SITE RESPONSE ANALYSIS FOR SEISMIC DESIGN OF A 48-STOREY TOWER BUILDING IN JAKARTA

Transcription

1 SITE RESPONSE ANALYSIS FOR SEISMIC DESIGN OF A 48-STOREY TOWER BUILDING IN JAKARTA I Wayan Sengara 1, Davy Sukamta 2 and Putu Sumiartha 3 1 Head, Geotechnical Engineering Laboratory, Engineering Center for Industry, Institut Teknologi Bandung, Jl. Ganesha 10, Bandung-40132, Indonesia, iws@geotech.pauir.itb.ac.id 2 Director, DavySukamta Consulting, Pondok Pinang Center Blok A/18 Jl. Ciputat Raya, Jakarta 12310, Jakarta, Indonesia, sukamta@rad.net.id 3 Research Assistant, Geotechnical Engineering Laboratory, Engineering Center for Industry, Institut Teknologi Bandung, Jl. Ganesha 10, Bandung-40132, Indonesia ABSTRACT: For accurate estimate on the seismic design criteria, site response analysis is performed for the proposed tower building. The site response analysis is based on probabilistic seismic hazard analysis (PSHA) previously conducted and refers to the uniform hazard spectra for 10% probability of exeedance in 50 years (475 years return period). The analysis takes into consideration various parameters in the form of seismic wave propagation from reference base-rock to the ground surface. The uniform hazard spectra was used as a basis in scaling various input motions from scenario earthquake of subduction representing far field earthquakes, and shallow crustal representing near field earthquakes. The input motions are scaled to short (T=0.2 seconds) and long periods (T=2 seconds) to cover earthquakes with potential different frequency contents considering the tower with relatively long natural period. Seismic wave propagation analysis using non-linear earthquake response analysis computer program was conducted considering dynamic soil properties at the site. Two seismic downhole test (SDT) to a depth of 50 meters was conducted. The SDT provided shear wave velocity (Vs) profile of the site. Results of the site response analysis provide response spectra with recommendation on design spectra at ground surface for structural design of the tower. The recommended design spectra are compared to the Indonesian Building Code and UBC97 for the range of the period of interest of the proposed tower. In addition, the recommended design spectra is also compared to that of IBC2006 based on uniform hazard spectra resulted from PSHA for 2% probability of exeedance in 50 years (2475 years return period). It is indicated that the recommended design spectra fall below that of Indonesian Building Code and slightly lower compared to the design spectral value of UBC1997 and IBC2006 for Soft site-class. 1. INTRODUCTION Plaza Indonesia II is a huge development, consisting of two towers, 42-storey and 48-storey, with 5-level basement. One of the tower, the Keraton Grand Hyatt Residence, will be the tallest building in Indonesia at 225 m when completed in Figure 1 shows the bird-eye of the project. The project is currently under construction and slated for completion in September Seismic design criteria for high-rise building can be simply referred from applicable building codes based on its level of seismic hazard or peak acceleration at referenced baserock (PBA). The other parameter that is needed is site classification. With this PBA and site-class, design response spectra representing seismic load to the building can be defined. PBA for city of Jakarta according to the current Indonesian building codes (SNI ) is 0.15g. This level of PBA is specified for 10% probability of exeedance (PE) in 50 years or equivalent to 475 years return period (RP) earthquake. Site-classification analysis of proposed 48 storey Keraton Grand-Hyatt Residence site in Jakarta suggests that the site class fall to S E (Soft-soil) category. This is due to the fact that the site consists of more than 3 m soil layer having undrained shear strength (s u ) less than 25 kpa, Plasticity Index (PI) > 20 and water content (w n ) > 40. The average shear wave velocity of the top 30 m, however indicate that the site fall to S D (medium) site class. Therefore, site-specific response analysis is conducted to recommend its design spectra. Back to Table of Contents 399

