FEMA

Port-to-Port Transportation Corridor Earthquake Vulnerability 

This study was accomplished as part of 
Project Impact - A Partnership between King and Pierce Counties, "Creating 
Disaster Resistant Communities" 

Note - this report is intended to be used with the project web site: 
www.rspa.dot.gov/oet, then select Special projects. The web site contains all the 
project result maps and graphics, none of which are included herein. However, this 
report contains all of the economic impact findings, none of which are not 
included on the web site. Alternatively, some of the project maps and graphics are 
contained in a 16-page color brochure entitled "Port to Port Transportation 
Corridor Vulnerability Mitigation". 


Executive Summary 
 
In the late 1990's, the Federal Emergency Management Agency (FEMA) initiated a 
nationwide program called Project Impact. The impetus for this effort was to 
promote public and private partnerships which would focus on pre-disaster 
mitigation to reduce the cost and impact of recurring natural disasters. Both King 
and Pierce Counties were separately awarded individual agency grants to promote 
disaster resistant communities in their jurisdictions. 

King and Pierce Counties agreed to work together beginning in 1998 on pre-disaster 
mitigation. This was done for two reasons: 1) Disasters do not recognize 
artificial jurisdictional boundaries; and 2) it was the Counties' intent to engage 
regional businesses in partnering in mitigation activities. Subsequently the 
Counties formed a legal partnership via an interlocal agreement for the purposes 
of collectively addressing pre-disaster mitigation. 

In the Fall of 1998 King and Pierce Project Impact jointly hosted a meeting with 
approximately over 40 community representatives from the public and private 
sectors to solicit their input on what hazards should be addressed, and which 
mitigation activities would be of value to our region. The greatest interest from 
the group was in investigating the seismic vulnerabilities of our transportation 
network within the Central Puget Sound Region. Transportation was seen as being 
vital to our economy, impacting every facet of our individual, business and 
community well-being. 

A significant amount of time was spent in a scoping exercise, in determining what 
segment of our transportation network could be examined based upon time and 
resources available. After much discussion and consultation with the partnering 
organizations the Counties determined to use the south-north I-5 corridor running 
from the Port of Tacoma to the Port of Seattle. The study area includes the main 
transportation routes of I-5, and Highways 99 and 167 and one short east-west 
segment of I-405. No major bridge structures or other routes were included because 
of the lack of resources to perform an accurate and in-depth study. 

Coordination meetings were held with partners on a routine basis with additional 
meetings conducted between consulting technical agencies and personnel. 
Significant contributions to the study came from: 

  United States Geological Survey (USGS)-who provided earthquake scenarios and 
projected ground motions based upon the three major sources of earthquakes in the 
region. 

  Washington State Department of Natural Resources (DNR) who provided soil 
analysis, including liquefaction information for the segment of the transportation 
network being studied. 

  Washington State Department of Transportation (WSDOT) provided construction data 
on the individual bridges that make up the road network. 

  ABS Consulting (formerly EQE International) pulled the above pieces together and 
provided the analysis to estimate individual bridge and route performance under 
six separate earthquake scenarios. 

  University of Washington Geography Department provided an economic impact 
assessment of transportation system outages to businesses and the community at 
large 

  Both King and Pierce Counties performed separate contingency planning scenarios 
to determine the impact of a studied route being made unavailable and how that 
would effect local arterials, and how detours might be managed to alleviate the 
expected traffic jams. 

Findings from the study are detailed within this document. In the worst case 
scenario of a Seattle Fault (a shallow earthquake) up to 40 bridges lying between 
the Port of Tacoma and the Port of Seattle have a significant risk of failing. 
Both deep earthquakes, of the type of the Nisqually Earthquake, and a subduction 
event do not have the same levels of risk, but there are lower percentages of 
bridges at risk within the study area. For a complete presentation on bridge 
performance, should go to http://www.rspa.dot.gov/oet/ and click on "Special 
Studies." 

The economic impact studied only one earthquake scenario which was a deep 
earthquake centered under the City of SeaTac. This therefore should not be 
considered a worst case scenario. However, the projected economic impact is 
severe. The earthquake's transportation related effects could mean 


 reduced business revenues of $3 billion; 


 loss of 39,000 jobs costing over $1 billion in income, 


 and tax losses to local government of $72 million. 

Trucking firms, port related businesses, "just-in-time" manufacturers and 
retailers depending on customer access to stores could be hard hit. Small 
businesses without financial reserves could face bankruptcy. Furthermore, these 
estimates of economic loss are based only on the study area, and not the entire 
region. 

The contingency planning efforts by both counties revealed significant challenges 
in redirecting up to 200,000 cars per day (I-5) onto surface streets. Simultaneous 
route outages could bring traffic to a standstill, with few arterial substitutes 
to carry daily traffic. Earthquake damage to multiple bridges would disable entire 
routes for up to three to six months. One of the major benefits of this multi-
jurisdictional contingency planning was that it brought together transportation 
planners for the first time to address these types of issues. 

This study is only a beginning. Expansion of this work is needed to give a 
complete picture of the seismic risks our region faces when we experience an 
earthquake. It is vital to the economic well-being of Washington State that the 
study be expanded to do the entire I-5 corridor from Olympia to Everett and 
include I-90, Highway 520, and all of I-405 in the study. Additional funding to 
accomplish this work is currently not programmed. 

King and Pierce County Emergency Management agencies are grateful to all our 
partners for their participation in conducting this study of our transportation 
system's seismic vulnerability. We have learned from one another and forged new 
public and private bonds that are being expanded upon in numerous areas of 
intergovernmental and public-private cooperation and coordination. Together we 
will continue to seek new ways to build disaster resistant communities right here 
in the Puget Sound Region. 

 1. Introduction 

1.1. Economic Importance 
Washington is the most trade-dependent state in the country, where one in four 
jobs rely on international trade. The airplane, software, financial, forest 
product, and biomedical industries are all key to the region's economic success. 
Two and a half million people (40% of the State population) reside along the 
north-south corridor supporting these industries. One hundred billion dollars in 
goods moves through the ports of Seattle and Tacoma annually. Much of the cargo, 
60 percent, moves through the region to inland domestic markets. The Puget Sound 
region accounts for seven percent of the nation's international trade, but is home 
to only one percent of the population. In the highly competitive import-export 
shipping business, disruption of service from a disaster could deal a terrible 
blow to the Puget Sound economy, shifting business to foreign and domestic West 
Coast ports. 

In this region, the I-5 corridor is funneled by the Cascade Mountains to the east, 
and the Puget Sound to the west, so there is minimal transportation "redundancy" 
in the event one of the major routes is damaged in an earthquake. The I-5 corridor 
moves about 200,000 vehicles per day north south between Tacoma and Seattle, where 
most of region's businesses are located. Parallel routes, including SR-167 and SR-
99, have a total capacity about half that of I-5, and are already jammed. Boeing 
(the nation's largest exporter) is dependent on moving airplane components along 
the corridor, as part of their manufacturing process. All employers depend on the 
corridor to get people to work to ship and receive goods, and for access of 
customers. 

1.2. Project Impact 
In 1998, Washington's King and Pierce counties combined their Project Impact 
funding to address regional issues. When they convened a group of public and 
private sector citizens and asked them to identify their greatest concern 
following a disaster, the overriding consensus was transportation. Not only is 
transportation critical to provide emergency response but it provides the 
lifeblood to maintain the long-term economic vitality of the region. And 
transportation systems in the Pacific Northwest are already at maximum capacity. 
With that direction the counties developed a project to evaluate the post-
earthquake reliability of a key section of the regional transportation system, and 
if it were not operational, to estimate the regional economic impact. Consulting 
with their partners, the Counties defined the study area as the "Port to Port" 
Corridor connecting the nation's fifth and sixth busiest container ports: Seattle 
and Tacoma. Ultimately, the counties are planning to use the project results to 
help spur development and implementation of effective mitigation strategies. 
Additionally, this project can serve as a model for similar vulnerability studies 
of other critical transportation corridors. 

1.3. Project Objectives 
Based on the background presented above, the project team established the 
following five project objectives: 
 1. Engage business and government participation. 
 2. Evaluate post-earthquake transportation system survivability. 
 3. Develop an emergency response and recovery plan. 
 4. Estimate the economic impact of transportation system outage. 
 5. Promote mitigation of high-risk bridges on critical lifeline corridors. 

These were used to formulate the project approach, and as guidance throughout the 
project. 

1.4. National Marine Transportation System 
One of the driving forces behind the project was dependence on the ports. On 
November 13, 1998, U.S. Congress directed the Secretary of Transportation to 
"...establish a task force to assess the adequacy of the Nation's marine 
transportation system (including ports, waterways, harbor approach channels, and 
their intermodal connections) to operate in a safe, efficient, secure, and 
environmentally sound manner." The Task Force was to consider the capability of 
the nation's marine transportation system, the adequacy of the depth of channels 
and harbors, and the cost to the Federal government of accommodating projected 
increases in foreign and domestic traffic over the next 20 years. 
The findings of the Task Force were published in September 1999 as "An Assessment 
of the U.S. Marine Transportation System" [ www.dot.gov/mts/report/ ]. The 
Executive Summary notes, "As the world's leading maritime and trading nation, the 
United States relies on an efficient and effective MTS to maintain its role as a 
global power. The MTS provides American businesses with competitive access to 
suppliers and markets in an increasingly global economy." 
This statement confirms the view of the King-Pierce Project Impact's Port to Port 
Transportation Corridor Seismic Vulnerability study team's thesis that our port's 
transportation corridors are, indeed, a critical component of a local, regional 
and national transportation system critical to our nation's economy and national 
security. 

1.5. Project Approach 
The project carried out a four-step evaluation process to estimate the earthquake 
risk associated with the transportation corridor. 
 1. 	 Hazard assessment - quantifying hazards, including ground motions and 
liquefaction 
 2. 	 Loss estimation - evaluating individual bridge vulnerability/reliability, 
route reliability, and recovery time. 
 3. 	 Economic analysis - estimating the regional economic impact if the corridor 
is disrupted. 
 4. 	 Contingency planning - developing plans to detour around key collapsed 
bridges. 

HAZUS was used as the platform to integrate the hazards and individual 
reliabilities. 
 HAZUS 99 (the version that was current when this study was performed) is an 
 earthquake loss estimation software developed and funded by FEMA to help 
 communities to prepare, plan, and build stronger and safer communities. 
 Documentation for HAZUS can be found as a link on this projects web site. Among 
many 
 HAZUS 99 uses are analysis of disaster-related damages; identification of 
vulnerable 
 areas; assessment of vulnerability of housing, essential facilities and 
lifelines; estimation 
 of potential losses; and aid in development of response and recovery plans. 
 
The HAZUS 99 earthquake loss estimation methodology for highway transportation 
 systems, and in particular for highway bridges is described herein. 
 
Highway transportation systems generally consist of roadways, bridges, and 
tunnels.
 Road damage occurs due to surface fault ruptures or extreme soil failure. Bridge 
 damage can occur due to extreme ground shaking and or site soil failure. Loss of 
bridge 
 function usually results in significant disruption to the transportation network, 
and thus is 
 a key component to reliability of these lifelines. The project focuses on the 
lost 
 estimation methodology utilized in HAZUS 99 for bridges. 
 
The reliability of individual bridge structures was estimated based on the local 
 earthquake hazards, the bridge structural design characteristics, and performance 
of 
 similar structures in previous earthquakes. Most of the bridges along the I-5 
corridor 
 
were constructed before the mid-1970s, when more rigorous seismic design codes 
were initiated. 
The reliability of the transportation corridor "system" was estimated by combining 
the reliabilities of the individual bridge structures. The reliability of each 
bridge in a linear system such as I-5 has a dramatic impact on the overall route 
reliability. In the Puget Sound region, there are limited redundant routes 
compared to the grid, or network of highways that were available in Los Angeles 
following the Northridge Earthquake. 
The potential regional economic impact was estimated considering the likely 
bridge/highway segment outage time, associated increased travel times, and the 
resulting impact on a cross section of the region's employers. 
Contingency planning was developed by stakeholders with interests in each county 
including the counties (public works and sheriffs), cities, the WSDOT, and the 
State Police. 

