Proposed Changes to the Geotechnical Earthquake Engineering Provisions of the Bangladesh National Building Code

of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information Report

Legal forms


Views: 25 | Pages: 8

Extension: PDF | Download: 0

The Bangladesh National Building Code (BNBC) has not been updated since its inception in 1993. Earthquake design provisions are important, since Bangladesh is located in a seismically active region not far from the boundary of the Indian Plate and
  Geotechnical Engineering Journal of the SEAGS & AGSSEA Vol. 43 No.2 June 2012 ISSN 0046-5828 Proposed Changes to the Geotechnical Earthquake Engineering Provisions of the Bangladesh National Building Code Tahmeed M. Al-Hussaini 1 , Tahsin R. Hossain 1  and M. Nayeem Al-Noman 1 1  Department of Civil Engineering, Bangladesh University of Engineering & Technology, Dhaka, Bangladesh    E-mail  :   ABSTRACT:  The Bangladesh National Building Code (BNBC) has not been updated since its inception in 1993. Earthquake design  provisions are important, since Bangladesh is located in a seismically active region not far from the boundary of the Indian Plate and Eurasian Plate. Many advances in earthquake engineering research have taken place over the last two decades. In 2010, a project has been taken up for upgradation of the building code to incorporate current knowledge and developments. Significant changes have been proposed in the seismic design provisions of the updated version of the building code. This paper describes the salient features of the proposed changes to geotechnical earthquake engineering provisions affecting computation of the seismic loads. Major changes have taken place with regard to the seismic zoning map, soil classification system, site-dependent response spectrum, and in defining seismic design category. 1. INTRODUCTION Bangladesh lies in a moderately seismic-prone region and historical evidence points to major earthquakes within or close to the country. The collision of the Indian plate, moving northeast, with the Eurasian plate is the cause of frequent earthquakes in the region comprising Bangladesh, North-East India, Nepal and Myanmar. Historically Bangladesh has been affected by five earthquakes of large magnitude (M) greater than 7.0 (Richter) within the last 150 years. Of them, the 1897 Great Indian earthquake (M=8.1 re-estimated by Ambraseys and Bilham, 2003) had an epicentral distance of only 230 km north of Dhaka, the capital. That earthquake caused extensive damages to masonry buildings in many  parts of Bangladesh. The 1885 Bengal earthquake (M=7.0) and 1918 Srimongal earthquake (M=7.6) had their epicentres within Bangladesh, they caused considerable damage locally. The 1762 Chittagong earthquake, also a local earthquake with estimated magnitude M=7.5, although not well documented, is reported to have caused major landmass changes in the coastline. In recent years, small to moderate earthquakes are regularly occurring (Al-Hussaini, 2005) with epicenters in neighbouring India and Burma (some within the country) which are being felt in many parts of the country, particularly in the southeast Chittagong region. Some of these earthquakes have caused some damage. Bangladesh National Building Code (BNBC-1993) was first  prepared in 1993, under directives issued by the Ministry of Works. On behalf of the House Building Research Institute (HBRI), the consulting firm Development Design Consultants (DDC) engaged Bangladesh University of Engineering and Technology (BUET) to take the leading role in developing the first building code of the country. BNBC-1993 (DDC, 1993) contains seismic design  provisions which were based on the knowledge of the time when it had been developed. Since then, major progress has been made in worldwide research in earthquake engineering and different building codes have been upgraded accordingly. Significant amount of research have also been carried out in Bangladesh. In recent years, HBRI has taken up an initiative to upgrade the existing building code (BNBC-1993) to incorporate the advances in knowledge and experience over almost two decades. A group of Consultants,  principally consisting of faculty members from BUET along with some external consultants, have been entrusted by HBRI to prepare an updated building code. Several editorial board members from within and outside BUET, have been engaged as reviewers of the updated code. The first and second authors have been responsible for updating the seismic design provisions of BNBC. Draft seismic design provisions of the building code have been submitted in December 2010. This paper presents briefly the basic philosophy for estimation of seismic load as per existing seismic design provisions of BNBC-1993. This is followed by some research findings and finally description of proposed changes in seismic design provisions that have been proposed in the updated code particularly in defining the ground motion. 2. BNBC-1993 2.1 Seismic Zoning Map The 1993 Bangladesh national building code has adopted a seismic zoning map (Ali and Choudhury, 1994) consisting of three seismic zones, with zone coefficients of 0.25 (Zone 3 in the north and north-east), 0.15 (Zone 2 in the middle, north-west and south-east) and 0.075 (Zone 1 in the south west), as shown in Fig.1. This zoning map is based on peak ground accelerations estimated by Hattori (1979) for a return period of 200 years. Figure 1 Seismic Zoning Map in BNBC-1993    Geotechnical Engineering Journal of the SEAGS & AGSSEA Vol. 43 No.2 June 2012 ISSN 0046-5828 The zone coefficient (Z) represents the peak ground acceleration (PGA) in units of g (acceleration due to gravity). 2.2 Seismic Base Shear The seismic base shear may be calculated as:  R ZICW V     (1) where, Z=zone coefficient, I=importance factor W= seismic weight of building R=reduction factor representing energy dissipation in structure, C=1.25ST 1.5  S= soil factor (Table 1), T=fundamental period of building (sec) Once base shear is calculated, seismic shear forces in other stories can be calculated. Site effect is characterized by soil factor S and response spectrum shape described later. The soil is classified into four classes S1, S2, S3 and S4 as described in Table 1. Soil type S4 produces largest seismic forces. Table 1 Soil Classification in BNBC-1993 2.3 Response Spectrum For buildings requiring dynamic analysis, response spectrum analysis or time history analysis may be necessary. Response spectrum analysis employs characteristic response spectrum depending on soil type. Figure 2 shows normalized response spectrum for soil class S1, S2 and S3. The ordinate represents spectral acceleration (S a ) normalized with respect to gZ. For soil class S4, site specific spectrum need to be determined. Figure 2 Response Spectrum for Soil Types S1, S2, S3 in BNBC-1993 3. RESEARCH STUDIES 3.1 Seismic Hazard Assessment Studies based on earthquake catalogues and historical data have revealed the significant seismic risk befalling this country. Hattori ’s (1979) seismic hazard assessment results for a return period of 200 years have been used in establishing the seismic zoning map of Bangladesh used in BNBC-1993. Ansary and Sharfuddin (2002) formed a homogeneous earthquake catalogue and performed  probabilistic seismic hazard assessment using attenuation law by Duggal (1989). Based on results for 200 year return period, they  proposed modified seismic zoning map with same zone coefficients  but with larger areas for Zone 3 and Zone 2, in other words increased seismic hazard in some areas. Their seismic hazard estimation methodology is based on the assumption that the PGA at a site maintains a recurrence frequency relationship similar to the Gutenberg-Richter magnitude-frequency relationship. The authors are of the opinion that this assumption is not generally justified for two reasons. The PGA at a site depends not only on the earthquake magnitude but also on the epicentral distance from the site, in addition different earthquake sources are most likely to possess different frequency characteristics. More recently, standard probabilistic seismic hazard assessment method using multiple source zones has been applied for determining the PGA values for various return periods ranging from 475 years to 2475 years. Al-Hussaini and Al-Noman (2010)  presented preliminary results from this analysis. The earthquake catalogue has been formed using various sources and including historical earthquakes. Main shock data is extracted from the earthquake catalogue after removing repeated data and by removing aftershocks using magnitude dependent time-window and space-window. Earthquakes are grouped into five magnitude classes (M = 4, 5, 6, 7, 8), each class represents a magnitude range M-0.5 to M+0.4. The period of completeness for different magnitude classes is determined by studying the cumulative earthquake occurrence during the entire period. Seismic source zones have been delineated considering Bolt’s (1987) source zones in addition to fault locations and cluster of major (M>5) earthquake epicentres affecting Bangladesh. A total of seven seismic source zones have been designated. Four seismic source zone models with some changes in source boundaries have been tried to take into account uncertainties in source boundaries. For each seismic source zone, the following Gutenberg-Richter relationship (Gutenberg and Richter, 1956) is assumed to be valid: bM a    log  (2) where ν is the cumulative annual frequency for earthquakes with  magnitude M and a, b are regression constants. The period of completeness is used to obtain the cumulative annual earthquake occurrence rate (ν) for each magnitude class  for each source zone. In the probabilistic seismic hazard assessment (PSHA) method, the ground motion at a site is estimated for a specified probability of  being exceeded in a given time period. The computational scheme involves: delineation of seismic source zones, characterization of the source zones, selection of appropriate attenuation laws with source-site distance, and a predictive model of seismic hazard. A key element is the variability of ground motion at a given site for a given earthquake scenario. The largest ground motions are controlled by the number of standard deviations above the median ground motion. Given the magnitude, distance and number of standard deviations for the ground motion, the ground motion is computed for each earthquake scenario using an attenuation law which represents the decay of ground motion with distance from source. The computer program CRISIS (UNAM 1999) is used to perform  probabilistic seismic hazard assessment (PSHA) studies for Bangladesh. In the absence of reliable attenuation laws for Bangladesh, recent well-established attenuation relations (one    S  p  e  c   t  r  a   l   A  c  c  e   l  e  r  a   t   i  o  n   /  g   Z  Geotechnical Engineering Journal of the SEAGS & AGSSEA Vol. 43 No.2 June 2012 ISSN 0046-5828 standard deviation above median) developed by various researchers for different regions (Western USA, Eastern USA, Iran, Europe and India) of the world have been used in the study. In addition a new attenuation relationship for Bangladesh srcinally developed by Islam et al. (2010), and later corrected for site effect, has been used. This local attenuation law is not based on ground motion measurements but is based on intensity based isoseismals of historical and recent earthquakes, and therefore employs intensity-PGA (peak ground acceleration) relationship as well. This law is found to be close to the attenuation law for Western USA developed  by Abrahamson and Silv a (1997). Also in Bolt’s (198 7) report, he stated that the attenuation in Bangladesh is expected to be similar to that in the Western USA. Results of this probabilistic hazard analysis for the attenuation law of Abrahamson and Silva (1997) are  presented here. Typically building codes specify the design earthquake to have a 10% probability of exceeding in 50 years (design life of building). This is equivalent to a return period of 475 years for earthquakes which are assumed to follow the Poissonian distribution. However, recently some codes are considering larger return periods to account for large earthquakes with long recurrence periods. The International Building code (ICA, 2006) considers the Maximum Credible Earthquake (MCE) to correspond to a return period of 2475 years which is equivalent to 2% probability of exceedance in 50 years. The Indian Code (BIS, 2005) is also using MCE motion in its seismic zoning map. Figure 3 shows results of PSHA studies for a return period of 2475 years for seismic source zone model no.2 and using the attenuation law of Abrahamson and Silva (1997). It is observed that the maximum PGA value is in the north-east (Sylhet/Mymensingh) amounting to 0.39g, while the PGA value in the port city of Chittagong in the south-east is 0.29g, the PGA value in the capital city of Dhaka in the centre is around 0.19g. These PGA values are for rock or firm soil and does not include local site effect. Figure 3 Predicted PGA values (in cm/sec 2 ) for return period of 2475 years for source zone model-2 and attenuation law of Abrahamson and Silva, 1997. 3.2 Site Effect A recent study by Al-Hussaini et al. (2007) addresses local site effects in Dhaka city which has a Z value of 0.15. Numerical simulations of ground motion amplification are studied for several sites in Dhaka city for a variety of ground motions using a modified version ‘SHAKE91’ (Idriss and Sun, 1992) of the srcinal one-dimensional wave propagation program ‘SHAKE’ (Schnabel et al., 1972). The spectral acceleration (S a ) values obtained for different sites for 5% damping ratio are normalized with respect to the gZ (PGA rock  ) value and then compared with the design response spectra curves (Figure 2) of existing building code. Figure 4 presents mean normalized spectral acceleration for 10 sites consisting of S3 type soil. Comparing with BNBC 1993 provision for Soil S3, it is observed that the normalized spectral acceleration values of the mean curves, in general, significantly exceed the corresponding values of the BNBC curve in the period range of 0.2 to 0.5 sec. These curves suggest that the peak Sa/gZ value should be increased to around 3.5 instead of existing 2.5 value for the early period range for S3 type soils. 024680 1 2 3 Period (Sec)    S   p   e   c   t   r   a   l   A   c   c   e   l   e   r   a   t   i   o   n   /   g   Z BHSGoranKeller MorhRampura(Banasree)SabujbaghUttaraWest RampuraBasundharaNiketonEskatonSoil S3, BNBC  Figure 4 Mean value of normalized spectral acceleration for 10 sites with S3 type soil and comparison with BNBC-1993 0240123 Period (Sec)    S  p  e  c   t  r  a   l   A  c  c  e   l  e  r  a   t   i  o  n   /  g   Z  AshkonaBasabo (central)Lalbagh BaddaNikunjaTikatuliSoil S2, BNBC  Figure 5 Mean value of normalized spectral acceleration for 6 sites with S2 type soil and comparison with BNBC-1993 Longitude (degree)    L  a   t   i   t  u   d  e   (   d  e  g  r  e  e   ) 384 234 133 234 83 234 133 184 184 284  Geotechnical Engineering Journal of the SEAGS & AGSSEA Vol. 43 No.2 June 2012 ISSN 0046-5828 Figure 5 presents mean normalized spectral acceleration for 6 sites of Dhaka city consisting of S2 type soils. Comparing with BNBC 1993 provisions for Soil S2, these curves suggest that the maximum Sa/gZ value should be increased to around 2.75 to 3 instead of existing 2.5 value. 4. PROPOSED CHANGES IN UPDATED BNBC This section deals with the proposed major changes in the seismic design provisions of the updated Bangladesh National Building Code which define the design ground motion and seismic shear forces in the building, as described by Al-Hussaini and Hossain (2010). 4.1 Seismic Zoning Map The concept of Maximum Credible Earthquake (MCE) with a return  period of 2475 years has been introduced in the updated building code. The concept of MCE has been adopted in the latest versions of International Building Code and Indian Building Code. Figure 5  presents the proposed seismic zoning map for Bangladesh based on PGA values for a return period of 2475 years. The country is divided into four seismic zones with zone coefficient Z equal to 0.12 (Zone 1), 0.2 (Zone 2), 0.28 (Zone 3) and 0.36 (Zone 4). The zone coefficient represents the PGA value on rock or very stiff soil site. Figure 5 Proposed seismic zoning map for Bangladesh based on a return period of 2475 years The design basis earthquake (DBE) ground motion is selected at a ground shaking level that is 2/3 of the maximum considered earthquake (MCE) ground motion. Comparison of the PGA values  between BNBC-1993 and Proposed BNBC is presented in Table 2 for some important cities and towns of Bangladesh. Due to introduction of four zones, the change of PGA between zones is more gradual than before. It is noted that there is almost no change for Sylhet, Mymensingh, Rajshahi, Khulna. However for Chittagong, Tangail there is significant change, the PGA increases from 0.15 to 0.19. Also for towns such as Pabna and Madaripur, the PGA increases from 0.075 to 0.13. For Dhaka, the PGA value is slightly reduced. For Bogra which is located very close to zone  boundary in 1993 zoning map, the PGA is reduced significantly. Table 2 Comparison of PGA Values in BNBC-1993 and Proposed BNBC for selected towns City/Town Proposed BNBC BNBC 1993 MCE DBE Dhaka, Dinajpur 0.2 0.13 0.15 Chittagong, Tangail 0.28 0.19 0.15 Bogra 0.28 0.19 0.25 Sylhet, Mymensingh, Kurigram, Kishoreganj 0.36 0.24 0.25 Khulna, Rajshahi 0.12 0.08 0.075  Natore, Pabna, Madaripur 0.2 0.13 0.075 4.2 Soil Classification Site will be classified as type SA, SB, SC, SD, SE, S 1  and S 2  adopting a similar system as in Euro-Code. Classification will be done in accordance with Table 3 based on the soil properties of upper 30 meters of the site profile. Table 3 Site Classification based on properties of top soil    S   i   t  e   C   l  a  s  s Description of soil profile up to 30 meters depth Average Soil Properties in top 30 meters    S   h  e  a  r  w  a  v  e  v  e   l  o  c   i   t  y    (  m   /  s   )   S   t  a  n   d  a  r   d   P  e  n  e   t  r  a   t   i  o  n   V  a   l  u  e ,    (   b   l  o  w  s   /   3   0  c  m   )   U  n   d  r  a   i  n  e   d  s   h  e  a  r  s   t  r  e  n  g   t   h ,    (   k   P  a   ) SA Rock or other rock-like geological formation, including 5 m of weaker material on top. > 800 -- -- SB Very dense sand, gravel, or very stiff clay 360  –   800 > 50 > 250 SC Dense or medium dense sand, gravel or stiff clay 180  –   360 15 - 50 70 - 250 SD Loose-to-medium dense cohesionless soil or  predominantly soft-to-medium stiff cohesive soil. < 180 < 15 < 70 SE Surface alluvium layer with V s  values of type SC or SD with thickness of 5 to 20 m, underlain by stiffer material with V s  > 800 m/s. -- -- -- S 1  Deposits containing a layer at least 10 m thick, of soft clays/silts with high  plasticity index (> 40) and high water content < 100 (indicative) -- 10 - 20 S 2  Liquefiable soils, or sensitive clays, or any other soil profile not included in SA to SE or S 1  -- -- -- Longitude (degree)    L  a   t   i   t  u   d  e   (   d  e  g  r  e  e   )  Geotechnical Engineering Journal of the SEAGS & AGSSEA Vol. 43 No.2 June 2012 ISSN 0046-5828 Let n  be number of soil layers in upper 30 m; d  i  , V   si  , N  i  represent thickness, shear wave velocity and Field (uncorrected) Standard Penetration Test (N) value respectively of layer i . Let also k   be number of cohesive soil layers in upper 30 m and d  ci ,  S  ui  represent thickness and undrained shear strength respectively of cohesive layer i . Then, average soil properties will be determined as given in the following equations:   =   =1     =1   (3)    =   =1     =1   (4)    =   =1     =1   (5) The site classification will preferably be done using average Vs, otherwise average N may be used. 4.3 Design Response Spectrum The earthquake ground motion for which the building has to be designed is represented by the design response spectrum. Both static and dynamic analysis methods are based on this response spectrum. This spectrum represents the spectral acceleration for which the  building has to be designed as a function of the building period, taking into account the ground motion intensity. The spectrum is  based on elastic analysis but in order to account for energy dissipation due to inelastic deformation and benefits of structural redundancy, the spectral accelerations are reduced by the response reduction factor R. For important structures, the spectral accelerations are increased by the importance factor I (Table 4). Table 4 Structure Importance Factor    I  m  p  o  r   t  a  n  c  e   C   l  a  s  s Type of Buildings    I  m  p  o  r   t  a  n  c  e   f  a  c   t  o  r   I I Ordinary buildings not belonging to other categories. 1.0 II Buildings whose seismic resistance is of importance in view of the consequences associated with a collapse, e.g. schools, assembly halls, important government buildings etc. 1.25 III Buildings whose integrity during earthquakes is of vital importance for civil protection, e.g. hospitals, fire stations, power plants, telephone exchange, police stations, designated emergency shelters, etc. 1.5 IV Monumental structures with cultural heritage. 1.5  NOTES: 1.   For buildings housing hazardous materials, higher importance factor may be established considering consequences in the event of serious damage or collapse. 2.   The designer may choose values of importance factor greater than those listed in this table. The spectral acceleration for the design basis earthquake (DBE) is given by the following equation:  sa  C  R ZI S  32   (6) where, S a  = Design spectral acceleration (in units of g)   = Coefficient used to provide lower bound (2/3*ZI*  ) for S a . Euro code recommended value for   is 0.2 Z = Seismic zone coefficient (Figure 5) I = Structure importance factor (Table 4) R= Response reduction factor which depends on the type of structural system. The ratio I/R cannot be greater than one. Values of R for different structural systems have been listed by Al-Hussaini and Hossain (2010) C s = Normalized acceleration response spectrum, which is a function of structure (building) period and soil type (site class) as defined by Eqs. 7a-d    )7(015.21  aT T  for  T T S C   B B s            )7(5.2  bT T T  for S C  C  B s        )7(5.2  cT T T  for  T T S C   DC C  s            )7(sec45.2 2  d T T  for  T T T S C   D DC  s          where, S = Soil factor (Table 5) T = Structure (building) period calculated using structural dynamics  procedures or using empirical equation η = Damping correction factor as a function of damping with a reference value of η=1 for 5% viscous damping.  The normalized acceleration response spectrum C s  depends on S and values of T B , T C  and T D , (Figure 6) which are all functions (Table 5) of the site class. The function C s  is defined in accordance with  provisions of Eurocode (ECS, 2004). Constant C s  value between  periods T B  and T C  represents peak spectral acceleration.          C       s Period T T B T C T D  Figure 6 Typical Shape of the Elastic Response Spectrum Coefficient C s   Table 5 Site dependent soil factor and other parameters defining elastic response spectrum Soil type  S T   B (s) T  C   (s) T   D  (s) SA 1.0 0.15 0.40 2.0 SB 1.2 0.15 0.50 2.0 SC 1.15 0.20 0.60 2.0 SD 1.35 0.20 0.80 2.0 SE 1.4 0.15 0.50 2.0
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!