Proposed site-specific response spectra for Surabaya-Madura Bridge.
Surabaya-Madura (Suramadu) Bridge is planned to be built on the northern part of East Java Province in Indonesia, spanning the Madura Channel, connecting Surabaya with Madura Island (Figure 1). In order to ensure the safety of bridge structure considering the seismological, geotechnical, and structural aspects, a site-specific seismic hazard analysis is required.
Past earthquakes have shown that damages induced in bridges can take many forms, depending on the ground motion, site conditions, structural configuration, and specific details of the bridge. Recent earthquakes, particularly the 1989 Loma Prieta and the 1994 Northridge earthquakes in California, the 1995 Hyogo-Ken Nanbu earthquake in Japan, the 1999 Jiji earthquake in Taiwan, and the 1999 Kocaeli earthquake in Turkey, have caused collapse of, or severe damage to, a considerable number of major bridges [1, 2]. Therefore, a thorough examination of the structure and its environment should be conducted in seismic design of bridge to determine the criteria selection including the seismic use group and importance factor, intended service/damage level, and performance objectives.
This paper presents a site-specific seismic hazard analysis for Suramadu Bridge. Two levels of ground motion hazard that represent 150 yr and 3300 yr return period of ground motion were analyzed in this paper using total probability theorem. Artificial time histories for site location were also developed using spectral matching analysis. Finally, ground response analysis was performed using 1-D shear wave propagation analysis to develop the design response spectra.
Overview of Seismotectonic Conditions
The seismic sources influencing Suramadu Bridge can be divided into subduction zone, transform zone, and diffuse seismicity zone. Subduction zone events occurred when an oceanic plate is being subducted under an island arc or continent. Thrust fault earthquakes along the interface, normal faulting events along the outer arc high and in the trench, and reverse and strike-slip faulting events in the upper plate fall within this classification as long as they are close to a convergent margin of the subduction zone.
The strike-slip events along the clearly defined faults in the frontal arc area such as Sumatera, Sorong and Mamberamo Fault are classified as transform fault zone events such as Palu-Koro and Matano Faults, Lengguru and Tarera Aiduna Faults.
Diffuse seismic zone includes all earthquakes that occur in areas where the seismicity is not associated with a single fault or fault type. Most of these diffuse seismic zones are found in back arc areas of collision zones, like Flores back-arc faulting behind the eastern end of Sunda arc and western end of Banda Arc. The seismic sources around the project location are shown in Figure 2.
Data Collection and Processing
Seismic hazard assessment requires a history as complete as possible of earthquakes in or near the region of interest. The earthquake catalog used in this study is based on compilation of several catalogs from local and international institutions, such as International Seismological Center (ISC), National U.S. Geological Survey (USGS), Bureau of Meteorology and Geophysics (BMG), Indonesia and several individual catalogs.
The combined catalog covers an area from 90[degrees]E to 125[degrees]E longitude and 10[degrees]S to 10[degrees]N latitude. From the catalog dating from 13 May 1897 to 31 December 2004, there are a total of 14119 earthquake events. The minimum magnitude is 5.0 and the maximum focal depth is 200 km.
Typical characteristic of earthquake catalogs are as follows: (1) the magnitude scales used in the catalogs are not uniform. This is due to the fact that the earthquake events were recorded using more than one type of instrument; (2) the earthquake catalogs contained both the main shock events and the accessory shock events (foreshock and aftershock events). Therefore, the data are not valid to be used when the temporal occurrence of earthquakes are analyzed using Poisson model; and (3) the small events are usually incomplete in earthquake catalogs. This is due to the limited sensitivity and coverage of the earth by seismographic networks.
[FIGURE 2 OMITTED]
The first problem is solved by choosing a consistent magnitude for seismic hazard assessment (SHA), and then the other magnitude scales are converted to the chosen magnitude scale by using empirical correlation [4,5,6]. In this research, a moment magnitude, [M.sub.w], is chosen as a measurement to quantify the size of earthquake. Other types of magnitude in the catalogs were then converted to [M.sub.w] by using empirical correlations.
The second problem is solved by declustering the catalog using time and distance windows criteria . The criteria proposed by Gardner and Knopoff  were used in this research because this criteria has considered the time and distance windows for magnitude above 8.0. The algorithm eliminated 6028 records from the catalog, and the result is shown in Figure 3.
