Seismic Hazard Assessment for a Characteristic Earthquake Scenario

  • Slides: 32
Download presentation
Seismic Hazard Assessment for a Characteristic Earthquake Scenario: Integrating Probabilistic and Deterministic Approaches Antonio

Seismic Hazard Assessment for a Characteristic Earthquake Scenario: Integrating Probabilistic and Deterministic Approaches Antonio Emolo with Vincenzo Convertito and Aldo Zollo Prague, March 18, 2005 Antonio Emolo 1

Table of Contents • Brief review of Probabilistic Seismic Hazard Analysis • • technique

Table of Contents • Brief review of Probabilistic Seismic Hazard Analysis • • technique Integration of Probabilistic and Deterministic approaches to seismic hazard Application to the September 26, 1997, 9: 40 GMT, Colfiorito (Central Italy) earthquake, MW=6. 0 Prague, March 18, 2005 Antonio Emolo 2

PSHA basic steps (Cornell, 1968) • • Seismogenetic zone Seismicity recurrence characteristics Earthquakes effects

PSHA basic steps (Cornell, 1968) • • Seismogenetic zone Seismicity recurrence characteristics Earthquakes effects Hazard evaluation Prague, March 18, 2005 Antonio Emolo 3

In our proposed approach we aim at applying the classical PSHA technique to the

In our proposed approach we aim at applying the classical PSHA technique to the single fault case by integrating PSHA with a statistical - deterministic technique for predicting strong ground motion parameters associated with a characteristic earthquake occurring on a given causative fault Prague, March 18, 2005 Antonio Emolo 4

For doing this we need • • the magnitude distribution the seismicity rate for

For doing this we need • • the magnitude distribution the seismicity rate for a single fault/magnitude earthquake a (deterministic) tool for evaluating earthquake effects a statistical description of (deterministic) earthquake effects Prague, March 18, 2005 Antonio Emolo 5

The Colfiorito earthquake: source parameters fault length, L 12 km fault width, W 7.

The Colfiorito earthquake: source parameters fault length, L 12 km fault width, W 7. 5 km bottom depth, zmax 8. 0 km strike, Φ 152° dip, δ 38° slip, λ -118° seismic moment, M 0 1. 0× 1018 Nm moment magnitude, MW 6. 0 rupture velocity, v. R 2. 7 km/s After Zollo et al. , 1999 Prague, March 18, 2005 Antonio Emolo 6

The Colfiorito earthquake: simulation facts • • number of simulated rupture processes: 150 investigated

The Colfiorito earthquake: simulation facts • • number of simulated rupture processes: 150 investigated area: 60× 60 km 2 number of receivers: 64 spacing between adjacent receivers: 5 km Prague, March 18, 2005 Antonio Emolo 7

The Colfiorito earthquake: simulation results Prague, March 18, 2005 Antonio Emolo 8

The Colfiorito earthquake: simulation results Prague, March 18, 2005 Antonio Emolo 8

The Colfiorito earthquake: simulation results Simulated PGAs vs. minimum distance from the surface fault

The Colfiorito earthquake: simulation results Simulated PGAs vs. minimum distance from the surface fault projection are compared with the Sabetta and Pugliese (1987) attenuation curve for a magnitude 6. 0 earthquake Prague, March 18, 2005 Antonio Emolo 9

Coming back to the hazard integral average seismicity rates: from simulation study we do

Coming back to the hazard integral average seismicity rates: from simulation study we do not both need! characteristic -1 c=0. 00204 yrs and exponential models exp=0. 155 yrs-1 b-value = 0. 8475 Prague, March 18, 2005 Antonio Emolo 10

Hazard maps in terms of PGA values having a fixed frequency of exceedance corresponding

Hazard maps in terms of PGA values having a fixed frequency of exceedance corresponding to three return periods: T 1=1, 000 yrs T 2=5, 000 yrs T 3=10, 000 yrs Prague, March 18, 2005 Antonio Emolo 11

Hazard maps – T=1, 000 yrs Prague, March 18, 2005 Antonio Emolo 12

Hazard maps – T=1, 000 yrs Prague, March 18, 2005 Antonio Emolo 12

Hazard maps – T=5, 000 yrs Prague, March 18, 2005 Antonio Emolo 13

Hazard maps – T=5, 000 yrs Prague, March 18, 2005 Antonio Emolo 13

Hazard maps – T=10, 000 yrs Prague, March 18, 2005 Antonio Emolo 14

Hazard maps – T=10, 000 yrs Prague, March 18, 2005 Antonio Emolo 14

Hazard curves for selected sites characteristic earthquake model Prague, March 18, 2005 Antonio Emolo

Hazard curves for selected sites characteristic earthquake model Prague, March 18, 2005 Antonio Emolo 15

Hazard curves for selected sites exponential magnitude distribution Prague, March 18, 2005 Antonio Emolo

Hazard curves for selected sites exponential magnitude distribution Prague, March 18, 2005 Antonio Emolo 16

