United Nations Educational, Scientific and Cultural Organization and International Atomic Energy Agency

Transcription

1 Available at: IC/2008/084 United Nations Educational, Scientific and Cultural Organization and International Atomic Energy Agency THE ABDUS SALAM INTERNATIONAL CENTRE FOR THEORETICAL PHYSICS EARTHQUAKE SOURCE AND LOCAL GEOLOGY EFFECTS ON THE SEISMIC SITE RESPONSE Mihaela Kouteva-Guentcheva 1 CLSMEE-BAS, 3 Acad. G. Bonchev str., 1113 Sofia, Bulgaria and The Abdus Salam International Centre for Theoretical Physics, Trieste, Italy, Ivanka Paskaleva CLSMEE-BAS, 3 Acad. G. Bonchev str., 1113 Sofia, Bulgaria and Giuliano F. Panza University of Trieste, DST, Via E. Weiss 4, Trieste and The Abdus Salam International Centre for Theoretical Physics, Trieste, Italy. MIRAMARE TRIESTE September Junior Associate of ICTP. mkouteva@geophys.bas.bg

2 Abstract Strong shallow and intermediate-depth scenario earthquakes for two major cites in Bulgaria are discussed. The contribution of the earthquake source and the local site geology to the seismic input is illustrated. Due to the lack of strong motion records a neo-deterministic seismic hazard assessment procedure is used to generate synthetic seismic signals. After some parametric analyses the computed signals are validated against the few available data. Prognostic estimates of the dynamic coefficient for the target sites are performed with respect to the defined scenario earthquakes and local site models, corresponding to the Eurocode 8 ground types A, B and C. The obtained results show that: (1) the seismic source influence on the seismic input at a given site is comparable with that of the local site geology; (2) the dynamic coefficients, computed for accelerograms (observed and computed) due to strong intermediate-depth Vrancea earthquakes overestimate significantly the values recommended by the Eurocode 8 (EC8) for periods T > 1s. 1

3 1. Introduction In the Bulgarian territory earthquakes play a major role among geological hazards, like landslides, erosion processes, liquefaction, loess potential collapse [Broutchev et al., 1994]. All the main Bulgarian cities are exposed to a significant earthquake hazard. Over the centuries, Bulgaria has experienced earthquakes with large epicentral intensities, I, e.g. the 1818, I=IX (MSK), and the 1858, I=IX (MSK), destructive events near Sofia. Some of the strongest earthquakes in Europe in the 20-th century, e.g. 4 April, 1904, occurred in SW Bulgaria I=X (MSK). The seismic hazard of Bulgaria is controlled by seismic sources located in country and also in the territory of the neighbouring countries (Romania, Greece, Turkey, Yugoslavia and Macedonia). Maximum intensity I=IX (MSK-64) is expected for the main Bulgarian city, Sofia. The seismic hazard in NE Bulgaria with the major town Russe, the biggest Bulgarian port on the Danube River, is controlled by the Vrancea seismic zone, located in Romania. The intermediate-depth Vrancea earthquake sources are of practical and scientific interest due to their social and economic impact on the territory of the adjacent countries. A brief analysis of the available instrumental records of the strong intermediate-depth Vrancea earthquakes, with Mw > 6.5 [Nenov et al, 1990; Ambraseys et al., 2002], has shown the significant effect of the earthquake source mechanism on the seismic input at sites, which clearly differ in local geological conditions and epicentral distances. An illustration of this Vrancea feature is shown in figure 1. Another Vrancea peculiarity is the much stronger frequency-dependent attenuation effect toward NW for higher frequencies (> 1Hz) than the attenuation toward SE [Radulian et al., 2006]. The unusually small attenuation at low frequency has important consequences on the seismic hazard assessment not only in Romania, but also in the neighboring countries (Bulgaria, Rep. of Moldova, Ukraine and even Russia). The main purposes of this study are to: provide strong scenario earthquakes ( Mw > 7.0), that can be used for prognostic estimates of the seismic input; supply seismic input for the chosen scenarios and to validate the synthetic seismic signals against the available data. 2. The Neo-Deterministic Seismic Hazard Assessment Procedure and the Modeling of the Seismic Input at Sofia and Russe To obtain the seismic input at Sofia and Russe a neo-deterministic procedure for earthquake ground motion modelling has been applied [Panza et al., 2001]. The major advantages of this procedure are: (1) the simultaneous treatment of the contribution of the seismic source and of the seismic wave propagation through inelastic media to the seismic motion at the target site/region and therefore (2) the application of this procedure does not require the use of any attenuation relation. 2

4 Figure 1. Dynamic coefficients of recorded Vrancea strong ground motions. Top: Vrancea 1977, March 4, event at Vrancioala (Romania) and NIS (Serbia) stations. Bottom: Vrancea 1990, May 30, event at Russe station (Bulgaria). Applying this procedure site response estimates are provided simultaneously in frequency and space domain. The irregular pattern of the site amplification in the frequency - space domain, obtained even when considering rather simplified geological settings, as for the Russe case study [e.g. Kouteva et al., 2004], confirms the complicated properties of the so-called site effect. They are due to the complex evolution of the seismic wavefield (while it propagates through the laterally heterogeneous, geological media) that cannot be captured by standard convolutive methods [e.g. Reiter, 1990]. The traditionally used attenuation relations, extracted from the available strong motion databases, represent the functional dependency of the random spectral acceleration on the random variables, magnitude, distance and measurement error, and thus the source of systematic error in the seismic hazard assessment, that might be introduced by the attenuation relations, is avoided [Klügel, 2007; Panza et al., 2008] when using the neo-deterministic approach. The problem how crustal properties affect the attenuation and the effect due to inelasticity is taken into account analytically by the neodeterministic procedure, using variational techniques [Panza et al., 2001 and references therein]. To model the seismic input in Sofia the hybrid neo-deterministic approach [Fäh et al., 1993; Panza et al., 2001] based on the modal summation technique applied in the bedrock, combined with the finite difference method for the target site is used. To model the seismic input in Russe the analytical neodeterministic approach based on the mode coupling technique, is applied [Romanelli et al., 1996; 1997; Panza et al., 2001]. 3. Scenario Earthquakes The scenario event represents different combinations of parameters, thus the scenario earthquakes can be different in what concerns source location, magnitude and parameters describing the geometry and the kinematics of the seismic source. Usually for an earthquake prone area, scenario earthquakes with different levels of severity are considered: moderate, severe and extreme earthquakes. Widely accepted in international practice in earthquake engineering analysis, including EC8, is the return period of 475 years. 3

5 3.1. SHALLOW LOCAL EARTHQUAKES - THE CITY OF SOFIA The return period of the maximum macroseismic intensity at Sofia, Io = IX (MSK), is about 150 years [Christoskov et al., 1982], i.e. it could correspond to the strong earthquake scenario. The shallow scenario earthquakes considered in this study are listed in Table 1. For more details see Paskaleva et al. [2008] and references therein. Table 1. Scenario Earthquakes local shallow quakes, Sofia City*. Scenario Geologic Strike Dip Rake Closest distance to Focal M profile* w fault depth Sce1_all M1, M2, M km 10 km Sce1_3a M km 10 km *City sketch and details on the geology of the considered geological profiles was published by Paskaleva et al. in STRONG INTERMEDIATE-DEPTH VRANCEA EARTHQUAKES - THE TOWN OF RUSSE Considering the specific natural conditions, the various categories of elements and systems of risk and the Vrancea earthquake record, the suitable scenario earthquakes, for this area, should correspond to return periods ranging from some 50 to 200 years, when severe or extreme magnitudes are considered, respectively. Suitable Vrancea scenario events can be considered the quakes in the magnitude range from 7.2 (severe earthquakes) to 7.8 (extreme earthquakes). The chosen scenario earthquakes are listed in Table 2. Table 2. Scenario Earthquakes - strong intermediate-depth Vrancea quakes. Scenario* Lat. Long. M w Focal depth Strike Dip Rake Sce_ o N o E km 240 o 72 o 97 o Sce_ o N o E km 225 o 60 o 80 o * Sce_1 seismic source corresponds to the 1986 Vrancea quake (August 30), [Dziewonsky et al. 1991] and Sce_2 corresponds to the Vrancea 1940, Nov. 10, earthquake [Radulian et al., 2000 and references therein; Lungu et al.2004]. 4. The Synthetic Strong-Motion Database 4.1. SHALLOW LOCAL EARTHQUAKES: CASE STUDY- SOFIA CITY Synthetic ground motions along three geological cross sections have been computed and validated by Paskaleva et al. [2004, 2008]. The signals have been grouped in three ranges of epicentral distances: km, km and km. For each group mean ground motion spectral quantities are computed. 4

6 Figure 2. Dynamic coefficients of the computed signals versus the EC8 recommended curves [Paskaleva et al., 2008]. The dynamic coefficients, computed from the synthetic seismic signals and the EC8 recommended curves, are plotted in figure 2. Figure 2 shows the influence of the soil conditions and of the seismic source on the site response STRONG VRANCEA INTERMEDIATE-DEPTH EARTHQUAKES: CASE STUDIES OF RUSSE (BULGARIA) AND NIŠ (SERBIA) Validation of the Seismic Input The frequency-time analysis of the available records [Nenov et al., 1990; Ambraseys et al., 2002] of strong intermediate-depth Vrancea earthquakes has shown that the frequencies up to 5 Hz have the major contribution to the seismic loading of practical importance. A typical example of the long period far reaching effect is the Vrancea 1977, March 4, earthquake. The available data on the Vrancea earthquakes of March 4, 1977 (VR77) and May 30, 1990 (VR901) are used to validate the computations at Niš (ep. distance Δ ~ 500 km) and Russe (Δ ~ km), respectively. Table 3. Strong Intermediate-Depth Vrancea earthquakes. Data used for Parametric Studies and for Validation of the Seismic Input. VR , March 4 Source * Latitude Longitude Mw Focal depth Strike Dip Rake CMT o N o E km 236 o 62 o 92 o CMT o N o E km 50 o 28 o 86 o OTHER o N o E km 225 o 70 o 110 o CMT1m o N o E km 236 o 62 o 92 o CMT2m o N o E km 50 o 28 o 86 o VR , May 30 CMT o N o E / o 63 o 101 o NIEP o N o E o 63 o 101 o * CMT1, CMT2, CMT [Global Centroid Moment Tensor Catalogue]; CMT1m, CMT2m correspond to CMT1, CMT2 respectively, focal depth H = 100 km considered for both cases; NIEP corresponds to Radulian et al. [2000]. 5

7 For both cases the same bedrock model is used [Paskaleva et al., 2001; Kouteva et al., 2004]. For the validation of the computed signals at Russe and Niš, three local models, corresponding to Eurocode 8 ground type C (V s,30 = 325 m/s) have been used: (a) deep model, top layer (of type C) 150 m thick, (b) intermediate model, top layer 60 m thick and (c) shallow model, top layer 30 m thick, for more details see in Kouteva et al. [2008]. The values, including their uncertainties, of the parameters describing the geometry and the motion at the earthquake source, are available from the GCMT Catalogue [ Radulian et al. [2000], Poiata and Miyake [2006] and the Romplus catalogue, [Oncescu et al., 1999]. The seismic input in Russe (VR901) and Niš (VR77) has been computed and validated using the parameters given in Table 3. The comparison of the elastic acceleration response spectra, computed for 5% damping, of the synthetic and observed signals, considering VR77 (Niš) and VR901 (Russe), are shown in figures 3a and 3b, respectively. Both figures clearly show the significant influence of the seismic source on the earthquake loading at the target sites. Figure 3a. Elastic acceleration response spectra, computed for5% damping. Synthetics against observation (solid grey line). The change of the focal depth of 84 km and 94 km (CMT1 and CMT2, plotted in the top line in figure 3a) to 100 km (CMT1, CMT2 - the bottom line in figure 3a) for the Niš case shows a shift of the maximum amplitudes of the acceleration response spectra to larger periods. A comparison of the obtained results for Niš, following Table 3, shows that among the considered parameters, the focal depth seems to have a controlling impact on the seismic input. The results of the theoretical modelling of the seismic input in Russe, including some parametric studies, are shown in figure 3b. Among the earthquake source parameters, the focal depth has, here too, the most significant influence on the seismic loading at the target site. 6

8 Figure 3b. Elastic acceleration response spectra, computed for 5% damping. Synthetics against observation (solid grey line) Scenario Estimates The results of the computations made considering the chosen scenario earthquakes (Table 2) and different local models are shown in figures 4a and 4b. Both scenarios, SCE1 (figure 4a) and SCE2 (figure 4b), show an obvious change, with varying epicentral distance, of the frequency content of the spectral site response, no matter which local model was considered. At epicentral distances Δ > 400 km, the dynamic coefficient at periods T > 1.5s appear visibly higher than the EC8 recommendations. For the considered frequency content, 0-5 Hz, the shallow local models give dynamic coefficients that are closer to the EC8 recommendations. 7

9 Figure 4a. Vrancea scenario earthquake, strong event, SCE1- Table Discussion Shallow and intermediate-depth scenario earthquakes are considered. The seismic input at a given site incorporates the coupled effects of the seismic source and of the inelastic media through which seismic wave propagates. Due to the lack of real strong motion records, synthetic seismic signals have been generated applying a neo-deterministic seismic hazard assessment procedure. The computed signals are validated against the few available observations. The major outcome of this study can be summarized as follows: the computed synthetic seismic input for shallow earthquakes is consistent with the Eurocode 8 requirements; for the intermediate-depth earthquakes, local models with a thin top layer of type C supply synthetic seismic signals, that are quite close to the observed ones; the site response due to both, shallow and intermediate-depth, earthquakes is significantly influenced by the earthquake source mechanism; the dynamic coefficients, computed for accelerograms (observed and computed) due to strong intermediate-depth Vrancea earthquakes exceed significantly the values recommended by the Eurocode 8 (EC8) for periods T > 1s. 8

10 Figure 4b. Vrancea scenario earthquake, extreme event, SCE2- Table 3. Acknowledgments The financial support from the NATO SfP Project N980468, INTAS-Moldova ; CEI Projects "Deterministic seismic hazard analysis and zoning of the territory of Romania, Bulgaria and Serbia" and "Geodynamical Model of Central Europe For Safe Development Of Ground Transportation Systems", and the CEI university network are gratefully acknowledged. This work was done within the framework of the Associateship Scheme of the Abdus Salam International Centre for Theoretical Physics, Trieste, Italy. References Ambraseys, N., Smit, P., Sigbjornsson, R., Suhadolc, P. and Margaris, B. (2002), Internet-Site for European Strong-Motion Data,European Commission, Research-Directorate General, Environment and Climate Programme. Broutchev, I. (Ed.). Geological hazards in Bulgaria, Bulgarian Academy of Sciences, Publishing House of the Bulgarian Academy of Sciences, 1994; Sofia. Christoskov, L., Georgiev, Tzv., Deneva, D., Babachkova, B. (1982) On the seismicity and seismic hazard of Sofia valley, Proc. Of the 4 th Int. Symposium on the Analysis of seismicity and seismic risk, Bechyne castle, CSSR, IX: Georgescu, E.S and Sandi, H. (2000). Towards Earthquake Scenarios under the Conditions of Romania. Proceedings of the 12 WCEE, January, 2000, Auckland, New Zealand, Pap. No Dziewonsky, A. M., Ekstrom, G., Woodhouse, J. H., Zwart, G. (1991), Centroid moment tensor solutions for April-June 1990, Physics of the Earth and Planetary Interiors, 66, Fäh, D., Iodice, C., Suhadolc, P., Panza, G.F. (1993) A new method for the realistic estimation of seismic ground motion in megacities: The case of Rome, Earthquake Spectra 9: Klügel, J. -U. (2007) Error inflation in Probabilistic Seismic Hazard Analysis, Engineering Geology 90 (2007) , available online at www. sciencedirect.com. Kouteva, M., Panza, G.F., Romanelli, F., Paskaleva, I. (2004) Modelling of the Ground Motion at Russe Site (NE Bulgaria) due to the Vrancea Earthquakes. Journal of Earthquake Engineering, 8, 2, pp Kouteva-Guentcheva, M.P., Paskaleva, I.P., Panza, G.F. (2008) Strong Intermediate-Depth Vrancea Earthquakes: Damage Capacity in Bulgaria, submitted to the 14WCEE, October 12-17, Beijing China, Ref

11 Lungu, D., Aldea,A., DemetriuI,S., Craifaleanu, I., 2004, Seismic strengthening of buildings and seismic instrumentation -two priorities for seismic risk reduction in Romania, CD, 1st Int. Conf. Science and Technology for Safe Development of Lifeline Systems, Natural Risks: Developments, Tools and Techniques in the CEI Area. Medvedev, S. V. (1977) Seismic Intensity Scale MSK 76, Publ. Inst. Geophys. Pol. Acad. Sc. 117, pp Nenov, D., Paskaleva, I., Georgiev, G., Trifunac, M. (1990) CATALOG of strong earthquake ground motion data in EQINFOS: Accelerograms recorded in Bulgaria between 1981 and Report No CE 90-02, Sofia and Los Angeles, 1990, 55 pages, published in Southern California University. Oncescu, M.C., Marza, V.I, Rizescu, M., Popa, M. (1999). The Romanian Earthquake Catalogue between , in "Vrancea Earthquakes: Tectonics, Hazard and Risk Mitigation", F. Wenzel, D. Lungu (eds.) & O. Novak (co-ed), Kluwer Academic Publishers, Dordrecht, Netherlands. Catalog under continuous update Panza, G.F., Romanelli, F., Vaccari, F. (2001). Seismic Wave Propagation in Laterally Heterogeneous Anelastic Media: Theory and Applications to the Seismic Zonation, Advances in Geophysics, Academic press; 43: Panza, G.F., Kouteva, M., Vaccari, F., Peresan A., Cioflan, C.O., Romanelli, F., Paskaleva, I., Radulian, M., Gribovszk,i K., Herak, M., Zaichenco, A., Marmureanu, G., Varga, P., Zivcic, M. (2008), Recent Achievements of the Neo-Deterministic Seismic Hazard Assessment in the CEI Region, Proc. of the 2008 seismic Engineering Conference, Reggio Calabria, Italy , Editors: A. Santini and N, Moraci, American Institute of Physics, Melville, New York, AIP Conf. Proc., Vol. 1020, Paskaleva, I., Kouteva, M., Panza, G.F., Evlogiev, J., Koleva, N., Ranguelov, B. (2001). Deterministic Approach of Seismic Hazard Assessment in Bulgaria; Case Study Northeast Bulgaria - The Town of Russe. The Albanian Journal of Natural & Technical Sciences, Vol. 10, Paskaleva, I., Panza, G.F., Vaccari, F., Ivanov, P. (2004) Deterministic modelling for microzonation of Sofia an Expected Earthquake Scenario. AGGH,Vol. 39(2-3), 2004, pp Paskaleva, I., Kouteva, M., Vaccari, F., Panza, G.F. (2008) Application of the Ne-Deterministic Seismic Microzonation Procedure in Bulgaria and Validation of the Seismic Input against Eurocode 8, Proc. of the 2008 seismic Engineering Conference, Reggio Calabria, Italy , Editors: A. Santini and N, Moraci, American Insitute of Physics, Melville, New York, AIP Conf. Proc., Vol. 1020, Radulian, M., Vaccari, F., Manderscu, N., Panza, G.F., Moldoveanu, C. (2000). Seismic Hazard of Romania: Deterministic Approach, PAGEOPH, Vol.157, no1/2, Radulian, M., Panza, G.F., Popa, M. and Grecu, B., (2006). Seismic wave attenuation for Vrancea events revisited. Journal of Earthquake Engineering, Vol. 10, N.3, Reiter, L. (1990). Earthquake Hazard Analysis, Columbia University Press, New York, 254 pp. Romanelli, F., Bing Z., Vaccari, F. and Panza, G. F. (1996). Analytical computation of reflection and transmission coupling coefficients for Love waves, Geophys. J. Int., 125, Romanelli, F., Bekkevold, J. and Panza, G. F. (1997). Analytical computation of coupling coefficients in nonpoissonian media, Geophys. J.Int., 129, CODES EUROCODE 8 Basis of Design and Actions on Structures, CEN