Development of Earthquake Assessment Methodology in Ncree

In order to promote researches in seismic hazard analysis, engineering structural damage assessment, and socio-economic loss estimation in Taiwan, the National Science Council started the HAZ-Taiwan project in 1998. The National Center for Research on Earthquake Engineering also develops the associated application software "Taiwan Earthquake Loss Estimation System" to integrate various inventory data and analysis modules. The analysis framework of TELES follows the approach of HAZUS, which has been developed in the United States. However, individual analysis modules and parameter values have been modified according to Taiwan local environment. There are three main functions in TELES. First, it helps to obtain reliable estimates of potential seismic hazards and losses soon after the occurrence of earthquakes. Second, it helps to simulate earthquake scenarios and to provide useful data when local governments propose their seismic mitigation plans. Third, it helps to provide seismic hazard assessment and catastrophic risk management tools, such as the seismic insurance policy for residential buildings. This paper focuses on the analysis modules used in early loss estimation sub-system including assessment of ground motion intensity, soil liquefaction potential, building damages and the induced casualties. BACKGROUND In general, risk can be defined by probability of occurrence of a seismic event, exposure of people and property to the event, and consequences of that exposure. Based on the previous definition of risk, an earthquake loss-estimation methodology, integrated with geographic information system (GIS) and designed to run on personal computers, has been developed in the United States. The methodology and associated application software are contained in HAZUS (RMS, 1997). Essentially, the HAZ-Taiwan project and the associated application software "Taiwan Earthquake Loss Estimation System" follow a similar approach used in HAZUS. However, TELES has made major modifications in analysis models and parameter values, not only to accommodate the special environment and engineering practices in Taiwan, but also to reflect the state-of-the-art technology after the introduction of HAZ-Taiwan project in 1998. Furthermore, TELES has functions to estimate automatically the induced disastrous regions and scales soon after the occurrence of strong earthquakes, and to integrate probabilistic seismic hazard analysis and loss estimation. HAZUS does not have the similar features of early loss estimation and seismic risk assessment. The results of the HAZ-Taiwan project help to plan and stimulate efforts to reduce risk from PDF wurde mit pdfFactory-Prüfversion erstellt. www.context-gmbh.de earthquakes, and to prepare for emergency response and recovery from an earthquake. It also provides a standard risk assessment and loss estimation methodology. Expected benefits of a standard methodology include consistency of approach, more economic use of available resources, improved sharing of knowledge, and more consistent performance measurement in hazard mitigation efforts, and providing effective means to set local, regional and national priorities. FRAMEWORK OF METHODOLOGY The HAZ-Taiwan project is mainly composed of three parts: collection of database; development or modification of analysis modules; and update of application software, as shown in Fig. 1. The input database consists of three types of data: inventory data with GIS information, earthquake hazard and geologic data maps, and analysis parameters. The analysis modules take the required inventory data and analysis parameters as inputs, conduct risk and loss estimations based on the site-specific outputs from hazard analysis, and output estimates in the result databases. The third part, integrated with commercial GIS software, is the PC-based application software to execute user's requests, to display input/output databases in both tabular and graphical forms, to generate summary reports, and so on. In ve nt or y D at a Earthquake Hazard and Geologic Data Maps A na ly si s P ar am et er s Direct Social/Economic Loss Estimation Physical Damage Assessment Induced Physical Damage Assessment Indirect Economic Loss Estimation Potential Earth Science Hazard Analysis I/O Inrface Fig. 1: Methodology framework of HAZ-Taiwan. The analysis part contains four groups of analysis modules, namely, potential earth science hazard (PESH) analysis, direct physical damage assessment, induced damage assessment, and social/economic loss estimation. These modules and sub-modules are interdependent. The output from one module acts as input to another. The modular approach allows estimates based on simplified models and limited inventory data. Addition or replacement of existing modules/data may be done without reworking the entire methodology. The modular approach