1.6. Project Partners 
One of the key aspects of Project Impact is the participation of project partners. 
King and Pierce Counties Project Impact partners included a significant list of 
active participants from both the private and public sector as shown below. 
Private 
ABS Consulting (formerly EQE) 
 American Red Cross 
 Bank of America 
 Boeing Employees Credit Union 
 (BECU) 
 Boeing 
 Burlington Northern Santa Fe 
 Railroad (BNSF) 
 Frank Russell Company 
 Gordon Trucking 
 KOMOABC4 
 Microsoft 
 Olympic Pipe Line (now BP 
 AMOCO) 
 Tacoma-Pierce County Chamber 
 of Commerce 
 Western States Seismic Policy 
 Council (WSSPC) 
 
Public 
Federal Emergency 
 Management Agency (FEMA) 
 King County 
 Pierce County 
 Port of Seattle 
 Port of Tacoma 
 Sound Transit 
 University of Washington (UW) 
 U.S. Department of 
 Transportation (USDOT) 
 U.S. Geological Survey (USGS) 
 Washington Military Department, 
 Emergency Management 
 Division (EMD) 
 Washington Department of 
 Community, Trade and 
 Economic Department 
 (WSDCTED) 
 Washington State Department 
 Natural Resources (DNR) 
 Washington State Department of 
 Transportation (WSDOT) 
 

1.7. Report Organization and Work Products 
This report is divided into nine sections as follows: 
 1. Introduction - this section. 
 2. 	 Hazard Assessment - the US Geological Survey (USGS) provided the earthquake 
scenario ground motions, the Washington State Department of 

Natural Resources (DNR) provided liquefaction and soils mapping, and ABS 
Consulting (formerly EQE) evaluated site amplification. 
 3. 	 Engineering Analysis - the Washington State Department of Transportation 
provided the bridge data and provided guidance on use of that data. ABS Consulting 
conducted the engineering analysis with the use of HAZUS to assist in data 
analysis. 
 4. 	 Economic Analysis - Dr. Stephanie Chang and Dr. William Beyers, both of the 
University of Washington, performed the economic analysis. 
 5. 	 Contingency Planning - King and Pierce Counties emergency management, public 
works, and sheriffs, and the Washington State Police developed contingency plans 
based on the outage scenarios. 

This report should be used in conjunction with two other documents: 
 1. 	 Port-to-Port Transportation Corridor Vulnerability Mitigation Brochure, a 16 
page color document showing many maps and graphics used in developing the project. 
 2. 	 Port to Port Transportation Corridor Vulnerability [www.rspa.dot.gov/oet/ 
(click on "special studies")]. - a web site that shows a comprehensive set of maps 
and graphics developed during the project. 



2. Hazard Assessment 
2.1. Introduction 
A hazard assessment was performed for six earthquake scenarios. Ground motion 
assumptions were developed, site amplification estimates applied, and liquefaction 
susceptibility mapping prepared. 
Ground motions were develop outside HAZUS due to the limitations of HAZUS 
generated scenarios. HAZUS does not allow site amplification of user supplied 
ground motions within the program, so "amplified" ground motions were input into 
HAZUS. 
The scenario ground motion approach was selected rather than using probabilistic 
ground motions. The premise of probabilistic ground motions is that they will not 
be exceeded over the associated return period. This may result in overestimating 
the impact as several different earthquakes may contribute to those ground 
motions. Ground motions for scenario were used to better approximate what may 
occur for a single event. 
Further information on the input hazard information for HAZUS can be found in 
Chapter 4 of the HAZUS Technical Manual, Potential Earth Science Hazards (PESH). 
Refer to the project web site for a link to the HAZUS web site. 

2.2. Pacific Northwest Earthquakes 
The Puget Sound is vulnerable to earthquakes from three sources each represented 
in the scenarios under evaluation. The USGS provided ground motions for six 
earthquake scenarios. 
The scenarios are based on the same ground-motion prediction relations used in the 
1996 national seismic hazard maps. The scenario maps are for a NEHRP B-C boundary 
site with an average shear-wave velocity in the top 30m of 760 m/sec. All 
magnitudes cited are moment magnitudes. The scenario maps were produced for median 
values of the peak horizontal ground acceleration and 0.3 and 1.0 sec spectral 
accelerations (5% of critical damping). The following is a description of each 
scenario. 
Crustal Earthquakes: Seattle, Tacoma, South Whidbey Island faults have been 
identified, each of which are capable of producing ground motions comparable to 
the Northridge and the Great Hanshin (Kobe, Japan) earthquakes. Paleoseismic 
evidence identified by the USGS, a Project Impact partner, recorded an event in 
approximately 900 AD that caused a vertical offset of 22 feet across Puget Sound. 
The scenarios modeled included: 


  Seattle Fault M6.5: The fault strikes east-west and has a length of 20 km and a 
downdip width of 12 km. The fault dips to the south at 20 degrees. The shallowest 
(northern) edge of the fault is at a depth of 7 km. For all crustal fault 
scenarios, we used the ground motions averaged from the relations of Boore et 
 al. (1997), Sadigh et al. (1997), and Campbell (1997). 


  Seattle Fault M7.0: The fault strikes east-west with a length of 40 km and 
downdip width of 19 km. The fault dips 45 degrees to south, with its shallowest 
(northern) edge at 3 km depth. 


  Tacoma Fault, M6.7: This fault is poorly known and its trace is highly 
uncertain. The length of the fault is 32 km, with a 12 km downdip width. The 
assumed fault had a 3 km minimum depth and dips to the north at 45 degrees. 

Cascadia Subduction Zone earthquake - an earthquake with an estimated magnitude of 
9 occurred along the Pacific coast 300 years ago, and produced "minutes" of 
shaking. The scenario modeled was: 


  Cascadia M9.0: We used the plate interface geometry of Fluck et al. (1997) for 
the rupture zone of this fault. The eastern edge of the rupture zone was taken to 

be the base of the elastic zone from their model. We used the Youngs et al. (1997) 
attenuation relations for interface earthquakes. 
Deep (Benioff Zone) earthquakes have guided the development of building codes. Six 
magnitude 6+ earthquakes have occurred over the last century, including the 
magnitude 6.8 Nisqually Earthquake on February 28, 2001. The two scenarios modeled 
were: 


  Benioff zone M6.5: We used the hypocenter of the 1965 earthquake: latitude 47.4 
N longitude 122.3 W, depth of 59 km. We applied the Youngs et al. (1997) 
attenuation for intraslab earthquakes. 



 Benioff zone M7.1: We used the same hypocenter as the 1965 earthquake. 

The contribution from each of these is being considered in evaluation of the 
earthquake risk. This new understanding of our regional seismicity is pushing 
earthquake hazard levels in the central Puget Sound, approaching those in Los 
Angeles and San Francisco. 

2.3. Site Amplification 
Site amplification for peak ground acceleration and 1-second spectral acceleration 
were taken as developed in NEHRP, and applied in HAZUS. The amplification values 
were renormalized to better represent the their original basis. 
2.3.1. Soils Map 
The soils map used as a basis for the site amplification was provided by the 
Washington State Department of Natural Resources. 

2.3.2. Peak Ground Acceleration 
The "original" site amplification values shown in Table 2.1 are based on a site 
amplification of 1 for site class B that has a mean shear wave velocity of 1130 
m/sec. The attenuation relationships used to estimate ground motion are from: 
Boore and 
Joyner, Sadigh, and Campbell (weighted equally). The shear wave velocities used in 
their attenuation relationships are 760, 620, and 420 m/sec, respectively. The 
geometric mean of the 3 is 580 m/sec, close the mean shear wave velocity of site 
class C (560 m/sec). Therefore the values were renormalized, by dividing the 
original amplification by the site class C amplification factor for the particular 
ground motion range. The results are shown in Table 2.1. 
The scenario peak ground accelerations were amplified by these values and used as 
input to HAZUS for calculation of liquefaction probability. 
Table 2.1 
SITE AMPLIFICATION FOR PEAK GROUND ACCELERATION 
Spectral Acceleration 
Ss<= 0.25 
Ss= 0.50 
Ss= 0.75 
Ss= 1.00 
Ss= 1.25 
Site Class\PGA 
PGA<= 0.10 
PGA=.20 
PGA=.30 
PGA=.40 
PGA=.50 
Original Amplification Values 
A 
0.8 
0.8 
0.8 
0.8 
0.8 
B 
1.0 
1.0 
1.0 
1.0 
1.0 
C 
1.2 
1.2 
1.1 
1.0 
1.0 
D 
1.6 
1.4 
1.2 
1.1 
1.0 
E 
2.5 
1.7 
1.2 
0.9 

Modified Amplification Values for Port to Port project 
A 
0.7 
0.7 
0.7 
0.8 
0.8 
B 
0.8 
0.8 
0.9 
1.0 
1.0 
C 
1.0 
1.0 
1.0 
1.0 
1.0 
D 
1.3 
1.2 
1.1 
1.1 
1.0 
E 
2.1 
1.4 
1.1 
0.9 

Note: PGA capped at 45% g in site class E soils 



2.3.3. 1 Second Spectral Acceleration 
The original site amplification values are based on a site amplification of 1 for 
site class B that has a mean shear wave velocity of 1130 m/sec. The attenuation 
relationships used to estimate ground motion are from: Boore and Joyner and Sadigh 
(weighted equally). The shear wave velocities used in their attenuation 
relationships are 760 and 
620 m/sec, respectively. The geometric mean of the 2 is 686 m/sec, approximately 
0.75 of the way between 1130 m/sec (site class B) and 560 m/sec (site class C mean 
shear wave velocity). The amplification values were renormalized by dividing the 
original amplification value by the (site class C amplification factor to the 0.75 
power) X (site amplification factor for site Class B to the 0.25 power [which is 
1] for the particular ground motion range. For example. for Ss<= 0.1 site class E, 
the new site amplification factor = 3.5/(1.7^0.75 * 1.0^0.25) = 2.4. The results 
are shown in Table 2.2. 
The scenario 1 second spectral accelerations were amplified by these values and 
used as input to HAZUS for calculation of bridge reliability. 
Table 2.2 
SITE AMPLIFICATION FOR 1 SECOND SPECTRAL ACCELERATIONS 
Spectral Acceleration/Site Class 
Ss<= 0.1 
Ss= 0.2 Ss= 0.3 Ss= 0.4 


Ss>= 0.5 
Original Amplification Values 
A 
0.8 
0.8 
0.8 
0.8 
0.8 
B 
1.0 
1.0 
1.0 
1.0 
1.0 
C 
1.7 
1.6 
1.5 
1.4 
1.3 
D 
2.4 
2.0 
1.8 
1.6 
1.5 
E 
3.5 
3.2 
2.8 
2.4 

Modified Amplification Values for Port to Port project 
A 
0.5 
0.6 
0.6 
0.6 
0.7 
B 
0.7 
0.7 
0.7 
0.8 
0.8 
C 
1.1 
1.1 
1.1 
1.1 
1.1 
D 
1.6 
1.4 
1.3 
1.2 
1.2 
E 
2.4 
2.2 
2.1 
1.9 




2.4. Generation of Liquefaction Hazard Maps for HAZUS 
The Washington State Department of Natural Resources provided liquefaction 
mapping. Liquefaction and lateral spread can have a dramatic impact on bridge 
foundations and highways segments. The Puget Sound region has a higher potential 
for liquefaction than many other areas, because of high water tables in river 
valleys with young geologic 
deposits, such as in the Kent and Duwamish river valleys. Liquefaction can result 
in loss of bearing, and lateral movement (spread) can occur, measured in meters. 
HAZUS uses liquefaction susceptibility map data in determining a conditional 
liquefaction probability used in subsequent damage-state calculations. The program 
translates a relative liquefaction susceptibility (ranging from very high to none) 
into a conditional liquefaction probability using a simplified version of a 
relationship presented by Liao and others (1988). The conditional probabilities 
are adjusted for earthquake magnitude (duration) effects and variation in the 
depth to groundwater. These probabilities are then used to calculate the expected 
permanent ground displacement both from lateral spreading and settlement. The 
permanent ground displacements then modify direct damage estimates. 
Liquefaction susceptibility mapping at a scale of 1:24,000 was available for the 
entire study area. Sources of these map data are summarized in Table 2.3, 
including reference citations. The individual maps were merged into a single GIS 
map coverage (originally ArcInfo and converted into MapInfo), and liquefaction 
susceptibility ratings were assigned to correspond to the HAZUS rating system. The 
relation between HAZUS liquefaction susceptibility and ground settlement amplitude 
(Table 3.15, HAZUS Technical Manual (1997), Volume I) was used in making these 
assignments. Table 2.4 summarizes the HAZUS liquefaction susceptibility with the 
geologic deposits delineated on the available liquefaction mapping. 
Table 2.3 
 LIQUEFACTION MAPPING DATA SOURCES 
 
