 C 142 76
Fine-Loamy 410.9 C 228 70
Fine-Silty 403.3 C 7 3
Fine 496.9 C 112 97
Loamy-Skeletal 587.1 C 135 16
Loamy-Skeletal 616.3 C 162 61
Medial-Skeletal ND C 1 0
Mixed 461.5 C 110 54
Sandy-Skeletal 403.2 C 30 5
Sandy-Skeletal 451.5 C 10 3
Torriorthents ND C 3 0
*ND = No Data

25

Table 5 Geo Units Class Averages
RockType1
Avg
Vs
NEHRP
Class
#
Objects
# Vs
Points
Alluvium 401 C 401 442
Andesite ND B 13 0
Anorthosite ND C 8 0
Argillite 469.5 C 56 4
Basalt 634.6 C 198 4
Conglomerate 439 C 149 1
Dune Sand 285.3 D 33 2
Felsic Volcanic
Rock
ND B 89 0
Gabbro 533.5 C 111 2
Glacial Drift ND C 3 0
Gneiss 521 C 389 12
Granite ND C 73 0
Granodiorite 692.6 C 587 23
Greenstone ND C 27 0
Int. Volcanic Rock ND B 12 0
Landslide ND C 6 0
Limestone ND C 104 0
Marble ND C 22 0
Melange ND C 1 0
Metavolcanic Rock ND C 1 0
Mica Schist 401 C 33 2
Mudstone 580.5 C 146 3
Orthoquartzite ND C 2 0
Pelitic Schist 501 C 10 1
Peridotite ND C 1 0
Plutonic Rock 579.8 C 177 7
Quartz Diorite 756 B 2 1
Quartz Monozite 927 B 11 1
Rhyolite 589 C 489 4
Sandstone 458.3 C 1249 36
Schist 893 B 224 2
Serpentinite 505 C 43 1
Siltstone ND C 1 0
Tephrite (basinite) ND B 6 0
Tolanite 654.7 C 120 12
Water 636.3 C 174 5
ND = No Data
26

A significant portion of the study region was classified as other than NEHRP Class D
(the default HAZUS soils class for the study area). Type B and C NEHRP soils were identified
throughout the study area, each class has higher Vs values than Class D, which attributes lower
shaking amplifications to these soils and a lower hazard level. These results for the NERHP
classifications for both STATSGO and GeoUnit maps (Figure 10 and Figure 11) confirm the
Neighbors et al. (2013) finding that hazard maps with local user-defined data decrease HAZUS
loss outputs due to the default soil class for HAZUS models being set as Class D.

Figure 10 STATSGO NEHRP Soil Class by Type
27

Figure 11 USGS CA Geological Units by NEHRP Class Type
Once all soil and geological units were NEHRP classified the datasets were formatted for
upload to the HAZUS hazard scenario data maps according to Appendix K in the HAZUS User
Manual (FEMA 2015a).
3.1.6. USGS Scenario ShakeMaps
Scenario ShakeMaps are hypothetical earthquake scenarios created by the USGS that
adhere to the known characteristics of a specific fault and display rupture length and ground
motion data as Peak Ground Acceleration (PGA), Peak Ground Velocity (PGV), Spectral
Acceleration at 0.3 seconds (SA0.3), and Spectral Acceleration at 1.0 seconds (SA1.0) (USGS
2016). ShakeMap datasets account for all necessary ground motion data needed for HAZUS
28

analysis. Epicenter, rupture length, depth, moment magnitude, and fault type data compiled by
USGS scientists are incorporated into this HAZUS scenario event data, removing significant
uncertainty and user-input mistakes regarding the hazard event. Two Scenario ShakeMaps were
selected and downloaded from the online USGS Earthquake Scenario Map (USGS 2017) to
represent a range of predicted magnitudes from UCERF3 for the area around Gorman, CA. One
M8.0 ShakeMap and one M7.4 ShakeMap with both epicenters a few miles apart on the southern
San Andreas Fault Carrizo Section which was identified as having the highest probability of a
M8.0 or greater event, shown below in Figure 12 and Figure 13. The event parameters from
these ShakeMaps, such as latitude and longitude location of the epicenter, fault section, and
magnitude were used to define the two earthquake scenarios for the other datasets tested in this
study. These ShakeMaps from the USGS Scenario Map were downloaded as shapefiles in
HAZUS format and were transferred to a .mdb HAZUS compatible database format for upload
to the hazard data maps.
29

Figure 12 M8.0 Scenario ShakeMap Peak Ground Acceleration in G-Force
30

Figure 13 M7.4 Scenario ShakeMap Peak Ground Acceleration in G-Force
3.1.7. FHWA National Bridge Inventory
Prepared, but not used in this analysis due to current technical problems with the HAZUS
CDMS. 2015 NBI data is updated bridge point data downloaded from the Federal High Way
Authorities’ National Bridge Inventory website. Highway bridge data is part of the HAZUS
Transport Systems module and must be uploaded through the CDMS before the study region,
and hazard scenario parameters are established. This data was intended to replace the default
HAZUS bridge data in the statewide dataset, which is based on older 2010 NBI data. HAZUS
bridge class designations were determined by querying and compositing several NBI fields, 20
different HAZUS bridge classes were identified for the study region (Figure 14) and 9 other NBI
31

fields were identified for field matching in the CDMS. One deficiency in user-supplied NBI data
is that replacement costs for bridge points are not a field in the NBI dataset but are defined in
HAZUS default data. However, uploading this data into the CDMS exposed a field mismatch in
the CDMS and made this dataset inaccessible for analysis runs until the HAZUS development
team can fix the problem in the next version of HAZUS/CDMS.

Figure 14 2015 National Bridge Inventory National Highway System Bridges by HAZUS Bridge
Class. Please see Appendix A for HWB definition table
3.2. HAZUS Earthquake Modeling using Soils Maps and ShakeMaps
HAZUS default data defaults all soils as in its inventory as NEHRP type D. Adding a
regional soils map to the analysis dataset will classify the various soils that the building and
32

infrastructure inventory overlays as one of five NEHRP soils classes designated as soil type A
thru E (fig 10). Different soil types transfer earthquake wave energy with varying intensity, with
soft-soils (class E) providing the greatest amplification. Classifying NEHRP types within
established soils units will lead to more accurate regional analysis when compared to the default
soil type ‘D.’ ShakeMaps provide historical or probable snapshots of earthquake ground motion
parameters and allow HAZUS to use these values to more accurately predict ground shaking.
Neighbors 2013 identified generally lower earthquake damage and ground motion output values
when using user-supplied data (Neighbors et al. 2013). Using soils and ShakeMaps in the
current HAZUS analysis will remove some model uncertainty with respect to ground motion and
soil amplification and provide more accurate bridge damage outputs from my infrastructure data.
3.2.1. Setting up HAZUS and loading User Supplied Datasets
Setting up HAZUS for analysis first involves downloading a state-specific dataset from
the FEMA website, unpacking it and designating the database path in the CDMS options menu.
Upon opening HAZUS, the user is asked to create or import a study area. The hazard mode
(Earthquake in this project) was selected, and the analysis aggregation scale was set to ‘County’
level, then individual counties for southern California were selected, the study region was then
generated from the statewide data and saved to be used for multiple run scenarios.
User-supplied data is uploaded to HAZUS through the ‘Data Maps’ option in the hazard
setup menu or through CDMS upload to the statewide dataset. NEHRP classified soil maps and
PGA, PGV, PSA0.3, and PSA1.0 data from Scenario ShakeMaps are input into ‘Data Maps’ in
the HAZUS Hazard drop-down menu. Once imported to the Data Maps menu, soils maps are
selected for the current hazard scenario by opening the scenario wizard and selecting ‘Define
Data Maps’ option and setting the soils drop-down menu to the desired map previously uploaded
33

to the Data Maps table. The software then applies this map to the study region and updates
inventory tables to reflect the soil type the inventory data overlays. Scenario ShakeMap data are
constructed by USGS for hypothetical earthquake events and are intended for planning purposes
only. Scenario ShakeMap data is input into a study region scenario through the scenario setup
wizard in the Hazard drop-down menu. A deterministic User-Defined scenario must be selected
to input the Data Maps established for the ShakeMap defined PGA, PGV, PSA0.3, and PSA1.0
contour maps. These maps account for all ground motion data to be calculated by the software,
which makes soil map data unnecessary (FEMA 2015b).
3.2.2. Setting Hazard Event Scenarios and Running Analysis
UCERF3 earthquake probabilities data for California was used to reference the fault
section with the highest probability for a great (M>8.0) earthquake in southern California. This
data showed the San Andreas-Carrizo fault section with an epicenter near Gorman, CA to have
the highest probability of a M8.0 or greater earthquake in all southern California. Setting the
hazard event scenarios for this project involved using the Scenario setup wizard to create two
deterministic User-Defined scenarios using ShakeMap datasets to define the hazards, two
deterministic Source scenarios using STATSGO and GeoUnit based NEHRP classified soils
maps, and two Source scenarios using only default HAZUS data and no user-supplied soils maps
or ShakeMaps. Two USGS ShakeMaps at M8.0 and M7.4 were selected from the USGS
website. The M8.0 ShakeMap has an epicenter in Gorman, CA and the M7.4 ShakeMap has
epicenter a short distance to the east in the town of Fraser Park, Ca. For accurate comparison,
the event parameters for magnitude, epicenter location, and source fault of these two ShakeMaps
were used to define the same event parameters for the Source scenarios. Once the hazard
scenarios were input, the relevant HAZUS analysis modules to use in the model were selected,
34

