Geoscience Reference
In-Depth Information
scenario,arupturealongtheentirelengthoftheHaywardFaultwouldclosealmost1500
streets and highways (ABAG 1997). While the ABAG study provides information on
street and bridge closures, it falls short of estimating the economic consequences for
these closures. Large number of closures that do not have a significant economic impact
would not beas critical as fewer closures witha great economic impact.
Inordertosystematicallyevaluatetheearthquakerisktotransportationsystems,itisnec-
essary to develop a methodology that includes the direct loss from damage to network
componentsandthelossfromlimitedfunctionalityofthenetworksystem.Severalrecent
studieshaveinvestigatedthelossesfromfailuresoftransportationnetworks.Forexample,
BasozandKiremidjian(1996)computetheriskofatransportationnetworkafteranearth-
quakeeventforemergencyplanningpurposes.Shinozuka(2000)studiedtheperformance
ofthehighwaynetworkintheLosAngelesareaafterthe1994Northridgeearthquakeand
developedaprobabilisticframeworktopredicttheeffectofbridgerepairsaftertheevent.
In another study for the same area, Shinozuka et al. (2003) used Monte Carlo simula-
tion to estimate the damage of bridges and its consequences to the performance of the
transportation network. Kiremidjian et al. (2006) assessed the damage of the Bay Area
transportation network bridges after four low probability but high consequence earth-
quake scenarios in the Bay Area and estimated the network delays for fixed and variable
post-event trip demand. The same study addressed the problem of post-event emergency
response planning and presented an example application for six hospitals located in the
East Bay. Moore et al., (2005) explored the economic impact of electric power loss in
the Los Angeles and Orange County area on the transportation network and the local
economy. Cho et al. (2003) studied the transportation network post-event performance
for variable demand and estimated thelosses based on the total delays afterthe event.
TheFederalEmergencyManagementAgencydevelopedHazardsU.S.(HAZUS),asoft-
wareforriskmitigationandplanning.ThemethodologiesinHAZUS(1999)estimatethe
structural and the downtime losses after natural disasters; however, these methodologies
do not have capabilities for network analysis of lifeline systems including transportation
systems.Untilrecently, very fewsoftwaretoolswereavailable fortheriskassessment of
networksystems.Inanefforttoovercomethelackoftools,theCaliforniaDepartmentof
Transportation is developing software for Risks from Earthquake Damage to Roadway
Systems(REDARS).REDARSisaseismicriskanalysissoftwarepackagethatestimates
the structural and operational losses of transportation network systems and is expected
toenabletheCaliforniaDepartmentofTransportation(Caltrans)toimproveitsabilityto
plan for and respond toearthquake emergencies.
Inallpreviousstudies,therisktothetransportationsystemisassessedundertheassump-
tion that the ground motion and damage tobridges inthe network areuncorrelated. Sim-
ilarly, the risk is computed either for specific scenario events or through Monte Carlo
simulationwithmanyoftheuncertaintiesinmodelparametersignored.Considerationof
thecorrelationandotherparameteruncertaintiesposesasignificantcomputationalchall-
enge when the riskassessment includes network analysis.