also facilitates the rapid transfer of information and technology between the academic/research communities and the end users. Specific regional analysis models and data can be incorporated in the framework. Another advantage of modular approach is that it enables users to limit studies to selected losses, which may be desirable because of limited budget and inventory constraints. In general, each module requires a comprehensive loss estimation study. However, the degree of required sophistication and associated cost varies greatly by user and application. It is necessary and appropriate that the modules accept multiple levels of detail and precision of input data. PDF wurde mit pdfFactory-Prüfversion erstellt. www.context-gmbh.de ESTIMATION OF GROUND MOTION INTENSITY The PESH module generates estimates of ground motion intensity and ground failure extent. Based on the source parameters of a scenario earthquake and the local site conditions, the ground motion demands are generally in terms of response spectra and peak values (PGA and PGV). The response spectra are simplified and mainly determined by as S and 1 a S , which represent the short-period and the long-period spectral accelerations, respectively. So far, the ground failure estimations consider only the effect of soil liquefaction and are in terms of occurrence probability and the induced permanent ground deformation (PGD). Other related earth science hazards, such as tsunami and inundation, may affect social/economic environment, but are not considered in the current framework of TELES. The first step in scenario simulation is to set source parameters. In general, there are three ways to define source parameters in a deterministic approach, that is, historical events, active faults and arbitrary events. The basic source parameters include event date, time, magnitude, epicenter location and focal depth. If the earthquake accompanies fault rupture, the fault mechanism (reverse, normal or strike), orientation of trace, inclination angle, length and width of rupture plane are required to define the source parameters. In case there is no information about the rupture fault, it may be assumed to be a rectangle passing through the hypocenter of scenario earthquake. If the active fault traces are known, they may compose of many line segments, though not necessarily continuous, to depict the traces realistically. Default length and width of the rupture plane are provided using Wells et al (1994) empirical formula; however, they can be customized to match the actual observation. Since different attenuation laws use different definitions for earthquake magnitude and source-to-site distance, TELES internally converts magnitude scale and selects proper definition for source-to-site distance. Both moment magnitude ( w M ) and local magnitude ( L M ) are generally used in Taiwan. They are converted by using the following equation (Wu et al, 2001), 091 . 2 ln 533 . 4 − = w L M M (1) For example, Jean (2001) uses local magnitude and shortest distance to fault plane, while Boore et al (1994) uses moment magnitude and shortest distance to the horizontal projection of fault plane. If the rupture length reduces to zero, the source-to-site distance automatically uses the definition of focal distance in the attenuation laws. TELES can consider the effect of seismogenic rupture zone when evaluating the source-to-site distance. Estimation of ground motion intensity due to a scenario earthquake may divide into three steps. Referring to Fig. 2, the first step uses the attenuation laws to predict the intensity at bedrock level. The second step uses the local site modification factors to obtain the intensity at ground surface. Finally, if the monitored data at nearby strong-motion stations are available in early seismic loss estimation, the local intensity can be updated accordingly. HAZUS classify the local site conditions into six categories by soil profiles and properties, such as shear velocity and SPT-N value. The site modification factors depend on the soil type as well as the ground motion intensity. However, topography and geology are very complex in Taiwan, basin effects or topographic conditions may influence the ground motion intensity significantly. To overcome the shortage of rough classification scheme of soil types, micro-zonation of site effect is necessary. Since the strong-motion stations installed by the Central Weather Bureau are dense enough, the site modification factors for each PDF wurde mit pdfFactory-Prüfversion erstellt. www.context-gmbh.de chun-li are determined from both the recorded accelerograms and local geology (Yeh et al, 2003). Location of Epicenter, Focal Depth, Magnitude, Fault Trace, Rupture Length, Width and Dip Angle Wave Propagation― Intensity Attenuation and Frequency Filtering Local Site-Soil Conditions Chun-li Site-Specific PGA, PGV and Response Spectrum A S