The third problem is solved by performing catalog completeness analysis. In this study, historical earthquake data occurred between 1900 and 2004 have been analyzed for completeness using Stepp method . Based on the catalog completeness analysis the earthquakes within interval 5.0 [less than or equal to] Mw [less than or equal to] 6.0 and 6.0 [less than or equal to] Mw [less than or equal to] 7.0 are completely recorded only for the last 27 and 31 years period of observation, respectively. While the earthquake catalog for the magnitude of 7.0 [less than or equal to] Mw [less than or equal to] 8.0 and Mw>8.0 are considered complete for the whole time length of the catalog.
There are three potential seismogenic sources addressed in this study, (1) Sunda Arc subduction fault (2) reverse thrusting fault and (3) shallow crustal fault. According to these seismotectonic features, seismic sources zone around site region are divided into several seismic zones. The location of those zones can be seen in Figure 4.
Based on the dip angle and location of the focal depth, subduction zones was divided into interplate (megathrust) and intraslab (benioff) zones. In this study, only seismic source zones within a radius of 500 km from the site were considered. Those outside this radius were assumed to have insignificant influence for the peak ground acceleration.
[FIGURE 3 OMITTED]
Seismic Hazard Parameters
b-value and Annual Rate
Temporal distribution of earthquakes is commonly assumed to follow frequency-magnitude relationship proposed by Gutenberg-Richter (G-R) . Several researchers have proposed alternative methods to obtain more reliable frequency magnitude relationship [10, 11, 12, 13, 14]. In this study, the seismicity rates of each source zones were calculated using the method proposed by Weichert , Kijko and Sellevoll [13, 14]. These methods considered the relationship between earthquake data and interval time when the catalogs are homogeneous (completeness time). The seismicity parameters for each source zone used in this analysis are shown in Figure 5.
Maximum Magnitude and Slip Rate
The maximum magnitude and slip rate are estimated based on the evaluation of available data and tectonic of the region. The procedure identifies a reasonable maximum magnitude for the given potential seismic source, and its most reasonable slip rate in the current tectonic environment. The values for maximum magnitude and slip rate of earthquake sources influencing the proposed site are listed in Table 1.
[FIGURE 4 OMITTED]
One of the critical factors in seismic analysis is to obtain or to select appropriate attenuation relationship. This formula, also known as ground motion relation, is a simple mathematical model that relates a ground motion parameter (i.e. spectral acceleration, velocity and displacement) to earthquake source parameter (i.e. magnitude, source to site distance, mechanism) and local site condition .
There has been a number of attenuation relations derived in the last two decades since the record of ground motions becomes more available. In general, they are categorized according to tectonic environment (i.e. subduction zone and shallow crustal earthquakes) and site conditions. In this study, the attenuation relationships for subduction zone at rock sites developed by Youngs  and that for shallow crustal developed by Boore, Joyner, Fumal  and Sadigh  for extension tectonic region were selected to estimate peak ground acceleration and response spectra at bedrock. The selection was based on previous study by Irsyam  for the development of the Indonesia seismic zone map, which indicated that these attenuation functions have a low variability compared to others.
Seismic Hazard Analysis
Epistemic uncertainty is included in the probabilistic seismic hazard assessment (PSHA) by explicitly including alternative hypotheses and models. The logic tree allows a formal characterization of uncertainty in the analysis by explicitly including alternative interpretations, models, and parameters that are weighted in the analysis according to their probability of being correct. Logic trees [20, 21, 22] are used in this study in order to allow uncertainty in selection of models for attenuation, recurrence rate, and maximum magnitude to be considered. The logic tree model is illustrated in Figure 6.
The PSHA was performed using the total probability theorem developed by McGuire . This theorem is based on the probability concept developed by Cornell , which assumed the earthquake magnitude M and the hypocenter distance R as a continuous independent random variable.
The analysis was conducted by using software EZFRISK  that has capability of 3-Dimension modeling for earthquake source. Using 3-Dimension model, it is expected that the results of hazard analysis became more accurate. The sequence of seismic hazard calculations for 3-D fault sources model can then be represented by a modified form of previous equation as follows:
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII.] (1)
Seismic Hazard Exposure
The result of PGA for difference return period for Suramadu Bridge location is shown in Figure 7. Seismic hazard for Suramadu Bridge location including the relative contributions from each source for PGA is shown in Figure 8.
