professor emeritus
MehmedČaušević1Emailmcausevic@gradri.uniri.hrassociate professor
MladenBulić1✉Emailmbulic@gradri.uniri.hr1Faculty of Civil EngineeringUniversity of RijekaR. Matejčić 351000RijekaCroatia
Mehmed Čaušević · Mladen Bulić
Mehmed Čaušević, professor emeritus
University of Rijeka, Faculty of Civil Engineering, R. Matejčić 3, Rijeka 51000, Croatia
E-mail: mcausevic@gradri.uniri.hr
Mladen Bulić, associate professor (corresponding author)
University of Rijeka, Faculty of Civil Engineering, R. Matejčić 3, Rijeka 51000, Croatia
E-mail: mbulic@gradri.uniri.hr
Abstract
The aim of the analysis and comments presented in this paper is to shed light on some doubts regarding the application of the second generation of Eurocode EN 1998-1-1 in the segment for determining the seismic action on buildings.
One of the most important parts of the EN1998-1-1 standard refers to the determination of seismic load for buildings using spectra. Here, these spectra of the second generation are presented and discussed, and how they should be defined in the National Annex by each EU member state, especially those in the Mediterranean, to meet the needs of extreme actions due to earthquakes. For comparison, the seismic action on buildings obtained by applying the European standards currently in force is taken.
If the value of the new spectrum
for the period
is determined in a probabilistic way, as presented in EN 1998-1-1, it would not correspond to the level of seismic load that we have applied thus far. This paper provides recommendations on how this problem can be avoided because the second-generation Eurocode 8 offers, in addition to the use of spectra, another way of determining the seismic load for buildings. Presentation in this paper is based on the fact that the intensity of the seismic load that we have obtained by applying the currently valid spectra of the first generation has been relatively well affected, which is reflected in a kind of verification of the behaviour of newer buildings in recent earthquakes in the EU.
Keywords
Second generation of Eurocode 8
Seismic load
Earthquake spectrum
Uniform hazard spectrum
Limit states
1 Introduction
Seismologists in Croatia and some other Mediterranean countries have been of the opinion from the beginning that the response spectra of Type 1 and Type 2 of the first generation EN 1998-1 (Eurocode 8 2004) are not entirely suitable for application (Fig. 1). Since this is a generally accepted conclusion in Europe, completely new spectra have been developed in the second-generation Eurocode EN 1998-1-1 (Eurocode 8 2022).
Here are some other reasons for this statement:
- The surface magnitude of
is surpassed in terms of its use. It is not an adequate measure of the strength of an earthquake. In the second generation EN 1998-1-1 (Eurocode 8
2022), the moment magnitude
is introduced instead of the surface magnitude.
- The elastic response spectra of Type 1 and Type 2 have only one point that is determined according to the probabilistic concept (PGA for T = 0), whereas all the other points of these spectra are defined deterministically. This had to be improved, which was done in the second generation of this Eurocode, but still, the application of the uniform hazard spectrum (UHS) was not adopted. The reasons for this are discussed later in this article.
The dilemmas of the Technical Committee CEN/TC 250/SC 8 when the final text of the second generation of the EN 1998-1-1 standard (Eurocode 8 2022) is accepted are discussed here. If UHS spectra are used, in which all points of the spectrum are defined probabilistically, lower values of seismic action for tall buildings would be obtained compared with the elastic spectrum of Type 1 of the first generation. This statement is illustrated by comparing the UHS spectrum with the Type 1 and Type 2 spectra for the city of Zagreb, Croatia (Čaušević et al. 2023), which was hit by a strong earthquake in 2020 (Atalić et al. 2021) (Fig. 2). That earthquake was a test for buildings designed according to the spectra of the first generation. The good behaviour of these buildings was noticed, and the conclusion is that the seismic loading while designing buildings according to the spectra of the second generation should not be less than the seismic loading according to the spectra of the first generation. For this reason, EN 1998-1-1 (Eurocode 8 2022) prescribes two ways of defining spectra.
One principle is proposed for most EU countries: to adopt such spectra for application on the basis of which the seismic loading on buildings will be determined, which will not be less than the loading obtained by applying the spectra of the first generation that have been valid thus far.
