This action might not be possible to undo. Are you sure you want to continue?
Vancouver, B.C., Canada August 1-6, 2004 Paper No. 3403
RELIABILITY OF EUROCODE 8 SPECTRA AND THE PROBLEMS OF THEIR APPLICATION TO CENTRAL EUROPEAN EARTHQUAKE REGIONS
Corina SCHOTT1 and Jochen SCHWARZ2
SUMMARY This paper will examine the recommended spectrum shapes and spectrum describing parameters of the final version of Eurocode 8 . The re-evaluated basic norm spectra are compared with results of statistical investigations of strong motion data for three different subsoil classes: rock, stiff soil and soft soil. A limited distance of less than 20 km and magnitude ranges staggered in half-magnitude units from Ms = 4.2 to 6.2 will be employed, thus meeting the relevant parameters of Central European design events. A new set of spectrum describing parameters will be derived, supporting the use of two different soil factors for the plateau and zero-period range. An additional aspect of this study is the description of qualitative differences to the concept of geological and subsoil dependent spectra prepared for the new draft of the German earthquake code DIN 4149 . INTRODUCTION In the currently revised version of Eurocode 8 , design spectra based on two different types of seismic action are specified (type 1 for regions with higher and type 2 for regions with lower seismicity), and a new subsoil classification scheme. Consequently, design spectra of type 2 are applicable for Central European earthquake regions. Compared to type 1 spectra for soft soil sites in particular, high soil factors S are defined that result in high amplifications, but in a narrow range of constant spectral acceleration (see Table 1). It must be stressed that most types of residential buildings will be affected by this tremendous increase of seismic design level. The proposed spectra are based on studies of different research groups that use a selected set of European strong motion data. The normalization and smoothing techniques applied to the statistically developed mean spectra do not follow a standardized procedure and are also not proved. Thus, harmonized principles for transferring results of statistical approaches to design parameters, whose emphasis is on the determination of suitable spectrum shapes and soil factors, must be investigated. This, then, is what this paper
Research Assistant at Earthquake Damage Analysis Center, Bauhaus-University Weimar, Germany, Email: email@example.com 2 Head of Earthquake Damage Analysis Center, Bauhaus-University Weimar, Germany, Email: firstname.lastname@example.org
revised version of EC 8. Ms. In this regard.30 is not available. Description of elastic design spectrum as proposed by EC 8  2 .” In the former. the Standard Penetration Test blow-count NSPT should be used for site classification. DESIGN SPECTRA ACCORDING TO EUROCODE 8 Background of seismicity conditions Two different types of spectrum shapes (type 1 and type 2) are considered for varying seismicity conditions. Parameter ag describes the design ground acceleration. ” No worst case scenario should be taken into account: ”In selecting the appropriate shape of the spectrum. The range between corner periods TB and TC constitutes the branch of constant spectral acceleration. These are soils which produce both irregular amplifications of ground motions and soil-structure interaction. The main distinguishing feature is the average shear wave velocity Vs. and also require special foundations due to their liquefaction potential.30 that exists in the uppermost 30 meters of the subsoil layers. Subsoil classification As shown in Table 1. Horizontal elastic response spectra Using equations (1) to (4) and the parameters of Table 2. provisions of Eurocode 8  state: ”If the earthquakes that contribute most to the seismic hazard defined for the site for the purpose of probabilistic hazard assessment has a surface wave magnitude. considering all the while the seismic action type for low seismicity regions in more detail. S is the soil factor.attempts to prove. and η represents the damping correction factor.0. In addition. maximum credible earthquake) defined for that purpose. constant spectral displacement starts at control period TD. Especially ambiguous and critical subsoil conditions that require separate measurements of site examinations are considered by classes S1 and S2. it is recommended that the Type 2 spectrum is adopted. Figure 1. If Vs.5. the threshold of surface wave magnitude between type 1 and type 2 spectra was specified as Ms = 6. five different subsoil classes (A-E) are defined. whereas periods TC and TD are the limits of the constant spectral velocity branch.g. consideration should be given to the magnitude of earthquakes that contribute most to the seismic hazard defined for the purpose of probabilistic a hazard assessment rather than on conservative upper limits (e. not greater than 5. The value cu describes the shear strength of the soil that has not been drained. elastic design spectra for different seismicity conditions and subsoil classes can be created (Figure 1).