I nearly forgot: are PGAs log-normally distributed? Prague, March 18, 2005 Antonio Emolo 17

I nearly forgot: are PGAs log-normally distributed? Prague, March 18, 2005 Antonio Emolo 17

In conclusion • we account for time variable (return period, time of • •

In conclusion • we account for time variable (return period, time of • • • interest, …) in deterministic scenarios; we account for source parameters (geometry, radiation pattern, directivity, …) in PSHA approach; due to the waveforms availability, we can consider any ground motion parameter both in time and in frequency domains we can easily include site effects in the modeling if specific transfer function was available Prague, March 18, 2005 Antonio Emolo 18

Before ending… I would like to thank the MAGMA center and all the people

Before ending… I would like to thank the MAGMA center and all the people who gave me the opportunity to spend a very useful period at the Charles University Prague, March 18, 2005 Antonio Emolo 19

And finally, that’s all Thank you very much for your kindly attention Prague, March

And finally, that’s all Thank you very much for your kindly attention Prague, March 18, 2005 Antonio Emolo 20

Seismogenetic zone • Each zone has uniform earthquake • potential The configuration could be

Seismogenetic zone • Each zone has uniform earthquake • potential The configuration could be point line area volume ZS 9 – Meletti and Valensise, 2004 Prague, March 18, 2005 Antonio Emolo 21

Seismicity recurrence characteristics Recurrence relationship (e. g. , Gutenberg and Richter, 1944) Prague, March

Seismicity recurrence characteristics Recurrence relationship (e. g. , Gutenberg and Richter, 1944) Prague, March 18, 2005 Antonio Emolo 22

Earthquakes effects The ground motion level at a given site and for a selected

Earthquakes effects The ground motion level at a given site and for a selected range of magnitude is generally evaluated through empirical attenuation relationships (e. g. , Joyner and Boore, 1981; Sabetta and Pugliese, 1986; …) Prague, March 18, 2005 Antonio Emolo 23

Hazard evaluation It consists in the computation of the probability of exceedance of different

Hazard evaluation It consists in the computation of the probability of exceedance of different levels of selected ground motion parameter A thorough the evaluation of the hazard integral frequency of exceedance seismic activity rate probability of of occurrence exceedance of aofgiven a threshold earthquake value A 0 For a given time of interest t, the probability of exceedance can be computed as of a given threshold (from catalogues) A 0 having atfor amagnitude distance given distance ininthe therange rrange and magnitude (r, (m, r+dr) m+dm) m Prague, March 18, 2005 Antonio Emolo 24

Hazard evaluation Prague, March 18, 2005 Antonio Emolo 25

Hazard evaluation Prague, March 18, 2005 Antonio Emolo 25

Hazard evaluation Prague, March 18, 2005 Antonio Emolo 26

Hazard evaluation Prague, March 18, 2005 Antonio Emolo 26

The characteristic earthquake model is based on the hypothesis that individual fault tend to

The characteristic earthquake model is based on the hypothesis that individual fault tend to generate similar size (i. e. , “characteristic”) earthquakes Characteristic earthquakes occur on a fault not at the exclusion of all other magnitude events, but with a frequency distribution which differs from the exponential one Several paleoseismic evidences in different tectonic environments support the idea that geometry, mechanism and average slip per event could be considered constant over a large time scale Prague, March 18, 2005 Antonio Emolo 27

The magnitude distribution Prague, March 18, 2005 Antonio Emolo 28

The magnitude distribution Prague, March 18, 2005 Antonio Emolo 28

The seismicity rate It can be evaluated both for the exponential model and for

The seismicity rate It can be evaluated both for the exponential model and for the characteristic earthquake model following the approach proposed by Youngs and Coppersmith (1985) Prague, March 18, 2005 Antonio Emolo 29

Earthquake effects Seismic radiation emitted by an extended rupturing fault is computed by solving

Earthquake effects Seismic radiation emitted by an extended rupturing fault is computed by solving the representation integral in high frequency approximation (Aki and Richards, 1980) The HF Green function is computed in a flat-layered velocity medium The slip function is approximated by a ramp A k-squared final slip distribution on the fault is assumed (Herrero and Bernard, 1996) Prague, March 18, 2005 Antonio Emolo 30

Statistical description of earthquake effects In the frame of a scenario simulation associated with

Statistical description of earthquake effects In the frame of a scenario simulation associated with a characteristic earthquake, some “low frequency” source characteristics can be considered constant over a large time scale in successive rupture episodes. However the single rupture process does not repeat the same style of nucleation, propagation and stopping even if it keeps the mean characteristics. With this in mind, we simulated a large number of rupture processes occurring on the same causative fault considering different positions of nucleation point and different final slip distributions. Synthetic seismograms are computed for each considered rupture process and ground motion parameters of interest are then evaluated through a statistical analysis Prague, March 18, 2005 Antonio Emolo 31

Statistical description of earthquake effects Prague, March 18, 2005 Antonio Emolo 32

Statistical description of earthquake effects Prague, March 18, 2005 Antonio Emolo 32