and the analysis run. With only the Transportation-Highway Bridges analysis selected (7
modules needed to analyze the data), each scenario for southern California only took several
minutes for HAZUS to process. Uploaded ShakeMaps data in the user-defined scenarios took
several minutes of processing time. Uploading a soils map in the define hazards wizard took at
least an hour of processing time for HAZUS and often crashed before applying the soil map
NEHRP values to inventory datasets. Uploading soils maps was the most problematic, as after
the wizard was done processing, the maps would show as ‘current’ in the ‘display current
hazard’ window, but the inventory data was not updated. After uploading it is necessary to
check the Inventory-Transportation Systems drop-down menu to confirm that the study region
bridge inventory soil class has been updated to the soil map and does not display only the default
HAZUS value of class ‘D’.
3.2.3. Exporting HAZUS Output Data
Once HAZUS had finished an analysis, individual output layers must be called to
ArcMap from the ‘Results’ tab Transportation Systems option. Bridge damage state probability
data were mapped from the results tab and exported as layer files to into a created file
geodatabase. One result layer for highway bridge damage will contain all damage state
probabilities in its attribute table. The damage state probabilities of ‘Exceeding Moderate,’
‘Exceeding Extensive,’ and ‘Complete’ damages are the fields of interest to this projects bridge
closure analysis.
3.3. Bridge Closure Analysis
For this project, HAZUS was used to output predicted bridge damage state probabilities
for each scenario (2) and dataset (4) modeled. Damage state probabilities are output as attribute
table fields in the results point data for bridges. These indicate the 0.0 to 1 probability that an
35

individual bridge will be in a discrete slight/moderate/extensive/complete damage state, as well
as cumulative damage states labeled as at least slight/moderate/extensive. HAZUS leaves the
significance threshold for damage state probabilities are open to user interpretation. To assess
regional bridge closures and restrictions, this research relied on previous studies significance
thresholds for reporting damage states, then used bridges meeting the stated damage state
probability threshold as actual damage states corresponding to closure and restriction criteria
suggested by Richardson et al. (2015).
3.3.1. Bridge Damage States to California Bridge Closures
Relative damage state probabilities, such as ‘moderate’ and ‘extensive’ are how HAZUS
and similar earthquake models output structural damage for building and lifelines like bridges.
Other research has used different levels of probability as thresholds for accepting that a structure
will reach this damage state for reporting losses. Kircher et al. (2006) discusses median demand
parameters for ground shaking and the effect on damage state variability. When comparing
predicted HAZUS building losses using ShakeMaps to actual data from the Northridge
earthquake the number of buildings with extensive or complete damage, the accepted probability
thresholds were not explicitly stated but might be implied in a an example fragility curve table at
50%. Barbat et al. (2008) examined predicted earthquake building losses in Barcelona and
identified probabilities in the 30-40% range as significant for severe damage states. Curtis
(2016) used a 20% probability threshold for moderate bridge damage states while using HAZUS
for transportation network analysis. The California Geological Survey in its June 2009 Project
Report on HAZUS Loss Estimation for California Earthquakes (CGS 2009) Reported moderate
bridge damage at 50% or greater damage state probabilities. Per the CGS report, this analysis
36

will use a 50% probability threshold for identifying ‘at least moderate’ and ‘at least extensive’
damage to bridges for means of identifying bridge closures and restrictions.
The Bridge Damage Index (BDI) method is how seismic engineers assess specific bridge
damage during field inspections (Park 2001; Shiraki 2007). Actual bridge closures and
restrictions are dependent on the relative rules and acceptable risk policies of transportation
departments in each state. Since earthquakes are common events in California, the bridge
infrastructure was constructed with high seismic risk considerations and the state DOT is more
liberal than other states when determining the minimum BDI for full bridge closure (Richardson
2015). Levels of closure and restriction are important for emergency response and economic
recovery, as even though the southern California road network is highly redundant, closing a
bridge that could handle restricted traffic in the hours, days, or weeks after this event will affect
access to casualties and shelter points as well as have significant indirect economic impact.
Richardson et al. suggests that for states like California, management strategies for moderately
damaged bridges can maintain limited levels of traffic capacity and allow restricted access, while
severe or extensive damage states would still pose an unacceptable risk and be closed to all
traffic. Its accepted that slightly damaged bridges will remain open with minimal risk and return
to full functionality days after the earthquake event. The bridge closure threshold for Richardson
et al. was set at 0.75 BDI, which corresponds to the ‘severe/extensive’ damage state. Traffic
restrictions begin at 0.33 BDI, which corresponds to ‘moderate’ damage states. For the current
analysis, Bridges with probabilities of 0.5 or greater for at least moderate or extensive damage
states were assigned corresponding traffic restrictions and closures by adding an attribute field
indicating C (closed) or R (restricted). The query for at least moderate damage excluded bridges
that also met probability requirements for at least extensive damage.
37

Chapter 4 Results
This chapter is divided into three parts. The first section provides a summary of the final results
from the bridge closure analysis. The second section provides results from each of the eight
HAZUS runs as well as a map for each run with outputs displayed for bridge damage
probabilities exceeding at least moderate damage states. The final section examines the eight
runs in terms of bridge closures and restrictions derived from the cumulative damage state
probabilities.
4.1. Final Results Summary
ArcMap layers were exported from HAZUS results according to each of the two
earthquake scenarios and the soils or ShakeMap dataset used in the analysis. Due to the rather
high levels of minimum damage states and probability thresholds required for analysis a small
percent of the total bridge inventory was identified for closures or restrictions in either scenario.
In general, those bridges closest to the fault rupture zone returned the highest probabilities for
damage to exceed moderate or extensive and therefore be closed or restricted. However, all 8
hazard scenario runs derived restricted bridges and 6 of 8 runs derived closed bridges. These
results are provided in detail in Table 6 at the end of section 4.3
As expected, the model runs using default data generally produced higher damage states
due to higher soil amplifications from the default HAZUS soil class of NEHRP type D. The one
unexpected exception is the M8.0 Scenario ShakeMap which produced several times the number
of bridge closures and restrictions over a much larger spatial extent as the M8.0 default data run.
38

4.2. Bridge Damage Probabilities to Exceed At Least Moderate Damage
The results in the sub-sections below indicate the range of probability a bridge point will
sustain at least moderate damage. These probability levels were chosen to illustrate bridges that
could be damaged by the scenario but did not meet the 50% or greater probability requirement to
be included in the bridge restrictions and closures analysis. Bridges not meeting the significant
probability requirement are displayed with ranges of 15-49.9% probability of at least moderate
damage. Only those bridge points in red indicate bridges with at least moderate damage
probabilities of 50% or greater and were determined to be restricted or closed based on the
probability for at least extensive damage also to be 50% or greater.
Describing the effects of each scenario on the whole dataset of 9,516 bridge points can be
challenging even with maps. The sum of predicted economic losses for the whole bridge
inventory is reported as well as the number of bridges that exceed 15% probability of at least
moderate damage. These values are summarized in Table 6 at the end of this section.

39

4.2.1. M7.4 Scenario ShakeMap
The M7.4 Scenario ShakeMap run results (Figure 15) produced 43 bridges with greater
than 15% probability of at least moderate damage. HAZUS predicts a total bridge economic loss
of $13,991,710.

Figure 15 M7.4 ShakeMap At Least Moderate Bridge Damage Probabilities
(Point Counts in Parenthesis)

40

4.2.2. M7.4 NEHRP STATSGO Data Map
The M7.4 STATSGO run results (Figure 16) produced 32 bridges with greater than 15%
probability of at least moderate damage. HAZUS predicts total bridge damage related economic
loss of $13,971,880.

Figure 16 M7.4 STATSGO At Least Moderate Bridge Damage Probabilities
(Point Counts in Parenthesis)

41

4.2.3. M7.4 NEHRP GeoUnit Data Map
The M7.4 Geological Units run results (Figure 17) produced 32 bridges with greater than
15% probability of at least moderate damage. HAZUS predicts total bridge damage related
economic loss of $10,545,840.

Figure 17 M7.4 Geologic Units At Least Moderate Bridge Damage Probabilities
(Point Counts in Parenthesis)
42

4.2.4. M7.4 No User-Supplied Data
The M7.4 Default data run results (Figure 18) produced 50 bridges with greater than 15%
probability of at least moderate damage. HAZUS predicts total bridge damage related economic
loss of $20,189,830.

Figure 18 M7.4 Default Data At Least Moderate Bridge Damage Probabilities
(Point Counts in Parenthesis)

43

4.2.5. M8.0 Scenario ShakeMap
The M8.0 Scenario ShakeMap results (Figure19) produced 707 bridges with greater than
15% probability of at least moderate damage. HAZUS predicts total bridge damage related
economic loss of $284,880,410.