or any other soil profile not included in types A-E or S1 < 100 10 -20 NSPT (blows/30cm) cu [kPa] A > 800 - - B 360 .250 D < 180 < 15 < 70 E S1 S2 Table 2.5 ⋅ ⎣ ⎢ ⎡ ⎦ ⎥ ⎤ Se (T ) = ag ⋅ S ⋅ η ⋅ 2. at least several tens of meters in thickness and characterized by a gradual increase of mechanical properties with depth Deep deposits of dense or medium-dense sand.30 values of type C or D. gravel.360 15 . or very stiff clay.800 > 50 > 250 C 180 . and thicknesses varying between 5 m and 20 m.5 − 1) TB Se (T ) = ag ⋅ S ⋅ η ⋅ 2. or stiff clay with thicknesses from several tens to many hundreds of meters Deposits of loose-to-medium noncohesive soil (with or without some soft cohesive layers).5 ⋅ ⎣ ⎢ ⎡ 2) TC T (3) TC ⋅ TD T2 (4) 3 . sensitive clays. Equations for elastic design spectra according to EC8  Period range 0s ≤ T ≤ TB TB ≤ T ≤ TC TC ≤ T ≤ TD TD ≤ T ≤ 4s Equation S e (T ) = a g ⋅ S ⋅ 1 + ⎦ ⎥ ⎤ Equation (1) T ⋅ (η ⋅ 2. Definition of subsoil classes according to EC 8  Parameters Subsoil Description of stratigraphic profile Vs. or of predominantly soft-to-firm cohesive soil Soil profile consisting of a surface alluvium layer with Vs.Table 1.30 [m/s] Rock or other rock-like geological formation.30 > 800 m/s Deposits consisting or containing a layer at least 10 m thick of soft clays/silts with high plasticity index (PI > 40) and high water content Deposits of liquefiable soils. underlain by stiffer materials with Vs.50 70 . gravel.5 ⎦ ⎥ ⎤ ⎣ ⎢ ⎡ Se (T ) = ag ⋅ S ⋅ η ⋅ 2. including at most 5 m of weaker material at the surface Deposits of very dense sand.
25 Period TD [s] Type 1 2.25 0.30 [m/s] > 800 360-800 180-360 < 180 Soil factor S Type 1 1. and epicentral distance Repi. soil factor S for subsoil class A (see Table 3) is fixed to 1. Regardless of this. recordings used for seismicity condition type 1 and type 2 are selected respectively above and below the threshold of magnitude.30 0.2 1.2 Development of soil factors in drafts of EC 8 It is common knowledge that soil factor S represents the ground motion amplification due to the presence of (soft) subsoil layers in relation to geological bedrock (half-space).5 0.4 0.20 0. For class C.6 Period TB [s] Type 1 0.10 0. the distribution of empirical data used to determine the control periods of EC8  is shown in Figure 3. a) Type 1 b) Type 2 Figure 2.05 0.0 1.0 1. It becomes clear from Figure 3 that the threshold value demarcating type 1 and type 2 spectra was defined as Ms = 6. the threshold of magnitude Ms is given for the two seismicity conditions. 4 .0 in .0 Type 2 1.35 1.0 2.2 1.15 1.0 2.8 1.25 0.5 Type 2 0. the factor was even lower than for class A in the former version of EC8 .0 2.35 1. In the course of time the soil factor for subsoil conditions with Vs. Parameters of elastic design spectra for different subsoil classes  Subsoil A B C D E Vs.6 0.10 0.20 0. the soil factor was modified mainly for subsoil classes B and C.8 0. During the investigations of the newly revised EC8 version.Table 3.05 0.15 0.4 Type 2 1.2 1.15 Type 2 0.0 2.25 0. Development of soil factors in the different drafts of EC8 EVALUATION OF STRONG MOTION DATA Investigation of earthquake data used in  In dependence on surface wave magnitude Ms. Hence.5 1.2 1.30 < 800 m/s has considerably changed.05 Period TC [s] Type 1 0.15 0.0. Furthermore.2 1.