2 Seismic design criteria proposed 48 storey Keraton Grand-Hyatt Residence site in Jakarta was needed for structural and foundation detailed engineering design. To provide representative seismic design criteria for the building, level of PBA is first verified through probabilistic seismic hazard analysis (PSHA) for city of Jakarta. PSHA for Jakarta was carried out for 10% and 2% PE in 50 years (475 and 2475 years RP) as described in Sengara et al. (2007). Further site-specific response analysis (SSRA) was conducted and presented in this paper to more accurately represent seismic design criteria in term of design response spectra. Figure 1 Bird eye of the 48-storey Keraton Grand Hyat Residence. 2. SRUCTURAL SYSTEM AND DESIGN CRITERIA Lateral system for this building is a dual system of RC corewall and ductile open frame. The seismic response modification factor R is 8.5. This put a high demand on the ductility of all structural elements. The corewall is designed according to the ACI 318 method which is adopted in the Indonesian national standard code for concrete (SNI ). The open frame is designed as special moment resisting frame and serves as second line of defense in resisting seismic forces. As such, the frame alone must be able to resist 25 percent of the total design base shear, which translates into heavier rebar in the beams at the lower portion of the building. One of the requirement from the client for this building is that the seismic design must consider 2% PE in 50 years (2475 years RP). Following this criteria, a special study on the seismic design by performing PSHA for Jakarta was conducted and site-specific response analysis for seismic design spectra of the tower building was then developed. 3. DYNAMIC SOIL PROPERTIES OF SITE Shear wave velocity data from field measurement for Plaza Indonesia site was obtained from two set of seismic down-hole test to a depth of 30 meter and 50 meter from the ground surface, respectively. Shear wave velocity data below this depth is approximated by using available formula that is correlated to N-SPT data and other soil properties. Reference baserock (S B with Vs Back to Table of Contents 400

3 = 760 m/s) is assumed at depth of 300m. Figure 2 shows shear wave velocity data as a function of depths. This profile was used as input to the SSRA. 0 Shear Wave Velocity, Vs (m/s) to 50 meter depth, Measurement Shear Wave Velocity Depth (meter) to 300 meter depth, Estimated Shear Wave Velocity Figure 2 Shear wave velocity profile for site response analysis for Plaza Indonesia site. There is some concern on basin effect (deep geologic structure) of Jakarta region to the amplification of ground motion. Based on Somerville et al. (2004) there is recommendation to multiply an amplification factor of 1.65 for periods of 4 seconds and longer. The assumption of deep geologic structure of Jakarta needs to be supported by geologic cross section through Jakarta region. This basin effect of Jakarta region is still debatable, since no comprehensive study has been done. Seismic hazard study by Seed et al. (1996) consider that basin effect would not be significant for two reasons: (1) increases amplitude of waves near edges of basin occurs over a more local scale than 30km distance between the site and edge of basin; (2) directionality of wave propagation of seismic sources is away from the otherwise potentially entrapping basin edge. 4. SITE-SPECIFIC RESPONSE ANALYSIS Site-specific response analysis (SSRA) is performed by considering appropriate input motions and dynamic soil properties of the site. There have been difficulties faced in SSRA since there are no strong-motion earthquake records available for Jakarta. Therefore, synthetic input motion generation has been conducted in this process. There are two steps in developing synthetic input motion. First step is to develop baserock target spectra. Baserock target spectra of particular dominant event could be developed based on the magnitude and distance parameters using appropriate attenuation functions. The second step is to use available strong-motion records having similar characteristics to the seismic source that its response spectra match the target spectra. In this process, spectral matching techniques proposed by Abrahamson that is built in the EZ-FRISK Computer Program (Risk Engineering, 2004) was conducted. Considering the relatively high natural period of the 48-storey building, then to recommend spectral acceleration for long period, the dominant earthquake events rock target spectra is scaled to T = 2 second of the uniform hazard spectra (UHS) resulted from PSHA of Jakarta (Sengara et al., 2007). Since spectral acceleration for whole range of periods are also needed, then to recommend spectral acceleration for shorter periods, rock target spectra of short period motions is also scaled to T = 0.2 second of that of the UHS. Result of PSHA indicated that PBA for city of Back to Table of Contents 401

4 Jakarta is 0.2g for 10% probability of exeedance (PE) in 50 years (475 years RP) and approximately 0.28g for 2% PE in 50 years (2475 years RP). UHS for 10% and 2% PE in 50 years is given in Figure 3. Deaggregation analysis provide controlling earthquakes for short and long periods motion. For short oscilatory period, it is indicated that the earthquake is domintaed from the shallow crustal earthquake source, whereas for long period the dominant earthquake is originated from subduction. Target spectra for these representative earthquakes were developed from the appropriate attenuation functions. There are two target spectra developed, one representing subduction earthquake and one representing shallow crustal earthquake. The target spectra are then scaled to the corresponding spectra values from UHS. This process is shown in Figure 4 and Figure 5, for scaling to T=0.2sec and T=2 sec, respectively. From scaled target response spectra, it is indicated that Megathrust event is dominating earthquake hazard for period greater than T = 2 second, as shown in Figure 5, whereas it is also indicated that shallow crustal event is dominating earthquake hazard for low period. Mean UHS for 475 and 2475-Year Return Period Year Return Period 2475-Year Return Period PGA at Base Rock (g) Period (s) Figure 3 Uniform hazard spectra for 10% and 2% PE in 50 years resulted from PSHA by Sengara et al., Benioff Megathrust Shallow Crustal 0.4 Spectral Acceleration (g) Figure 4 Target spectra of dominant events scaled to T = 0.2 second. Back to Table of Contents 402

5 Benioff Megathrust Shallow Crustal 0.6 Spectral Acceleration (g) Figure 5 Target spectra of dominant events scaled to T = 2.0 second. Five input motions, that are scaled benioff, scaled megathrust and two scaled shallow crustals fault have been generated from existing strong-motion records of appropriate earthquake mechanisms. These input motions were used as input in the SSRA 4.1 Base-rock Condition Geological survey and geotechnical investigation data was collected to know the subsurface conditions of Jakarta. There is limited information on geological information at the site. Geotechnical South-North profile was generated from this limited subsurface information. Based on the interpolation on base-rock South-North profile for Jakarta (Geological Research and Development Center, 1997), the depth of base-rock at Plaza Indonesia is estimated about 300 meter, and this is the basis of subsurface geometry used as an input in the SSRA. The soil profile is assumed to be horizontal (soil column) with no basin effect. 4.2 Seismic Wave Propagation Analysis Seismic wave propagation analysis was conducted using NERA computer program (Bardet and Tobita, 2001). This program applies time-domain approach of non-linear soil properties where its shear modulus decreases as a function of increasing strain, while damping increases as a function of increasing strain. The wave propagation analysis using NERA computer program indicated that the peak acceleration for return period 475 years is amplified from 0.2g at the base-rock to values that vary from 0.23g to 0.31g at ground surface. The PGA is influenced by the shear wave velocity profile, frequency content and duration of input motions that have resulted in variations in the PGA. 4.3 Ground Surface Response Spectra Result of the wave propagation analysis for 5% damped spectral response at ground surface is as shown in Figure 6. The wave propagation analysis was initially conducted using input motions consisting of that scaled to long period (T=2.0 second), considering the natural periods of the tower building which is relatively high (T s =5-6 seconds). Since peak accelerations for whole range of periods are required in the dynamic structural analysis, then wave propagation analysis using input motion scaled to T=0.2 second was also conducted and response spectra resulted from this analysis representing short period motions is presented. Entire response spectra resulted from both short and long-period input motions are plotted and an envelope of design response spectra is recommended as also shown in Figure 6. The recommended design response spectra was not averaged from series of response spectra as resulted from wave propagation analysis, since the Back to Table of Contents 403

6 input motions are scenario earthquakes, each of them with its different characteristics that would provide different response. The recommended design spectra resulted from this study is then compared to that of SNI and UBC97 for the range of the period of interest of the proposed tower (5-6 seconds) as shown in Figure 7 and Figure 8, for PBA of 0.15g and 0.19g, respectively. It is indicated that response spectra resulted from this analysis fall within the range of S D and S E site class of both SNI and UBC97. Spectra Acceleration (g) Megathust (1) Megathrust (2) Shallow Crustal (5) Shallow Crustal (6) Beniof (3) Beniof (4) Megathrust (1) 2 Sec Megathrust (2) 2 Sec Shallow C (5) 2 Sec Shallow C (6) 2 Sec Beniof (3) 2 Sec Beniof (4) 2 Sec Figure 6 Recommended design spectra and ground surface response spectra based on time-domain analysis for the range of the period of interest of the proposed tower, 10% PE in 50 years SNI g-Soft SNI g-Soft 0.25 SNI g-Medium 0.25 SNI g-Medium UBC g-Soft UBC g-Soft Spectral Acceleration (g) UBC g-Medium Recommended, This Study 2005 Spectral Acceleration (g) UBC g-Medium Recommended, this study, Figure 7 Comparison design spectra of 475 years return period resulted from this study, SNI and UBC97 for the range of the period of interest of the proposed tower (5-6 seconds), PBA=0.15g Figure 8 Comparison design spectra of 475 years return period resulted from this study, SNI and UBC97 for the range of the period of interest of the proposed tower (5-6 seconds), PBA=0.19g It is essential to consider the S MS, maximum considered earthquake spectral response acceleration both for short-period (0.2 second), and S D1, long-period (1-second) adjusted for site-class effect as formulated in International Building Code 2006 (IBC2006). Therefore, we calculated the S MS and S M1 based on UHS for short and long-period spectral values, resulted from our PSHA for 2% PE in Back to Table of Contents 404

7 50 years (2475 years RP). Furthermore, we also calculated the design earthquake spectral response acceleration for both periods and plot its design spectra. In this case, according to IBC2006, the design earthquake spectral response acceleration is S DS =(2/3)S MS and S D1 = (2/3)S M1. Both the maximum considered and design earthquake response spectra is plotted and compared to the recommended design spectra resulted from this study, as shown in Figure 9. Spectra Acceleration (g) SNI-2002, 0.20g, Soft Soil SNI-2002, 0.20g, M edium Soil UBC-1997, 0.20g, Soft Soil UBC-1997, 0.20g, M edium Soil This Study, PBA 0.19g (475 yr RP PSHA) IBC-2006 M ax Values, 2475 yr RP PSHA, Soft Soil IBC-2006 M ax Values, 2475 yr RP PSHA, M edium Soil IBC-2006 Design Values, 2475 yr RP PSHA, Soft Soil IBC-2006 Design Values, 2475 yr RP PSHA, M edium Soil Period (second) Figure 9 Comparisons of recommended design spectra with that of some seismic building codes. It can be seen from Figure 9 that the recommended design spectra resulted from the SSRA is significantly lower compared to that of Soft site-class of the SNI (with reference to the same PBA= g), and slightly lower compared to both Soft site-class of the UBC1997 and IBC2006. It is considered that the recommended design spectra for the tower building resulted from the SSRA fall within the range of the maximum Indonesian Building Codes and design spectral values of International Building Codes with amplified between medium to soft site-class. 5. CONCLUSIONS Site-specific response analysis (SSRA) for developing design response spectra for Keraton Grand- Hyatt Residence has been conducted. The SSRA has provided the structural engineer with the recommended design spectra in comparison with that obtained from the building codes. The SSRA is based on the PSHA that has previously been conducted. Essential input parameters in the analysis are the shear wave velocity profile of the site and seismic input motions. The seismic input motions used in the analysis are based on the uniform hazard spectra as well as deaggregation showing the dominant or controlling earthquake to the hazard as resulted from PSHA. The seismic input motions were generated with scaling of the target spectra to T=0.2 sec and T=2 sec. The scaling to T=2 sec is to anticipate long period motions from far distance subduction earthquakes. Two points seismic downhole test up to depth of 50m have provided representative shear wave velocity profile of the site. Results of SSRA provide ground surface response spectra for earthquake scenarios representing both subduction and shallow crustals earthquakes. It is indicated that the recommended design spectra fall within the S D (medium) and S E (soft) site class of either SNI , UBC97, or IBC2006. It has been shown that the recommended design spectra resulted from the SSRA is significantly lower compared to that of Soft site-class of the SNI , and slightly lower compared to the either Soft site-class of the UBC1997 and IBC2006. The design spectra fall within spectral values of the medium and soft site-class Back to Table of Contents 405

8 International International Conference Conference Earthquake on Earthquake Engineering Engineering and Disaster and Mitigation, Disaster Mitigation Jakarta, April , 2008 SSRA is essential to identify and verify the critical earthquake events and to recommend spectral acceleration associated with dynamic properties of the site as well as with natural period of the building. It is recommended to apply UHS with PBA=0.2g corresponding to 475 years return period, with maximum spectral values corresponding to 2475 years RP, for highrise buildings in Jakarta and use input motions scaled to particular spectral values considering natural period of the building. 6. ACKNOWLEDGEMENTS The authors thankful to Dr. M. Irsyam, Mr. E. Kertapati, Mr. Hendarto, and Mr. N. Wijayanto for support on seismic hazard input to the analysis. Seismic downhole test and data analysis supported by Dr. G. Handayani is appreciated. The authors also thank to owner of Keraton Grand-Hyatt Residence management and PT Davy Sukamta Consultant that have facilitated so that this analysis could be presented. 7. REFERENCES Bardet, J.P. and Tobita, T. (2001). NERA, A Computer Program for Nonlinear Earthquake Site Response Analyses pf layered Soil Deposits, University of Southern California. International Building Codes (IBC) (2006). Code and Commentary, International Code Council, Volume 2. Kertapati, E. (1999). Probabilistic estimates of the seismic ground motion hazard in Indonesia, Proceeding of National Conference on Earthquake Engineering, Bandung. Kramer (1996). Geotechnical Earthquake Engineering, Prentice-Hall Inc. Upper Suddle River, New Jersey. McGuire, R.K. (1990). Computation Seismic Hazard, Risk Engineering Inc., Golden, Colorado, USA Risk Engineering (2004). EZ-FRISK (Software for in-depth Seismic Hazard Analysis) User s Manual, Risk Engineering Inc., Golden, Colorado, USA. Seed, R.B., Abrahamson, N. and Rudianto, S. (1996). Seismic Design Criteria Jakarta Tower, PT. Indocitra Grahabawana. Seismological Society of America, U.S. Geological Survey (1997). Seimological research letter, National Seismic Hazard Mapping Project, Volume 68, Number 1. Sengara, IW., Irsyam, M., Merati, W. and Aswandi, (1999). Seismic microzonation and site response analysis for Jakarta, National Conference in Earthquake Engineering, Earthquake Engineering Research Center, Bandung Institute of Technology. Sengara, IW., Hendarto, Kertapati, E., Sukamta, D. and Sumiartha, P. (2007). 3-Dimensional source zones probabilistic seismic hazard analysis for Jakarta and site specific response analysis for Plaza Indonesia II building, Proceeding of Indonesian Society of Structural Engineers Seminar, Jakarta. SNI (2002). Indonesian Seismic Building Codes, Indonesian Department of Public Work. SOFOCO (1995), Second Phase Soil Investigation Report for Plaza Indonesia II Project. SOFOCO (2002), Soil Investigation Report for Entertainment Center. Somerville, P., Collins, N., Graves, R. and Pitarka, A. (2004). An engineering ground motion model for basin generated surface waves, Proceeding, 13th World Conference on Earathquake Engineering, Vancoever, British Columbia, Paper N Uniform Building Code (UBC) (1997). Volume 2, structural engineering design provisions, International Conference of Building Officials. Back to Table of Contents 406