Here are some reasons to apply the proposed principle:
- In the current standard, the elastic response spectra of Type 1 and Type 2 have only one point that is determined according to the probabilistic concept (PGA for T = 0), whereas all the other points of these spectra are defined deterministically.
- In the second generation of the standard (Eurocode 8
2022), several points of the spectrum are introduced, which are determined probabilistically: these are the points “on the plateau” of the spectrum and the point of the spectrum for
, which define the newly introduced values of
and
for site category A of the location of the building.
- European hazard maps for
and
are introduced, both for a return period of 475 years. The reference values of these parameters,
and
, are determined by seismologists, who produce national hazard maps for each EU for the respective limit states and certain consequence classes, CC.
There are 9 return periods according to the second generation of the Eurocode (Eurocode 8 2022, Čaušević and Bulić 2020), as shown in Table 1, and they are determined depending on the selected limit state (limit state, LS) and the consequence class (consequence class, CC). For example, for the limit state named the significance damage (SD) and CC2, the return period is 475 years (1), as well as in Eurocode 8 2004:
𝑇𝑟𝑒𝑓 = 𝑇𝑆𝐷,2 = 475 godina (1)
Table 1
Limit states (LS) | Consequence classes (CC) |
|---|
CC1 | CC2 | CC3-a | CC3-b |
|---|
NC | 800 | | 2500 | 5000 |
|---|
SD | 250 | 475 | 800 | 1600 |
DL | 50 | 60 | 60 | 100 |
Seismic hazard maps should be prescribed by the National Committee of each EU country. The European Seismic Hazard Maps are derived from the deliverable ESHM20 of the SERA research project (Seismology and Earthquake Engineering Research Infrastructure Alliance for Europe), which received funding from the EU Horizon 2020 research and innovation program to provide updated information on seismic hazards in Europe (Danciu et al. 2021) (Fig. 3).
2 Spectrum of the second generation of Eurocode 8 for buildings: selective application
In the spectrum of the second generation, a compromise is adopted, and a classic deterministic spectrum is prescribed, in which several of its points are determined in a probabilistic way (Fig. 4). It has already been stated that the seismic forces for the design of buildings according to the second generation of the standard should not be less than the forces obtained according to the first generation standard.
The second-generation spectrum is defined deterministically, with several of its points determined in a probabilistic way (Fig.
4). The curve in Fig.
4 for
"falls" too steeply. To meet the abovementioned principle that seismic forces should not be less than those obtained according to the first-generation standard (Eurocode 8
2004), the
value should be increased in the manner prescribed in Article 5.2.1. (Eurocode 8
2022).
In Fig.
4,
represents the ratio of
with respect to the zero-period spectral acceleration (
).
3 Interpretation and application of Article 5.2.1 (Eurocode 8 2022)
The provision of Art. 5.2.1 in EN 1998-1-1(Eurocode 8 2022) reads as follows:
(4) For the derivation of
, a) or b) may be used:
(a) Map the territory in terms of
and obtain
by applying Formula (5.3):
is the value of
for site category A and for the return period
, where
for low and very low seismicity levels;
for moderate seismicity levels;
for high seismicity levels.
The values of
may increase in the National Annex.
b) Obtain
from a seismic hazard study of the territory under consideration and map it concurrently with
.
In accordance with the second generation of Eurocode 8, nine earthquake return periods are introduced (there have been only 2), which means that as a rule, 18 maps should be made (9 return periods and each with two spectral values,
and
), as well as at least one map, the one for
. The number of these maps can therefore be reduced according to the needs of designers of buildings and allowed by the National Annex.
4 Numerical confirmation of the advantages of applying Article 5.3.1 (Eurocode 8 2022)
Therefore, civil engineers object to the acceleration value obtained for the probabilistically defined point of the spectrum for
. If it remained like this, we would obtain much less seismic force for tall buildings than for the Type 1 elastic spectrum from the first-generation standard, which should be avoided. The second-generation standard EN 1998-1-1 (Eurocode 8
2022) also provides a way to avoid this.
From (5.3), it can be concluded that for high seismicity levels:
where
- Topography amplification factor
- Intermediate period (
) site amplificationfactor
The values for
can even increase when
and
are greater than 1.
If the hazard mapping takes the form of seismic zones, these zones should be delimited, and their seismicity levels should be qualified according to Table
2, where
is the reference spectral acceleration calculated for a return period of 475 years.
Table 2
Range of
values to define seismicitylevels
Seismicity level | |
|---|
Very low | |
|---|
Low | |
|---|
Moderate | |
High | |
If a different reference return period is used and
is not available,
values may be obtained by applying the following equation:
5 Conclusions
- It can be concluded that the spectra of the second generation for periods approximately 1 s are significantly ‘lowered’ compared to the spectra that have been applied thus far.
- To obtain seismic loading on buildings by applying a standard of the second generation that is not less than those obtained by applying the first-generation standards, we should not adopt the values of loading by applying a seismic hazard map but rather by using Formula (5.3), as offered by EN 1998-1-1. The reason for this is a kind of verification of the behaviour of buildings in recent earthquakes (Atalić et al. 2021). Buildings designed according to the first-generation Eurocodes have performed well in recent earthquakes.
- EN1998-1-1 allows the level of seismic loading for buildings to increase relative to the seismic loading on structures obtained from the seismic hazard map for
.
- Not too much attention should be given to saving the amount of reinforcement, but the focus should be on the safety of structures. It is well known that 1 euro invested in prevention is worth many times over in the rehabilitation of structures after an earthquake.
- In Mediterranean countries, the calculation of seismic loading on buildings should be applied as defined in Part a) of Article 5.2.1 (Eurocode 8
2022). This means that
can be calculated from the hazard map for
multiplied by the multiplier
, which depends on the strength of the expected earthquakes.
- The earthquake engineering community is aware of the discrepancy between research achievements and practical applications. With the suggestions presented in this paper, it is believed that this discrepancy should be reduced.
- To support the proposals presented in this paper, the following example of construction behaviour in the Kobe earthquake, Japan, 1995, M = 7.3 is given in Fig. 6. Although Japan has high-level knowledge of seismology and excellent regulations and construction is of high quality, it is possible that the seismic load in this case (Fig. 6) was taken from seismic hazard maps, as prescribed by seismologists. Apparently, that was not sufficient.
Acknowledgments
The research presented in this paper was funded by the University of Rijeka, Croatia, as part of Research Project No. uniri-iskusni-tehnic-23-237.
References
Atalić J, Uroš M, Šavor Novak M, Demšić M, Nastev M (2021) The Mw5.4 Zagreb (Croatia) earthquake of March 22, 2020: impacts and response. Bull Earthq Eng 9:3461–3489. https://doi.org/10.1007/s10518-021-01117-w
Čaušević M, Bulić M (2020) Proposal of response spectra in the second generation of Eurocode EN1998-1-1 for seismic areas and comparison with the existing standard EN 1998-1: 2004. Građevinar 72:895–904. https://doi.org/10.14256/JCE.2838.2019
Čaušević M, Mitrović S, Bulić M (2023) Determination of Seismic Load for Buildings using Different Response Spectra and Application on Different Methods of Analysis. Cogent Eng 10:2220494. https://doi.org/10.1080/23311916.2023.2220494
Danciu L, Nandan S, Reyes C, Basili R, Weatherill G, Beauval C, Rovida A, Vilanova S, Sesetyan K, Bard P-Y, Cotton F, Wiemer S, Giardini D (2021) The 2020 update of the European Seismic Hazard Model: Model Overview. EFEHR Technical Report 001, v1.0.0. https://doi.org/10.12686/a15
Eurocode 8 (2004) EN 1998-1:2004 + AC:2009, Design of structures for earthquake resistance – Part 1, General rules, seismic actions and rules for buildings. European Committee for Standardization CEN, Brussels
Eurocode 8 (2022) prEN 1998-1-1:2021, Design of structures for earthquake resistance, Part 1–1: General rules and seismic action. European Committee for Standardization CEN/TC 250/SC 8 N 1141, Document date 2022-01-11, Brussels
Statements and Declarations
A
Author Contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Mehmed Čaušević and Mladen Bulić. The first draft of the manuscript was written by Mehmed Čaušević and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
A
Data Availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.