Figure 19 M8.0 ShakeMap At Least Moderate Bridge Damage Probabilities
(Point Counts in Parenthesis)

44

4.2.6. M8.0 NEHRP STATSGO Data Map
The M8.0 STATSGO run results (Figure 20) produced 85 bridges with greater than 15%
probability of at least moderate damage. HAZUS predicts total bridge damage related economic
loss of $56,776,890.

Figure 20 M8.0 STATSGO At Least Moderate Bridge Damage Probabilities
(Point Counts in Parenthesis)

45

4.2.7. M8.0 NEHRP GeoUnit Data Map
The M8.0 Geological Units run results (Figure 21) produced 74 bridges with greater than
15% probability of at least moderate damage. HAZUS predicts total bridge damage related
economic loss of $33,805,660.

Figure 21 M8.0 Geo Unit At Least Moderate Bridge Damage Probabilities
(Point Counts in Parenthesis)

46

4.2.8. M8.0 No User-Supplied Data
The M8.0 default data run results (Figure 22) produced 102 bridges with greater than
15% probability of at least moderate damage. HAZUS predicts total bridge damage related
economic loss of $80,541,180.

Figure 22 M8.0 Default Data At Least Moderate Bridge Damage Probabilities
(Point Counts in Parenthesis)

47

Table 6 Economic Losses and >15% Probability of At Least Moderate Damage

Bridges
Sum of
Economic
Losses
# with >
15% prob
of At
Least
Moderate
Damage
M7.4
GeoUnit
$10,545,840 32
M7.4
STATSGO
$13,971,880 32
M7.4
ShakeMap
$13,991,710 43
M7.4
Default Data
$20,189,830 50
8.0 GeoUnit $33,805,660 74
M8.0
STATSGO
$56,776,890 85
M8.0
ShakeMap
$284,880,410 707
M8.0
Default Data
$80,541,180 102

4.3. Bridge Restrictions and Closures Analysis Results
This map series depicts bridge restrictions and closures derived from damage state
probabilities according to Richardson et al. (2015). These are determined by greater than 50%
probability of at least moderate damage for restrictions and greater than 50% probability of at
least extensive damage for full closure. A number of either restrictions or closures is reported,
this data is displayed in Figure 23 through Figure 30 below, and is summarized in Table 7 at the
end of this section.

48

4.3.1. Regional Bridge Closures
The M7.4 ShakeMap run, returned four restrictions and one closure, all just south of the
epicenter at the junction of Interstate 5 and Hwy 138 (Figure 23).

Figure 23 M7.4 ShakeMap Restrictions and Closures
(Point Counts in Parenthesis)

49

The M7.4 STATSGO run, returned three restrictions and no closures, all just south of the
epicenter at the junction of Interstate 5 and Hwy 138 (Figure 24).

Figure 24 M7.4 STATSGO Restrictions and Closures
(Point Counts in Parenthesis)

50

The M7.4 Geological Unit run, returned three restrictions and no closures, all just south
of the epicenter at the junction of Interstate 5 and Hwy 138 (Figure 25).

Figure 25 M7.4 Geological Unit Restrictions and Closures
(Point Counts in Parenthesis)

51

The M7.4 Default data run, returned seven restrictions and three closures, several south
of the epicenter at the junction of Interstate 5 and Hwy 138, and several north of Gorman on the
I-5 and Fraser Mountain Park Rd (Figure 26).

Figure 26 M7.4 Default Data Restrictions and Closures
(Point Counts in Parenthesis)
The M8.0 ShakeMap run, returned one hundred two restrictions and forty-nine closures
(Figure 27). Mostly along the faultline of the southern San Andreas with a large cluster near the
city of San Bernardino, one outlier restriction on the I-10 westbound on ramp from Baldwin Ave
in Rosemead, CA, a cluster near the Gorman epicenter, and one closure and three restrictions in
San Luis Obispo County. The M8.0 ShakeMap modeled far more extensive damage than the
52

other scenarios and it is important to note that closures were produced at chokepoints on the I-5,
I-15, and I-10, cutting off or greatly restricting both main North-South arteries and the East-West
artery into southern California. Potentially reducing access for emergency response to the North
and East of the region.

Figure 27 M8.0 ShakeMap Restrictions and Closures
(Point Counts in Parenthesis)

53

The M8.0 STATSGO run, returned seven restrictions and one closure, all around the
epicenter near Gorman, Fraser Park, and Lebec (Figure 28).

Figure 28 M8.0 STATSGO Restrictions and Closures
(Point Counts in Parenthesis)

54

The M8.0 Geological Units run, returned seven restrictions and one closure, similar to the
STATSGO run, all around the epicenter near Gorman, Fraser Park, and Lebec (Figure 29).

Figure 29 M8.0 GeoUnit Restrictions and Closures
(Point Counts in Parenthesis)
55

The M8.0 default data run, returned seven closures and twenty-two restrictions, around
the epicenter near Gorman, Fraser Park and Lebec as well as east of east of Castaic Lake on San
Francisquito Rd (Figure 30).

Figure 30 M8.0 Default Data Restrictions and Closures
(Point Counts in Parenthesis)
Table 7 below summarizes the number of restrictions and closures for each model run.
For the M7.4 earthquke scenairos, the model run with HAZUS default data produced the greatest
number of closures and restrictions. This was followed by the M7.4 ShakeMap, then the
GeoUnit and STATSGO soil maps which produced the same restricted bridges. The M8.0
earthquake scenario runs returned a greater difference of closures and restrictions between them.
56

The M8.0 GeoUnit and STATSGO runs produced the same number of restrictions and fewer
closures than the M7.4 defaut data run. The M8.0 ShakeMap run returned the greatest number of
closures and restrictions over a larger spatial extent compared to the default data or either soil
map run. The individual bridges identified for closures or restrictions are summarized in
Appendix A.
Table 7 HAZUS Scenario Runs and Number of Bridges Restricted or Closed
Bridges
M7.4
GeoUnit
M7.4
STATSGO
M7.4
ShakeMap
M7.4
Default
Data
M8.0
GeoUnit
M8.0
STATSGO
M8.0
ShakeMap
M8.0
Default
Data
#
Restricted
3 3 4 7 7 7 103 22
# Closed 0 0 1 3 1 1 49 7

4.4. Summary
Except for the M8.0 ShakeMap run, the default data runs generally displayed higher
damage state probabilities and greater numbers of closures and restrictions. The reason for the
difference in the M8.0 ShakeMap runs may be due to the default HAZUS calculations for ground
motion data in the Source Scenario compared to a User-defined Scenario. When entering
ShakeMap datasets into a User-Defined Scenario HAZUS will use the values and spatial extents
defined by those datasets, including the rupture lengths. Defining a Source Scenario involves
selecting the desired fault section, in this case the San Andreas Carrizo section, then selecting an
epicenter along that segment from which HAZUS will calculate equal distances of rupture length
from the epicenter that follow the contours of the faultline. If the line segment for the fault
section reaches its endpoint, HAZUS will stop the rupture length at the endpoint (FEMA 2015b).
The sections of the San Andreas are connected on the HAZUS Source fault map and large
rupture lengths from one section should continue into adjacent sections, as this would reflect the
57

actual continuity of the fault. But the difference between the spatial extent of damages from the
User-Defined (ShakeMaps data) and Source (soil map and default data) Scenarios can be
explained by the rupture lengths of the M8.0 Source scenarios being contained within the
endpoints of the San Andreas Carrizo fault section. This difference in damage is not observed
between the M7.4 runs because the rupture length from that magnitude is much shorter and did
not exceed the endpoints of the Carrizo section. It is not known if or why shortened rupture
lengths are intended by HAZUS programmers in Source scenarios for large continuous faults
like the San Andreas, but this should be important to note for users intending to model great
earthquakes with large rupture lengths using Source Scenario fault selection. Arbitrary scenarios
will apply the full rupture length predicted from the magnitude level, but only as a straight line
with a user-defined orientation value measured in degrees from north. (FEMA 2015b).
Damage probability returns between GeoUnit and STATSGO soil maps were similar,
with STATSGO producing slightly higher values, probably due to more NEHRP type D soils
located along the rupture line of the fault and in the Los Angeles Basin. Both ShakeMap runs
produced higher damage probability returns than runs using either a STATSGO or GeoUnit
NEHRP classified soil map. Three bridges south of Gorman, CA at the Interstate 5 junction to
highway 138 were closed or restricted in every single run and were identified as the bridges of
greatest importance in this study. These had the HighwayBridgeId CA20662, CA20668, and
CA20674.