Geographic Area 
Map Data Sources 
Seattle urban area 
Grant and others (1998) 
Greater Eastside (Bellevue-Redmond) 
Palmer and others (2001, in press) 
Kent Valley (Renton-Auburn-Sumner) 
Palmer and others (1994) - northern Kent Valley Palmer and others (1995) - 
southern Kent Valley Palmer (1995) - Sumner area 
Tacoma urban area - Puyallup Valley 
Shannon and Wilson, Inc. (1993), Palmer and others (in preparation)



3. Engineering Analysis 
The engineering methodology is described in this section. Refer to the associated 
project web site for the results (www.rspa.dot.gov/oet then click on special 
studies). 
3.1. HAZUS Highway Bridge Damage Functions 
3.1.1. General Model 
Earthquake damage can be estimated for various transportation components based on 
anticipated ground accelerations and ground deformation. The required data to 
estimate bridge damage includes: 
-     Geographical location 
-     Longitude and latitude 
-     Bridge classification (structure type) 
-     Site spectral accelerations at 0.3 seconds and 1.0 seconds and permanent-
ground deformation (PGD) 
-     Peak-ground acceleration (for PGD-related calculations) 
HAZUS 99 (refer to the project web site for links to the HAZUS 99 reference 
documents) classifies bridges into 28 categories based on the following structural 
characteristics: 
-     Seismic design 
-     Number of spans 
-     Structure type and material 
-     Pier type 
-     Abutment type 
-     Span continuity 
General bridges with good seismic design features can accommodate relatively 
higher seismic input and allowable drift limits. The general fragility curves for 
each of the 28 bridge classes can be further refined in HAZUS 99 using bridge 
specific data such as: bridge length/span length/number of spans, bridge width, 
and skew. The effects of these parameters for several bridge types were 
investigated using sensitivity spreadsheet analyses. 

3.1.2. Definitions of Damage States 
HAZUS 99 defines five bridge damage states that are related to the damage ratio 
(i.e. the repair-to-replacement cost) for evaluation of direct economic loss. The 
basic damage states are: none (DS1), slight or minor (DS2), moderate (DS3), 
extensive (DS4), and complete (DS5). These represent the following types of damage 
(descriptions modified by WSDOT, but consistent with HAZUS): 
-     Slight damage (DS2) - minor cracking or spalling to concrete bridge 
elements. Bridge remains structurally sound. 
-     Moderate damage (DS3) - Any column experiencing moderate cracks but 
remaining structurally sound, moderate superstructure displacement (< 2 inches), 
any damaged connections, bearing failure or moderate (< 6 inches) settlement of 
approach. Requires temporary repair and/or capacity or functionality reduction. 
-     Extensive damage (DS4) - Any columns degrading without collapse - shear 
failure (column structurally unsafe), significant permanent displacement at 
connectors or major settlement (6 inches or greater) of an approach, or 
differential structural alignment. 
-     Complete damage (DS5) - Any column collapsing or span losing all bearing 
support that may lead to imminent span collapse, tilting of structure due to 
foundation failure. 

3.1.3. Damage Algorithms for Bridges 
The HAZUS 28 primary bridge classes are defined for the above damage states as a 
function of ground motion and ground displacement. These are median damage 
functions with a dispersion of 0.4 for ground shaking and 0.2 for ground failure. 
The assumptions in the development of these damage algorithms were reviewed and 
verified for applicability in this project. Discussion of this verification 
procedure is presented in the subsequent section. The verification focused on the 
most vulnerable bridge types that generally consist of simple-span bridge 
structures lacking modern seismic design features. A typical plot of a family of 
highway bridge fragility curves as the function of spectral acceleration is 
presented in Figure 3.1. For ground deformation, HAZUS considers incipient 
unseating and collapse as the possible types of damage due to ground failure. 
Initial damage to bearings, which correspond to slight damage from ground failure, 
is not considered. 
Restrainers do not have a significant effect on the shape of the fragility curve 
for ground motion but will modify the expected performance of bridges when 
subjected to liquefaction/lateral spread. 
For this project, damage states 4 and 5 were of primary interest since they result 
in loss of functionality. Figure 3.1 shows probabilities that a given bridge 
structure will not be extensively damaged, i.e., damage state will be less than 
DS4 (extensive damage) or DS5 (complete damage). 

Figure 3.1 
HIGHWAY BRIDGE FRAGILITY - PRESTRESSED CONCRETE BRIDGE MULTI-COLUMN BENT, SIMPLE 
SUPPORT 


3.2. Bridge Damage Function Evaluation 
3.2.1. Evaluation Considerations 
The important questions to ask when applying HAZUS-based bridge vulnerability 
functions to Project Impact port-to-port corridor study are the following: 
 1. 	 Is the bridge data currently in HAZUS database accurate relative to bridge 
location, physical characteristics, and key descriptors important for seismic 
performance? 
 2. 	 Do damage functions, particularly for the most vulnerable bridges, 
accurately reflect Washington State bridge stock and design practice? 
 3. 	 What modifications need to be made to appropriately model corridor bridges 
and assess the vulnerability of the lifeline network? 

The following paragraphs describe our approach to bridge data verification and 
analysis of the assumptions in the current HAZUS bridge model relative to 
Washington State bridge design and construction practice. Recommendations are made 
for modeling modifications for bridges "driving" lifeline segment reliability 
results. 

3.2.2. Bridge Inventory Verification 
It is important that the database used for HAZUS analysis has accurate information 
relative to bridge locations, bridge type and classification, span length, and 
construction date. In addition, it is important to update that information by 
examining Washington State Department of Transportation's (WSDOT) bridge database 
and seismic upgrade database. This verification process had several components. 
First, the existing HAZUS geographic database for the port-to-port corridor was 
examined to see whether bridges are accurately located. Second, the HAZUS database 
and WSDOT bridge database fields were mapped against each other to verify 
consistency of the key descriptors such as bridge type, bridge length, span 
length, number of spans, construction date, etc. Bridges with data discrepancies 
were discussed with WSDOT staff to establish an accurate input data for those 
structures. Furthermore, new structures, retrofitted bridges, and bridges with 
unique seismic features were also discussed to appropriately classify these within 
the available 28 HAZUS bridge classes. Specific issues addressed included: 



 11 bridges that were originally unclassified were classified. 



 Bridge database was updated for retrofits. 


  Bridge classifications were modified by date of construction in accordance with 
WSDOT discussions or use of seismic design standards. 


  Use of large diameter columns in Washington than in California - minimum 3' 
diameter, majority 4' diameter, most using 1% steel. Ultimate shear capacity 2 
times what is required. Shear failure not addressed until 1970's. 


  Starting in 1960, WSDOT designed bridges for 0.10 g lateral force. Before then, 
0.03 g was used in Eastern Washington and 0.05 g in Western Washington. 


  Starting in 1972 - as a result of San Fernando event - WSDOT design criteria 
included increased confinement (significant code change). 


  "Current" AASHTO seismic design requirements were published in 1983, and 
generally adopted in 1990. However, WSDOT began using these provisions in 1982. 



 Liquefaction considerations were incorporated into the design in mid-1980's. 


3.2.3. Preliminary Bridge Screening 
Once the input database was verified, it was appropriate to examine, on a 
preliminary basis, the resulting database content for bridge type, bridge function 
(i.e., ramp, overcrossing, or main line bridge), and by route. The purpose of this 
screening was to determine the common bridge types within a segment, or route, and 
to observe which bridge classes appeared on a preliminary basis to be most 
vulnerable, thus potentially controlling the reliability of a given segment. The 
results of this screening is presented in Table 3.1. The table presents bridge 
count within each category, the percentage for all bridges, the bridge's lateral 
load capacity (in percent of 1 second spectral acceleration), and its capacity to 
accommodate lateral displacement caused by liquefaction/lateral spreading. 
Table 3.1 
 SUMMARY OF BRIDGE CLASSES AND ASSOCIATED CAPACITIES 
 Bridge Class 
Number/ Percent of Total 
Lateral Capacity (g) 
PGD Capacity (in) 
Bridge Type 
HWB10 
76 
36% 
1.05 
3.9 
Continuous Concrete 
HWB17 
68 
32% 
0.44 
3.9 
Multi-Col. Bent, Simple Support-P/T Concrete 
HWB23 
23 
11% 
1.05 
3.9 
Continuous - Prestressed Concrete 
HWB22 
12 
6% 
1.05 
3.9 
Continuous - Prestressed Concrete 
HWB11 
11 
5% 
1.05 
23.6 
Continuous Concrete 
HWB3 
9 
4% 
1.1 
3.9 
Single Span 
HWB4 
4 
2% 
1.1 
3.9 
Single Span 
HWB12 
4 
2% 
0.44 
3.9 
Multi-Col. Bent, Simple Support- Steel 
HWB15 
3 
1% 
0.76 
3.9 
Continuous Steel 
Other 
4 
2% 
varies 


Total 
214 





Based on the tabulated data the m*ost vulnerable bridge types in the inventory are 
single-span or simple support multi-span bridges. A relatively high percentage of 
these bridges is located along Routes 5 and 167, and many are mainline structures. 
Consequently, it is appropriate to focus on the vulnerability assumptions for 
simple-span bridges. The following paragraphs discuss this topic. 

3.2.4. Screening of Vulnerable Bridges 
A significant number of corridor bridges consist of simple-span structures. 
Traditionally these types of bridges consist of single or multiple spans placed in 
a series. The span supports are alternately fixed and free to allow for thermal 
movements. As demonstrated in recent earthquakes, these types of structures suffer 
from the lack of redundancy and ductility. 
Simple-span bridge collapses observed in recent earthquakes in California, Japan, 
and Central and South America, can be attributed to unseating of spans and 
inadequate reinforcement in the supporting piers and columns. Causes of collapses 
were due to severe ground shaking, ground rupture, or excessive ground movement 
(liquefaction or landsliding.) Many simple-span Washington State bridges were 
built in the 1960s and may exhibit similar concerns. In summary, damage to bridges 
built prior to the early 1970's was due to: 
 1. Low seismic design forces 
 2. Higher than expected deflection 
 3. Inadequately located hinge locations 

Since for simple-span bridges the critical elements are unseating of span and/or 
pier column failure, bridge vulnerability curves in HAZUS for these bridges are 
based on shear capacity of the critical pier and additional capacity provided by 
the deck arching action. For simple-span bridges there are three subclasses of 
bridges based on seismic detailing incorporated in the design: 
 1. 	 Seismic design. Modern ATC-6 and 1980s AASHTO seismic design provisions. 
 2. 	 California design prior to 1975. This is a lesser quality bridge predating 
AASHTO 1973 seismic design provisions. 
 3. 	 Non-California Pre-1990. This is the most vulnerable category assumed to be 
lacking the fundamental bridge seismic design detailing. 

3.2.4.1. Age of Construction 
Based on our discussions with WSDOT Bridge and Structures engineers, it is 
apparent that WSDOT has closely followed California DOT (Caltrans) seismic design 
practice. 
Consequently, for this project it is appropriate to modify the use of these three 
curves as follows: 
 1. 	 "Seismic design" category will be applied to all Washington State bridges 
built during or after 1982. WSDOT adopted ATC-6 provisions at that time. 
 2. 	 The second category, originally intended to correspond to pre-1975 
construction in California, will be applied to WSDOT bridges built between 1973 
and 1982 according to AASHTO 1973 provisions. 
 3. 	 The third category, intended for those bridges which lacked key seismic 
features or were designed to very low seismic forces, will be applied to WSDOT 
bridges built in Washington state prior to 1973. 

A family of these vulnerability curves is shown in Figure 3.2. The plots indicate 
reliability (onset of extensive damage) along the vertical axis as a function of 
spectral acceleration. Various damage states for pre-1973 WSDOT bridges is shown 
in Figure 3.1. 

3.2.4.2. Drift Limitations 
The loss of function or failure of a simply supported bridge can occur either due 
to loss of span or pier column failure. Based on our calculations, it appears that 
the span seat width typically used in WSDOT bridges appears sufficient to prevent 
unseating prior to column failure. Consequently, it appears that drift limits and 
permanent ground displacements used in HAZUS appear appropriate. 
The vulnerability curves for permanent ground displacement are calculated as a 
function of column height times 0.025. This corresponds to extensive damage in 
columns with low ductility (confinement). 