For physical interpretation of the results from PSHA and to take certain engineering decisions, de-aggregation analysis was performed in this study. The results of de-aggregation were used to identify the mean magnitude and distance of earthquakes that control the ground motions at a particular response spectral frequency. Figure 9 shows an example of the result of de-aggregation hazard for Suramadu Bridge.
[FIGURE 8 OMITTED]
Acceleration time-histories are required in the analysis of shear wave propagation in soil deposits. Selection of time-histories appropriate for specific geological and seismological conditions plays an important role for obtaining accurate results. Since there are no strong motions detected near the site, the ground motion was selected based on the controlling earthquake with specific magnitude and distance for each source and various return periods from the de-aggregation result. For each source mechanism, response spectra were generated using appropriate attenuation functions. These response spectra were scaled to the T=2.0 second value as shown in Figure 10. Scaled time histories were then generated using spectral matching analysis. An example of time histories for subduction event is shown in Figure 11.
Local Site Effect
The local soil effect analysis in this study was carried out using 1-D shear wave propagation theory to obtain the estimate of ground motion parameter, i.e. surface spectral acceleration and time histories. The Geotechnical software, NERA , was employed for this procedure.
In this study, the soil dynamic property for the site location was based on the result of shear wave velocity measurement which can be found in the report of Seismic Ground Motion Parameters at Suramadu Bridge Site Indonesia conducted by Institute of Geophysics, China Earthquake Administration . The shear wave velocity data is available until 90 m depth. For this analysis, it was assumed that the bedrock at the site location is at 250m depth, since from the 90 m deep bored-hole results there was no bedrock found yet. The shear wave velocity profile for the depth below 90 m to 250 m was estimated. Shear wave velocity profile used in this analysis is shown in Table 2
The result of local site effect analysis using 1-D shear wave propagation analysis from bedrock to ground surface is shown in Figure 12. The computed surface spectral accelerations were obtained using scaled bedrock time histories. Based on these spectral accelerations, the average and average plus one standard deviation response spectral acceleration were calculated.
Development of Site-Specific Design Response Spectra
The recommended design spectra for this study are presented in Figure 13 for return period of 150 years and 3300 years. The recommended response spectra were determined based on mean plus one standard deviation. In this project, the elastic surface response spectra for 50% and 3% probability of exceedance (PE) in design time period of 100 years (corresponding to return period of approximately 150 and 3,300 years, respectively) were used to develop the expected earthquake (EE) and the maximum considered earthquake (MCE) spectra, respectively. This MCE response spectrum was taken as the design spectra for designing an inelastic structural response, whereas the elastic design spectra were obtained using EE response spectra.
[FIGURE 9 OMITTED]
[FIGURE 10 OMITTED]
[FIGURE 11 OMITTED]
[FIGURE 12 OMITTED]
The site-specific seismic hazard assessment for Suramadu Bridge has been described in this paper. The analysis used was a probabilistic approach for return periods of 150 yr and 3300 yr of earthquake hazard. The controlling earthquakes were derived by employing deaggregation process. Finally, the design response spectra at ground surface for 5 percent damping were developed. The design spectra obtained from this study could be used for further design procedures.
Note: Discussion is expected before November, 1st 2008, and will be published in the "Civil Engineering Dimension" volume 11, number 1, March 2009.
Received 25 January 2008; revised 26 January 2008; accepted 5 June 2008.
[1.] Moehle, J.P. and Eberhard, M. O., Earthquake Damage to Bridges, in Bridge Engineering Handbook, Chen, W.F. and Duan, L., Eds., CRC Press, Boca Raton, FL, chap. 34, 2000.
[2.] Yashinsky, M., Earthquake Damage to Structures, in Structural Engineering Handbook CRCnetBase 2000, Chen, W.F. and Duan, L., Eds., CRC Press, Boca Raton, FL, chap. 29a. 2000.
[3.] Firmansjah, J. and Irsyam, M., Development of Seismic Hazard Map for Indonesia, Prosiding Konferensi Nasional Rekayasa Kegempaan di Indonesia, Institut Teknologi Bandung, Indonesia, 1999, pp, I-29-I-41.