58

Chapter 5 Conclusion
HAZUS showed there is potential for significant closures and restrictions from any scenario
tested. The HAZUS runs also showed significant output differences between different user
supplied datasets and default HAZUS data. This study shows the degree to which the
earthquake scenarios could damage bridge infrastructure and how the different datasets used can
change these outputs. This research also investigated significant damage state probabilities and
how different cumulative damage states may translate into bridge restrictions and closures.
The first research question asked to what degree will a major earthquake damage
southern California bridges. This was answered using the studies two earthquake scenarios to
represent the range of magnitudes for a major to great earthquake. Different instances of bridge
damage for each scenario were evaluated with default HAZUS data as well as user-supplied soils
maps and ShakeMaps. Bridges reaching a cumulative damage state of ‘at least moderate’ or ‘at
least extensive’ were identified at the 50% or greater damage state probability level.
The second research question asked how different user-supplied datasets for ground
motion will change bridge damage outputs compared to default HAZUS data. This was
evaluated by each scenario runs; number of bridges in the dataset with greater than 15%
probability of at least moderate damage, the total bridge damage related economic loss, and the
total number of bridges determined to be closed or restricted.
The final research question posed how relative damage state outputs returned by HAZUS
could correspond to bridge closures and restrictions. This was examined using a damage state
acceptance criterion of at least 50% per the reporting limits of the California Geological Survey
Project Report: HAZUS Loss Estimation for CA Scenario Earthquakes (CGS 2009).
59

Correspondence thresholds for relative damage states to closures and restrictions were set using
the research of Richardson et al. (2015).
Overall HAZUS showed itself to be particularly useful for assisting planners in limiting
economic losses and informing emergency preparedness. Implications for the study are
discussed below, followed by limitations of the implemented methodology and the possibilities
for future research.
5.1. Implications
Running HAZUS with either earthquake scenario and any user-supplied dataset showed
that damage to highway bridge infrastructure could be significant, even with the high seismic
engineering standards of California bridges. The degree and spatial extent to which the
earthquake scenarios could damage southern California bridge infrastructure varies between
scenarios and between soil, ShakeMap, and default datasets. The dataset used can change the
outputs for either scenario, and it was unexpected that the M8.0 ShakeMap exceeded the damage
estimates of the M8.0 default data run as this was not the case in the M7.4 scenario runs. The
default data run for the M7.4 scenario estimated higher damage state probabilities than the other
three runs because all soils are left at NEHRP Type D. HAZUS developers likely left the default
soils to Type D so the software will output more liberal estimates of damages to avoid liability
and not promote a false sense of security for users performing analysis with default datasets by
outputting an underestimation of damages. The M7.4 scenario runs are in line with the findings
from Neighbors (2013) that user-supplied hazard maps will return decreased damage values
compared to the default data. Even the M7.4 ShakeMap estimated lower values for ground
motions and damages compared to the default HAZUS values. The M8.0 ShakeMap run may
have been a more accurate assessment of damages than the M8.0 soil maps and default data runs
60

due to rupture lengths with those datasets possibly being cut short by endpoints of the fault
section in the Source scenarios.
The NEHRP soil classification methodology was applied by other researchers to the scale
of an average sized city and not to regional scale. Ground truth Vs data is intended to inform
local site conditions and the averages of many Vs points over large mapping units, such as some
alluvium units in the CA Geological Units map, may have skewed the Vs values for those units
creating less variability than a larger scale map such as SSURGO might have made.
ShakeMaps, in general, showed greater damage returns than soil maps, the M7.4 Default
data run had the largest and most expansive damage returns of that scenario, while the M8.0
ShakeMap run covered a much larger geographic area than the M8.0 Default data run and had
the largest number of bridges damaged. As stated in the results summary, this is most likely due
to the User-Defined ShakeMap hazard scenario that implements a longer rupture length than the
Carrizo fault segment selected in the Source scenario used for the other datasets. The Source
scenarios were selected to mimic the contour of the San Andres fault and rupture line used by the
ShakeMaps in the User-Defined scenarios. There is no other option for selecting particular faults
in a deterministic scenario and the user cannot select multiple connected fault segments in an
effort to define a more expansive fault hazard. This difference between User-Defined ShakeMap
scenarios and Source scenarios suggests Scenario ShakeMaps, if available for the fault the user
desires to model, are a better source for regional ground motion data.
Bridge closure and restrictions classifications were based on identified relationships to
BDI and bridge damage states established in Richardson (2015) and Shiraki (2009). The criteria
for significant probability of a damage state could be lowered to include a larger set of at least
61

moderately and at least extensively damaged bridges in the closures analysis, but these damage
state probabilities would be more likely not to reach the cumulative damage states mentioned.
Several of the bridges at the junction of the I-5 and state highway138 were closed or
restricted in every scenario run. These are inconveniently located in mountainous terrain that
would be difficult to bypass and may create a significant impediment for emergency responders
and supplies attempting to reach the Greater Los Angeles region from central and northern
California. The M8.0 scenario ShakeMap also produced this chokepoint effect on the I-5 as well
as on the I-15 and I-10, restricting arteries to the north and east of the regions most populated
areas. Appendix A contains a table for each scenario run, listing individual bridge restrictions
and closures.
5.2. Limitations
Some soil data units had no Vs data from any similar unit to average NEHRP classes to.
These were assigned values by me (not a geologist or soils scientist) according to class
classification guides. Generalized soil classifications based on soil taxa or geologic lithography
are not recommended due to differing compaction and other site-specific criteria (Medves 2009).
The default NBI highway bridge inventory was used in this analysis. Compared to the
2015 dataset I downloaded and cut to the study area, there are about three hundred bridges built
after 2010 omitted from the study area. Newer bridges may have less wear and stress, and be
built to higher current seismic standards, but a more complete data set may have informed model
returns better. The CDMS upload functionality for highway bridges was not working during this
study, the 2015 data prepared for HAZUS threw upload errors in the HAZUS CDMS likely due
to mislabeled CDMS data matching field for statewide census tracts.
62

As mentioned above there may be differences in PGA distributions and rupture lengths
between User-defined Scenario (ShakeMaps) and Source Fault scenarios (Soil Maps) using
default HAZUS calculations for rupture length constrained by the endpoints of a fault section.
5.3. Future Research
HAZUS development plans to fix the highway bridge CDMS upload issue in the next
version of HAZUS set to be released sometime in 2018. Future research should incorporate the
latest 2015 NBI datasets for analysis, the FHWA updates the NBI every five years.
A significant number of Vs points in the USGS Vs dataset used in this study are located
along the 605 freeway in Los Angeles. (Most of these were averaged into the alluvium GeoUnit
that covers much of LA.) A large geographic scale study focused on the bridges and segments of
the 605 using SSURGO soils units may be very informative.
There is a large database of Scenario ShakeMaps that could be used to compare HAZUS
outputs to ShakeMaps generated from historical events, this could inform variability or
sensitivity between the datasets.

63

References
Barbat A.H., L.G. Pujades, and N. Lantada. "Seismic Damage Evaluation in Urban Areas Using
the Capacity Spectrum Method: Application to Barcelona." Soil Dynamics and
Earthquake Engineering 28, no. 10 (2008): 851-65.

Bird, P. (2009), Long-term fault slip rates, distributed deformation rates, and forecast of
seismicity in the western United States from joint fitting of community geologic,
geodetic, and stress direction data sets, J. Geophys. Res., 114, B11403,
doi:10.1029/2009JB006317.

CA Geological Survey. CGS. CA Emergency Management Agency. 2009. “CGS project
Report: HAZUS Loss Estimation for CA Scenario Earthquakes.”
ftp://ftp.consrv.ca.gov/pub/dmg/rgmp/CalEMA2009/Report%20CalEMA%20HAZUS%2
009%20Final.pdf Accessed 7/8/2017

Curtis, C. 2016. “Preparing for Earthquakes in Dallas-Fort Worth: Applying HAZUS and
Network Analysis to Assess Shelter Accessibility.” Masters Thesis. University of
Southern California. Los Angeles, CA.

FEMA. 2017. “FEMA's Hazus program.” Federal Emergency Management Agency. Accessed
June 10, 2017. https://www.fema.gov/hazus

Federal Emergency Management Agency (FEMA). 2015a. “Multi-hazard loss estimation
methodology: Earthquake model Hazus-MH 2.1 user manual.” Mitigation Division
(January) 1-863.

Federal Emergency Management Agency (FEMA). 2015b. “Multi-hazard loss estimation
methodology: Earthquake model Hazus-MH 2.1 technical manual.” Mitigation Division
(January) 1-718.

Federal Emergency Management Agency (FEMA). 2013a. “CDMS Data Dictionary.” Federal
Emergency Management Agency

FHWA. 2015. “National Bridge Inventory Data Dictionary.” Federal Highway Authority.
Accessed on August 10, 2017. http://nationalbridges.com/nbiDesc.html

Fialko, Y. 2006. “Interseismic strain accumulation and the earthquake potential on the southern
San Andreas fault system.” Nature 441, 968-971. Accessed May 27, 2017
doi:10.1038/nature04797.Daniel Faigin. 2015 “The history of Southern California
Freeway Development.” CA Highways.org. http://www.cahighways.org/maps-sc-
fwy.html. Accessed on 7/9/2017.