“ For each record the envelope of the 5% damped horizontal acceleration spectra is determined and normalized to the larger value of PGA. response spectra for a damping of 5% were calculated. STATISTICAL INVESTIGATIONS OF EMPIRICAL DATA Assessment of proposed spectrum shape with empirical data A total of four different statistical approaches for deriving the proposed spectrum shape by empirical data are described below. only three were unavailable. It can clearly be seen that waveform no. Magnitude-epicentral distance relationship of evaluated strong motion database An overview of the selected earthquakes. type 2. Plausibility check of strong motion data The time histories and critically damped (ξ = 5 %) response spectra of the strong motion data listed above were examined with the following criteria to determine their plausibility: shear wave velocity (subsoil conditions. Off all the strong motion records used here. Problems with strong motion data that repeatedly occur are caused by nonlinear effects within the time histories and by missing information on the pre-event time segments. capability of amplification. The average of each set of spectra is then determined and compared with the proposed spectral shape for the corresponding subsoil class. Since the availability of only three earthquake records used for statistical investigation of control periods for classes D1 and D2 was constricted. reliability of derived results may not be given.Figure 3. nonlinear effects of the subsoil. 0055 possesses a much higher value of peak ground acceleration and a higher amplification than all other recordings of class A. frequency content. and the number of strong motion records that are used for the evaluation is given in Table 4. Aside from this. The above mentioned strong motion data used to derive the spectrum shape in  will be denoted as “empirical data” for the rest of the paper. Method 1: envelope of horizontal components Henv The first approach is taken over by the procedure applied in . Examples illustrating these effects are given in Figures 4 and 5. Irrespective of the method used for the single time histories.” 5 . their regions. magnitude and distance conditions. the time history of Figure 5 shows an unusually long-period content for a rock station.
Gr = Greece.Table 4. Evaluated strong motion data Subsoil class Spectrum type (No. Yu = Yugoslavia. 0055) with longperiod contents 6 . Montenegro (4) Kalamata (1) Plati (1) Ionian (1) Dursunbey (1) Urmiya (1) Umbria (4) Lazio Abruzzo (1) Killini (1) aftershock Spitak [It] [It] [Yu] [Yu] [Gr] [Gr] [Gr] [Tr] [Ir] [It] [It] [Gr] [Ar] D1 / D2 (3) (1) Gazli (1) Basso Tirreno (1) 2) [Uz] [It] E1 (5) (5) Campano Lucano [It] 1) 2) It = Italy. Unscaled response spectra for subsoil class A. of earthquakes) name [region] A1 (5) (3) aftershock Friuly (2) Campano Lucano [It] [It] A2 (9) (2) Friuly (7) aftershock Friuly [It] [It] B1 (8) (2) Campano Lucano (2) aftershock Friuly (2) Montenegro (2) aftersh. spectrum type 2 Figure 5. Time history (waveform no. of earthquakes) name [region]1) Subsoil class Spectrum type (No. Tr = Turkey. Uz = Uzbekistan data for this earthquake are not included in ESMD 2000  Note: records which were not available are italicized Figure 4. Ar = Armenia. Montenegro [It] [It] [Yu] [Yu] B2 (24) C1 (11) (3) Tabas (2) Alkion (2) Ionian (1) Spitak (3) Manjil [Ir] [Gr] [Gr] [Ar] [Ir] C2 (10) (1) Friuly (9) aftershock Friuly (1) Montenegro (8) aftersh. Ir = Iran.
type 2 b) Class B. 2 2 S a . type 2 Figure 6. It can be seen that determining control periods does not depend on the type of method used. type 2 c) Class C. Hres (T) = max t S a . a) Class A. Figure 6 shows the mean response spectra in comparison to the proposed design spectra for seismicity condition type 2. Method 4: horizontal component Hres Here a horizontal resultant response spectrum is calculated according to Equation 5. Finally. The SRSS method is applied to superimpose both horizontal components H1 and H2.0 m/s² and used to calculate both the mean and the median spectra for each subsoil class.0 m/s² and used to calculate both the mean and the median spectra for each subsoil class. the response spectra for both horizontal components H1 and H2 are normalized to PGA = 1. were analyzed with respect to the techniques mentioned above. t ) (5) Mean spectra of empirical data The empirical strong motion data of each subsoil class and spectrum type. these response spectra are also normalized to PGA = 1.0 m/s² and used to calculate both the mean and the median spectra for each subsoil class.Method 2: horizontal components H1+H2 For this statistical investigation. as used in  to describe the shape including the control periods of the recommended design spectra. H1 (T. t ) + S a . Only small deviations occur between the four different methods. Method 3: horizontal component H1 Only the horizontal component having the larger value of PGA (H1) is normalized to PGA = 1. Mean response spectra determined by different methods 7 . H 2 (T. type 2 d) Class D.
