Encyclopedia of Earthquake Engineering

Living Edition
| Editors: Michael Beer, Ioannis A. Kougioumtzoglou, Edoardo Patelli, Ivan Siu-Kui Au

Structural Design Codes of Australia and New Zealand: Seismic Actions

  • George P. KouretzisEmail author
  • Mark J. Masia
  • Clive Allen
Living reference work entry
DOI: https://doi.org/10.1007/978-3-642-36197-5_120-1

Synonyms

Introduction

The design of structures to resist earthquake effects in Australia and New Zealand follows the pertinent Australian Standard AS 1170.4:2007 and New Zealand Standard NZS 1170.5:2004 provisions. Both Standards are based on the common Australian/New Zealand Standard AS/NZS 1170:2002 on structural design actions; and although they share elements such as the site subsoil classification system, they incorporate certain provisions such as the near-fault factor in NZS 1170.5:2004 to account for the different seismotectonic regime of the two countries. Indeed, Australia is an area of generally low seismicity, with the most catastrophic recent event being the 1989 Newcastle earthquake of magnitude M = 5.6, which resulted in 13 casualties and significant damages in the wider Newcastle area. On the other hand, New Zealand is located on the boundary between the Indo-Australian and Pacific Plate and suffers from frequent strong earthquakes. In fact, New Zealand is one of the first countries to account for seismic actions in a building standard; as early as 1935, and following the 1931 Hawke’s Bay earthquake that claimed 256 lives, the NZS 95 provided seismic loads to be considered in design (McRae et al. 2011). Despite constant advances and amendments in the seismic standards over the years, the recent 22 February 2011 Christchurch earthquake of magnitude M = 6.3 resulted in 185 casualties, numerous injuries, and extensive damages in the central business district of Christchurch and its eastern suburbs.

The purpose of this entry is to provide an outline of the seismic design actions currently considered in the abovementioned Standards and a short discussion on the rationale behind their adaptation, in the light of other widely used modern seismic codes provisions, such as the Eurocode 8 EN-1998-1 and the American ASCE/SEI 7-10. It covers the determination of elastic and inelastic design response spectra to be used together with static and dynamic analysis methods, as well as scaling of strong motion recordings according to NZS 1170.5:2004 to perform time history analyses. Emphasis is put on the determination of the input rather than on the provisions about the performance of equivalent static, modal, and time-domain analyses, which can be found elsewhere, including the Standards themselves.

Elastic Response Spectra

The elastic response spectra for horizontal loading, C(T), can be described by the following generic equation, compatible with both Standards:
$$ C(T)={C}_h(T)\cdot \left(Z\cdot R\right)\cdot N\left(T,D\right) $$
(1)
where C h (T) is the spectral shape factor, ZR is the product of the hazard factor Z times the factor R that depends on the annual probability of exceedance of the design earthquake and the design limit state (ultimate or serviceability) under consideration, and N(T, D) is the near-fault factor, applicable only to the NZS 1170.5:2004. The parameters comprising the elastic response spectra equation are presented in the following paragraphs.

Spectral Shape Factor for Different Subsoil Classes

The spectral shape factors C h (T) to be used for equivalent static analyses and for dynamic (modal and time history) analyses are plotted separately in Fig. 1, while the relevant functions are provided in Table 1 for the different site subsoil classes. These spectral shape factors correspond to 5 % structural damping, and a correction factor to convert them to other damping values is not provided as, e.g., in EN-1998-1. However, such a provision would be of value only for special structures such as tanks containing liquids, which are not within the scope of the specific Australian and New Zealand Standards.
Fig. 1

Spectral shape factor (5 % damping) plots for (a) AS 1170.4:2007/equivalent static analyses, (b) AS 1170.4:2007/dynamic analyses, (c) NZS 1170.5:2004/equivalent static analyses, (d) NZS 1170.5:2004/dynamic analyses

Table 1

Spectral shape factor (5 % damping) for different site subsoil classes

Site subsoil class

Equivalent static method

Dynamic analyses methods

Structure period, T (sec)

AS 1170.4:2007

NZS 1170.5:2004

Structure period, T (sec)

AS 1170.4:2007

NZS 1170.5:2004

A

0 < T ≤ 0.1

2.35

1.89

0 < T ≤ 0.1

0.8 + 15.5 T

1.0 + 1.35(T/0.1)

0.1 < T < 0.3

0.704/T ≤ 2.35

0.1 < T < 0.3

0.704/T ≤ 2.35

2.35

0.3 ≤ T < 0.4

0.3 ≤ T ≤ 1.5

1.60(0.5/T)0.75

0.4 ≤ T ≤ 1.5

1.60(0.5/T)0.75

1.5 < T ≤ 3.0

1.056/T 2

1.05/T

1.5 < T ≤ 3.0

1.056/T 2

1.05/T

3 < T

3.15/T 2

3 < T

3.15/T 2

B

0 < T ≤ 0.1

2.94

1.89

0 < T ≤ 0.1

1.0 + 19.4 T

1.0 + 1.35(T/0.1)

0.1 < T < 0.3

0.88/T ≤ 2.94

0.1 < T < 0.3

0.88/T ≤ 2.94

2.35

0.3 ≤ T < 0.4

0.4 ≤ T ≤ 1.5

1.60(0.5/T)0.75

0.3 ≤ T ≤ 1.5

1.60(0.5/T)0.75

1.5 < T ≤ 3.0

1.32/T 2

1.05/T

1.5 < T ≤ 3.0

1.32/T 2

1.05/T

3 < T

3.15/T 2

3 < T

3.15/T 2

C

0 < T ≤ 0.1

3.68

2.36

0 < T ≤ 0.1

1.3 + 23.8 T

1.33 + 1.60(T/0.1)

0.1 < T < 0.3

1.25/T ≤ 3.68

0.1 < T < 0.3

1.25/T ≤ 3.68

2.93

0.3 ≤ T < 0.4

0.4 ≤ T ≤ 1.5

2.0(0.5/T)0.75

0.3 ≤ T ≤ 1.5

2.0(0.5/T)0.75

1.5 < T ≤ 3.0

1.874/T 2

1.32/T

1.5 < T ≤ 3.0

1.874/T 2

1.32/T

3 < T

3.96/T 2

3 < T

3.96/T 2

D

0 < T ≤ 0.1

3.68

3.0

0 < T ≤ 0.1

1.1 + 25.8 T

1.12 + 1.88(T/0.1)

0.1 < T < 0.3

1.98/T ≤ 3.68

0.1 < T < 0.3

1.98/T ≤ 3.68

3.0

0.3 ≤ T < 0.56

0.3 ≤ T < 0.56

0.56 ≤ T ≤ 1.5

2.4(0.75/T)0.75

0.56 ≤ T ≤ 1.5

2.4(0.75/T)0.75

1.5 < T ≤ 3.0

2.97/T 2

2.14/T

1.5 < T ≤ 3.0

2.97/T 2

2.14/T

3 < T

6.42/T 2

3 < T

6.42/T 2

E

0 < T ≤ 0.1

3.68

3.0

0 < T ≤ 0.1

1.1 + 25.8 T

1.12 + 1.88(T/0.1)

0.1 < T < 0.3

3.08/T ≤ 3.68

0.1 < T < 0.3

3.08/T ≤ 3.68

3.0

0.3 ≤ T < 1.0

0.3 ≤ T < 1.0

1.0 ≤ T ≤ 1.5

3.0/T 0.75

1.0 ≤ T ≤ 1.5

3.0/T 0.75

1.5 < T ≤ 3.0

4.62/T 2

3.32/T

1.5 < T ≤ 3.0

4.62/T 2

3.32/T

3 < T

9.96/T 2

3 < T

9.96/T 2

As mentioned in the Introduction, Australian and New Zealand Standards share a common site classification scheme, which is preferably based on the predominant site period (or low amplitude natural period) of the site T s , estimated as:
$$ {T}_s=4{H}_s/{V}_s $$
(2)
where H s is the depth to the seismic bedrock of the site and V s is the shear wave velocity of the overlying soil layer. In the absence of sufficient data to estimate the predominant site period, the site can be classified to one of the five subsoil classes (A to E) using (in the order of most preferred to least preferred): borehole log data together with in situ and laboratory tests, seismological methods including strong motion recordings (Nakamura 1989; Bouckovalas et al. 2002), qualitative borehole log descriptions, and geological information for the site. For layered sites, the site period can be estimated via an averaging method described in AS 1170.4:2007 and NZS 1170.5:2004.
The five different subsoil classes are defined in the Standards as follows (AS 1170.4:2007, NZS 1170.5:2004):
  • Class A: Strong rock. Strong to extremely strong rock, with:
    1. (a)

      Unconfined compressive strength greater than 50 MPa, and

       
    2. (b)

      An average shear wave velocity over the top 30 m V s,30 > 1500 m/s, and

       
    3. (c)

      Not underlain by materials having a compressive strength less than 18 MPa or a shear wave velocity less than 600 m/s

       
  • Class B: Rock. Rock, with:
    1. (a)

      Unconfined compressive strength between 1 MPa and 50 MPa, and

       
    2. (b)

      An average shear wave velocity over the top 30 m V s,30 > 360 m/s, and

       
    3. (c)

      Not underlain by materials having a compressive strength less than 0.8 MPa or a shear wave velocity less than 300 m/s

       

    A surface layer of no more than 3 m depth of highly weathered or completely weathered rock or soil material may be present in a class B site. It should be mentioned here that, based on the average shear wave velocity over the top 30 m criterion, other seismic codes such as EN-1998-1 or ASCE/SEI 7-10 would classify a site with V s,30 > 360 m/s to dense/stiff soil deposits, rather than rock (class B according to EN-1998-1, which corresponds to 360 < V s,30 < 800 m/s, and class C according to ASCE/SEI 7-10, which corresponds to 366 < V s,30 < 762 m/s). This suggests that class B in AS 1170.4:2007 and NZS 1170.5:2004 is rather broad, covering a range of subsoil conditions that may not exhibit similar behavior during an earthquake rich in high- to medium-frequency content (Bouckovalas and kouretzis 2001).

  • Class C: Shallow soil sites. Sites that are not classified as class A, class B, or class E:
    1. (a)

      The predominant site period estimated as above is T s ≤ 0.6 s, or

       
    2. (b)
      Soil depth does not exceed the maximum values listed in Table 2.
      Table 2

      Maximum depth limits for site subsoil class C (After AS 1170.4:2007, NZS 1170.5:2004)

      Soil type and description

      Representative undrained shear strength, S u (KPa)

      Representative SPT N-values

      Maximum depth of soil (m)

      Cohesive soils

      Very soft

      <12.5

      0

      Soft

      12.5–25

      20

      Firm

      25–50

      25

      Stiff

      50–100

      40

      Very stiff to hard

      100–200

      60

      Cohesionless soils

      Very loose

      <6

      0

      Loose dry

      6–10

      40

      Medium dense

      10–30

      45

      Dense

      30–50

      55

      Very dense

      >50

      60

      Gravels

      >50

      100

       
  • Class D: Deep or soft soil sites. Sites that are not classified as class A, class B, or class E; and:
    1. (a)

      The predominant site period estimated as above is T s > 0.6 s, or

       
    2. (b)

      Soil depth exceeds the maximum values listed in Table 2, or

       
    3. (c)

      Sites that are underlain by less than 10 m of cohesive soil with undrained shear strength S u < 12.5 KPa or cohesionless soil soils with SPT values N < 6.

       
  • Class E: Very soft soil sites:
    1. (a)

      More than 10 m of very soft cohesive soil with undrained shear strength S u < 12.5 KPa, or

       
    2. (b)

      More than 10 m of very loose cohesionless soils with SPT values N < 6, or

       
    3. (c)

      More than 10 m depth of soft-loose soils with shear wave velocity values V s < 150 m/s, or

       
    4. (d)

      More than 10 m combined depth of soils with properties described in (a), (b), and (c) above.

       
A comparison between the spectral shape factors of AS 1170.4:2007 and NZS 1170.5:2004 is attempted in Fig. 2, for the two extreme cases of rock (A and Β) and soft soil (Ε) sites. The New Zealand Standard, which generally applies to earthquakes with higher magnitudes (M ≥ 6.5), features lower spectral factor values in the low structural period range (T < 1 s) and higher spectral factor values in the high period range (T > 1 s). This is compatible with the fact that soil nonlinearity effects alter the spectral content of stronger earthquakes with magnitude M > 5.5, which in addition have a richer low-frequency content attributed also to near-fault effects.
Fig. 2

Comparison of AS 1170.4:2007 and NZS 1170.5:2004 dynamic spectral shape factor plots for rock (A, A and B) and soft soil (E) types (5 % damping)

Hazard Factor and Annual Probability of Exceedance

The hazard factor Z of a particular location corresponds to the peak ground acceleration (in g’s) for site class B (AS 1170.4:2007) or classes A and B (NZS 1170.5:2004), considering a design earthquake with return period of 500 years, i.e., a 10 % probability of exceedance during a 50-year design life. The hazard factors in the Australian Standard have not been updated since its previous version AS 1170.4:1993 and are based on the work of Gaull et al. (1990), who used data of the Australian Geological Survey Organisation dated back to 1856. All areas in Australia are assumed to be seismically active for design purposes, with the hazard factor generally ranging between Z = 0.05 and Z = 0.13, with the exception of the Meckering region in Western Australia, where the hazard factor ranges between Z = 0.14 and Z = 0.22, and the Macquarie Island, located halfway between New Zealand and Antarctica, where the hazard factor is equal to Z = 0.60.

In NZS 1170.5:2004, the hazard factor Z corresponds to 0.5 times the spectral acceleration (5 % damping) for a structure period T = 0.5 s and soil subclass C, considering a design earthquake with a return period of 500 years (Fig. 1c, d). The minimum value across New Zealand is Z = 0.13, to ensure no collapse of structures even in areas of low seismicity, i.e., every structure is designed to survive the 84-percentile strong motion of a magnitude M = 6.5 normal-faulting earthquake at source-to-site distance of 20 km, a seismic scenario corresponding to the low-seismicity areas of New Zealand. The hazard factor at the high-seismicity major cities (e.g., Wellington, Napier, Hastings) is of the order of Z = 0.40, with the maximum value considered in the Standard being Z = 0.60.

The probability factor (AS 1170.4:2007) or return period factor (NZS 1170.5:2004) R, listed in Table 3, is used to scale the hazard factor to the required annual probability of exceedance, for the design limit state under consideration. In AS 1170.4:2007, the annual probability of exceedance is correlated with the design working life and the importance level of the structure, with the latter provided in AS/NZS 1170.0:2002 (Appendix F). The design annual probability of exceedance applies to the ultimate limit state only: According to AS 1170.4:2007, structures of importance levels 1–3 (i.e., all structures except ones carrying critical post-disaster functions and exceptional structures) de facto satisfy the serviceability limit state requirements when designed in accordance to the Standard. A special study needs to be carried out for critical post-disaster structures only (importance level 4), to ensure that they remain operational after a seismic event equivalent to the ultimate limit state design event for ordinary importance level 2 structures. The design annual probability of exceedance for structures and buildings (importance level 2) with a working life of 50 years is 1/500 and corresponds to a 10 % probability of exceedance during the design life of the structure. For major structures affecting crowds and associated with “very great” economic, social, and environmental consequences of failure (importance level 3), the annual probability of exceedance is reduced to 1/1,000 for structures with a design working life of 50 years and to 1/2,500 for structures with a design working life of 100 years.
Table 3

Probability factor (AS 1170.4:2007) or return period factor (NZS 1170.5:2004)

Annual probability of exceedance

R

1/2,500

1.8

1/2,000

1.7

1/1,500

1.5

1/1,000

1.3

1/800

1.25

1/500

1.0

1/250

0.75

1/200

0.70

1/100

0.50

1/50

0.35

1/25

0.25

1/20

0.20

The required annual probability of exceedance for structures designed according to NZS 1170.5:2004 is provided again in AS/NZS 1170.0:2002 (Table 3.3 of AS/NZS 1170.0:2002); however, in the New Zealand Standard, it is correlated with the design (serviceability or ultimate) limit state: The annual probability of exceedance for the common serviceability limit state SLS1 (requirement for no repairs on structural and nonstructural components) is 1/25 for all structures. The special serviceability limit state SLS2 applies to structures carrying critical post-disaster functions only (importance level 4) and is associated with an annual probability of exceedance of 1/500 for design working life of 50 years. A special hazard study is required to define the SLS2 actions for structures of importance level 4 with design working life of 100 years or more.

As far as the ultimate limit state is concerned, the design annual probability of exceedance is the same as in AS 1170.4:2007, for structures classified to importance levels 2 and 3. Note that the product Z∙R in NZS 1170.5:2004 need not exceed Z∙R = 0.70 for ultimate limit state analyses but must be higher than Z∙R = 0.20 when an annual probability of exceedance 1/2,500 is considered. These Z∙R bounds correspond to the higher and lower seismicity regions of New Zealand, respectively: The upper bound matches the 84-percentile near-fault strong motion due to a M = 8.1 event from the major Alpine Fault that runs along the South Island, divided by a margin of safety equal to 1.5 likely to result from applying the code design provisions.

Near-Fault Factor

The near-fault factor N(T,D) is introduced in NZS 1170.5:2004 to account for near-source effects on the strong motion during the activation of a strike-slip fault: a) forward directivity, resulting in high peak velocities and displacements, and b) polarization of the long-period motions in the near-source region, where medium- to long-period pulses tend to be stronger in the direction perpendicular to the strike of the fault. Near-fault effects in NZS 1170.5:2004 are considered for eleven (11) major strike-slip faults classified as “class A” faults according to the Californian Category A fault criteria (Petersen et al. 2000), capable of producing earthquakes with magnitude M = 7.0 or greater and having slip rates of 5 mm/year of greater. The near-fault factor N(T,D) must be considered during the derivation of the medium- to long-period band of the elastic response spectra for sites located within a distance D less than 20 km from the traces of the major strike-slip faults listed in NZS 1170.5:2004 and applies to annual probability of exceedance less that 1/250:
$$ N\left(T,D\right)={N}_{\max }(T)\kern2.25em \mathrm{for}\kern1em D\le 2\;\mathrm{km} $$
(3.1)
$$ N\left(T,D\right)=1+\left({N}_{\max }(T)-1\right)\cdot \left(20-D\right)/18\kern1em \mathrm{for}\kern1.25em 2\mathrm{km}<D\le 20\;\mathrm{km} $$
(3.2)
$$ N\left(T,D\right)=1.0\kern2.25em \mathrm{for}\kern1em D>20\;\mathrm{km} $$
(3.3)
where the factor N max(T) is provided as a function of the structure period T in Table 4 and Fig. 3a. Linear interpolation functions are provided for intermediate period values, as proposed in NZS 1170.5:2004. In addition, a comparison of the effect of near-fault correction (Eqs. 3.1, 3.2, and 3.3) on the spectral shape factors for dynamic analyses is illustrated in Fig. 3b, for different distances from the fault.
Table 4

Maximum values of the near-fault factor N max(T) for different structure periods

NZS 1170.5:2004

Linear interpolation functions

Structure period, T (sec)

N max(T)

Structure period, T (sec)

N max(T)

≤1.5

1.0

≤1.5

1.0

2

1.12

1.5 < T < 4

0.24 T + 0.64

3

1.36

4

1.60

4 ≤ T < 5

0.12 T + 1.12

≥5

1.72

≥5

1.72

Fig. 3

(a) Variation of the near-fault factor N max(T) with the structure period, T, and (b) comparison of spectral shape factors corrected for near-fault effects for a rock site (soil subclass A and B) at a distance D ≤ 2 km, D = 10 km, and D > 20 km from a major active fault (5 % damping)

Vertical Design Actions

According to AS 1170.4:2007, vertical earthquake actions generally need not to be considered in design. For the design of mechanical and electrical components, the vertical earthquake forces are taken equal to 50 % of the horizontal forces. However, a provision is included stating that when the dynamic analysis requires the consideration of vertical earthquake forces, both upward and downward motions need to be considered and the vertical spectral shape factor shall be taken equal to
$$ {C}_v(T)=0.5\cdot {C}_h\left({T}_v\right)\kern1em \left(\mathrm{AS}\ 1170.4:2007\right) $$
(4)
where T v is the vertical period of vibration. In NZS 1170.5:2007 on the other hand, when vertical design actions need to be considered in the analyses (e.g., time history analyses), the elastic response spectrum for vertical loading is taken equal to 70 % of the corresponding horizontal spectrum, as
$$ {C}_v(T)=0.7\cdot {C}_h\left({T}_v\right)\kern1em \left(\mathrm{NZS}\ 1170.5:2004\right) $$
(5)
where T v = 0 for the design of the structure as a whole, or equal to the vertical period of the element under consideration, for the design of parts and components. Note that the vertical component of strong earthquake motion is usually rich in high-frequency content; thus, peak ground acceleration and spectral values in the near-source region may exceed the horizontal values (NZS 1170.5 Supp 1:2004, Niazi and Bozorgnia 1992; Ambrasseys and Simpson 1996). In light of this, NZS 1170.5 Supp 1:2004 recommends that when the distance of the site from the fault trace is less than D = 10 km, the vertical design actions be considered equal to the horizontal design actions for structure period T ≤ 0.30 s. Strong motion recordings from recent earthquakes, such as the Christchurch 2011 earthquake, provided further evidence in support of the above, as will be discussed later in this entry.

Structural Ductility and Structural Performance Factor-Inelastic Design Spectrum

Derivation of the (inelastic) design spectrum requires the correction of the elastic response spectrum to account for the ability of the particular structure to dissipate seismic energy via nonlinear response. Apart from the structural ductility factor, μ, which is related to the level of inelastic demand that can be reliably sustained by the structure, AS 1170.4:2007 and NZS 1170.5:2004 introduce the structural performance factor S p to reduce the elastic design seismic actions. The structural performance factor is employed to quantify a number of effects that are not explicitly taken into account in state-of-practice structural analysis procedures by simply scaling the design loads. Such effects include (NZS 1170.5 Supp 1:2004):
  1. (a)

    Excitation effects: The estimated seismic loads correspond to a peak ground acceleration value, which may be reached only during a single loading cycle and therefore is unlikely to lead to significant damage (“effective” ground acceleration concept).

     
  2. (b)

    Individual structural elements typically feature a higher capacity than modeled during the analysis of the structure, due to higher material strength, strain hardening, strain rate effects, etc.

     
  3. (c)

    The total structural capacity is typically higher than predicted, due to redundancy effects or the contribution of nonstructural elements (e.g., in fill walls) which is not taken directly into account in typical analysis models.

     
  4. (d)

    The energy dissipation of the structure is typically higher than assumed, due to damping introduced from nonstructural elements and soil-foundation interaction effects.

     
The ductility factor μ for the ultimate limit state is calculated in accordance with the appropriate material standard, where such data are provided. Contrary to NZS 1170.5:2004, where reference is made to material standards only and a special study is required otherwise, maximum ductility factors for typical structural systems are provided directly in AS 1170.4:2007 (Table 5). Alternatively, for a specific structure, μ and S p can be determined via a nonlinear static pushover analysis. Note also that AS 1170.4:2007 applies only to structures with ductility factor μ ≤ 3, and when a higher ductility factor is considered, the design must be performed in compliance with NZS 1170.5:2004 provisions (Table 5).
Table 5

Ductility factor μ and structural performance factor S p for different structural systems and specific structure types (After AS 1170.4:2007)

Structural system

Description

μ

S p

Steel structures

Special moment-resisting frames (fully ductile)a

4

0.67

Intermediate moment-resisting frames (moderately ductile)

3

0.67

Ordinary moment-resisting frames (limited ductile)

2

0.77

Moderately ductile concentrically braced frames

3

0.67

Limited ductile concentrically braced frames

2

0.77

Fully ductile concentrically braced framesa

4

0.67

Other steel structures not defined above

2

0.77

Concrete structures

Special moment-resisting frames (fully ductile)a

4

0.67

Intermediate moment-resisting frames (moderately ductile)

3

0.67

Ordinary moment-resisting frames

2

0.77

Ductile coupled walls (fully ductile)a

4

0.67

Ductile partially coupled wallsa

4

0.67

Ductile shear walls

3

0.67

Limited ductile shear walls

2

0.77

Ordinary moment-resisting frames in a combination with limited ductile shear walls

2

0.77

Other concrete structures not listed above

2

0.77

Timber structures

Shear walls

3

0.67

Braced frames (with ductile connections)

2

0.77

Moment-resisting frames

2

0.77

Other wood- or gypsum-based seismic-force-resisting systems not listed above

2

0.77

Masonry structures

Close-spaced-reinforced masonryb

2

0.77

Wide-spaced-reinforced masonryb

1.5

0.77

Unreinforced masonryb

1.25

0.77

Other masonry structures not complying with AS 3700:2001

1

0.77

Specific structure types

Tanks, vessels, or pressurized spheres on braced on unbraced legs

2

1.0

Cast-in-place concrete silos and chimneys having walls continuous to the foundation

3

1.0

Distributed mass cantilever structures, such as stacks, chimneys, silos, and skirt-supported vertical vessels

3

1.0

Trussed towers (freestanding or guyed), guyed stacks, and chimneys

3

1.0

Inverted pendulum-type structures

2

1.0

Cooling towers

3

1.0

Bins and hoppers on braced or unbraced walls

3

1.0

Storage racking

3

1.0

Signs and billboards

3

1.0

Amusement structures and monuments

2

1.0

All other self-supporting structures not otherwise covered

3

1.0

aThe design of structures with μ > 3 should be in accordance to NZS 1170.5:2004

bValues from AS 3700:2001

While NZS 1170.5:2004 does not refer explicitly to ductility factor values for the ultimate limit state, for serviceability limit state analyses of common structures (SLS1), it is provisioned that 1.0 ≤ μ ≤ 1.25, and for critical post-disaster structures (SLS2), it is provisioned that 1.0 ≤ μ ≤ 2.0. Furthermore, the ductility factor for vertical actions is always considered to be μ = 1.0.

Note that in NZS 1170.5:2004, the inelastic horizontal design spectra are not derived by directly multiplying the elastic spectra by (1/μ). Instead, to account for the transition between equal displacement theory (which is valid for longer structure periods) and equal energy theory (valid for shorter structure periods), a transition point is defined at T 1 = 0.7 s for soil subclasses A to D and at T 1 = 1.0 s for soil subclass E, where T 1 is the largest translation period of vibration along the direction under consideration. So, the factor to derive the inelastic response spectra is estimated as:

For subsoil classes A, B, C, and D:
$$ {k}_{\mu }=\mu \kern2em \mathrm{for}\kern1em {T}_1\ge 0.7\;\mathrm{s} $$
(6.1)
$$ {k}_{\mu }=1+\left(\mu -1\right){T}_1/0.7\kern2.25em \mathrm{for}\kern1em {T}_1<0.7\;\mathrm{s} $$
(6.2)
For subsoil class E:
$$ {k}_{\mu }=\mu \kern2.25em \mathrm{for}\kern1em {T}_1\ge 1.0\;\mathrm{s}\kern0.5em \mathrm{or}\kern0.5em \mu <1.5 $$
(6.3)
$$ {k}_{\mu }=1.5+\left(\mu -1.5\right){T}_1\kern2.25em \mathrm{for}\kern1em {T}_1<1.0\;\mathrm{s}\ \mathrm{and}\kern0.5em \mu \ge 1.5 $$
(6.4)
For the purpose of calculating k μ , the largest translation period of vibration T 1 is not taken less than 0.4 s.
The structural performance factor S p in AS 1170.4:2007 is provided in Table 5. In NZS 1170.5:2004, the structural performance factor is taken equal to S p = 0.7 for ultimate limit state analyses, except for low-ductility structures with 1.0 < μ < 2.0, where S p is calculated as:
$$ {S}_p=1.3-0.3\mu $$
(7)
For ultimate limit state verification of structures against sliding or overturning, S p is taken equal to S p = 1.0. For serviceability limit state SLS1 and SLS2 analyses, S p is taken equal to S p = 0.7, except otherwise dictated by the relevant material standard.
In accordance to the above, the inelastic design response spectrum in AS 1170.4:2007 is defined as
$$ {C}_d(T)=\frac{C(T){S}_p}{\mu } $$
(8)
and in NZS 1170.5:2004 as
$$ {C}_d(T)=\frac{C(T){S}_p}{k_{\mu }}\;\left(\mathrm{for}\ \mathrm{horizontal}\ \mathrm{actions}\right) $$
(9.1)
$$ {C}_d(T)={C}_v(T){S}_p\;\left(\mathrm{for}\ \mathrm{vertical}\ \mathrm{actions}\right) $$
(9.2)

Actions for Dynamic Time History Analyses

In addition to actions applicable to common equivalent static and modal dynamic analyses, AS 1170.4:2007 and, mainly, NZS 1170.5:2004 include certain provisions for the determination of the appropriate design actions when dynamic time-domain analyses are to be performed. While in AS 1170.4:2007 there is a vague requirement that the response spectra of the actual acceleration time histories used shall “approximate” the design spectrum of Eq. 8, NZS 1170.5:2004 prescribes a more elaborate and explicit procedure for the selection and scaling of ground motion records.

Ground motion accelerographs to be used for time-domain analyses according to NZS 1170.5:2004 shall consist of both horizontal components of the recording; the vertical component may have to be considered too, for the analysis of structures sensitive to vertical strong ground motion. The above imply that the procedure refers to three-dimensional analyses, yet an adaption to two-dimensional models is feasible, by considering only one horizontal component of each ground motion record as, e.g., in ASCE/SEI 7-10. A “family” of no less than three records must be employed in each time history analysis.

The selected acceleration records must be consistent with the local seismotectonic regime (expected magnitude, fault rupture mechanism, source-to-site distance, etc.). When the site is located near a major fault, i.e., N(T, D) > 1 (Eqs. 3.1, 3.2, and 3.3), then one set of the records must have a forward directivity component. Simulated, artificial accelerographs may be used too, when appropriate real acceleration recordings are not available. Each record shall be scaled by a “record scale factor” k 1 to match the target design spectrum over the period range of interest and by a “family scale factor” k 2 to ensure that the energy content of at least one record in the family exceeds that of the design spectrum over the period range of interest. The target design spectrum is defined as
$$ S{A}_{\mathrm{target}}=\left(\frac{1+{S}_p}{2}\right)C(T) $$
(10)
where the spectral shape factor C(T) is calculated from Eq. 1 and S p is the structural performance factor defined in the section “Structural Ductility and Structural Performance Factor-Inelastic Design Spectrum.”
In order to present the procedure of scaling ground motion records according to NZS 1170.5:2004, the horizontal components of three typical strong ground motion recordings from the 2011 Christchurch M = 6.3 earthquake were used. These records were scaled to match the target ultimate limit state spectrum of NZS 1170.5:2004 for a common structure (importance level 2) located in Christchurch, founded on subsoil class D (soft soil). According to NZS 1170.5:2004, for a structure of importance level 2 in Christchurch with a design life of 50 years, Z · R = 0.22 g and N(T,D) = 1, i.e., no near-fault effects, need to be taken into account due to possible rupture of the Port Hills Fault that is believed to have caused the 2011 earthquake. Assuming for simplicity that S p = 1, the target spectrum is plotted in Fig. 4.
Fig. 4

(a) Comparison of the 2011 Christchurch strong motion recordings elastic response spectra (5 % damping, horizontal components) with the design NZS 1170.5:2004 elastic design spectra for the city of Christchurch and soil subclass D, (b) scaled elastic response spectra of the same strong motion recordings to match the NZS 1170.5:2004 design spectra for largest structure translation period T 1 = 1 s and S p = 1

Strong motion recordings during the Christchurch 2011 earthquake exceeded by far the design peak ground acceleration at the ground surface of class D soft soils implied by NZS 1170.5:2004 (0.246 g) and reached maximum values of 1.67 g in the horizontal direction and 2.20 g in the vertical direction at the Heathcote Valley Primary School Station (HVSC) (www.geonet.org.nz). Here a family of three near-field recordings on soft to very soft soil are used, with peak horizontal ground accelerations ranging from 0.34 to 0.67 g (Table 6).
Table 6

Characteristics of the typical strong motion recordings from the Christchurch 22 February 2011 M = 6.3 earthquake (Source: www.geonet.org.nz)

Site code

Subsoil class acc. to NZS 1170.5:2004

Epicentral distance

PGA-horiz.1 (g)

PGA-horiz.2 (g)

PGA-vert. (g)

CBGS

D

7

0.553

0.452

0.360

PRPC

E

6

0.669

0.595

1.88

CHHC

D

6

0.345

0.364

0.601

The elastic response spectra of the horizontal components of these three recordings are plotted in Fig. 4a, in comparison with the design response spectra of NZS 1170.5:2004. A band-pass filter in the frequency ranges of 0.10–0.25 Hz and 24.5–25.5 Hz was applied to the time histories used to derive the spectra. It is clear that the spectral values of these typical records are above the design spectra across practically the whole range of important structural periods, an indication of the severity of the Christchurch earthquake.

Scaling of ground motion records to match the target design spectrum via the scale factor k 1 was performed while taking into account a structure period range that depends on the largest translational period of the structure in the direction of interest, T 1. The period range of interest is defined as 0.4 T 1T ≤ 1.3 T 1, with the product 0.4 T 1 not taken less than 0.4 s. Notice that this band of significant periods is considerably narrower compared to the corresponding one prescribed in EN-1998-1, which is 0.2 T 1T ≤ 2.0 T 1, or even ASCE/SEI 7-10, where the corresponding range is 0.2 T 1T ≤ 1.5 T 1. For the case at hand, it was assumed that T 1 = 1 s, and the period range of interest is denoted by the yellow band in Fig. 4. The best-fit scale factor k 1 of each horizontal ground motion component was determined as the value minimizing the function log(k 1 SA component/SA target) over the period range of interest, in a least mean square sense. In other words, the aim was to find via an interactive procedure the k 1 value that minimizes the sum:
$$ {{\displaystyle \underset{0.4{T}_1}{\overset{1.3{T}_1}{\int }}\left[ \log \left(\frac{k_1S{A}_{\mathrm{component}}}{S{A}_{\mathrm{target}}}\right)\right]}}^2\mathrm{d}T= \min $$
(11)
The selected recordings must satisfy a “similarity” criterion 0.33 < k 1 < 3.0. Moreover, in order to verify that each selected record reasonably matches the target spectrum over the period range of interest, the following inequality quantifying the mean (over the period of interest) square difference of the scaled over the target spectral values must be satisfied:
$$ {D}_1=\sqrt{\frac{1}{\left(1.3-0.4\right){T}_1}\cdot {\displaystyle \underset{0.4{T}_1}{\overset{1.3{T}_1}{\int }}{\left[ \log \left(\frac{k_1S{A}_{\mathrm{component}}}{S{A}_{\mathrm{target}}}\right)\right]}^2\mathrm{d}T}}\le \log (1.5) $$
(12)
Although not a normative criterion, a stricter fit D 1 ≤ log(1.3) is proposed for most cases, according to NZS 1170.5 Supp 1:2004.
The k 1 and D 1 factors for the records at hand are listed in Table 7, where it is depicted that the specific records are compatible with NZS 1170.5:2004 requirements. The component of each record with the lower k 1 value was nominated as “principal” (Fig. 4, Table 7), and this value of k 1 was used as the record scale factor.
Table 7

Scaling factors for the considered typical strong motion records

Site code

Component

k 1

D 1

k 2

CBGS

N89W (principal)

0.608

0.070

1.05

S01W

0.789

0.105

PRPC

W (principal)

0.672

0.081

S

0.785

0.097

CHHC

N01W

0.683

0.064

S89W (principal)

0.598

0.058

The record family scale factor, k 2, ensures that the principal component of at least one record spectrum (after being scaled by k 1) exceeds the target spectrum over the period range of interest. It is estimated as the maximum value of the ratio SA target/max(SA principal) ≥ 1.0 within the period range of interest, where max(SA principal) is the maximum spectral value between all the principal components of the family, at each period step considered for the derivation of the spectra. To confirm the selection of the principal and secondary components, the record family scale factor must be within the range 1.0 < k 2 ≤ 1.3. If k 2 > 1.3, then either a different record must be selected or the selection of the principal/secondary components may be reversed, aiming to minimize the product k 1 k 2. This may be the case when all three principal components in a family feature low spectral values within a particular period band, while one of the secondary components is rich in frequency content within the same band; thus, although the scale factor k 1 may be greater for that particular strong motion component, the product k 1 k 2 may be lower. In the case at hand however, the family factor k 2 is within the desirable range (Table 7), a fact that can be confirmed visually from Fig. 4.

The above scaling procedure must be followed for all directions of interest, corresponding to different translational periods of the structure, T 1, and thus a different period range of interest. Although more elaborate than the analogous procedure of EN-1998-1 or ASCE/SEI 7-10 for the time history representation of the seismic action, where it is simply required that no value of the mean elastic spectrum of all time histories used is less than 90 % (EN-1998-1) or 100 % (ASCE/SEI 7-10) of the corresponding value of the elastic response spectrum, the procedure prescribed in NZS 1170.5:2004 results in a more rational scaling of strong motion records, and in a closer match to the design spectrum, as it does not impose restrictions related only to lower-than-target spectral acceleration values found in EN-1998-1 and ASCE/SEI 7-10. As a result of the above, if one embraced the more conservative EN-1998-1 and ASCE/SEI 7-10 provisions regarding the average response spectra from the three principal components, the specific scaling factors applied would not be acceptable, as their average spectrum is below 100 % (ASCE/SEI 7-10) and marginally below 90 % (EN-1998-1) of the target spectrum (Fig. 5).
Fig. 5

Comparison of the mean response spectrum (5 % damping) of the three principal components listed in Table 7 with the NZS 1170.5:2004 target spectrum for soil subclass D and S p = 1. The period of interest band according to NZS 1170.5:2004 (0.4 T 1T ≤ 1.3 T 1) and according to ASCE/SEI 7-10 (0.2 T 1T ≤ 1.5 T 1) is also drawn

Finally, note that NZS 1170.5:2004 (as also ASCE/SEI 7-10) bases the accelerograph scaling procedure on spectral values within the range of interest only and does not impose restrictions on the zero period spectral response acceleration values of the strong motion, as, e.g., EN-1998-1.

Summary

Seismic design action definitions in the Australian AS 1170.4:2007 and New Zealand NZS 1170.5:2004 Standards were presented in a concise way, focusing on the determination of the elastic and inelastic design response spectra to be used together with static and dynamic methods of analysis. A short discussion against provisions of other modern seismic codes, such as EN-1998-1 and ASCE/SEI 7-10, was attempted to point out certain key differences in particular clauses. Although Australian and New Zealand Standards embrace most of the recent developments on earthquake engineering, such as accounting for near-source effects on strong motion (but not topography effects as, e.g., EN-1998-1), any seismic code is not a “static” document, but rather a “dynamic” one, and evolves with lessons learned from recent major earthquakes that are included in future revisions as, e.g., the increase of the hazard factor Z for the Christchurch area from Z =0.22 g to Z = 0.30 g in the latest compliance document of the New Zealand Building Code (Department of Building and Housing 2011).

Cross-References

References

  1. Ambrasseys NN, Simpson KA (1996) Prediction of vertical response spectra in Europe. Earthq Eng Struct Dyn 25:401–412CrossRefGoogle Scholar
  2. AS 1170.4:2007. Structural design actions. Part 4: earthquake actions in Australia. Standards AustraliaGoogle Scholar
  3. AS 3700:2001. Masonry structures. Standards AustraliaGoogle Scholar
  4. AS/NZS 1170:2002. Structural design actions. Standards Australia/standards New ZealandGoogle Scholar
  5. ASCE/SEI 7-10. Minimum design loads for buildings and other structures. ASCE Standard, American Society of Civil Engineers, Structural Engineering InstituteGoogle Scholar
  6. Bouckovalas GD, Kouretzis GP, Kalogeras IS (2002) Site-specific analysis of strong motion data from the September 7, 1999 Athens, Greece Earthquake. Nat Hazards 27:105–131CrossRefGoogle Scholar
  7. Bouckovalas G, Kouretzis GP (2001) Stiff soil amplification effects in the 7 september 1999 Athens (Greece) earthquake. Soil Dyn Earthq Eng 21(8): 671–687Google Scholar
  8. Department of Building and Housing (2011) Compliance document for New Zealand building code, clause B1, structures, Amendment 11, August. New Zealand GovernmentGoogle Scholar
  9. EN-1998-1. Eurocode 8: design of structures for earthquake resistance – Part 1: general rules, seismic actions and rules for buildings. European Committee for StandardizationGoogle Scholar
  10. Gaull BA, Michael-Leiba MO, Rynn JMW (1990) Probabilistic earthquake risk maps of Australia. Aust J Earth Sci 37:169–187CrossRefGoogle Scholar
  11. GeoNet-Earthquake Commission-GNS Science. website: www.geonet.org.nz. Accessed Sept 2013
  12. McRae G, Clifton C, Megget L (2011) Review of NZ building codes of practice. Report to the royal commission of inquiry into the building failure caused by the Christchurch earthquakes ENG.ACA.0016.1Google Scholar
  13. Nakamura Y (1989) A method for dynamic characteristics estimation of subsurface using microtremors on the ground surface. RTRI Q Rep 30(1):25–33Google Scholar
  14. Niazi M, Bozorgnia Y (1992) Behaviour of near-source vertical and horizontal response spectra at SMART-1 array. Earthq Eng Struct Dyn 21:37–50CrossRefGoogle Scholar
  15. NZS 1170.5:2004. Structural design actions. Part 5: earthquake actions-New Zealand. Standards New ZealandGoogle Scholar
  16. NZS 1170.5 Supp 1:2004. Structural design actions. Part 5: earthquake actions. New Zealand-commentary. Standards New ZealandGoogle Scholar
  17. NZS 95. New Zealand standard code of building by-laws. Wellington, Standards Association of NZGoogle Scholar
  18. Petersen MD, Toppozada TR, Cao T, Cramer CH, Reichle MS, Bryant WA (2000) Active fault near-source zones within and bordering the state of California for the 1997 Uniform Building Code. Earthq Spectra 16(1):69–83CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • George P. Kouretzis
    • 1
    Email author
  • Mark J. Masia
    • 1
  • Clive Allen
    • 1
  1. 1.Faculty of Engineering and Built EnvironmentThe University of NewcastleCallaghanAustralia