Field, E.H., G.P. Biasi, P. Bird, T.E. Dawson, K.R. Felzer, D.D. Jackson, K.M. Johnson, T.H.
Jordan, C. Madden, A.J. Michael, K.R. Milner, M.T. Page, T. Parsons, P.M. Powers, B.E.
Shaw, W.R. Thatcher, R.J. II Weldon, and Y. Zeng. 2013. Uniform California earthquake
64

rupture forecast, version 3 (UCERF3)—The time-independent model: U.S. Geological
Survey Open-File Report 2013–1165, 97 p., California Geological Survey Special Report
228, and Southern California Earthquake Center Publication 1792,
http://pubs.usgs.gov/of/2013/1165/.

Jennings, C.W., Strand, R.G., and Rogers, T.H. 1977 Geologic map of California: California
Division of Mines and Geology, USGS.
https://mrdata.usgs.gov/geology/state/state.php?state=CA

Jones, L.M., and M. Benthien. 2011. “Putting down roots in earthquake country.” Southern
California Earthquake Center Fall edition: 1-32.

Kircher, C.A., R.V. Whitman, and W.T. Holmes. 2006. “HAZUS earthquake loss estimation
methods.” Natural Hazards Review no. 7 (May): 45-59.

Ludington S., B.C. Moring, R.J. Miller, K.S. Flynn, P.A. Stone, D.R. Bedford. (2005)
‘Preliminary integrated databases for the United States - Western States: California,
Nevada, Arizona, and Washington.’ USGS Open File Report
http://pubs.usgs.gov/of/2005/1305/

Medves, E. 2009. “Use of Enhanced NEHRP Soils Maps for HAZUS-MH analysis in
Charleston, SC.” Masters Thesis. College of Charleston. Charleston, SC.

Michigan Tech. 2017. “UPSeis. An educational Site for budding Seismologists.” Michigan
Technical University. Accessed June 14, 2017.
http://www.geo.mtu.edu/UPSeis/magnitude.html

Moehle J.P., M. Eberhard. “Chapter 2: Earthquake Damage to Bridges” In Principles and
Applications in Engineering: Bridge Engineering: Seismic Design, edited by J Chen,
Baton Rouge: CRC Press, 2003.

Neighbors, C.J., E.S. Cochran, Y. Caras, and G.R. Noriega. 2013. “Sensitivity analysis of FEMA
HAZUS earthquake model: case study from King County, Washington.” Natural
Hazards Review 14 (May) 134-146.

NORSAR/ICG 2010. “The SELENA-RISe Open Risk Package.” NORSAR/ICG. Accessed on
1/3/2018. http://selena.sourceforge.net/index.shtml

Olsen, K.B., S.M. Day, J.B. Minster, Y. Cui, A. Chourasia, M. Faerman, R. Moore, P.
Maechling, and T. Jordan. 2006. “Strong shaking in Los Angeles expected from southern
San Andreas earthquake.” Geophysical Research Letters 33 (April): 1-4. doi:
10.1029/2005GL025472.

Park, S., Stubbs, N., Bolton, R., Choi, S., & Sikorsky, C. (2001). “Field Verification of the
Damage Index Method in a Concrete Box Girder Bridge via Visual Inspection.”
Computer Aided Civil and Infrastructure Engineering, 16(1), 58-70.
65

Porter, K., L. Jones, D. Cox, J. Goltz, K. Hudnut, D. Mileti, S. Perry, D. Ponti, M. Reichle, A.Z.
Rose, C. Scawthorn, H.A. Seligson, K.I. Shoaf, J. Treiman, and A. Wein. 2011. "The
ShakeOut Scenario: A Hypothetical Mw7.8 Earthquake on the Southern San Andreas
Fault" Earthquake Spectra vol. 27, no. 2 (May) 2 239-261.
http://research.create.usc.edu/published_papers/189

Richardson H.W., Q. Pan, J. Park, J.E. Moore. 2015 Regional Economic Impacts of Terrorist
Attacks, Natural Disasters and Metropolitian Policies. Switzerland: Springer
International Publishing.

SCEC. 2015. “UCERF3 .kml data”. Southern California Earthquake Center. Accessed on July
11, 2017. https://files.scec.org/s3fs-public/ucerf3_timedep_30yr_probs.kmz.

Shiraki, N., M. Shinozuka, J.E. Moore, S.E. Chang, H. Kameda, and S. Tanaka. 2007. “System
risk curves: Probabilistic performance scenarios for highway networks subject to
earthquake damage.” Journal of Infrastructure Systems 13 (1): 43-54.

Soil Survey Staff (2016) “Web Soil Survey.” Natural Resources Conservation Service, United
States Department of Agriculture. Accessed August 1, 2017.
http://websoilsurvey.nrcs.usda.gov/.

Thitimakorn, T. and T. Raenak. 2016. “NEHRP Site Classification and Preliminary Soil
Amplification Maps of Lamphun City, Northern Thailand.” Open Geosciences. 8(1):
538-547. Retrieved 1 Aug. 2017, from doi:10.1515/geo-2016-0046

US Census Bureau. 2016. “Quick Facts – Los Angeles County, CA.” Census Burau. Accessed
January 1, 2018.
https://www.census.gov/quickfacts/fact/table/losangelescountycalifornia/PST045216

USDA. 2017. “Description of STATSGO2 Database.” United States Department of Agriculture.
Accessed on September 1, 2017.
https://www.nrcs.usda.gov/wps/portal/nrcs/detail/soils/survey/geo/?cid=nrcs142p2_0536
29.

USDA. 1999. Soil Taxonomy. A Basic System of Soil Classification for Making and
Interpreting Soil Surveys. USDA Natural Resources Conservation Service

USGS. 2017. “USGS Earthquake Scenario Map.” United States Geological Service. Accessed
8/10/2017.
http://usgs.maps.arcgis.com/apps/webappviewer/index.html?id=14d2f75c7c4f4619936da
c0d14e1e468

USGS. 2016. “ShakeMap Documentation. 3.3.3 ShakeMap Scenarios” United States Geological
Service. Accessed 8/10/2117.
http://usgs.github.io/shakemap/manual3_5/shakemap_archives.html#generating-
earthquake-scenarios
66

USGS Staff. 2002. ‘Geologic Map Unit Classification, ver. 6.1.’ United States Geological
Survey

Yeats, R. 2013. “Living with Earthquakes in the Pacific Northwest. Ch13 The Federal
Government and Earthquakes.” Oregon State University epub. Accessed on June 6,
2017.
https://oregonstate.edu/instruct/oer/earthquake/15%20chapter%2013_epub_noninteractiv
e.html

Yong, A., Thompson, E.M., Wald, D., Knudsen, K.L., Odum, J.K., Stephenson, W.J., and
Haefner, S., 2015. “Compilation of VS30 Data for the United States: U.S. Geological
Survey Data Series”. 978, 8 p., http://dx.doi.org/10.3133/ds978.

67

APPENDIX A TABLES

Table 8 HAZUS Highway Bridge Types (FEMA 2015b)
CLASS
NBI
Class

State

Year
Built
#
Spans
Length
of
Max.Span
(meter)

Length
less
than
20 m
Design Description
HWB1
All

Non-
CA
<
1990 > 150 N/A

Conventional
Major Bridge - Length
> 150m
HWB1
All CA
<
1975 > 150 N/A

Conventional
Major Bridge - Length
> 150m
HWB2
All

Non-
CA
>=
1990 > 150 N/A Seismic
Major Bridge - Length
> 150m
HWB2
All CA
>=
1975 > 150 N/A Seismic
Major Bridge - Length
> 150m
HWB3
All

Non-
CA
<
1990 1 N/A

Conventional Single Span
HWB3
All CA
<
1975 1 N/A

Conventional Single Span
HWB4
All

Non-
CA
>=
1990 1 N/A Seismic Single Span
HWB4
All CA
>=
1975 1 N/A Seismic Single Span
HWB5 101-
106

Non-
CA
<
1990 N/A

Conventional
Multi-Col. Bent
Simple Support -
Concrete
HWB6 101-
106 CA
<
1975 N/A

Conventional
Multi-Col. Bent
Simple Support-
Concrete
HWB7 101-
106

Non-
CA
>=
1990 N/A Seismic
Multi-Col. Bent
Simple Support-
Concrete
HWB7 101-
106 CA
>=
1975 N/A Seismic
Multi-Col. Bent
Simple Support-
Concrete
HWB8 205-
206 CA
<
1975 N/A

Conventional
Single Col. Box
Girder -Continuous
Concrete
HWB9 205-
206 CA
>=
1975 N/A Seismic
Single Col. Box
Girder -Continuous
Concrete
HWB10
201- < N/A Continuous Concrete
68

CLASS
NBI
Class

State

Year
Built
#
Spans
Length
of
Max.Span
(meter)

Length
less
than
20 m
Design Description
206 Non-
CA
1990 Conventional
HWB10
201-
206 CA
<
1975 N/A

Conventional Continuous Concrete
HWB11 201-
206

Non-
CA
>=
1990 N/A Seismic Continuous Concrete
HWB11
201-
206 CA
>=
1975 N/A Seismic Continuous Concrete
HWB12 301-
306