98 ± 5. C.92 10.95 24. Equation 6 was applied to the H1 component of each earthquake recording.56 11.22 5. For classes B.73 7. and t1 is the entire duration of the time history. and D do not cover the empirical response spectra shape (Figure 6).22 ± 0.12 5. In contrast. a constant value of 2.84 4.87 ± 0. the recommended control periods TC (right margin of constant spectral acceleration range) for subsoil classes B.84 ts.70 ± 10. The final draft of EC8 also indicates that the duration should be consistent with magnitude and important features of the proper seismic event.95 Type 1 12. the formulation of modified control periods is necessary.84 4. and D.5 for the amplification of peak ground acceleration seems to be suitable.30 ± 1.90 Strong motion duration ts ± Standard deviation σ ts. which can be used for design purposes.57 ± 4. No data was evaluated for subsoil class E and spectrum type 2 within the scope of . corner period TB has increased and only changed for classes B and D.04 14.23 ± 13.75 and ts. a minimum duration of 10 seconds should be used.76 1.66 3.67 6. Two different investigations are applied to assess the end of strong motion duration tE (75%) and tE (90%).17 ± 4.02 4. Nevertheless.87 ± 0. Strong motion duration ts.95 ± 5.90 [s] ± σ Type 2 2. This is certainly the consequence of only three evaluated earthquake recordings.24 ± 0.49 30. t t1 0 Subsoil class Type 1 A B C D E Artificially generated accelerograms.40 ± 2. Strong motion duration To calculate the strong motion duration. If no detailed information is available. It is recommended that the strong motion duration in dependence on the seismicity conditions be considered as well. where E(t) is the energy-integral. The analysis in Table 5 indicates a rather big difference in strong motion duration for type 1 and type 2 spectra.52 5. CONTROL PERIODS OF PROPOSED DESIGN SPECTRA Validity of existing control periods Except for subsoil class A. Figure 7 shows the existing periods TB and TC (filled marks) compared with the suggested ones (blank marks) for all subsoil classes and spectrum type 2. C.75 [s] ± σ Type 2 1. The results of the energy-integral are listed in Table 5.83 ± 1. values of corner period TC increasingly changed.20 ± 14. should comply with the given design spectra of EC8. mean spectra of class D2 show variations in terms of a smoothed progression. The values of control periods TB (left margin of constant spectral acceleration range) can be slightly shifted to longer periods.Compared with the other classes. The beginning of strong motion duration is defined as tA (5%). ∫ 0 E( t ) = a 2 ( t )dt ∫ a 2 ( t )dt Table 5.78 ± 1.24 ± 0.12 ± 5.49 (6) 8 . Formulation of modified control periods Because of the deficiencies mentioned above.05 ± 15.
Variations of control periods TB and TC for spectrum type 2 Evaluation of periods with statistical investigations of the European Strong Motion Database (ESMD 2000) A triply-logarithmic diagram is used to compare the newly proposed spectrum shapes with mean response spectra of empirical data (component H1+H2) and statistical investigations of European Strong Motion Database (ESMD 2000) .7 6. data was selected according to magnitude and distance ranges compatible for German earthquake regions.7 ≤ Ms ≤ 5.Figure 7. The evaluation of various ESMD 2000 data sets shows smoothed mean spectra of higher quality than investigations of empirical data from  do. data groups for an epicentral distance Repi ≤ 20 km were assembled from ESMD 2000  in conjunction with the following: a) b) c) d) 3. Figure 8 contains the newly defined shape of design spectra compared with mean response spectra of empirical data and analyses of ESMD 2000 in dependence on the subsoil class. stiff soil. The mean response spectra of data groups (b) and (c) are displayed for three different subsoil classes in this paper.2 4. Only strong motion recordings of free field stations were included in the extensive statistical investigations.7 4.2 ≤ Ms ≤ 5. As a result.2 ≤ Ms ≤ 6. In  and . and soft soil.7 ≤ Ms ≤ 4. 9 . since the input parameters are applicable for comparison with EC8 type 2 spectra. The position of control periods can be assessed with certainty.2 The subsoil conditions are limited to rock.