Figure 3.2 
HIGHWAY BRIDGE FRAGILITY - PRESTRESSED CONCRETE BRIDGE MULTI-COLUMN BENT, SIMPLE 
SUPPORT 


3.2.5. Simple-Span Bridge Analyses and Results 
Given that simple-span bridges tend to dominate the results and overall lifeline 
segment vulnerability, it is appropriate to examine specific bridges of this type 
and verify that the bridge configuration and design assumptions are consistent 
with HAZUS approach. In addition, some of these structures originally built in the 
1960s and 1970s were widened. The effect of bridge widening in seismic response is 
of interest. 
A nonlinear static pushover analysis for typical transverse bridge bents for 
selected bridges along State Route 167 was performed. This was of interest since 
column behavior governs WSDOT bridge response for simple span bridges. A pushover 
analysis is used to determine the load-deformation behavior of a structure prior 
to failure. Failure is defined as that point at which the structure becomes 
unstable resulting in collapse. The analysis was performed using a two-dimensional 
finite element model of a 
transverse bridge bent. The lateral force can be related to the spectral 
acceleration, and further normalized to spectral acceleration at one-second period 
to compare with existing HAZUS bridge fragility models. 
Five WSDOT bridges originally built between 1966 and 1972 were analyzed to 
determine load-deformation behavior and spectral acceleration characteristics (see 
Table 3.2). The following sections provide a brief description of each bridge. 
Table 3.2 
SUMMARY OF PUSHOVER ANALYSES 
Bridge 
_ (in) 
Period (sec) 
Sa (g) 
Sa norm (g) 
Railroad 
3.7 
1.00 
0.38 
0.38 
212th Street 
2.8 
0.83 
0.42 
0.35 
Meeker (West) 
2.2 
0.66 
0.52 
0.34 
Meeker (East) 
3.5 
0.79 
0.57 
0.45 
277th Street 
3.5 
0.91 
0.43 
0.39 

3.2.5.1. 212th Street Bridge 
 The 212th Street bridge is a four-span 1972 vintage State Route 167 mainline 
structure. 
 The original structure has three bridge bents with 105-foot deck spans. Columns 
are 
 approximately 18 feet tall and spaced 18 feet apart. In 1991 the existing 
structure was 
 widened. Extension of the existing cap beam and addition of a circular column was 
 performed at each bent. Concrete-filled steel piles support the columns. Bent-
frame 
 member data is summarized below: 
 
212th Street Bridge Bent Member Data 
Member 
Dimensions 
Longitudinal Steel 
Steel Ties 
Original Capbeam 
48" x 48" Square 
18 #11 Grade 40 ksi 
#5 at 5"o.c. 
Widened Capbeam 
48" x 48" Square 
18 #8 Grade 40 ksi 
#4 at 4"o.c. 
Original Columns 
36" Dia 
12 #11 Grade 40 ksi 
#4 at 12"o.c. 
Widened Column 
36" Dia 
18 #10 Grade 40 ksi 
#4 at 3"o.c. 

3.2.5.2. 277th Street Bridge 
The 277th Street bridge is a three-span 1966 vintage structure along State Route 
167. The original structure has two bridge bents with 70 feet of deck span 
tributary to each bent. The original columns are approximately 18 feet tall and 
spaced 25 feet apart. In 1994 the existing cap beam lengthened 16 feet, with a 
circular column added 16' from the nearest existing column. Cast-in-place concrete 
piles underlay the column footings. 
277th Street Bridge Bent Member Data 
Member 
Dimensions 
Longitudinal Steel 
Steel Ties 
Original Capbeam 
48" x 48" Square 
18 #11 Grade 40 ksi 
#5 at 5"o.c. 
Widened Capbeam 
48" x 48" Square 
18 #8 Grade 40 ksi 
#4 at 4"o.c. 
Original Columns 
36" Dia 
12 #11 Grade 40 ksi 
#4 at 12"o.c. 
Widened Column 
36" Dia 
18 #10 Grade 40 ksi 
#4 at 3"o.c. 


3.2.5.3. Meeker Bridge East and West 
The 1966-vintage East Meeker Street bridges include two bents with a tributary 
deck span of 55 feet to each bent. The original bents had three columns, each 
approximately 16 feet tall and spaced on average 18 feet apart. In 1994 each 
bridge bents cap beam was ltngthened by 16 feet with an additional circular column 
located 16 feet away from an existing column. Timber piles underlay the column 
footings. 
Meeker Street Bridge Bent Member Data 
Member 
Dimensions 
Longitudinal Steel 
Steel Ties 
Original Capbeam 
48" x 39" Rect. 
20 #10 Grade 40 ksi 
#5 at 5"o.c. 
Widened Capbeam 
48" x 39" Rect. 
18 #9 Grade 60 ksi 
#5 at 10"o.c. 
Original Columns 
36" Dia 
12 #11 Grade 40 ksi 
#4 at 12"o.c. 
Widened Column 
36" Dia 
18 #10 Grade 40 ksi 
#4 at 3"o.c. 


3.2.5.4. Railroad Bridge 
The 1966-vintage Railroad Bridge has two bridge bents with tributary deck spans of 
113 feet. The original bents have three columns approximately 30 feet tall and 
spaced 18 feet apart. The existing cap beams were lengthened on one side of the 
existing bridge with an additional column 18 feet away from existing columns. 
Concrete piles underlay the column footings. 

Railroad Bridge Bent Member Data 
Member 
Dimensions 
Longitudinal Steel 
Steel Ties 
Original Capbeam 
42" x 36" Rect. 
16 #11 Grade 40 ksi 
#6 at 18"o.c. 
Widened Capbeam 
48" x 36" Rect. 
16 #9 Grade 60 ksi 
#5 at 7"o.c. 
Original Columns 
36" Dia 
14 #11 Grade 40 ksi 
#4 at 12"o.c. 
Widened Column 
36" Dia 
12 #8 Grade 60 ksi 
#5 at 3"o.c. 

. 


3.2.6. Bridge Modeling 
"Pushover" analyses were performed on the typical transverse bents for the above 
bridges. The purpose of the analysis was to validate that the WSDOT bridges were 
representative of the assumed fragility curves. A simplified diagram of a bent 
model is shown in Figure 3.3. 
Bridge Deck Weight Lateral Load 
Plastic Hinge 
Original Bent Widened Bent 
Figure 3-3 
Typical Bridge Bent Model 

Pushover analysis included the following assumptions: 



 Model transverse bridge bent 



 Use cracked concrete section properties 



 Include "widened" section properties 



 Include gravity "deck" loads 



 Model soil stiffness (linear springs) 


  Failure of bridge bent controlled by column plastic moment capacity (Mp) - 
verified 

Pushover analysis procedure includes: 



 Increase lateral load until Mp is developed at column. 



 Determine bent stiffness based on load and deflection prior to failure. 


  Calculate spectral acceleration (Sa) at failure given bent stiffness and the 
weight of the structure. 


  Compare to HAZUS bridge fragility curves for non-ductile detailing (pre-1973 
WSDOT bridges). 

The maximum calculated displacements for sample bridges range from 2.2" to 3.7" 
(this is just prior to the last hinge forming), as summarized in Table 3.2. At 
this displacement level, a bridge is extensively damaged, but has not yet failed. 
The corresponding HAZUS value is 3.9", which is in reasonable agreement. The 
"failure" displacement of HAZUS is 13.9" - i.e., the bridge can probably 
accommodate additional displacement after the full hinge mechanism is formed. This 
corresponds to some reduced moment capacity (i.e. non-zero moment strength) in the 
columns, after the hinges have formed. 

To determine the effect of the bridge widening a separate analysis was performed 
neglecting the additional columns and cap beam extension or assuming the new 
columns and cap beam were built similar to the original structure. The results are 
similar to those reported in Table 3.2. 


3.2.7. Earthquake Duration Effects in Bridge Vulnerability 
It is important to consider long-duration earthquakes, such as those originating 
on the Cascadia Subduction zone, and their effects on bridge performance. Long-
duration effects apply at 0.23 g lateral force representing the median value for 
onset of cracking in bridge columns. 
Bridge vulnerabilities are based on: 


 90% probability of failure is at 0.40 g 

 90% probability of cracking is at 0.28 g 

Thus, if a bridge responds in the "plastic" range (above 0.28 g), in a a long 
duration event, then it could degrade and fail. Therefore, the vulnerability of 
bridges under these conditions is reduced by a factor of 1.4 (0.4 g/0.28 g). 


3.3. Highway Segment Reliability and Recovery 
Highway segment reliabilities immediately following the earthquake were calculated 
by multiplying together the reliabilities of individual bridges in the segment. 
The results on shown on the project web site. Bridge reliabilities are combined 
using this approach immediately after the earthquake because the probabilities of 
failure of each bridge are not influenced by restoration management decisions. The 
reliability of highway segments after restoration is based on the restoration of 
the individual bridge that HAZUS estimates has the longest restoration time. 
Restoration time will be influenced by the availability of resources. The general 
approach assumes that for "low" damage category earthquakes, there will be 
adequate resources 
to pursue restoration of all bridges immediately following the earthquake. For 
"moderate" and "large" damage category earthquakes (refer to Table 3.3), it is 
assumed that bridge restoration will be delayed for lower priority bridges. 
The HAZUS restoration curve shows that on the "average" bridge will be about 67 
percent restored within 3 months. It is assumed that many of the resources that 
would be used in the initial stages of bridge restoration would be freed up after 
3 months, and could be applied to other bridges. For high damage category 
earthquakes, bridges are divided into 3 priorities for restoration, with delays 
for starting restoration of 3 and 6 months respectively for 2nd and 3rd priority 
highway segment bridges. 

In order to establish the bridge damage categories in Table 3.3, bridge damage 
results are used to estimate the total number of bridges across the state that 
will have at least extensive damage. This done by multiplying the average 
probability of being in the damage state times the total number of bridges. This 
approach is applied to main line bridges, as those are the bridges required to 
resume near full traffic volumes. The estimated number of bridges with at least 
extensive damage is shown in Table 3.3 for each earthquake scenario. 

The study area only encompasses a portion of the interstate bridge system that 
would likely be impacted by an earthquake. The WSDOT would be responsible for 
restoration of all state owned bridges. Therefore an order of magnitude estimate 
is provided of the total number of interstate bridges that would be damaged as 
shown in Table 3.3. This estimate takes into account the location, type, and 
expected distribution of ground motions of each earthquake, and the location of 
bridges in areas with significant site amplification and liquefaction 
susceptibility. In general, the study area (this project) encompasses the largest 
liquefiable areas in the region. To the north, the next large liquefiable area is 
the Snohomish River valley/delta just north of Everett. To the south, the next 
significant liquefiable are is the Nisqually River valley/delta. Each earthquake 
scenario affects bridges differently. Those affects are described in Table 3.3. 
The result is the total number of bridges with at least extensive damage for the 
entire region, as shown in Table 3.3. 

The six earthquake scenarios are subdivided into 3 groups for low, moderate, and 
high levels of earthquake bridge damage, based on the total number of bridges with 
at least extensive damage. High levels of damage are expected for the Cascadia 
Subduction M9.0 earthquake and the Seattle M7.0 earthquake. Moderate levels of 
damage are expected for the Seattle M6.5 and the Tacoma M6.7 events. Low levels of 
damage are expected for the Benioff earthquakes. For the high level of damage 
category, it is assumed that bridges will be restored in 3 priorities: 1) I-5, 2) 
SR-167 north of SR-18 to I-405, and SR-518, and 3) all other highways. For 
moderate levels of damage, 2 priorities are identified: 1) I-5, and 2) all others. 
For low levels of damage, it is assumed that restoration will start on all bridges 
at the same time. 


4. Economic Analysis 
4.1. Overview 
This portion of the study aimed to gain insights into how highway transportation 
disruption in the Project Impact earthquake scenarios would impact local 
businesses and the regional economy.1 Our approach consisted of two steps. First, 
we interviewed local businesses to learn about potential impacts and developed 
case studies from these interviews. Second, we ran a model of the regional economy 
to estimate overall effects based, in part, on what we learned from the 
interviews. We anticipate that the case studies will encourage local businesses to 
think systematically about the role of transportation in their operations, their 
own potential losses in an earthquake, and the importance of the region's 
investing in seismic strengthening of the highway system. We anticipate that the 
regional economic impact estimates will help inform decision-makers who can 
influence investment in earthquake mitigation and preparedness. Due to the limited 
scope of the pilot study, our results represent preliminary findings only. 