Non-
CA
<
1990 No

Conventional
Multi-Col. Bent
Simple Support -Steel
HWB13
301-
306 CA
<
1975 No

Conventional
Multi-Col. Bent
Simple Support- Steel
HWB14 301-
306

Non-
CA
>=
1990 N/A Seismic
Multi-Col. Bent
Simple Support Steel
HWB14
301-
306 CA
>=
1975 N/A Seismic
Multi-Col. Bent
Simple Support- Steel
HWB15 402-
410

Non-
CA
<
1990 No

Conventional Continuous Steel
HWB15
402-
410 CA
<
1975 No

Conventional Continuous Steel
HWB16 402-
410

Non-
CA
>=
1990 N/A Seismic Continuous Steel
HWB16 402-
410 CA
>=
1975 N/A
Seismic
Continuous
Steel
HWB17 501-
506

Non-
CA
<
1990 N/A

Conventional
Multi-Col. Bent
SimpleSupport -
Prestressed Concrete
HWB18 501-
506 CA
<
1975 N/A

Conventional
Multi-Col. Bent
SimpleSupport -
Prestressed Concrete
HWB19 501-
506

Non-
CA
>=
1990 N/A Seismic
Multi-Col. Bent
SimpleSupport -
Prestressed Concrete
HWB19 501-
506 CA
>=
1975 N/A Seismic
Multi-Col. Bent
SimpleSupport -
Prestressed Concrete
HWB20 605-
606 CA
<
1975 N/A

Conventional
Single Col. Box
Girder -Prestressed
Continuous Concrete
69

CLASS
NBI
Class

State

Year
Built
#
Spans
Length
of
Max.Span
(meter)

Length
less
than
20 m
Design Description
HWB21 605-
606 CA
>=
1975 N/A Seismic
Single Col. Box
Girder -Prestressed
Continuous Concrete
HWB22 601-
607

Non-
CA
<
1990 N/A

Conventional Continuous Concrete
HWB22
601-
607 CA
<
1975 N/A

Conventional Continuous Concrete
HWB23 601-
607

Non-
CA
>=
1990 N/A Seismic Continuous Concrete
HWB23
601-
607 CA
>=
1975 N/A Seismic Continuous Concrete
HWB24 301-
306

Non-
CA
<
1990 Yes

Conventional
Multi-Col. Bent
Simple Support - Steel
HWB25
301-
306 CA
<
1975 Yes

Conventional
Multi-Col. Bent
Simple Support - Steel
HWB26 402-
410

Non-
CA
<
1990 Yes

Conventional Continuous Steel
HWB27
402-
410 CA
<
1975 Yes

Conventional Continuous Steel
HWB28

All other bridges that
are not classified

70

Table 9 M8.0 ShakeMap Bridge Closures
Highway
BridgeID
HAZUS
Bridge
Class Name
PDs
Exceed
Moderate
PDs
Exceed
Extensive
Econ
Loss
Thou Status
CA020662 HWB10 INTERSTATE 5 SB 0.71 0.58 856.19 C
CA020668 HWB10 ROUTE 5 SB 0.68 0.55 768.38 C
CA020674 HWB10
S5-E138
CONNECTOR 0.71 0.58 517.16 C
CA022964 HWB10 AVENUE S 0.68 0.55 423.18 C
CA025533 HWB11 GREENSPOT ROAD 0.65 0.52 1,194.72 C
CA016483 HWB13
MOONSTONE
BEACH DR 0.85 0.76 261.28 C
CA022704 HWB13
PEARBLOSSOM
HWY 0.77 0.67 644.13 C
CA024060 HWB13 HIGHLAND AVE 0.88 0.81 2,914.82 C
CA024084 HWB13
N215-E10 RAMP
CONN 0.62 0.50 906.24 C
CA024087 HWB13
INTERSTATE 215
SB 0.62 0.50 1,627.02 C
CA024090 HWB13
INTERSTATE 215
NB 0.62 0.50 1,160.74 C
CA024574 HWB13 STATE ROUTE 138 0.74 0.63 1,397.45 C
CA025101 HWB13 WATERMAN AVE 0.78 0.68 633.15 C
CA025102 HWB13 WATERMAN AVE 0.70 0.58 1,112.33 C
CA025103 HWB13 E FIFTH ST 0.92 0.86 573.23 C
CA025116 HWB13 MT VERNON AVE 0.73 0.62 1,104.95 C
CA025121 HWB13 MT. VERNON AVE 0.80 0.70 514.64 C
CA025317 HWB13 NINTH ST 0.89 0.83 1,243.89 C
CA025326 HWB13 RIALTO AVE 0.72 0.61 773.98 C
CA025447 HWB13 CENTRAL AVE 0.73 0.62 799.06 C
CA025494 HWB13 ORANGE ST 0.88 0.80 298.21 C
CA027074 HWB13 ROUTE 60 0.74 0.63 377.42 C
CA023972 HWB15 STATE ROUTE 38 0.66 0.66 408.64 C
CA024005 HWB15 ROUTE 66 (5TH ST) 0.51 0.51 561.97 C
CA024071 HWB18
ORANGE SHOW
ROAD 0.70 0.58 1,433.81 C
CA024078 HWB18
INTERSTATE 215
SB 0.62 0.50 826.18 C
CA024080 HWB18
INTERSTATE 215
NB 0.62 0.50 1,209.04 C
CA024331 HWB18 16TH STREET 0.67 0.55 596.56 C
CA024529 HWB18
STATE ROUTE 259
SB 0.81 0.71 558.04 C
CA024531 HWB18
STATE ROUTE 259
NB 0.81 0.71 558.04 C
71

Highway
BridgeID
HAZUS
Bridge
Class Name
PDs
Exceed
Moderate
PDs
Exceed
Extensive
Econ
Loss
Thou Status
CA024652 HWB18 INTERSTATE 15 SB 0.76 0.66 819.13 C
CA024653 HWB18 INTERSTATE 15 NB 0.76 0.66 819.13 C
CA025117 HWB18 WATERMAN AVE 0.66 0.55 641.81 C
CA025146 HWB18 ANDERSON ST 0.66 0.54 688.37 C
CA025158 HWB18 BASE LINE ROAD 0.74 0.63 1,077.87 C
CA025159 HWB18 BARTON RD 0.82 0.73 1,545.07 C
CA025312 HWB18 G ST 0.83 0.74 1,045.64 C
CA027454 HWB18
CHERRY VALLEY
BLVD 0.62 0.50 749.18 C
CA027456 HWB18 SINGLETON ROAD 0.64 0.52 569.88 C
CA027458 HWB18
SANDALWOOD
DRIVE O 0.62 0.50 671.77 C
CA024541 HWB22
W30-S259
CONNECTOR 0.73 0.61 2,564.30 C
CA025343 HWB27 SECOND STREET 0.65 0.65 321.10 C
CA025186 HWB28 5TH ST 0.72 0.61 174.87 C
CA025353 HWB28 DEL ROSA AV 0.66 0.55 181.07 C
CA025443 HWB28 ALABAMA ST 0.70 0.59 355.44 C
CA022771 HWB6 VALYERMO RD 0.83 0.74 398.69 C
CA024181 HWB6
LITTLE LEAGUE
DR 0.82 0.72 416.91 C
CA025164 HWB6 CAJON BLVD 0.62 0.50 101.95 C
CA028070 HWB6 14TH ST 0.62 0.50 129.58 C
CA016163 HWB10 STATE ROUTE 46 0.60 0.46 172.45 R
CA018491 HWB10 INTERSTATE 5 0.52 0.38 181.67 R
CA018492 HWB10 INTERSTATE 5 0.57 0.44 234.26 R
CA018493 HWB10 0.57 0.44 148.07 R
CA020664 HWB10 INTERSTATE 5 NB 0.60 0.47 592.32 R
CA020670 HWB10 ROUTE 5 NB 0.59 0.46 554.33 R
CA020673 HWB10
RAMP/CONNECTOR
138 0.52 0.39 490.87 R
CA023494 HWB10
BARREL SPRINGS
RD 0.54 0.41 328.33 R
CA024069 HWB10 INTERSTATE 10 WB 0.56 0.43 651.66 R
CA024070 HWB10 INTERSTATE 10 EB 0.56 0.43 651.66 R
CA024150 HWB10 INTERSTATE 215 0.59 0.46 254.03 R
CA024151 HWB10 INTERSTATE 215 0.59 0.46 254.03 R
CA024163 HWB10 DEVORE ROAD 0.52 0.39 524.92 R
CA024246 HWB10
INTERSTATE RTE
10 0.51 0.38 1,637.55 R
72