[4.] Geller, R.J., Scaling Relations for Earthquake Source Parameters and Magnitudes, Bulletin of the Seismological Society of America, Vol. 66, 1976, pp, 1501-1523.
[5.] Rong, Y, Evaluation of Earthquake Potential in China, University of California Los Angeles, PhD. Thesis, 2002.
[6.] McGuire, R.K., Seismic Hazard and Risk Analysis, Oakland, California, EERI, 2004.
[7.] Gardner, J.K., and Knopoff, L., Is the Sequence of Earthquakes in Southern California, with Aftershocks removed, Poissonian?, Bulletin of the Seismological Society of America, Vol. 64, No. 5, 1974, pp, 1363-1367.
[8.] Stepp, J.C., Analysis of the Completeness of the Earthquake Hazard Sample in the Puget Sound Area, NOAA Technical Report, ERL 267-ESL 30, May 1973, Boulder, CO, pp, 16-28. 1973.
[9.] Gutenberg, B., Richter, C. F., Seismicity of the Earth, Princeton University Press, Princeton, New Jersey, 1954.
[10.] Dong, W. M., Bao, A. B., and Shah, H. C, Use of Maximum Entropy Principle in Earthquake Recurrence Relationship, Bulletin of the Seimological Society of America, Vol. 74, No. 2, 1984, pp, 725-737.
[11.] Bender, B., Incorporating Acceleration Variability into Seismic Hazard Analysis, Bulletin of the Seismological Society of America, Vol. 74, No. 4, 1984, pp, 1451-1462.
[12.] Weichert, D.H., Estimation of the Earthquake Recurrence Parameters for Unequal Observation Periods for Different Magnitudes, Bulletin of the Seismological Society of America, Vol. 70, No.4, 1980.
[13.] Kijko, A. and Sellevol, M.A., Estimation of Earthquake Hazard Parameters from Incomplete Data Files Part I. Utilization of Extreme and Complete Catalog with Different Threshold Magnitudes, Bulletin of the Seismological Society of America, Vol. 79, No. 3, 1989, pp, 645-654.
[14.] Kijko, A., Sellevol, M.A., Estimation of Earthquake Hazard Parameters from Incomplete Data Files Part II. Incorporation of Magnitude Heterogeneity, Bulletin of the Seismological Society of America, Vol. 82, No. 1, 1992.
[15.] Campbell, K.W., Prediction of strong ground Motion using the Hybrid Empirical Method and its use in the Development of Ground-Motion (Attenuation) Relations in Eastern North America, Bulletin of the Seismological Society of America, Vol. 93, 2003, pp, 1012-1033.
[16.] Youngs, R.R., Chiou, S.J., Silva, W.J., Humphrey, J.R. Strong Ground Motion Attenuation Relationships for Subduction Zone Earthquake, Seismological Research Letters, Vol. 68, No. 1, 1997.
[17.] Boore, D.M., Joyner, W.B., and Fumal, T.E., Equation for Estimating Horizontal Response Spectra and Peak Acceleration from Western North America Earthquakes: A Summary of Recent Work, Seismological Research Letters, Vol. 68, No. 1, 1997.
[18.] Sadigh, K., et al. Attenuation Relationships for Shallow Crustal Earthquakes Based on California Strong Motion Data, Seismological Research Letters, Vol. 68, No. 1, 1997.
[19.] Irsyam, M. Development of Earthquake Microzonation and Site Specific Response Spectra to Obtain More Accurate Seismic Base Shear Coefficient. Final Report for University Research for Graduate Education (URGE) Project. Department of Civil Engineering, Institute of Technology Bandung, 2001.
[20.] Power, M. S., Coppersmith, K. J., Youngs, R. R., Schwartz, D. P., Swan, R. H., Seismic Exposure Analysis for the WNP-2 and WNP-1/4 Site: Appendix 2.5K to Amendment No. 18, Final Safety Analysis Report for WNP-2, Woodward-Clyde Consultants, 1981.
[21.] Kulkarni, R.B., Youngs, R.R., Coppersmith, K.J. Assessment of Confidence Intervals for Results of Seismic Hazard Analysis, Proceeding, 8th World Conference on Earthquake Engineering, San Francisco, 1984, pp, 90-98.