The interviews sought an in-depth understanding of how businesses might be 
impacted by the highway disruption scenarios. Section 4.2 below describes the 
earthquake scenario information we presented to the businesses. We conducted 21 
in-person interviews with local business owners and managers from February through 
May, 2001. Interviews typically lasted one to one and a half hours. In designing 
the study, we were guided by previous research, particularly studies of business 
impacts in the 1994 Northridge and 1995 Great Hanshin (Kobe, Japan) earthquakes. 
In addition to the potential magnitude of business losses, we were interested in 
finding out what are the main factors influencing loss, how impacts might differ 
between types of businesses, and 1 We would like to express our appreciation to 
all those who assisted us in this study, including Washington State Department of 
Transportation, Washington State Emergency Management Division, and businesses 
that we interviewed. To protect confidentiality, we do not name these businesses 
(exceptions are noted in the text). 

what firms have been doing to prepare themselves for disasters. We also inquired 
into the firms' experiences in the February 28, 2001, Nisqually earthquake. We 
supplemented the in-person interviews with telephone interviews to 5 additional 
businesses, primarily to gather information on the impacts of the Nisqually event. 

Interviewees were selected non-randomly with the intent of covering a cross-
section of the regional economy. The 21 businesses interviewed in person included 
3 manufacturing firms, 3 transportation service companies, 6 retail and wholesale 
trade concerns, 6 in finance and insurance, and 3 transportation and 
communications infrastructure providers. (The 5 telephone interviews were 
conducted with transportation firms and infrastructure agencies.) Half are major 
employers while 4 of the 21 are small businesses. Some had invested heavily in 
disaster mitigation and preparedness; others had given it no thought at all. All 
but 5 had their sole or main regional facility in the Port-to-Port transportation 
corridor. 

Section 4.3 below summarizes findings from the interviews in aggregate statistical 
terms as well as through a series of descriptive case studies. Several caveats are 
warranted. First, with one exception, we developed composite case studies in order 
to protect confidentiality. These composite case studies depict "typical" 
businesses rather than specific firms and capture what we consider to be the 
essence of business vulnerability. The exception consists of infrastructure 
service providers where disguising their identity would have rendered the case 
study meaningless; these providers agreed to be identified individually. Second, 
the scope of the pilot study limited our efforts to a small, non-random sample. 
The statistical results therefore pertain to the sample of businesses we 
interviewed and cannot be directly extrapolated to the entire population of 
regional businesses. We suspect, for example, that the large employers we 
interviewed constitute a biased sample of businesses that are unusually well-
prepared for disasters. Similarly, it is difficult to delineate a "typical" 
business from a small sample. A third caveat pertains to the lack of major 
disaster experience by businesses in the region, even including the Nisqually 
event, where transportation disruptions were mainly confined to short-term air 
transport problems. The extent to which our interviewees fully anticipated what 
would actually happen in a major disaster is unknown. Finally, the scope of the 
pilot study limited the discussion to the Port-to-Port highway corridor, 
considering only highway damage and then only to a small portion of the regional 
transportation network. This pilot study therefore lays the groundwork for a full 
study that could involve conducting a scientific survey of businesses, considering 
transportation impacts in the context of other earthquake effects, estimating 
travel time effects using a traffic model, and including disruption to the entire 
regional transportation system. 

The Washington State Input-Output Model was used to estimate the impacts of the 
M7.1 Benioff Zone earthquake. The model is driven by final demands; it has at its 
heart a system of multipliers that translate changes in final demands into changes 
in sales or output. The model also tracks labor income and the level of employment 
associated with output in each industry included in the model. For the purposes of 
this analysis a 28-sector version of the model was utilized. In undertaking the 
impact analyses reported in section 4.4, we modified the structure of this model, 
to take into account types of impacts identified by people we interviewed. In 
effect, an earthquake of the type hypothesized would alter the structure of 
production, which means that the multiplier structure would be affected. In 
particular we developed a model in which it was possible to adjust truck transport 
costs (both own-account and purchased) and labor costs. We developed from the 
detailed Washington input-output model a highway construction cost vector, which 
we used to model the impacts of rebuilding the damaged highway system. We also 
developed a flexible system for modeling these various parameters, which allowed 
sector-specific levels of impact. The simulations undertaken here were based on 
the state structure of production, as no data are available that define the 
composition of industrial output in the study region in terms of the sectors 
included in the input-output model. The impacts that were estimated should be 
regarded as indicative of the levels of impact that would occur due to an event of 
the type modeled. More precise estimates of impact would require the development 
of sector-specific output measures for the region under study; resources were not 
available in this pilot study to make such measures. 

4.2. Highway Outage Scenarios and Estimated Travel Times 
Each business interview focused on two of the earthquake scenarios described 
earlier in this report: (1) the M 7.1 Benioff Zone earthquake, and (2) either the 
M 6.5 Seattle Fault or the M 6.7 Tacoma Fault earthquake, depending on the 
business' location in the study area. We conducted a pre-test interview to 
determine how to effectively communicate the highway outage scenarios to 
businesses. This indicated that the scenarios should be presented in 
deterministic, rather than probabilistic, terms, should distinguish between short- 
and medium-term timeframes, and should include estimates of associated travel time 
increases. 

We therefore developed map scenarios for use in the interviews in which highway 
segments are coded either open, closed, or partial capacity (red, green, or 
yellow). These deterministic scenarios were based on Monte Carlo simulation using 
the probabilistic bridge damage results. Specifically, random numbers were 
generated for each bridge and compared with its damage probability to simulate 
either an open or closed state. If one or more bridge were closed on a highway 
segment, the entire segment was considered to be closed. If bridge closure 
affected traffic in only one direction, partial capacity was indicated. Businesses 
were asked to consider the Benioff Zone event at 72 hours after the earthquake and 
at 1 month afterwards. It was assumed that in the 1 month scenario, detours and 
lane re-striping would be implemented for partial capacity segments, and travel 
times would be somewhat improved. Additional questions pertained to the 
Seattle/Tacoma Fault event at 1 month after the earthquake. 

In consultation with representatives of Washington State Department of 
Transportation, very rough travel times were estimated for use in the interviews. 
The Port-to-Port corridor was divided into four north-south portions, which, under 
normal rush hour conditions, typically each take about 15 minutes to traverse. 
Depending on which particular highway route segments were open or closed, these 
travel times were increased judgmentally for the various earthquake scenarios. For 
example, in the Benioff Zone event at 1 month, it was estimated that the normal 1-
hour travel time between Seattle and Tacoma would be lengthened to 4.5 hours. 
These travel time estimates were indicated on the scenario maps used in the 
interviews. In practice, businesses required only approximate travel time impact 
information to help them visualize the scenarios. 

4.3. Business Interviews and Results 
The interview questions focused on obtaining several types of information. We 
gathered background data on the business itself; for example, number of employees, 
facility locations, scope of business, etc. We also inquired generally into modes 
of business resiliency. That is, we were interested in the kinds of options 
available to particular business types to "work around" disaster situations. Such 
options as flexible scheduling of employee work, operating overtime, and sending 
work to other company locations are more applicable to some types of businesses 
than to others. We asked about the Nisqually earthquake and other recent, non-
earthquake experiences of business interruption. 

We then spent a good share of time on the first scenario, the M 7.1 Benioff Zone 
event at 72 hours after the earthquake. Key questions pertained to the degree of 
disruptiveness of highway outage for various business functions including 
employees' ability to get to work, ability to receive supplies and ship products, 
and customers' ability to reach the business location. We inquired as to which 
highway-dependent business functions were most critical for the operations of the 
business. We also asked about anticipated revenue losses and cost increases 
associated with the scenario. 

The final portion of the interviews concerned determining the range of revenue and 
cost impacts that might occur. Businesses were asked to consider whether the 
situation would be better or worse after one month, and why. They were asked about 
likely revenue and cost impacts in the Seattle or Tacoma Fault event, this time 
considering not only highway damage but building and other types of damage that 
would be likely to occur. For this purpose, we used a regional loss estimate for 
the Seattle Fault event that the Washington State Emergency Management Division 
had developed for us using 

FEMA's HAZUS software tool. Findings are presented below in two parts: in 
aggregate statistical form, and in a series of case studies for various sectors of 
the economy. 


4.3.1. Aggregate Findings 
Table 4.1 reports on the expected disruptiveness of highway outage to business 
functions in the Benioff Zone event at 72 hours. The vast majority of businesses 
would find the situation "disruptive" or "very disruptive" to employees' ability 
to get to work. Of the businesses with major facilities in the Port-to-Port 
corridor, 44 percent would find the situation very disruptive to employee 
commuting. Many businesses, however, anticipated that other business functions 
such ability to make shipments or customer access would be even more impacted. In 
terms of overall business functionality, 76 percent of the interviews considered 
that the scenario would be "very disruptive". 


4.3.2. Case Studies 
This section presents several case studies of potential business impact in the 
earthquake scenarios described above. The case studies represent "typical" firms 
in the industry. While based on interviews with actual companies, they do not 
depict any specific firms. 

4.3.2.1. Infrastructure Service Providers 

Background 
Urban infrastructure services are critical to the vitality of a region's economy. 
Without them, businesses would be unable to operate. In addition to highways, 
infrastructure includes other transport modes and utility systems that support the 
activities of households, businesses, and government - systems that, themselves, 
rely on the viability of the highway network. This section considers how highway 
damage might impact these other infrastructure services and, in turn, cause 
substantial economic losses to ripple through the economy. 
Infrastructure service disruption was one of the most important sources of 
business interruption loss in the Nisqually earthquake. Examples include temporary 
closure of SeaTac Airport and King County International Airport (Boeing Field), 
localized power outage, and loss of phone service due to congestion from excessive 
call volumes. Indeed, in most of our interviews, business owners and managers 
cited loss of phone service as a main reason - sometimes the only reason - for 
business interruption in the disaster. 

Our case studies here consider the Ports of Seattle and Tacoma, respectively, and 
Qwest, a major telephone service provider. We focus on how highway damage might 
impact the operations of these infrastructure providers and, in turn, cause losses 
to businesses that they serve. 


Port of Seattle 
The vast majority of import and export goods are transported via seaports. Ports 
act as transportation hubs, a convergence zone for trucks, railroad cars, and 
ocean vessels. More than $32 billion of foreign trade goods made their way through 
the Port of Seattle in 2000, much of it originating from or destined for markets 
east of the Rocky Mountains. The Port of Seattle also owns and operates SeaTac 
International Airport, which served 28.4 million passengers in 2000. It has been 
estimated that maritime and air transportation activities at the Port of Seattle 
generate 83,700 direct jobs in the region in cargo handling, trucking, airlines, 
and the like. 

While the Port of Seattle's expected costs and revenue losses from an earthquake 
will depend on the particular event, a few generalizations can be made. Four 
factors are critical to operations: 1) the efficient functioning of the highway 
system; 2) efficient functioning of the railway network; 3) availability of 
electric power, and 4) functioning of piers and cranes. Only the last factor can 
be directly influenced by port mitigation activities. 

Highway damage in the M 7.1 Benioff Zone earthquake scenario would be very 
disruptive to the port's ability to receive and distribute cargo. When the Alaskan 
Way Viaduct was closed for safety inspections after the Nisqually earthquake, port 
managers observed that road congestion made it very difficult to get truck freight 
moved to and from the port. The Benioff Zone scenario would also severely hamper 
port employees' ability to get to work. 

Highway damage in the scenario earthquake would also impede port customers' 
ability to use the port efficiently. A number of warehousing customers in the Kent 
Valley would have difficulty reaching their storage facilities on East Marginal 
Way and at the port itself. Moreover, since freight is moved from ship to rail via 
truck at the Port of Seattle, impeded access to the rail yards would have 
ramifications for shippers' ability to move cargo to the mid-west and eastern U.S. 
by rail. 

One sector that could be hard hit is the Alaskan fishing industry. The fleet docks 
at the Port of Seattle's Fisherman's Terminal near Ballard, as well as other 
facilities, then transports fish to cold storage and processing facilities 
throughout the Seattle metropolitan region. A large share of fish processing, for 
example, takes place in the Kent Valley. Depending on the season, major highway 
damage could be very problematic for this industry. 