Highway
BridgeID
HAZUS
Bridge
Class Name
PDs
Exceed
Moderate
PDs
Exceed
Extensive
Econ
Loss
Thou Status
CA024270 HWB10 INTERSTATE 10 WB 0.56 0.43 552.67 R
CA024272 HWB10 INTERSTATE 10 EB 0.56 0.43 552.67 R
CA024315 HWB10
INTERSTATE RTE
10 0.50 0.37 1,329.38 R
CA024543 HWB10 E STREET OC 0.51 0.38 855.52 R
CA024545 HWB10 ARROWHEAD AVE 0.52 0.39 940.70 R
CA024547 HWB10 MT VIEW AVE OC 0.54 0.41 552.99 R
CA024549 HWB10 MT VIEW AVE OC 0.54 0.41 552.99 R
CA024551 HWB10 SIERRA WAY OC 0.54 0.41 1,004.49 R
CA024555 HWB10 STATE ROUTE 18 0.54 0.41 1,112.59 R
CA025493 HWB10 BOULDER AVE 0.59 0.46 241.34 R
CA025526 HWB10 KENDALL DR 0.52 0.39 90.97 R
CA025530 HWB10 BOULDER AVE 0.62 0.49 304.61 R
CA027038 HWB10 INTERSTATE 10 0.51 0.38 814.71 R
CA027272 HWB10 INTERSTATE 10 0.59 0.46 267.61 R
CA027398 HWB10 INTERSTATE 10 0.51 0.38 696.40 R
CA016372 HWB11 PALO PRIETO RD 0.60 0.47 195.33 R
CA024953 HWB11
STATE ROUTE 30
WB 0.54 0.41 149.06 R
CA024955 HWB11
STATE ROUTE 30
EB 0.54 0.41 140.50 R
CA024973 HWB11
STATE ROUTE 30
WB 0.62 0.49 342.97 R
CA024974 HWB11
STATE ROUTE 30
EB 0.62 0.49 362.03 R
CA016654 HWB13
WHEELER RIDGE
RD 0.54 0.42 412.42 R
CA024055 HWB13
INTERSTATE RTE
10 0.57 0.45 808.55 R
CA024056 HWB13
INTERSTATE RTE
10 0.57 0.45 862.69 R
CA024405 HWB13
N15-N395
CONNECTOR 0.61 0.49 434.42 R
CA024519 HWB13 STATE ROUTE 18 0.53 0.41 245.77 R
CA024520 HWB13 STATE ROUTE 18 0.53 0.41 471.55 R
CA024594 HWB13 STATE ROUTE 18 0.57 0.45 202.12 R
CA024595 HWB13 STATE ROUTE 18 0.53 0.41 177.83 R
CA024175 HWB18 PEPPER AVE 0.53 0.41 456.43 R
CA024185 HWB18 RIVERSIDE AVE 0.53 0.41 844.82 R
CA025137 HWB18 MT VERNON AVE 0.57 0.45 318.79 R
CA025325 HWB18
SAN TIMOTEO CYN
RD 0.54 0.42 317.15 R
73

Highway
BridgeID
HAZUS
Bridge
Class Name
PDs
Exceed
Moderate
PDs
Exceed
Extensive
Econ
Loss
Thou Status
CA025379 HWB18 MAPLE AVE 0.52 0.40 224.39 R
CA027337 HWB18 STATE ROUTE 60 0.59 0.47 476.67 R
CA027452 HWB18 BROOKSIDE AVE 0.56 0.44 454.68 R
CA027483 HWB18
SAN TIMOTEO
CANYON 0.53 0.41 486.16 R
CA027540 HWB18 STATE ROUTE 111 0.52 0.40 273.41 R
CA027541 HWB18 STATE ROUTE 111 0.52 0.40 273.41 R
CA025538 HWB19
WATERMAN
AVENUE 0.53 0.32 383.19 R
CA021741 HWB22 ROUTE 14 SB 0.52 0.39 551.91 R
CA021743 HWB22 ROUTE 14 NB 0.52 0.39 551.91 R
CA022963 HWB22 AVENUE T 0.63 0.50 435.55 R
CA024590 HWB22 INTERSTATE 15 NB 0.56 0.43 959.60 R
CA024596 HWB22 VALENCIA AV OC 0.54 0.41 1,071.89 R
CA024602 HWB22
STATE ROUTE 30
WB 0.59 0.46 505.05 R
CA024604 HWB22
STATE ROUTE 30
EB 0.59 0.46 505.05 R
CA024606 HWB22
STATE ROUTE 30
WB 0.60 0.47 489.28 R
CA024608 HWB22
STATE ROUTE 30
EB 0.60 0.47 489.28 R
CA024853 HWB23
STATE ROUTE 30
WB 0.51 0.38 825.98 R
CA024854 HWB23
STATE ROUTE 30
EB 0.51 0.38 804.28 R
CA024951 HWB23 STATE ROUTE 138 0.54 0.41 367.45 R
CA024952 HWB23 STATE ROUTE 138 0.62 0.49 594.76 R
CA024969 HWB23
STATE ROUTE 30
WB 0.60 0.46 680.15 R
CA024972 HWB23
STATE ROUTE 30
EB 0.60 0.46 839.59 R
CA024979 HWB23 CENTRAL AVENUE 0.57 0.44 612.01 R
CA024981 HWB23 PALM AVENUE 0.58 0.45 1,294.43 R
CA024983 HWB23 BASELINE ROAD 0.57 0.44 1,117.44 R
CA024985 HWB23
STATE ROUTE 30
WB 0.57 0.44 748.34 R
CA024986 HWB23
STATE ROUTE 30
EB 0.57 0.44 794.26 R
CA024992 HWB23 STATE ROUTE 330 0.59 0.46 889.44 R
CA024998 HWB23 ORANGE STREET 0.57 0.44 672.40 R
CA025009 HWB23 STATE ROUTE 330 0.57 0.44 587.69 R
74

Highway
BridgeID
HAZUS
Bridge
Class Name
PDs
Exceed
Moderate
PDs
Exceed
Extensive
Econ
Loss
Thou Status
CA025011 HWB23
E30-N330
CONNECTOR 0.57 0.44 721.86 R
CA016371 HWB28
CHOLAME VALLEY
RD 0.59 0.47 60.77 R
CA025203 HWB28 HIGHLAND AV 0.50 0.38 86.07 R
CA025356 HWB28 PUMALO ST 0.50 0.38 47.89 R
CA025373 HWB28 THIRD ST 0.62 0.50 101.33 R
CA025490 HWB28 40TH ST 0.61 0.49 831.16 R
CA025524 HWB28 LYNNWOOD DR 0.50 0.38 60.69 R
CA025531 HWB28 HIGHLAND AVE 0.50 0.38 96.24 R
CA024537 HWB3 STATE ROUTE 259 0.55 0.43 373.26 R
CA024539 HWB3 STATE ROUTE 259 0.55 0.43 373.26 R
CA024553 HWB3 STATE ROUTE 30 0.50 0.38 186.24 R
CA024554 HWB3 STATE ROUTE 30 0.50 0.38 186.24 R
CA024598 HWB3
STATE ROUTE 30
WB 0.50 0.38 281.62 R
CA024600 HWB3
STATE ROUTE 30
EB 0.50 0.38 289.45 R
CA025107 HWB3 ORANGE ST 0.50 0.38 51.19 R
CA025143 HWB3 30TH STREET 0.61 0.49 513.56 R
CA025384 HWB3 GILBERT ST 0.58 0.46 141.70 R
CA025444 HWB3 21ST ST 0.53 0.41 122.96 R
CA025448 HWB3 LYNWOOD DR 0.50 0.38 115.65 R
CA027263 HWB3
INTERSTATE RTE
10 0.53 0.41 1,402.42 R
CA024975 HWB4
STATE ROUTE 30
WB 0.51 0.39 254.05 R
CA024977 HWB4
STATE ROUTE 30
EB 0.51 0.39 273.10 R
CA024994 HWB4
HIGHLAND
AVENUE 0.50 0.38 562.03 R
CA025007 HWB4
S330-E30
CONNECTOR 0.50 0.38 190.89 R
CA025106 HWB4 BASELINE ROAD 0.55 0.43 180.36 R
CA025498 HWB4 HIGHLAND AVE 0.51 0.39 579.42 R
CA019362 HWB6 I 10 WB 0.56 0.44 612.38 R