[22.] Coppersmith, K. J., Youngs, R. R. Capturing Uncertainty in Probabilistic Seismic Hazard Assessment with Intraplate Tectonic Environments, Proceedings, 3rd U. S. National Conference on Earthquake Engineering, Charleston, South Carolina, Vol. 1, 1986, pp, 301-312.
[23.] McGuire, R.K., FORTRAN Computer Program for Seismic Risk Analysis, US Geological Survey Open File Report: 67-76, 1976.
[24.] Cornell, C.A. Engineering Seismic Risk Analysis, Bulletin of the Seismological Society of America, Vol. 58, 1968, pp, 1583-1606.
[25.] Risk Engineering Inc. EZ-FRISK version 7.20 Software for Earthquake Ground Motion Estimation. User's Manual, 2006.
[26.] Bardet, J.P., Tobita, T., A Computer Program for Non-linear Earthquake Response Analyses of Layered Soil Deposits, Department of Civil Engineering, University of Southern California, 2001.
[27.] Institute of Geophysics, China Earthquake Administration, Seismic Ground Motion Parameters at Suramadu Bridge Site, Indonesia, November 2005.
Masyhur Irsyam (1,2), Donny T. Dangkua (1), Hendriyawan (1), R. Bambang Boediono (1), Dyah Kusumastuti (1), Engkon K. Kertapati (3)
(1) Faculty of Civil Engineering and Enviromental, Institut Teknologi Bandung, Indonesia
(2) Corresponding author, Owner Expert at Suramadu Bridge Project
(3) Geological Research and Development Center, Indonesia
Table 1. Estimated maximum magnitude and slip rate for earthquake sources influencing the site location No. Seismic Source Source No. Source [M.sub.max] Slip Rate Type (mm/yr) A.1 Subduction Faults-Megathrust 1 West Center Z011 R 8.5 77 Java Domain 2 East Java Z021 R 8.5 77 Domain 3 Bali Domain Z031 N 8.5 77 4 Sumba Fragment Z041 R 8.5 77 Domain A.2 Subduction Faults-Benioff (50<Depth[less tthan or equal to]100) 1 West Center Z072 N 8.5 77 Java Domain 2 East Java Domain Z082 N 8.5 77 3 Bali Domain Z092 R 8.5 77 4 Sumba Fragment Z102 N 8.5 77 Domain B. Reverse Thrusting Faults 1 Kendeng BackArc Z133 R 7.5 1.0 Thrust 2 Bali-Lombok Z143 R 7.5 2.0 Backarc Thrust 3 Flores Backarc Z153 R 7.5 2.0 Thrust C. Shallow Crustal Faults 1 Bumiayu Fault Z174 R 6.8 2.0 2 Semarang Fault Z184 SS 6.8 0.2 3 Lasem Fault Z194 R 6.8 0.2 4 Melange Fault Z204 R 6.8 0.2 5 Kemirian Fault Z214 R 6.8 0.1 6 Sepanjang Fault Z224 R 6.8 0.3 7 Kangean Fault Z234 R 6.8 0.1 8 Doang Fault Z244 N 6.8 0.1 Table 2. Shear wave velocity at Suramadu Bridge  Layer Description Depth (m) Thickness VS (m/s) Order of Soil (m) 1 Clay 0.0-4.0 4.0 130 2 Sandy Clay 4.0-9.6 5.6 180 3 Silty Clay 9.6-11.5 1.9 200 4 Clayey Silt 11.5-16.00 4.5 360 5 Sandy Silt 16.0-23.1 7.1 330 6 Clayey Silt 23.1-34.6 11.5 345 7 Silty Clay 34.6-48.0 13.4 375 8 Clayey Silt 48.0-54.0 6.0 350 9 Clayey Silt 54.0-58.0 4.0 320 10 Clayey Silt 58.0-61.0 3.0 310 11 Clayey Silt 61.0-65.0 4.0 350 12 Clayey Silt 65.0-68.0 3.0 290 13 Clayey Silt 68.0-78.0 10.0 310 14 Clayey Silt 78.0-84.0 6.0 300 15 Clayey Silt 84.0-90.0 6.0 390