The extent of economic disruption to the Port of Seattle from highway damage would 
depend largely on the duration of outage. If regional highways were to be severely 
impaired for a month, for example, shipping companies would certainly divert 
freight to other seaports. For example, some lines that routinely call at Seattle 
and Vancouver, British Columbia, could drop Seattle from their itinerary, at least 
temporarily. 

In the M 7.1 Benioff Zone scenario, the Port of Seattle estimates that it may lose 
$2.5 million in revenue per month of severe highway damage. As a general rule of 
thumb, extra expenses related to dealing with the disaster might cost an 
additional 60 percent of that amount. For example, to counter-act highway 
congestion and delays, the Port of Seattle may have to extend current 12-hour 
operations to 24-hour service, thereby incurring substantial overtime labor costs. 
These costs would probably be borne largely by the port's customers and local 
trucking companies. 

In a catastrophic disaster such as a M 6.5 Seattle Fault earthquake, losses would 
be even more severe. Because 80 percent of the cargo moving through the port is 
discretionary - meaning that it does not have to transit through Seattle - the 
Port of Seattle is very vulnerable to diversion of this business to its 
competitors. It could suffer as much as a 75 percent reduction in traffic due to 
the regional highway situation. Damage at the port itself might reduce capacity by 
50 percent. Large numbers of customers could be lost to Vancouver, BC, and the 
major southern California ports. It is likely that, as with the Port of Kobe after 
the 1995 earthquake, it would be difficult for the Port of Seattle to recover from 
such a disaster. 

Port of Tacoma 
Like the Port of Seattle, the Port of Tacoma is a major container port and 
economic engine in the region. In 2000, $19.8 billion in imports and exports 
flowed through the Port of Tacoma. It has been estimated that over 10,000 direct 
jobs and 22,000 total jobs in Pierce County are related to port activity. All of 
the ports along the West Coast compete heavily with each other. In a major 
earthquake, both the Port of Tacoma and the Port of Seattle are likely to be 
similarly impacted and to lose customers to competitors outside the region. The 
displacement of cargo would likely result in substantial congestion at these other 
ports, as excess capacity is insufficient to absorb the combined volumes from 
Seattle and Tacoma. This would hamper commerce throughout a broad sector of the 
U.S. Highway damage as estimated for the M 7.1 Benioff Zone earthquake scenario 
would be very disruptive to truck access to the Port of Tacoma. About 30 percent 
of cargo at the port uses truck transport. Half of this may be lost in such a 
disaster, amounting to a 15 percent loss of traffic due to highway problems, even 
if no damage were suffered to the port's own facilities or to railroad 
connections. Railroad failures would cause even more disruption than highway 
damage. 

The port would probably try to compensate for traffic congestion by increasing 
hours of operation. This could increase labor costs by as much as 50 percent. The 
scenario would also be disruptive to employees' ability to get to work and the 
port's ability to receive needed supplies, such as crane parts. 
In addition to regional impacts, disruptions at the Port of Tacoma would quickly 
cause problems to the economy of the state of Alaska. Some 70 percent of 
northbound cargo to Alaska passes through the Port of Tacoma, including much of 
the food and beverages consumed there. A large portion is handled through 
warehousing in the Seattle-Tacoma corridor. 

Cargo would undoubtedly be shifted to competing West Coast ports in the U.S. and 
Canada. In the 1989 Loma Prieta earthquake that struck the San Francisco Bay Area, 
the Port of Tacoma apparently benefited from such cargo diversion from the Port of 
Oakland. 

One of the port's steamship line customers confirmed that if port and regional 
transportation disruptions were to last a week or more, the company would divert 
cargo to other ports. If the Port of Seattle were also experiencing problems, the 
company would look first to Vancouver, BC, then to Portland and Oakland. About 
half of the company's cargo through the Port of Tacoma uses regional trucking. 
In a catastrophic disaster such as a M 6.7 Tacoma Fault earthquake, the situation 
would be compounded by damage at the port itself. Revenue losses of $20 million 
(one-third of current annual revenues) are possible. As with the Port of Seattle, 
some 70 percent of Tacoma's cargo is bound for and originating from outside the 
Pacific Northwest. Since it does not need to transit through Tacoma, portions of 
the diverted cargo may not return. A catastrophic earthquake would likely lead to 
long-term loss of business at the port. 

Highway disruption would also impede military readiness. In the event that 
military action is called for in the Pacific, plans are for Fort Lewis to move, 
wholesale, out of the Port of Tacoma. Highway system disruption would prevent a 
rapid deployment. 


Qwest 
Qwest is the major provider of local telephone service in the region. As an 
infrastructure provider, it has invested in disaster preparedness and mitigation 
activities; for example, Qwest maintains emergency operation centers supporting 
each state that it serves. In addition, if customer-care centers and business 
office functions in Seattle become inoperable, customer calls can be automatically 
rerouted to Portland or other cities where Qwest customer support groups are 
located. In a truly catastrophic disaster, the company has an option of shifting 
business office functions to Portland while damage is being repaired in the 
Seattle area. 

Qwest's main concern in an earthquake would be to ensure adequate service on the 
telephone network. Electric power failures, for example, are a major concern. 
While highway disruption would have little direct impact on phone service, it 
could impede the company's ability to make rapid damage assessments and repairs to 
the telecommunications network. The M 7.1 Benioff Zone scenario would be 
disruptive to repair capabilities. To get around congestion problems, Qwest would 
probably set up temporary garages near repair sites and have workers stay at the 
site, if necessary. Such activities would entail additional costs. 


 4.3.2.2 Transportation Industry 

Background 
Any disruption to the regional highway system would cause immediate and direct 
losses to firms in the transportation industry. Highway damage would not only 
disrupt the business of goods movement, it would also cause losses to other 
economic sectors that depend on these goods being transported regionally, 
nationally, and globally. 

The Nisqually earthquake caused temporary closures and reduced capacity at SeaTac 
airport, Boeing Field, the Alaskan Way Viaduct and other transportation 
infrastructure. While short, this disruption led to significant dislocations in 
the transportation industry. For example, one air cargo carrier using Boeing Field 
estimated that it incurred over $600,000 in extra expenses from having to work 
around the 2-week closure of the airport, despite suffering virtually no damage to 
its own facilities. 

Trucking is particularly vulnerable to highway damage in earthquakes because 
profit margins are very low in this industry, fixed costs are high, and the vast 
majority of firms are small with low financial reserves. Indeed, as demonstrated 
recently when fuel prices soared, a single bad quarter can put a trucking firm out 
of business. 

According to the 1999 Washington State Data Book, there are 78,000 jobs in the 
transportation, communications, and utilities sector in King and Pierce Counties. 
In this profile, we focus on two important types of businesses involved in goods 
transportation: trucking companies and air cargo shipping firms. 


Case A: Trucking Company 
Company A is a Renton-based trucking firm that has been in business for over 40 
years. With 150 office employees and 400 drivers, it is considered a large firm in 
this industry. Company A hauls industrial and consumer products such as paper, 
clothing, and beverages within a service area spanning the Pacific Northwest and 
California. 

The highway damage estimated for the M 7.1 Benioff Zone earthquake scenario would 
be devastating to Company A's ability to conduct business. Trucking in the Puget 
Sound would be crippled by congestion and the company would be unable to bring 
cargo into the region. 

Because of weight restrictions on roads and through certain cities, Company A 
would have difficulty diverting its trucks off the main highways even in an 
emergency. For the most parts, trucks would have to sit in traffic. Local drivers 
are paid by the hour, so this would translate directly into productivity and 
profit losses for the company. 

Revenues would be immediately impacted because perhaps one quarter of the fleet 
would be inoperable. This translates into a daily loss of some $85,000 in revenue 
loss per day. In an industry characterized by very slim profit margins, loss of 
even one day's worth of revenue could potentially wipe out profits for the month. 
Any losses that Company A suffers from a major earthquake would have repercussions 
for its customers. Some of the cargo feeds into just-in-time production processes 
that would shut down if the goods did not arrive on time. More generally, for a 
disruption lasting more than a few days, the company would probably impose an 
emergency surcharge to pass on unanticipated costs to its customers. For 
disruptions of a month or more, the company would need to resort to such actions 
to survive. 

In the M 6.5 Seattle Fault scenario, the situation would be further complicated by 
the need for employees to take care of their home situations first. With fewer 
available drivers and legal limits on the number of hours per day that a driver 
can be on the road, Company A would have to walk away from any business that it 
couldn't handle. This could easily lose the company accounts, particularly from 
customers outside the disaster region. Even though it is currently in good 
financial health, in a prolonged highway disruption scenario, Company A would face 
prospects of failing altogether. 

Case B: Air Shipping Company 
Company B is a major shipping company that provides overnight package delivery 
service to businesses and consumers. While headquartered outside of Washington 
State, it maintains a major regional office in downtown Seattle that houses 1,000 
employees as well as a critical mainframe computer. The company uses Boeing Field 
for all its deliveries going from or to the Pacific Northwest. An inbound package 
would be flown into Boeing Field and trucked to one of the 5 distribution stations 
to be sorted and delivered to its ultimate destination. 

In an earthquake, Company B has identified three critical factors governing 
business impacts: performance of highways, airports, and the company's mainframe 
facility. Should access to the mainframe computer be compromised, the firm has 
contingency plans to activate a "hotsite" in Texas where a duplicate, temporary 
facility could be rapidly set up to take over information processing functions. 
This would require transporting over 1,000 backup tapes out of Seattle through a 
regional airport to the hotsite and would therefore also rely on the viability of 
the regional transportation system. 

The old rule of thumb was that a shipping company must recover within 3 days in 
order to avoid noticeable losses in revenue. However, in today's world of online 
sales and dot.com business, losses now accrue in as little as 3~5 hours. 
As a large business, Company B is able to respond flexibly to disasters to try to 
minimize business interruption losses. It has invested in a backup electric power 
generator at its Seattle office, a telecommuting program available to its key 
employees, and the capacity to automatically reroute incoming telephone calls to 
other call centers in Portland and elsewhere outside the region. 

Nonetheless, highway disruption of the severity estimated for the M 7.1 Benioff 
Zone earthquake scenario would be very disruptive to Company B's ability to 
deliver packages in a timely fashion. The company would probably have to suspend 
regional overnight delivery service, lose service fees, and incur substantial 
overtime labor costs. In the long run, it would lose customers, particularly in 
the arena of online businesses requiring next-day delivery, to any competitors 
that performed better during the crisis. 

Disruptions to shipping firms like Company B would cause ripple effects throughout 
the Seattle regional economy. Many small manufacturers use these firms rather than 
maintaining their own fleet of trucks and planes. Businesses operating just-in-
time production processes require immediate transportation functionality and would 
be quickly impacted. 


4.3.2.2. Manufacturing Industry 

Background 
After transportation services, manufacturing may be the industry that is most 
reliant on the regional highway infrastructure. Manufacturing processes typically 
involve shipping raw materials and other inputs to a plant, usually by truck on 
interstate highways. Manufactured goods must then be similarly transported to 
distribution centers or directly to customers, generally outside the region. 
For instance, the regional highway network plays an important role in the 
manufacture of commercial airplanes by the Boeing Company. In the Seattle 
metropolitan area, Boeing operates major commercial airplane and aerospace plants 
at Everett, Renton, Auburn, Kent, Frederickson, and Boeing Field. Parts for 
aircraft production, such as wing sub-assemblies, are continuously being moved by 
truck between these plants in a just-in-time production process. 

Many manufacturing firms are large businesses that tend to have more resources for 
disaster preparedness, mitigation, response, and recovery than smaller businesses. 
Manufacturers interviewed for this study were confident of their ability to 
weather a major earthquake disaster, provided the production plants themselves 
survived. Highway damage alone would be challenging and possibly costly, but not 
insurmountable. The case study below illustrates how a "typical" manufacturing 
firm might be impacted by highway disruption in a disaster. 

It should be noted, however, that dependence on the highway network varies 
tremendously from one manufacturing company to the next. The degree of dependence 
corresponds to three main factors: the nature of the product (e.g., its 
perishability), the nature of the production process (e.g., single- or multi-
plant, origin of raw materials, production lead times), and the nature of the 
market for the product (e.g., competition). 


Case C: Large Manufacturer 

Company C is a world-renowned producer of gourmet chocolates and confectioneries. 
Over the last century, it has grown from a small, family-run business to a large 
corporation with a global market. It sells about half its products to specialty 
food distributors and the other half directly to consumers through its own chain 
of retail stores located throughout the country. 