75

Table 10 M8.0 Default Data Bridge Closures
Highway
BridgeID
HAZUS
Bridge
Class Name
PDs
Exceed
Moderate
PDs
Exceed
Extensive
Econ
Loss
Thou Status
CA017090 HWB10
FRAZIER MTN
ROAD 0.64 0.51 248.21 C
CA020662 HWB10 INTERSTATE 5 SB 0.74 0.62 928.70 C
CA020664 HWB10 INTERSTATE 5 NB 0.64 0.51 652.96 C
CA020668 HWB10 ROUTE 5 SB 0.74 0.62 883.01 C
CA020670 HWB10 ROUTE 5 NB 0.66 0.53 650.76 C
CA020674 HWB10
S5-E138
CONNECTOR 0.74 0.62 557.45 C
CA016748 HWB13 LEBEC RD 0.73 0.63 914.68 C
CA016569 HWB10 INTERSTATE 5 0.59 0.46 761.10 R
CA017091 HWB10
FRAZIER MTN
ROAD 0.55 0.42 322.62 R
CA017092 HWB10
FRAZIER MTN
ROAD 0.62 0.48 375.08 R
CA018491 HWB10 INTERSTATE 5 0.52 0.39 184.44 R
CA018492 HWB10 INTERSTATE 5 0.57 0.44 237.58 R
CA018493 HWB10 0.57 0.44 150.04 R
CA020623 HWB10 INTERSTATE 5 0.52 0.39 243.43 R
CA020624 HWB10 INTERSTATE 5 0.52 0.39 243.43 R
CA020625 HWB10 INTERSTATE 5 0.55 0.42 345.48 R
CA020626 HWB10 INTERSTATE 5 0.55 0.42 345.48 R
CA020627 HWB10 TEJON PASS OC 0.52 0.39 390.24 R
CA020666 HWB10 INTERSTATE 5 0.51 0.38 506.86 R
CA020667 HWB10 INTERSTATE 5 0.51 0.38 506.86 R
CA020672 HWB10
RAMP/CONNECTOR
138 0.52 0.39 699.44 R
CA020673 HWB10
RAMP/CONNECTOR
138 0.59 0.46 585.61 R
CA022965 HWB10 LANCASTER RD 0.53 0.39 268.47 R
CA016654 HWB13
WHEELER RIDGE
RD 0.56 0.44 433.79 R
CA018490 HWB13 0.55 0.43 661.65 R
CA023607 HWB28 PEACE VALLEY RD 0.50 0.38 57.72 R
CA022730 HWB6
SAN FRCSQUTO CA
RD 0.59 0.46 33.87 R
CA022731 HWB6
SAN FRCSQUTO CA
RD 0.55 0.43 51.07 R
CA022732 HWB6
SAN FRCSQUTO CA
RD 0.54 0.42 49.97 R

76

Table 11 M8.0 STATSGO Bridge Closures
Highway
BridgeID
HAZUS
Bridge
Class Name
PDs
Exceed
Moderate
PDs
Exceed
Extensive
Econ
Loss
Thou Status
CA020662 HWB10 INTERSTATE 5 SB 0.63 0.50 716.13 C
CA017090 HWB10
FRAZIER MTN
ROAD 0.52 0.39 194.55 R
CA020664 HWB10 INTERSTATE 5 NB 0.52 0.39 479.34 R
CA020668 HWB10 ROUTE 5 SB 0.63 0.50 684.93 R
CA020670 HWB10 ROUTE 5 NB 0.54 0.41 486.24 R
CA020674 HWB10
S5-E138
CONNECTOR 0.63 0.50 432.40 R
CA016654 HWB13
WHEELER RIDGE
RD 0.56 0.44 433.79 R
CA016748 HWB13 LEBEC RD 0.56 0.43 619.37 R

Table 12 M8.0 GeoUnit Bridge Closures
Highway
BridgeID
HAZUS
Bridge
Class Name
PDs
Exceed
Moderate
PDs
Exceed
Extensive
Econ
Loss
Thou Status
CA020662 HWB10 INTERSTATE 5 SB 0.63 0.50 716.13 C
CA017090 HWB10
FRAZIER MTN
ROAD 0.52 0.39 194.55 R
CA020664 HWB10 INTERSTATE 5 NB 0.52 0.39 479.34 R
CA020668 HWB10 ROUTE 5 SB 0.63 0.50 684.93 R
CA020670 HWB10 ROUTE 5 NB 0.54 0.41 486.24 R
CA020674 HWB10
S5-E138
CONNECTOR 0.63 0.50 432.40 R
CA016654 HWB13
WHEELER RIDGE
RD 0.56 0.44 433.79 R
CA016748 HWB13 LEBEC RD 0.56 0.43 619.37 R

Table 13 M7.4 ShakeMap Bridge Closures
Highway
BridgeID
HAZUS
Bridge
Class Name
PDs
Exceed
Moderate
PDs
Exceed
Extensive
Econ
Loss
Thou Status
CA020662 HWB10 INTERSTATE 5 SB 0.65 0.52 742.44 C
CA020664 HWB10 INTERSTATE 5 NB 0.53 0.40 500.20 R
CA020668 HWB10 ROUTE 5 SB 0.62 0.48 661.52 R
CA020670 HWB10 ROUTE 5 NB 0.53 0.40 467.36 R
CA020674 HWB10
S5-E138
CONNECTOR 0.58 0.45 382.33 R
77

Table 14 M7.4 STATSGO Bridge Closures
Highway
BridgeID
HAZUS
Bridge
Class Name
PDs
Exceed
Moderate
PDs
Exceed
Extensive
Econ
Loss
Thou Status
CA020662 HWB10
INTERSTATE 5
SB 0.51 0.38 518.94 R
CA020668 HWB10 ROUTE 5 SB 0.51 0.38 506.80 R
CA020674 HWB10
S5-E138
CONNECTOR 0.51 0.38 319.95 R

Table 15 7.4 GeoUnit Bridge Closures
Highway
BridgeID
HAZUS
Bridge
Class Name
PDs
Exceed
Moderate
PDs
Exceed
Extensive
Econ
Loss
Thou Status
CA020662 HWB10 INTERSTATE 5 SB 0.51 0.38 518.94 R
CA020668 HWB10 ROUTE 5 SB 0.51 0.38 506.80 R
CA020674 HWB10
S5-E138
CONNECTOR 0.51 0.38 319.95 R

Table 16 M7.4 Default Data Bridge Closures

Highway
BridgeID
HAZUS
Bridge
Class Name
PDs
Exceed
Moderate
PDs
Exceed
Extensive
Econ
Loss
Thou Status
CA020662 HWB10 INTERSTATE 5 SB 0.67 0.55 790.86 C
CA020668 HWB10 ROUTE 5 SB 0.67 0.54 747.47 C
CA020674 HWB10
S5-E138
CONNECTOR 0.67 0.54 471.89 C
CA016569 HWB10 INTERSTATE 5 0.51 0.38 624.27 R
CA017090 HWB10
FRAZIER MTN
ROAD 0.56 0.42 211.56 R
CA017092 HWB10
FRAZIER MTN
ROAD 0.53 0.40 309.42 R
CA020664 HWB10 INTERSTATE 5 NB 0.56 0.43 538.82 R
CA020670 HWB10 ROUTE 5 NB 0.58 0.45 537.08 R
CA020673 HWB10
RAMP/CONNECTOR
138 0.51 0.38 474.26 R
CA016748 HWB13 LEBEC RD 0.62 0.50 711.88 R

Abstract (if available)

Abstract The San Andreas Faultline is the largest fault in California. On average, this fault has produced a major earthquake every 150 years, the last to strike the southern section was the magnitude 7.9 Fort Tejon earthquake of 1857. Today the Greater LA Area is one of the largest urban agglomerations in the world and the second largest metropolitan region in the U.S. The area is well known for its urban sprawl and expansive highway system. The weak points of any highway system are the bridges and overpasses. Modeling the effects of an earthquake on this infrastructure will help inform emergency planning and speed economic recovery. ❧ Experts have predicted a major earthquake, from magnitude 7.0 to 8.0, will strike the fault within the next 30 years. The goal of this project was to examine enhanced HAZUS hazard datasets to assess potential earthquake damage to highway bridges and how this may correspond to bridge closures. NEHRP soils maps were created by joining shear wave velocity data to STATSGO and Geological Unit data, then classifying each soil unit by the NEHRP class shear wave velocity range. USGS Scenario ShakeMaps at M7.4 and M8.0 were selected near the area of fault section with the highest probability of a great earthquake. Eight scenarios were modeled with this data. Results of this work show that user-supplied datasets for ground motion generally reduce HAZUS bridge damage outputs, that all earthquake scenarios will significantly damage southern California bridge infrastructure, and how relative damage state outputs translate into bridge restrictions and closures.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Preparing for earthquakes in Dallas-Fort Worth: applying HAZUS and network analysis to assess shelter accessibility

PDF

A comparison of two earthquake events in the City of Downey: the Puente Hills and Whittier faults at 7.0 and 6.8 magnitudes

Asset Metadata

Creator Cover, Drew Ryan (author)

Core Title Preparing for the next major southern California earthquake: utilizing HAZUS with soils maps and ShakeMaps to predict regional bridge damage and closures

School College of Letters, Arts and Sciences

Degree Master of Science

Degree Program Geographic Information Science and Technology

Publication Date 02/08/2018

Defense Date 01/24/2018

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag BDI,bridge damage,closures,damage state,earthquake,EQ,geological units,hazard map,HAZUS,HAZUS-MH,highway bridges,NBI,OAI-PMH Harvest,restrictions,Scenario ShakeMap,ShakeMap,shear wave velocity,SoCal,soils maps,source scenario,Southern California,STATSGO,UCERF3,user-defined scenario,Vs30

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Ruddell, Darren Martin (committee chair), Loyola, Laura Cyra (committee member), Swift, Jennifer N. (committee member)

Creator Email dcover@usc.edu,drew.cover@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-472090

Unique identifier UC11266598

Identifier etd-CoverDrewR-6021.pdf (filename),usctheses-c40-472090 (legacy record id)

Legacy Identifier etd-CoverDrewR-6021.pdf

Dmrecord 472090

Document Type Thesis

Rights Cover, Drew Ryan

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA