Current Development of Seismic Design Code to
Consider the Near-fault Effect in Taiwan
J.-F. Chai 1) W.-I Liao 1) T.-J. Teng 1) C.-H. Loh 1)
1) National Center for Research on Earthquake Engineering, Taipei, Taiwan, R.O.C.
ABSTRACT
The object of this paper is to develop the design methods for structures against the near-fault ground motions. The first method is to consider the near-fault seismic demand at the design level, and hence, the near-fault design response spectrum is developed to intensify the resistance capacity of structures directly. The other alternative method is called the two-level design method. No near-fault effect is considered at the primary force-based design level, but the additional capacity checking level is requested to limit the ultimate capacity of the designed structure to exceed the maximum considered seismic demand caused by the near-fault effect.
INTRODUCTION
In recent years, people have learned that near-fault ground motions have many different characteristics from the far-field ones, and the near-fault ground motions will cause much more damage. In fact, the associated high PGA and the pulse-like velocity waveform of the near-fault ground motion will destroy structures with short and long structural periods, respectively [1~3]. Prior to the 1994 Northridge earthquake, the near source effects were particularly addressed by SEAOC for the UBC97 [4]. In UBC97, the near-source factors are incorporated in Seismic Zone 4, which is intended to recognize the amplified ground motions occurring close to the
fault. Two near-source factors defined for the short period (acceleration control) and long period (velocity control) domains are needed because the effect is substantially greater at longer periods. After the 1999 Chi-Chi (Taiwan) earthquake, the normalized near-fault design response spectrum for sites near the Chelungpu Fault was developed based on the current Taiwan seismic design code and the near-fault attenuation functions for the spectral acceleration demands, which was regressed from the near-fault ground motions observed during the Chi-Chi earthquake [5].
The current design requirement for structures in Taiwan is based on the seismic hazard defined at a uniform 10
percent probability of exceedance in 50 years (return period of 475 years). Therefore, in order to consider the near-fault effect at the same seismic hazard level, the near-fault attenuation law corresponding to the magnitude of an event with expected recurrence rate of 475 years should be considered. However, it is much more difficult to estimate the magnitude of event with an expected recurrence rate of 475 years than to estimate the maximum potential magnitude. Then, to consider the near-fault effect, the so-called maximum considered earthquake (MCE) with 2 percent probability of exceedance in 50 years (return period of 2,500 years) should be taken into account [6].
For an interesting near-fault zone, both the probabilistic analysis based on the seismic hazard analysis at a return period of 2,500 years and the deterministic analysis based on the attenuation law corresponding to the maximum potential magnitude of the fault are implemented to define the seismic demands at the MCE level. The seismic demand should be dominated by the fault effect in the region with smaller distance from the fault where the seismic demand determined by the attenuation law is larger than that determined by the probabilistic analysis. Otherwise, the fault effect can be ignored and the seismic demand should be dominated by other potential sources. Therefore, as shown in Fig. 1, the required spectral acceleration demand at the MCE level can be defined for both near-fault sites and general sites. On the other hand, there is a specific ratio between the seismic demands determined by the seismic hazard analysis at return periods of 2,500 years and 475 years for this seismic zone. Hence, the near-fault spectral acceleration demand atthe MCE
Fig. 1 Near-fault design spectral acceleration demand reduced from the MCE level
level can be reduced to the design level on the basis of the same specific ratio, and then to develop the near-fault design response spectrum for designing structures against the near-fault ground motions.
On the other hand, it has been believed that it is not feasible to design a building structure to remain elastic under intense ground motions. The seismic design has aimed to ensure that (a)the structure should not suffer any structural damage from frequently minor earthquakes, (b)the repaired structure should be usable after an infrequent earthquake of major intensity, and (c)the structure should not collapse (life safety limit state) for the safety of occupants during the largest possible earthquake at the construction site. Therefore, in addition to the development of near-fault design response spectrum, an alternative method called two-level design method is developed. As shown in Fig. 2, no near-fault effect is considered at the primary force-based design level, and the seismic demand determined by the seismic hazard analysis at a return period of 475 years is adopted to develop the design response spectrum for both the general sites and near-fault sites. However, an additional capacity checking level is requested to limit the ultimate capacity of the designed structure to exceed the maximum considered seismic
demand, which is defined by the required spectral acceleration demand at the MCE level. It means that the maximum considered seismic demands should be defined by the attenuation law corresponding to the maximum potential magnitude of the fault for a near-fault site and defined by the seismic demand determined probabilistically at a return period of 2,500 years for general sites, respectively. Therefore, based on the two-level design method, the near-fault effect is reflected indeed at the ultimate capacity checking level even though it is not considered at the primary force- based design level.
In the following sections, the develop- ment of seismic design base shear, the near-fault spectral acceleration demands at both the MCE level and design level, and the ultimate capacity checking requirements which are developed in the current revised seismic design code are introduced briefly.
SEISMIC DESIGN BASE SHEAR
For the current development of seismic design code in Taiwan [7,8], the elastic seismic demand is represented by the design spectral response acceleration SaD corresponding to a uniform seismic
hazard level of 10% probability of exceedance within 50 years (return period of 475 years). Based on the uniform hazard analysis, the mapped design 5% damped spectral response acceleration at short periods (SSD) and at 1 second (S1D) are determined and prepared for each administration unit of village, town or city level. These spectral response acceleration parameters should be modified by site coefficients to include local site effects, and the site adjusted spectral response acceleration at short periods (SDS) and at 1 second (SD1) are expressed as
D v D D S a DS F S S F S S = ; 1= 1 (1)
where site coefficients Fa and Fv are defined in Tables 1 and 2, and they are functions of the soil type and the mapped spectral response acceleration parameters, SSD for Fa and S1D for Fv, respectively.
Based on the soil structures in the upper 30 meters below the ground surface, the site can be classified into three classes by using V -method, s
N -method or s -method as shown in u Table 3. The site class parameters V s
and N are defined as the averaged
shear wave velocity and averaged standard penetration resistance for all
Table 1 Values of site coefficients Fa Values of Fa Site Class SS ≤ 0.5 SS = 0.6 SS = 0.7 SS = 0.8 SS ≥ 0.9 S1 (Hard Site) 1.0 1.0 1.0 1.0 1.0 S2 (Normal site) 1.2 1.2 1.1 1.1 1.0 S3 (Soft Site) 1.4 1.3 1.2 1.2 1.1 Note: SS may be SSD, SSM, or for different cases, and
straight- line interpolation for intermediate values of S
D B S AS N , NASSM,B S is used Table 2 Values of site coefficients Fv
Values of Fv Site Class S1 ≤ 0.3 S1 = 0.35 S1= 0.4 S1= 0.45 S1 ≥ 0.5 S1 (Hard Site) 1.0 1.0 1.0 1.0 1.0 S2 (Normal site) 1.5 1.4 1.3 1.3 1.2 S3 (Soft Site) 1.7 1.6 1.5 1.5 1.4 Note: S1 may be S1D, S1M, or for different cases, and
straight-line interpolation for intermediate values of S
D B VS
N 1, NVS1M,B
S is used Table 3 Site classification
s
V -method N-method su-method
Site Class s V (m/s) N Nch su (kPa) S1 (Hard site) Vs> 360 N> 50 Nch> 50 su> 100 S2 (Normal site) 180 ≤Vs≤ 360 15 ≤ N≤ 50 15 ≤ Nch≤ 50 50 ≤su≤ 100 S3 (Soft site) Vs< 180 N< 15 Nch< 15 su< 50
Note: If the su-method is used and the Nch and su criteria differ,
select the category with the softer soils
soil layers in the top 30m, respectively, and they are determined by
∑
∑
∑
∑
= = = = = = n i i i n i i n i si i n i i s N d d N V d d V 1 1 1 1 ; (2)where Vsi is the shear wave velocity, Ni is the standard penetration resistance not to exceed 100 as directly measured in the field without corrections, and di is the thickness of any layer with
∑
=
n i 1 di =
30m. On the other hand, if the u
s -method is adopted, the averaged
standard penetration resistance N for ch cohesionless soil layers (PI < 20) and averaged undrained shear strength s u for cohesive soil layers (PI > 20) in the top 30m can be determined by
∑
= = m i i i s ch N d d N 1( / )∑
= = m i 1di ds(cohesionless soils)∑
= = k i i ui c u s d d s 1( / )∑
= k i 1di =dc (cohesive soils) (3) where ds and dc are the total thickness of cohesionless and cohesive soil layers in the top 30m (ds + dc = 30m), respectively, and sui is the undrained shear strength not to exceed 250kPa.Based on the site adjusted spectral response acceleration parameters SDS and SD1, the design spectral response acceleration SaD can be developed by
> ≤ < ≤ < ≤ + = 0 0 0 1 0 0 0 0 5 . 2 ; 4 . 0 5 . 2 ; / 2 . 0 ; 2 . 0 ; ) / 3 4 . 0 ( T T S T T T T S T T T S T T T T S S DS D DS DS aD with DS D S S 1 0 = T (4)
where T is the structure period in the unit of second, and the shape of design response spectrum is shown in Fig. 3.
The structure system ductility capacity R for some basic types of seismic-force-resisting system can be found in the seismic design code, and further, the allowable ductility capacity Ra can be defined by 5 . 1 / ) 1 ( 1+ − = R Ra (5)
It implies that only two-third of the ultimate inelastic deformation capacity is permitted to be utilized. Based on the equal displacement principle between elastic and elastic-plastic systems for long period range and equal energy principle for short periods, the structure system seismic reduction factor Fu can be defined by the allowable ductility capacity Ra and structure period T as
≤ − × − − + − ≤ ≤ − ≤ ≤ − × − − + − ≥ = 0 0 0 0 0 0 0 0 0 0 2 . 0 ; 2 . 0 2 . 0 ) 1 1 2 ( 1 2 6 . 0 2 . 0 ; 1 2 6 . 0 ; 4 . 0 6 . 0 ) 1 2 ( 1 2 ; T T T T T R R T T T R T T T T T T R R R T T R F a a a a a a a u (6)
As shown in Eq. (6), the structural period larger than T0 is defined as the long period range with T0 being the corner period of the design response spectrum as defined by Eq. (4). On the other hand, the constant acceleration range is divided into two equal parts, the structural period in the range of 0.2T0 to 0.6T0 is defined as the short period range, and the linear interpolation is defined for the other part (0.6T0 to T0) between short and long period ranges. Furthermore, the linear interpolation is also defined for structural period less than 0.2T0, such that the reduction factor Fu will be equal to one when the structural period becomes zero, because no ductility can be considered for a rigid body.
Finally, the seismic design base shear can be expressed as
; Buildings) (for 4 . 1 F W S I V m u aD y α = Bridges) (for 2 . 1 F W S I V m u aD y α = (7) and > ≤ < + ≤ = 9 . 0 ; 75 . 0 9 . 0 4 . 0 ; 18 . 0 55 . 0 4 . 0 ; u aD u aD u aD u aD u aD u aD m u aD F S F S F S F S F S F S F S (8)
herein, I is the important factor, W is the total gravity load of the structures, αy is the first yield seismic force amplification factor that is dependent on the structure types and design method. The constant 1.4 (for buildings) or 1.2 (for bridges) means the over strength factor between the ultimate and first yield forces, and it is dependent on the redundancy of the structural system. The modified ratio of (SaD/Fu)m is defined to reduce the seismic demand because the higher damping ratio (5% ~ 10%) will be caused by the soil-structure interaction for short period structures. The procedures to determine the seismic design base shear are outlined in Fig. 4.
NEAR-FAULT DESIGN
RESPONSE SPECTRUM
To consider the effect of near-fault ground motions in seismic design, both the probabilistic analysis based on the seismic hazard analysis at a return period of 2,500 years and the deterministic analysis based on the attenuation law corresponding to the maximum potential magnitude of the fault are implemented. Based on the uniform hazard analysis at a return period of 2,500 years, the mapped spectral response acceleration parameters SSM and S1M can be determined for each administration unit near the fault of interest. Furthermore,
the averaged demand S and
can be determined, and they are recognized as the lower limit of seismic demand in the near-fault zone at the MCE level caused by other potential sources. On the other hand, the attenuation relations S
M B
S, S1M,B
S,Att(r) and S1,Att(r) for the median 5% damped spectral acceleration demands at short periods (e.g., 0.3 second period) and at 1 second can be developed on the basis of the maximum potential magnitude of the specified active fault. For example, the Chi-Chi earthquake with a magnitude of ML = 7.3 can be recognized as the maximum potential earthquake of Chelungpu fault. Compared with the lower limit of the spectral response acceleration parameters at the MCE level, the near-fault factors NA(r) and NV(r) can be defined by M B Att V M B S Att S A S r S r N S r S r N , 1 , 1 , , ( ); ( ) 1.5 ( ) 5 . 1 ) ( = = (9)
The factor of 1.5 implies the consideration of 1σ deviation of
uncertainty of fault movement and the component effect (fault-normal). Therefore, the required spectral response acceleration at short periods (S ) and at 1 second (S ) for the near-fault zone at the MCE level can be defined by
M NF S, M NF , 1 M B V M NF M B S A M NF S r N r S S r N r S S , ( )= ( ) , ; 1, ( )= ( ) 1, (10)
and they are functions of the distance from the fault.
In order to determine the reduced factor from the MCE level to design level, the mapped spectral response acceleration parameters SSD and S1D for each administration within the same near-fault zone should be averaged to define the lower limit of seismic demand at the design level, and denoted by S and , respectively. Based on the reduced factors, which are defined by
and , the
required spectral response acceleration at short periods ( D B S, D B S1, D B S S , M B S s S R = / , D NF S1, M B D B S S R1= 1, / 1, NF D S, S ) and at 1 second ( ) for the near-fault zone at the design level can be defined from Eq. (10) as D B V D NF D B S A D NF S r N r S S r N r S S , ( )= ( ) , ; 1, ( )= ( ) 1, (11)
Similar to Eq. (1), the site-adjusted near- fault spectral response acceleration parameters SDS and SD1 can be determined by D B V v D D B S A a DS F N S S F N S S = , ; 1= 1, (12)
It is noted that the associated site coefficients Fa and Fv should be evaluated from Tables 1 and 2 on the basis of the near-fault spectral response
acceleration parameters and
, respectively. Therefore, substituting Eq. (12) into Eq. (4), the
D B S AS N , D B VS N 1,
required near-fault design response spectrum can be developed, and then to determine the seismic design base shear for designing structures. Because the near-fault effect is considered at the design level, it can intensify the resistance capacity of structures against near-fault ground motions directly.
TWO-LEVEL DESIGN METHOD
In addition to the development of near-fault seismic design base shear for designing structures, an alternative two-level design method is developed to consider the near-fault effect. At the primary force-based design level, only the mapped design spectral response acceleration parameters (SSD and S1D) corresponding to a return period of 475 years are adopted to determine the design base shear for both the general sites and near-fault sites. Then, an additional capacity checking level is requested to limit the ultimate capacity of the designed structure to exceed the maximum considered seismic demand defined at the MCE level.
For determining the maximum seismic demand at the checking level for general sites, the site-adjusted spectral response acceleration at short periods (SMS) and at 1 second (SM1) can be defined by the mapped spectral response acceleration parameters SSM and S1M at the MCE level as
M v M M S a MS F S S F S S = ; 1= 1 (13)
herein, the associated site coefficients Fa and Fv should be evaluated from Tables 1 and 2 on the basis of SSM and S1M, respectively. For near-fault sites, the site-adjusted spectral response acceleration at short periods (SMS) and at 1 second (SM1) should be defined by the near-fault spectral response acceleration
parameters, and , at the
MCE level as M B S AS N , NVS1M,B M B V v M M B S A a MS F N S S F N S S = , ; 1= 1, (14) and the associated site coefficients Fa and Fv should be evaluated from Tables 1 and 2 on the basis of the near-fault spectral response acceleration
parameters and , respectively. M B S AS N , M B VS N 1, Then, the required spectral response acceleration SaM at the checking level can be developed as > ≤ < ≤ < ≤ + = M MS M M M M M MS M M MS aM T T S T T T T S T T T S T T T T S S 0 0 0 1 0 0 0 0 5 . 2 ; 4 . 0 5 . 2 ; 2 . 0 ; 2 . 0 ; ) / 3 4 . 0 ( with MS M M S S 1 0 = T (15)
Furthermore, at the ultimate capacity checking level, the ductility demand is allowed to reach its capacity R instead of the allowable ductility capacity Ra as defined for the primary force-based design level. Therefore, the structure system seismic reduction factor FuM at the checking level can be defined by
≤ − × − − + − ≤ ≤ − ≤ ≤ − × − − + − ≥ = M M M M M M M M M M uM T T T T T R R T T T R T T T T T T R R R T T R F 0 0 0 0 0 0 0 0 0 0 2 . 0 ; 2 . 0 2 . 0 ) 1 1 2 ( 1 2 6 . 0 2 . 0 ; 1 2 6 . 0 ; 4 . 0 6 . 0 ) 1 2 ( 1 2 ; (16)
The criteria for the ultimate capacity check is that the allowable lateral capacity Pa should exceed the maximum shear force demand, i.e.,
W I F S P m uM aM a > (17)
herein, the modified ratio (SaM/FuM)m is defined by > ≤ < + ≤ = 9 . 0 ; 75 . 0 9 . 0 4 . 0 ; 18 . 0 55 . 0 4 . 0 ; uM aM uM aM uM aM uM aM uM aM uM aM m uM aM F S F S F S F S F S F S F S (18)
It is noted that the two-level design for bridges at near-fault sites is considered in the current revised code in Taiwan. The seismic demand caused by the near-fault effect at checking level, the estimation of ultimate capacity of a RC bridge pier and the checking requirements are developed in the current revised seismic design code for bridges [8]. On the other hand, because the allowable lateral capacity Pa for buildings can hardly be evaluated, the two-level design method is simplified by defining the seismic design base shear as
α α = W F S I W F S I V m uM aM y m u aD y 1.4 , 4 . 1 max (19)
and the second ultimate capacity checking process can be dropped. It is noted that, based on the current revised seismic design code for buildings, Eq. (19) should be considered for both near-fault sites and general sites.
CONCLUSIONS
Based on the uniform hazard analysis at a return period of 475 years, the mapped design 5% damped spectral response acceleration at short periods and at 1 second are prepared for the specified administration unit. Furthermore, by considering the local site effect, the site-adjusted design spectral response acceleration parameters can be defined through the site coefficients and then to develop the design spectral response acceleration. Together with the system reduction factor and the first yield amplification factor, the seismic design base shear for buildings and bridges can be well defined.
For near-fault sites, based on the seismic hazard analysis at a return period of 2,500 years and the attenuation law corresponding to the maximum potential magnitude of the fault, the required spectral response acceleration parameters for the near-fault zone at the MCE level can be defined. Then, it can be either reduced to the design level to develop the near-fault design response spectrum for designing structures against near-fault ground motions directly, or utilized to define the required spectral response acceleration demand at the checking level for the two-level design method.
The two-level design for bridges at near-fault sites is considered in the current revised code in Taiwan. The seismic demand caused by the near-fault effect at checking level, the estimation of ultimate capacity of a RC bridge pier and the checking requirements are developed in the current revised seismic design code. For buildings, based on the current revised seismic design code, the design base shear defined at the design
level should be limited to be larger than that defined at the MCE level for both the near-fault sites and general sites, and the second ultimate capacity checking process can be dropped.
REFERENCES
1. Hall, J.F., Heaton, T.H., Halling, M.W. and Wald, D.J. (1995). “Near-source ground motion and its effects on flexible buildings,” Earthquake Spectra, Vol. 11, No. 4, pp. 569–605.
2. Somerville, P.G., Smith, N.E., Graves, R.W. and Abrahamson, N.A. (1995). “Accounting for near-fault rupture directivity effects in the development of design ground motions,” Pressure
Vessels and Piping Division, ASME,
PVP319, pp. 67–82.
3. Iwan, W.D. (1997). “Drift spectrum: measure of demand for earthquake ground motions,” J. Struct. Engng, ASCE, Vol. 123, No. 4, pp. 397–404. 4. ICBO (1997). Uniform Building Code,
1997 Edition, Whittier, CA.
5. Chai, J.F., Loh, C.H. and Chen, C.Y. (2000). “Consideration of the near- fault effect on seismic design code for sites near the chelungpu fault,”
Journal of the Chinese Institute of Engineers, Vol. 23, No. 4, pp. 447–454.
6. BSSC (1997). NEHRP Recommended
Provisions for Seismic Regulations for New Buildings and Other Structures, 1997 Edition, Part 1: Provisions and Part 2: Commentary, prepared by the
Building Seismic Safe Council for the Federal Emergency Management Agency (Report Nos. FEMA 302 and 303), Washington, D.C.
7. ABRI (2000). Research on the
Modification of Taiwan Building Seismic Design Code (Draft), MOIS-892040,
Architecture and Building Research Institute, Ministry of Interior Affair, Taiwan (in Chinese).
8. MOTC (2001). The Modification of
Seismic Design Provision and Commentary for Highway Bridges
(Drafts), Ministry of Transportation and Communication, Taiwan (in Chinese).