Company C maintains two main facilities in the region: a headquarters building in 
Bellevue and a manufacturing plant in the Kent Valley. Approximately 1,200 
employees work at these two locations. The Bellevue facility houses marketing, 
payroll, and other important support functions. 

The company has taken some actions to deal with the earthquake risk in the region. 
They are undergoing seismic retrofit of their headquarters building and have 
invested in some key nonstructural mitigation such as bolting down computer 
servers. They also have earthquake insurance for property damage and some business 
interruption. The insurance is subject to the standard 2 percent deductible, which 
comes to over $1 million. Plans to build a second plant on the East Coast will 
provide the added benefit of diversifying the location of production. 

Regional highway disruption would principally impact Company C by hindering its 
ability to get materials and products in and out of the Kent Valley plant. The M 
7.1 Benioff Zone earthquake scenario of highway damage would be disruptive - 
possibly very disruptive - to the functioning of the plant. Since the chocolates 
and confections are produced with about a one-month lead time, any short-term 
disruptions could be made up through working overtime. This would involve extra 
labor costs but ensure that to its customers, there would be no interruption in 
business. 

The Benioff Zone scenario of highway damage would also be very disruptive to 
employees' ability to get to work. If necessary, the company would put up key 
employees in nearby hotels and incur other similar costs to work around the 
congestion situation. 

If this kind of highway disruption were to last a month, it would probably cause 
revenue losses. Not all of the production losses could be made up through overtime 
work. The company would also have to delay introducing some new products to the 
market. The M 6.5 Seattle Fault scenario would be extremely disruptive to the 
company's ability to do business. The biggest impact would derive from the 
crippling of distribution capacity from the Kent Valley plant. Because Company C 
enjoys high brand name recognition and loyalty, and moreover does not face serious 
competition for its products, it is not especially worried about suffering long-
term loss of business in such a scenario. 



4.3.2.3. Retail Trade 

Background 
Generally speaking, retail trade businesses are less dependent on the highway 
network than firms in transportation services or manufacturing, for whom being 
able to move goods by truck is vital. Losses in retail would derive primarily from 
impeded passenger mobility in the region - if customers cannot reach the stores, 
they cannot buy from them. However, unlike trucks, automobiles can use local 
streets to bypass highway closures and congested areas. 

While retailers are less dependent on highways, this does not mean they will fare 
better in the event of an earthquake. Transportation disruptions alone could 
potentially result in layoffs and lost revenue. Indeed, of the businesses 
interviewed for this study, retailers anticipated some of the highest revenue 
losses from highway disruption. 

The Nisqually earthquake demonstrated how even local road damage can be very 
disruptive to retail trade businesses. Closure of the Fourth Avenue Bridge in 
Olympia reduced access to downtown from the west side. This reportedly cost some 
stores and restaurants sales losses of up to 25~70 percent. Several weeks after 
the earthquake, retailers in Seattle's earthquake-damaged Pioneer Square district 
were most concerned that the closure of Jackson Street was turning away potential 
customers. 

It can be argued that one business' loss may be another's gain. If indeed 
consumers simply make their purchases elsewhere, then while specific local 
businesses and business districts may suffer, there would be no net loss from a 
regional standpoint. It should be kept in mind, however, that disasters can reduce 
overall consumption levels. If regional traffic congestion is severe and 
protracted after a disaster, consumers may very well reduce non-essential shopping 
trips. There may also be net out-migration of population. 

Retail trade is highly vulnerable to losses in disasters for a variety of reasons. 
First, businesses tend to be small or medium-sized. They therefore lack the 
resources available to large businesses - resources to invest in mitigation and 
preparedness before a disaster, as well as to respond to and recover from the 
event once it happens. Second, retailers generally serve local markets that would 
be impacted in the same disaster. In exporting to other regions, manufacturers do 
not face the same potential loss of demand for their products. Additionally, some 
types of retail sales loss cannot be made up through overtime work. 

Of course, some retailers such as franchises may have access to the resources of a 
national or multi-national corporation. They would be able to better weather the 
potential impacts of a disaster than small, single-location businesses. Case D 
below presents an example of a large retailer of this type. 

Small businesses are perhaps the most vulnerable of all businesses to disaster. 
Because revenues are often low and debt relatively high, profit margins are thin. 
Competition for customers is often intense. Small businesses often cannot spare 
the time or resources to plan for disasters ahead of time. They often cannot 
afford earthquake insurance. In a disaster, they have generally have few options 
and little financial reserves to draw upon. Previous studies have found that many 
small business owners will attempt to finance disaster losses using personal 
credit card debt. Case E below depicts how a typical small retailer would be 
impacted in the scenario earthquakes. 



Case D: Large Retailer 

Company D is a well-known home electronics "superstore" located in south Seattle. 
Some 40,000 shoppers visit the store each week. Although operating under the name 
of a national chain, it is owned by two partners in a franchise-type arrangement 
with the parent company. The store employs 400 people, most of whom live close by. 
While Company D earns the vast majority of revenue from the sales floor, it has 
recently begun a home delivery service and maintains a small fleet of trucks this 
purpose. The store serves a market area extending from Mount Vernon to Portland. 
Most of its customers, however, live north of the store in the Seattle-Bellevue 
area. 

Company D has not given much thought to disaster preparedness beyond setting up 
emergency evacuation plans. Fortunately, as a franchise, the store participates in 
the parent company's national insurance plan and can expect some financial and 
technical support from the parent company in the event of a disaster. Even the 
worst-case earthquake scenario is therefore unlikely to force Company D into 
bankruptcy. 

However, the store could suffer severe short-term business losses if 
transportation systems were disrupted in a disaster. On a recent winter day when 
streets were paralyzed by heavy snow, the store managed to open but made only 25 
percent of its usual daily sales. Experience has shown that up to two-thirds of 
lost sales may be made up later, as consumers simply defer making their purchases. 
The remaining third, being spontaneous shopping, is never recaptured. The actual 
magnitude of permanently lost sales would also depend upon the duration and 
extensiveness of the transportation disruption. 

In the event of highway disruption, store managers would be most concerned about 
two things: first, the ability of customers to reach the store, and second, the 
ability to receive inventory shipments by truck. The M 7.1 Benioff Zone scenario 
would be very disruptive to both these critical functions. 

Company D relies on trucks to bring in its entire inventory. Some of the stock is 
delivered from warehouses in California and elsewhere, mostly traveling from the 
south via Interstate 5. The remainder arrives by ship at the Port of Seattle from 
overseas manufacturers. These goods also need to be trucked to the store. Company 
D normally replaces stock sold in one weekend by the next weekend. It maintains 
about 3 weeks' worth of inventory in the store, so delays in restocking would 
quickly affect its bottom line. 

Store managers anticipate substantial cost increases in the scenario earthquakes. 
The trucking companies used by the business, such as Company A (see above), would 
certainly pass on the costs of disaster-related expenses. In addition, Company D 
would probably incur extra advertising and promotional expenses. It would expand 
its home delivery service, possibly "eating" the extra cost rather than charging a 
fee for the service. It would have to pay overtime to its delivery truck drivers 
and office personnel, which could amount to as much as 10 to 20 percent of 
revenues. Even with the extra hours, the drivers might not be able to make as many 
trips as needed due to heavy congestion. 

Revenues, moreover, would take an enormous hit. People would not be thinking about 
buying electronics in the aftermath of a disaster. In the first week of the M 7.1 
Benioff Zone earthquake scenario, the store might suffer as much as a 75 percent 
loss of revenue. As people adjust their travel behavior, subsequent weeks might 
see this reduced to a 50 percent loss. If the situation did not improve much, 
layoffs might be necessary. Layoffs would have a multiplier effect on the regional 
economy, as the lost income would not be spent. 

The M 6.5 Seattle Fault scenario would be even more devastating, since it would be 
a direct hit on Company D's main customer base. After one month of disruption, 
revenue losses might still be as high as 70 percent. 

Like other retailers, Company D is "very much at the mercy of the local economy." 
The impact of an earthquake would depend largely upon how the market responded. 
There could be a silver lining to the disaster, as households would need to 
replace damaged home electronics. However, in the atmosphere of thrift that would 
likely prevail, this boon might not extend to the luxury product lines from which 
the store gains much of its profits. 




Case E: Small Business 

Company E is a small, independently owned women's apparel store in South Center 
Mall. It has been in business for less than a year and employs 4 people, including 
the manager. The store takes in average revenues of $1,000 per day. As a new 
business still paying off a loan, Company E is struggling to stay on its feet and 
maintains no cash reserves. If all goes well, the owner would like to open up a 
second store in the future. Company E's main competitors are the chain, brand-name 
apparel stores in the mall. 

If regional highways were disrupted in an earthquake, customer access to the store 
would be very disrupted. This would be the manager's main concern. The store's 
ability to receive supplies would also be very disrupted, but this would be of 
lesser concern if there were fewer customers in the event. Since most of the 
employees live very close by, commuting difficulties would also be a lesser 
problem. 

Highway disruption in the M 7.1 Benioff Zone scenario earthquake could cause 
revenue losses of some 50 percent in the short-term. If the situation continued 
for a month, people might adapt their travel patterns to deal with the traffic 
congestion, so the number of shoppers might recover somewhat. However, the 
business situation for Company E would deteriorate further. The store would not 
order any new stock, but instead just try to sell what was on the racks in order 
to stay in the black. Some employees might be laid off. Company E would be in 
serious financial trouble if the situation lasted for more than a few weeks. 


4.3.2.4. Finance and Insurance 

Background 
Transportation disruption would cause less overall disruption to businesses in the 
finance and insurance industry, as compared with other sectors profiled in this 
report, but the impacts of a major disaster would still be serious. 

Businesses in this industry are primarily dependent upon the highway network for 
employee commuting. All of those interviewed for this study - even those whose 
major facilities are located outside the Port-to-Port study corridor - affirmed 
that highway disruption in the M 7.1 Benioff Zone earthquake disaster would be 
"disruptive", if not "very disruptive", to employees' ability to get to work. 
Insurance companies are unique in that the major portion of their losses would 
come from claims from policy-holders. Since they interact with their customers via 
telephone or electronically, transportation disruption alone would have few 
impacts other than on commuting. Highway congestion would cause inconveniences for 
field claims adjusters, though this would not necessarily delay paying out 
insurance claims. Being able to maintain acceptable levels of service would be 
important to avoid losing new customers and customers outside the disaster region, 
who may not be as understanding of the local disaster situation. 
Banks, however, also rely on the highway network to transport checks and cash 
between facilities in the region. Checks need to be processed and returned to 
their originating bank, and access to the airports is important for this function. 
Delays in processing checks would result in loss of interest revenue that could 
add up to significant financial losses for a bank. 

In the Nisqually earthquake, temporary closures of SeaTac Airport and Boeing Field 
caused some disruption in check processing functions at regional banks. The 
physical flow of paper had to be diverted to other airports, in some instances to 
Everett or Portland, from where they were trucked in at some delay. One 
institution diverted a small portion of its Seattle region check processing work 
to an out-of-state facility. 

In the aftermath of a disaster, just when the public would be demanding cash for 
emergency purchases, banks may have difficulty stocking ATMs due to traffic 
conditions. Banks prefer to stock ATMs on a just-in-time basis. Machines typically 
get serviced about once a week, although some busier ones get restocked almost 
once a day. 

Impaired customer access to banks could also be an issue. However, if customers 
could not reach their bank in person, conducting business by Internet or telephone 
could be a viable alternative. 

Businesses in finance and insurance tend to be large firms that have engaged in 
some degree of disaster mitigation and preparedness. Parts of the industry are 
required by regulatory agencies to have business recovery plans. Firms often 
invest in backup electricity generators at critical sites, can automatically 
reroute customer calls to out-of-state locations, and subscribe to earthquake 
insurance. 

Partial relocation is typically an option, though a costly one. Many have 
contracted with vendors for emergency work locations, including hotsites where 
data centers could be moved. Activating a hotsite and moving a data center there 
could cost up to $1 million. One company that was interviewed had suffered 
extensive damage to a leased building in the 1994 Northridge earthquake. They were 
able to quickly relocate the entire office to another facility in the L.A. region. 
Some businesses have contracts with hotels near the offices, where they could put 
employees up in an emergency. 

Other responses are possible that would make a business more resilient to 
transportation disruption. Telecommuting and flexible scheduling of work hours are 
options for some employees, particularly in the insurance industry. Given the 
range of adjustments that would take place, businesses typically anticipate that 
losses would decrease over time, even if the highway network remained impaired. 
Actual losses in finance and insurance - and indeed, in other services - will 
depend greatly on the overall impacts to the regional economy and the severity of 
losses suffered to the employment and customer base. 




Case F: Banking Services 

Company F is a banking firm with offices throughout the western and central U.S. 
It employs 900 people in King and Pierce counties, where facilities include 
branches and a regional headquarters building in Tacoma. The latter houses a call 
center, a data center with mainframe computers, a cash vault, and a check 
processing center. Incoming checks and outgoing cash letters need to be 
transported between the firm's main office facility and Boeing Field, SeaTac 
Airport, and the Federal Reserve clearinghouse in downtown Seattle. 

Highway damage in the M 7.1 Benioff Zone scenario earthquake would be very 
disruptive, particularly for check processing. Delays could cost the firm several 
thousands of dollars a day in forgone interest. The bank could also face 
surcharges if it were unable to return checks back to a sending bank within a 
certain period of time. Employee commuting would be severely impacted, but people 
would learn to adjust their travel patterns. Up to one third of the employees 
would be able to get around transportation problems by working at other locations, 
at home, or on flexible work schedules. 

If the dislocation were to last a month or more, Company F would start serving its 
western Washington banking out of its smaller Portland office. This would incur 
additional expenses for relocating employees and hauling checks and related paper. 
In a catastrophic disaster such as a M 6.7 Tacoma Fault event, Company F would 
expect serious damage to its regional headquarters building, as well as widespread 
infrastructure failures. It would probably relocate its data center to its out-of-
state hotsite for up to a month, at a cost of $400,000. Some employees would also 
be relocated to other offices. Overall losses and expenses could exceed a million 
dollars. However, the company does not expect to suffer any long-term loss of 
business once regional facilities and infrastructure are repaired. 



4.4. Regional Economic Impact 
In approaching the impacts of the M 7.1 Benioff Event we were guided by the 
results of the interviews, as described above. We assumed that there would be 
relatively severe impacts for the first 3 months, the time we assumed it would 
require the regain functionality in the highway network. However, we also assumed 
that after this time period there would still be impacts due to the ongoing 
reconstruction efforts, which we assumed would take one year to complete. We 
assumed that the cost of highway reconstruction would be $292 million, a figure 
based on analyses of the cost of replacing bridges in the Northridge earthquake. 
The impact of this reconstruction activity is described in Table 4.4, which 
indicates business activity in the regional economy would total $654 million, 
creating 8,247 jobs, yielding $214 million in labor income, and almost $15 million 
in tax revenues. We assumed that the funds for this reconstruction activity 
would come from outside the region, thereby not redistributing funds that would 
otherwise go towards other transportation projects. 

4.5. Economic Analysis Summary 
The economic impact portion of this study investigated how the Project Impact 
earthquake scenarios of highway outage would impact local businesses, as well as 
the regional economy as a whole. The approach consisted of conducting a series of 
business interviews, developing case studies from these, and running a regional 
economic impact model to estimate total impacts. In-person interviews were 
conducted with 21 local businesses selected to represent a cross-section of the 
economy, from manufacturing to trade to financial services. A few critical 
infrastructure service providers were also interviewed. The businesses ranged from 
a 2-person family operation to major employers accounting for thousands of 
regional jobs. Important findings from the interviews include the following: 

  The highway outage situation anticipated for the M 7.1 Benioff Zone earthquake 
scenario would be "very disruptive" for 76 percent of the businesses interviewed. 

  Revenue losses in this scenario would be minimal for some types of firms but 
very great for others. Retail trade businesses appear to be particularly 
vulnerable, with respondents anticipating up to an immediate 75 percent loss of 
revenue. 

  Even businesses that could avoid revenue losses would suffer significant cost 
increases. Many cited the need to pay for overtime labor as an important disaster-
related expense. 

  Certain sectors would be especially hard hit by the highway disruption. Examples 
include trucking and other transportation firms, sectors requiring access to the 
ports (e.g., warehousing, Alaskan fishing fleet), just-in-time manufacturers, and 
retailers dependent on customer access to the stores. 

  Trucking firms would find the M 7.1 Benioff Zone scenario "devastating" and pass 
on cost increases to their customers. As a result, regional transportation costs 
will rise significantly. 

  Many small businesses have low financial reserves and could face bankruptcy in 
the event of prolonged regional transportation disruption. 

Insights such as these were incorporated into the regional economic impact 
modeling effort. This analysis made use of a modified version of the Washington 
State input-output model, customized to handle cost changes and output changes 
similar to those described above from our interviews. We explicitly considered 
increased transportation costs (purchased and own-account), as well as lost 
output, increased labor costs (related to overtime), as well as the costs of 
reconstructing the highway system. The stimulatory effect of spending for 
reconstruction is more than offset by the losses incurred by businesses due to the 
hypothesized earthquake. It is estimated that sales or output would decrease by 
$2.8 billion, or 4.1%, job reduction would be 39 thousand (a 4.7% decrease), labor 
income would decline by $1.1 billion (4.6%), and sales and B&O tax revenues would 
decline by $72 million (4.6%). 




5. Contingency Planning 
5.1. Introduction 
The Puget Sound highway transportation system is currently at/or near capacity for 
most of the day. Accidents, major and minor repairs or maintenance or other 
unforeseen short-term closures can cause significant delays, often lasting for 
hours. 

What would be the effect on the region to a longer-term major outage in the system 
do to a significant earthquake, and what could be put into place to quickly help 
address the problem? 



5.2. Transportation Contingency Planning 

Project Impact of King/Pierce Counties through the "Port to Port" Transportation 
Study decided to develop a pilot contingency planning program for two specific 
locations, one in each County. Working independently of one another, each County 
selected a location to work through the contingency planning process. 
In King County the selected site was Interstate 5 and the junction of 405 near the 
Seattle/Tacoma Airport. The Pierce County location was the Puyallup River Bridge 
on Interstate 5 at the City of Tacoma and the City of Fife border. This site is 
near the Port of Tacoma. 

In both cases Transportation Planners and Engineers, Public Works Officials, Law 
Enforcement, Emergency Management Response personnel , and representatives of the 
Ports of Seattle and Tacoma were brought together from the effected Cities, the 
County and the State to begin the discussion process. In addition a representative 
of the United States Department of Transportation monitored the effort in Pierce 
County. 



5.3. Methodology 
Historically, while their has been some cooperative efforts among jurisdictions to 
coordinate transportation issues, nothing like this contingency planning for a 
specific major transportation outage has ever been undertaken in either County. 
The first question to answer was given this exact geographical site being 
affected, where do the vehicles go when this option is taken away? Transportation 
experts had to educate the other representatives on the new transportation 
computer modeling programs that provide estimates on the effects on the 
surrounding road infrastructure. Maps were then produced utilizing Geographic 
Information Systems (GIS) that showed the traffic alternatives and effects on the 
collateral road systems. 
From that information the alternative routes were identified by responsible 
jurisdiction, analyzed for capacity and signaling capabilities, and placarding, 
route directional signage, and road barricade needs identified. 




5.4. The Plan 
The emergency response plan for failure at either site consists of the following 
element; 

 1. 	 An emergency response team consisting of personnel from the Washington State 
Department of Transportation, Washington State Patrol, the Public 
Works/Transportation and Law Enforcement Departments from the affected Counties 
and Cities will operate from the County Emergency Operations Center and they will 
immediately activate the emergency plan. 

 2. 	 Preplanned traffic signal changes, road direction signage, road blockades 
will be implemented as specified under the plan. In addition WSDOT will temporally 
restripe one-lane exit ramps on the Interstate that have suitable width, into two 
lane exit ramps. 
 3. 	 The County's Public Information Plan will be activated along with the Joint 
Information Center Plan to ensure that complete and adequate information is 
disseminated to the media and the public. 
 4. 	 Damage assessment and time outage estimations will be quickly made to 
determine the possible closure length of time and to analyze if there needs to be 
a transition from immediate or temporary closure to a more permanent long-term 
situation. This may require additional detours or signage issues. 



5.5. Conclusion 
There have been many positive developments from the contingency planning 
initiative that has been undertaken in the "Port to Port" study area. 
The first multi-jurisdictional transportation emergency response plan is being 
developed and exercised by agencies that historically have not worked together. 
New public safety planning and working relationships have been formed between 
transportation officials and public safety personnel that did not previously 
exist. Identification of roles and responsibilities among the group, and the 
education of each other's function and capabilities have been extremely valuable. 
The new emergence of disaster transportation planning, and the economic impact 
that is possible from not doing such planning, has been a revelation to those 
involved in this process. Given the fragile state and capacity issues involving 
our road infrastructure system, it is clear that this kind of effort needs to be 
quickly expanded. 




References 
 
Earthquake Scenarios 

Fluck, P., Hyndman, R.D., and Wang, K.. (1997). Three-dimensional dislocation 
model for great earthquakes of the Cascadia subduction zone, J. Geophys. Res., 
 v. 102, pp. 20,539-20,550. 
 The ground-motion attenuation references are from Seis. Res. Letts. 
 (all are in Seismological Research Letters, vol. 68, no. 1, Jan/Feb 1997): 
 Boore, D.M., Joyner, W.B., and Fumal, T.E. (1997). Equations for estimating 
 horizontal response spectra and peak acceleration from western North 
 American earthquakes: a summary of recent work, Seis. Res. Letts., v. 68, 
 no. 1, pp. 128-153. 
 Campbell, K. W. (1997). Empirical near-source attenuation relationships for 
 horizontal and vertical components of peak ground acceleration, peak ground 
 velocity, and pseudo-absolute response spectra, Seis. Res. Letts., v. 68, 
 no. 1, pp. 154-179. 
 Sadigh, K., Chang, C.Y., Egan, J.A., Makdisi, F., and Youngs, R.R. (1997). 
 Attenuation relationships for shallow crustal earthquakes based on 
 

California strong motion data, Seis. Res. Letts., v. 68, no. 1, pp. 
180-189. 
Youngs, R.R., Chiou, S.J., Silva, W.J., and Humphrey, J.R. (1997). Strong 
ground motions attenuation relationships for subduction zone earthquakes, 
Seis. Res. Letts., v. 68, no. 1, pp. 58-73. 
Liquefaction 
Grant, W. P.; Perkins, W. J.; Youd, T. L., 1998, Evaluation of liquefaction 
potential in Seattle, Washington; in Rogers, A. M.; Walsh, T. J.; Kockelman, W. 
J.; Priest, G. R., editors, Assessing earthquake hazards and reducing risk in the 
Pacific Northwest: U.S. Geological Survey Professional Paper 1560, v. 2, p.441-
473. 
Palmer, S. P. 1995, Liquefaction analysis of soil deposits found in the Sumner 
quadrangle, in Dragovich, J. D.; Pringle, P. T., 1995, Liquefaction susceptibility 
for the Sumner 7.5-minute quadrangle, Washington: Washington Division of Geology 
and Earth Resources Geologic Map GM-44, 1 sheet, scale 1:24,000, with 26 p. text. 
Palmer, S. P.; Schasse, H. W.; Norman, D. K., 1994, Liquefaction susceptibility 
for the Des Moines and Renton 7.5-minute quadrangles, Washington: Washington 
Division of Geology and Earth Resources Geologic Map GM-41, 2 sheets, scale 
1:24,000, with 15 p. text. 
Palmer, S. P.; Walsh, T. J.; Logan, R. L.; Gerstel, W. J., 1995, Liquefaction 
susceptibility for the Auburn and Poverty Bay 7.5-minute quadrangles, Washington: 
Washington Division of Geology and Earth Resources Geologic Map GM-43, 2 sheets, 
scale 1:24,000, with 15 p. text. 
Palmer, S. P.; Evans, B. D., Schasse, H. W., 2001 (in preparation), Liquefaction 
susceptibility of the greater Eastside area, King County, Washington: Washington 
Division of Geology and Earth Resources Geologic Map GM-48, 1 sheet, scale 
1:36,000. 
Shannon and Wilson, Inc., 1993, Evaluation of liquefaction potential Tacoma, 
Washington; Final technical report under U.S. Geological Survey Contract No. 14-
08-0001-G-1978: Shannon & Wilson, Inc., 1 v.