10 . a comparison with the approach for determining soil factors proposed in ATC 35-1 is applied (Figure 2). but since only the shape is explored and since the factor S scales the ordinates at all periods equally. Mean response spectra. “ The exercise presented here is useful to explore the validity of the control periods established for each subsoil class. this investigation does not serve to confirm the proposed values for S. the strong motion data used are not comprehensible.a) Class A b) Class B c) Class C b) Class D Figure 8. Nevertheless.” Approaches with new data sets are conducted independently to develop soil amplification factors. investigations of different methods for validity of existing and developing of new soil factors are employed. Application of different methods for developing soil factors Approach 1: Classification of proposed soil factors referring to the procedure of ATC 35-1 To classify the suggested soil factors in EC8. Therefore. type 2 (component Hres) SPECTRAL SOIL AMPLIFICATION FACTORS As stated in . empirical data is only used to determine the control periods.
Thus their application for own investigations cannot be realized. Equation 8). To derive Fa and Fv.5. 5.5. making further investigations necessary. The results of applying the Housner-Intensity method to these data are shown (Table 6) in comparison with proposed soil factors. Comparison of soil factors with provisions given in ATC 35-1 In order to investigate the local geological conditions and estimates of amplification factors Fa and Fv in ATC 35-1.a) Type 1 b) Type 2 Figure 9.0. constant spectral velocity) is made.01 is much higher than S(EC8) = 1. either a discrete or a continuous function can be used. As it can be seen in Figure 9a (EC8 .30 = 180-360 m/s is rather unusual. S could not be approved. They also used a data set of strong motion recordings provided by ESMD 2000. Figure 9 shows a decrease of the soil amplification factor (for 0.4-2.0. Unlike the procedure used in EC8. only strong-motion recordings of Loma Prieta earthquake are used. Approach 2: Intensity based soil factors with data of ESMD 2000 The final draft of EC8 includes soil factors recommended by .5 I A . 6. In fact.5<M<6.5 s.0 s.05 ∫ R ⋅S a (T)dt Ic IA (7) Sc = (8) The ESMD 2000 data base was searched for recordings with certain magnitude ranges (Table 6) and limited epicentral distance REpi ≤ 20 km in order to meet the relevant conditions of Central European earthquake regions.0<M<4. is not given. For stiff soil S(ESMD) = 0. Nevertheless. constant spectral acceleration) and mid-period ranges (0. this fact is reflected in EC8 with different soil factors for two different seismicity conditions.0<M). this decrease is stronger than for mid-period motion. 11 . whereas a distinction between short-period ranges (0. 2.Type 1).35.1 g) with increasing shear wave velocity (ATC 35-1) . the dependence of the soil factor on peak ground acceleration is shown for the continuous function method. A detailed documentation of the applied data. the Housner spectrum intensity according to Equation 7 was calculated and finally related to geological condition rock (cf. no difference in the seismicity conditions is implied in ATC 35-1 for the creation of response spectra. 4. which is staggered in five different magnitude ranges (4. whereas for soft soil S(ESMD) = 2. By using the average normalized spectral curves for each subsoil class. The higher the peak ground acceleration.5. the lower the soil amplification factor. B.5<M<5. a break between Vs.98 is considerably lower than S(EC8) = 1. Since the achieved results are not continuously and specified soil factors.0<M<5.1-0. For short-period motion.C = SB = IB IA 0. 5.
0 5. Results of predicted soil factors on the basis of Housner spectrum intensity Subsoil rock (A) Magnitude 4.73 13.5 4.5 4.0 stiff (B) soft (C) Approach 3:Relations between mean spectra of empirical data The third approach was carried out on the basis of empirical data . a) Class B/A b) Class C/A c) Class D/A Figure 10. is shown in Figure 10.0 5.5s) (ESMD) 10.74 21. and compared with appropriate soil factors S of EC8.01 1. Mean response spectra of empirical data in relation to rock.0 ≤ Ms ≤ 5.45 8.Table 6.62 S (ESMD) 1.07 1.72 1.5 ≤ Ms ≤ 5.5 ≤ Ms ≤ 5. The relation of the average values of spectral acceleration in dependence on the period between soil and rock conditions.52 1.41 13.0 ≤ Ms ≤ 5.04 41.37 3.5 I(T=2.98 1.35 S (ESMD) S  I(T=2.37 0.5 10.5s) (ESMD) 11.25 0.0 S (EC8) Type 2 1.16 1.0 ≤ Ms ≤ 5.0 5.31 2. For each subsoil class of type 2 spectra. the mean acceleration response spectra of raw data were calculated for component H1+H2.67 16. type 2 (component H1+H2) 12 .5 ≤ Ms ≤ 5.15 1.02 1.
Scaling factors as generated using the attenuation relationship of Ambraseys et al.2 4. Mean response spectra of ESMD 2000 data related according to rock (component H1+H2) 13 . While the curves show a much more balanced progression. a scaling factor is generated (Table 7).7.1 ASB 1996 Sa (T=0.2 ≤ Ms ≤ 5.5 11. As the author`s recommendation is to adopt two different soil factors S1 for period range 0.2 4.2 ≤ Ms ≤ 5.Approach 4: Evaluation of varying soil factors for different spectral ranges with statistical investigations of European Strong Motion Data (ESMD 2000) For this type of soil factor.04s) [m/s²] 0.7 5. large differences between S1 and S2 can be recognized.7 4. By means of attenuation relationships as proposed by Ambraseys et al. As the results of Approach 3 have already shown.9417 0. they display a nearly constant amplification level for periods larger than TC.7 5.01 ≤ T ≤ TB and S2 for period range TB ≤ T ≤ TC.5630 0.  Subsoil Magnitude 4.7 ≤ Ms ≤ 5. Since mean values of amplification factors do not tend to increase continuously from stiff to soft soil. a unique scaling of magnitude and distance conditions (of raw data) is inevitable.1 REpi [km] 11.7491 0.7 5.9 12. Ms = 4.9570 stiff (B) soft (C) The mean spectra of ESMD 2000  and  were standardized by the PGA -values given in Table 7. From period TB to TC.7142 0.2 11. The soil spectra were related to rock spectra (Figure 11) thereafter.1 4. The mean value of surface wave magnitude Ms and epicentral distance Repi was detected for ranges Ms = 4. .7 12.5 11.7 Mean value ESMD 2000 Ms rock (A) 4. the results of Figure 11 were analyzed for distinct period ranges (Table 8). evaluation data of ESMD  is consulted again. and Repi ≤ 20km. the PGA -values of respective attenuation relationship response spectra (ASB 1996) were calculated.2 ≤ Ms ≤ 5.7 ≤ Ms ≤ 5.2-5. Based on these values and the corresponding subsoil classification.7 ≤ Ms ≤ 5.7-5.7 4. a nearly linear increase of the soil amplification can be observed. Table 7.2.2 4.2 4. a) Class B/A b) Class C/A Figure 11.7932 0.
7 4.24 TB ≤ T ≤ TC 1. 2. Shear wave velocity of the near-surface layers Vs. 2. Soil material types are represented by three different soil condition classes: 1. 14 . SC 2: 350m/s ≤ Vs. Finally it became evident during these examinations that consistent data sets are imperative for statistical investigations.2 4.7 ≤ Ms ≤ 5. SC 3: Vs. mostly Quaternary alluvial layers reaching depths between 100 and 1000 m. Geological subsoil class A can be characterized by missing or overlying sediments with a maximum thicknesses of 25 m. Geological subsoil class B represents the transition zones between classes A and C. it could be shown that control periods TB. Four different methods were introduced to calculate the mean response spectra.25 ≤ 800m/s.25 can principally be regarded as the main distinguishing parameter between soil condition classes 1. B.Table 8. