Very Low Seismicity Areas

Transcription

1 Vol. XIX, 2011, No. 3, 1 9 M. Sokol, L. Konečná, M. Čuhák, M. Dallemule Very Low Seismicity Areas in Slovakia and Comparison of Seismic Risks in Central European Countries Milan SOKOL milan.sokol@stuba.sk Research field: dynamics of structures, seismic engineering, modelling of structures Lenka KONEČNÁ lenka.konecna@stuba.sk Research field: dynamics of structures, seismic engineering, modelling of structures Address: Department of Structural Mechanics Marek ČUHÁK marek.cuhak@stuba.sk Research field: concrete structures Address: Department of Concrete Structures and Bridges Marián DALLEMULE marian.dallemule@stuba.sk Research field: steel structures Address: Department of Steel and Timber Structures Faculty of Civil Engineering Radlinského Bratislava, Slovakia ABSTRACT KEY WORDS A parametric study concerning the importance of very low seismicity for Slovakia is introduced. Analyses were performed for ordinary buildings such as family houses, residential brick buildings, residential panel buildings and high-rise buildings. Both horizontal wind and seismic actions were analysed and compared in detail for all the regions and conditions in Slovakia. Very low seismicity, wind action, seismic action, peak ground acceleration, effective ground acceleration, seismic hazard. 1. INTRODUCTION The effects of all possible actions should be taken into account during the design of a structure. Because an analysis of a structure that assumes seismic effects is a very sophisticated and demanding process, there are times when a seismic analysis is not necessary. These limits are specified by a seismic action level which is called a very low seismicity area. These limits are especially defined in [2]. In other words, it is necessary to specify under what conditions it would not be required to consider seismic effects. The paper provides an answer to the question of whether it is necessary to take into account seismic effects in cases of the common types of buildings in Slovakia (family houses, residential buildings, residential multi-storey buildings and high-rise buildings). These representative buildings were examined on all types of soil which could significantly be affected by a seismic response. If there is a construction site which could not be described as a very low seismicity area, there is one more possible simplification of a seismic analysis if the seismic effects are not very great. In this case the seismic analysis does not have to be done in so much detail. These sites are called low seismicity areas. This means that only a few restrictions in such areas are necessary for an area to be approved. In many cases that means that only detailed rules are necessary and that there is no need to perform a seismic analysis at all. 2. SEISMICALLY VULNERABLE AREAS Due to the simplicity of designing a seismic design procedure, there are only a few basic parameters needed to describe the seismic vulnerability of a location. In Eurocode 8 [2], peak ground acceleration (PGA) on a rocky soil is the main representative value defining the seismic vulnerability of an area. The map of a country 2011 SLOVAK UNIVERSITY OF TECHNOLOGY 1

2 describing this value provides the answer as to where the most vulnerable seismic areas are. In some countries such as Slovakia, the characteristic value of such a map is defined in a slightly more appropriate manner and is called effective peak acceleration (EPA), which more precisely describes the possibility of structural damage or collapse at a site (Fig. 7a). EPA is defined as a spectral acceleration during a period of T 1 =0.5s divided by a factor of 2.5. Period T 1 is assumed to be the first vibration period of an ordinary structure. 3. VERY LOW SEISMICITY AREAS The basic principle of defining very low seismicity areas is based on the assumption that seismic effects (seismic design situations) are smaller than the effects of other design situations, i.e., effects caused by a dead load, wind load, etc. (basic design situations). So both seismic and basic design situations are compared, and if the basic design situation is unfavourable, such a site is called a very low seismicity area. But these two design situations are quite different, even in their philosophy. A seismic design situation is described by a limit state when a structure after a seismic event can be damaged to a certain extent but has not yet collapsed. Plastic joints can develop in a specific part of a structure. They can gradually extend at such locations on the structure where the earthquake s energy has dissipated. So this means that the analysis of the potential structural response should be performed by a non-linear dynamic analysis. This is quite a demanding process and cannot be done in ordinary practice, so simplified quasi-linear solutions should be performed as a reference code procedure. On the other hand, the effects caused by wind are also influenced by many factors (terrain, orientation of the building, etc.). An appropriate tool for indicating how a seismic design situation can be compared to a basic design situation is to define the factor expressing the ratio between the horizontal design force and vertical dead load: where S h - total design effect caused by horizontal force (wind or seismic) in the corresponding design situation for a specific action orientation, G - total dead load of the structure. The horizontal force effect is combined with the dead load effect, taking into account the partial safety factor γ G = 1.0, which is valid in the case of a seismic design situation. Other variable actions such as wind and snow actions are not taken into account during (1) the seismic design situation. Only the quasi-permanent value of the variable action is considered using a combination factor of ψ = 0.3 according to [2]. In the case of different local conditions of a wind load, the basic wind velocity, terrain category, etc., should be taken into account. Representative buildings were examined in all the wind-specific regions in Slovakia (Fig. 7b). 4. BUILDINGS ANALYZED Common types of buildings, such as a family house, a residential brick building, a residential panel building and a high-rise building, were taken into account. Family house The structure examined (Fig. 1) was a one-story family house. Ceramic ceilings with a thickness of about 200mm, along with bearing (350mm) and non-bearing (100mm) masonry walls were assumed. A timber roof with ceramic tiles was also assumed. The house s total dimensions were: 10 m 12 m 7.5 m (width length height). The total weight was 216 tons (240 kg/m 3 ). The soilstructure interaction assuming discrete springs in all three directions acting between the foundation and soil was taken into account. Fig. 1 Family house. 4-story residential building A 4-story residential masonry building (Fig. 2) was chosen as a second commonly used structure in Slovakia. The ceilings were made of reinforced concrete with a thickness up to 200 mm; the bearing and non-bearing walls were made of masonry bricks. A timber roof as a part of the fourth storey was assumed. The building s total dimensions were: 12.4 m 48.9 m 12.5 m. The total weight was 2,102 tons (277 kg/m 3 ). 8-story R/C residential building The next structure (Figs. 3 and 4) was an 8-story building. The structural system is assumed to be a precast R/C wall system. 2

3 2011/3 PAGES 1 9 The ceilings were made of 200 mm thick reinforced concrete; the bearing and non-bearing walls were also made of concrete of various thicknesses. The total dimensions were: 12.6 m 30.0 m 24 m. The total weight equaled 2,800 tons (310 kg/m3). 22 storey high-rise building Fig. 2 Residential building. The structure examined (Figs. 5 and 6) is a 22-story building with 2 additional underground levels. The main vertical structural system consists of two R/C walls at the utmost sides of the building and one longitudinal wall inside the building. The lift cores are also integrated into the internal wall (Fig. 6). The ceilings consist of 220 mm thick concrete. The total dimensions are: 14.6 m 45.6 m 74.2 m. The total weight equals 19,571 tons (396 kg/m3). Fig. 3 8-story R/C residential building ground plan view. Fig. 5 High-rise building structural model. Fig. 4 8-story R/C residential building structural system. Fig. 6 High-rise building detail of modelling. 3

4 5. ASSUMPTIONS OF ANALYSES Total dead load None of the underground levels are included in the calculation of the total weight because the variety of underground levels can considerably influence the results. Basic design situation (including the wind effect) The procedure is precisely defined in [3]. Only the horizontal actions of the wind were taken into account. The National Annex to the Eurocode [3] divides the area of Slovakia into two wind regions (Fig. 7b), where the basic values of the wind speed are defined as v =24 m/s and v =26 m/s, respectively. Further, there are four terrain categories. We only considered category II. (flat areas with low vegetation), III. (villages) and IV. (cities). Wind actions were calculated in both longitudinal and transverse orientations, because there was no predominant orientation of the wind actions. However, in a later comparison, only the wind direction was used in which higher design effects for seismic action were expected. where a g - design ground acceleration on type A ground (rock) λ - correcting factor according to [2] λ = spectral acceleration m - total mass of the structure q - behaviour factor. From Tables 1, 2 it can be seen that if the k w factor is greater than k s, then the location concerned is a site of very low seismicity, because the wind load actions are greater than the seismic ones. The wind effects are summarized in Tables 1 and 2. The values from Table 1 are higher than those in Table 2, so they were taken into account. The values in Tables 3 and 4 are presented here for purposes of comparison. Note: To have terrain category IV surrounding high-rise buildings is very rare. In most cases around 95% of the buildings were located in the I, II, and III categories. (3) Seismic design situation For the seismic analyses and initially for defining the design response spectrum, 5 different soil categories (A to E) and all four different areas of seismic risk (Fig. 7a) were assumed. In the preliminary stage of preparing the National Annex to the Eurocode 8 [2] for Slovakia, there was quite a long discussion about the value of a very low seismicity from a g S = 0.03g up to 0.05g. The design ground acceleration on the type A ground (rock) is: a g = a gr γ l = 0.3 m/s 2 (2) where a gr - reference peak ground acceleration on type A ground (rock), γ l - importance factor. In this paper only locations with the lowest seismic acceleration and with a building importance factor of γ l = 1.0 are considered. All the soil categories starting from class A to class E were taken into account. The first periods were analysed for all the assumed structures; due to the need for generalization, the precise values were not considered, but the intervals of these periods were enlarged, e.g., for the family house, the periods between T 1 = 0 s to s, for the 4-story residential brick building the periods between s, for the 8-story residential panel building periods between s, and for the 22-story highrise building periods between s were assumed. The most unfavourable value / a g (see Fig. 8 trough Fig. 12) was taken into account. The total seismic shear force is: Fig. 7 a) Slovak seismic risk zones, b) maximum wind speed locations (26m/s, 24m/s). 4

5 Tab. 1 Factor for the larger windward structure orientation. Wind speed 24 m/s 26 m/s Terrain category Building 1-storey 4-storeys 8-storeys 22-storeys II III IV II III IV Tab. 2 Factor side. Wind speed 24 m/s 26 m/s Terrain category for the orientation of the lower windward Building 1-storey 4-storeys 8-storeys 22-storeys II III IV II III IV Tab. 3 Behaviour factors and mean periods assumed in the analyses. Number of storeys of building Behavioural factor q T[s] 1-storey building storey building storey building storey building Fig. 8 Spectrum of flexible response for A soil category. Fig. 9 Spectrum of flexible response for B soil category. Tab. 4 Soil categories assumed in the analyses. Soil category A B C D E S Behavioural factor The resistance and capacity for energy dissipation depend on to what extent the non-linear response can be used up. In practice, the balance of the resistance and capacity for dissipated energy is characterized by the behavioural factor q. The seismic action as defined by the design response spectrum can be approximately divided by this factor. The determination of this factor is quite a complicated process. The principles for how to choose a value for a structure are provided in codes [1, 2]. Fig. 10 Spectrum of flexible response for C soil category. 5

6 2011/3 PAGES 1 9 Fig. 11 Spectrum of flexible response for D soil category. Fig. 12 Spectrum of flexible response for E soil category. Structures in which the dissipation of energy is expected only to a very small extent, e.g., masonry structures, belong to the category of limited ductile structures with a minimum behaviour factor of q=1.5. On the other hand, the lower limit of the behavioural factor q for steel and reinforced concrete structures is roughly 2 or higher, and the upper limit is almost 5 or more. The assumed behavioural factors for our types of structures are listed in Table 3. The minimum possible values were taken into account in order to be on the safer side of a design. The input values according to Table 3 were considered in the analyses. 7. COMPARISON OF SEISMIC ZONE MAPS IN CENTRAL EUROPE All the values needed for a comparison of the seismic and basic design situations are summarized in Table 5. Of course, there are many parameters that influence the problem of assessing very low seismicity. One of the most important is a seismic risk map. Discussions about the differences at the borders of adjacent countries in the Central European region have recently started. Problems arise because the maps were not compared in detail during their publication phase. When comparing, e.g., the maps from the National Annexes of countries such as Austria, Slovakia and Hungary, one can see that the differences at borders are still significant. The value of the reference ground acceleration on rock soil agr is not similar considering, e.g., the location near Bratislava, the capital of Slovakia, where we can get the following values: 0.6 m/s2 from the Slovak map (Fig. 7a), Fig. 13 Austrian seismic zone map [7]. Fig. 14 Hungarian seismic zone map [5]. 6. RESULTS 6

7 Tab. 5 Comparison of k w and k s factors. a gr = 1 γ I = 1 a g = 0.3 λ = 0.85 Sub-soil A = 1-storey storey storey storey Sub-soil B = 1-storey storey storey storey Sub-soil C = 1-storey storey storey storey Sub-soil D = 1-storey storey storey storey Sub-soil E = 1-storey storey storey storey m/s 2 from the Austrian map (Fig. 13), 1.4 m/s 2 from the Hungarian map (Fig. 14). A map from Poland either is not available or has not yet been published, not even in [7] where maps from all the European countries are listed. The Slovak map (Fig. 16) from this source [7] is not the original one published in the National Annex in [4]. This discrepancy should also be explained in the future. Possibly some countries (e.g., Hungary) have published peak ground acceleration values (PGA) that are contrary to the effective peak ground acceleration values (EPA) used in Slovakia, Austria and Czech Republic. In [7] it is noted that in Austria, EPA values are calculated as 70% of the PGA value, but the differences cannot be explained just for this reason. The EPA values in the literature are calculated a little bit differently [8]. 7

8 Fig. 15 Seismic zone map of the Czech Republic [7]. For a given return period, the effective peak ground acceleration (EPA) is determined by dividing the corresponding 5% damping short period spectral acceleration value by 2.5 as follows:, (4) where T R is between 0.1 and 0.5 s, which are the first periods of the most commonly occurring structures. Countries such as the Czech Republic, Hungary, Austria, Poland and Slovakia should therefore harmonize their maps at the borders. 8. CONCLUSION From the results in Table 5 it can be seen that the value of very low seismicity (a g ) for the territory of Slovakia can be set as: Fig. 16 Seismic zone map of Slovakia published in [7] but not in [4]. because in this case the seismic effects are always smaller than the wind effects k s <k w. This statement is valid even if the maximum wind velocity is only 24 m/s. This result is important for assessing the value of the very low seismicity areas in Slovakia according to clause (5) in [4]. The Central European countries should harmonize their seismic zone maps not only because they use different values such as, e.g., PGA or EPA, but also because the reference acceleration of the borders at some locations differs greatly. ACKNOWLEDGEMENT We acknowledge the research program VEGA No. 1/1119/11 granted by the Scientific Grant Agency of Slovak Republic. (5) 8

9 REFERENCES [1] STN Seismic action of structures. SÚTN [2] EN Design of structures for earthquake resistance General rules, seismic actions and rules for buildings. Brussels [3] EN Actions on structures. Part 1-4: General actions wind actions. Brussels [4] STN EN /NA. Design of structures for earthquake resistance General rules, seismic actions and rules for buildings. Slovak National Annex. SÚTN Bratislava [5] MSZ EN :2008 NA nemzeti melleklet. Budapest [6] CSN P ENV , National Application Document. [7] A review of the seismic hazard zonation in national building codes in the context of Eurocode 8. JRC Scientific and Technical Reports [8] MATHEU, E.E.; YULE, D.E.; KALA, R.V. Determination of Standard Response Spectra and Effective Peak Ground Accelerations for Seismic Design and Evaluation. US Army Corps of Engineers. December 2005, ERDC/CHL CHETN- VI-41, pp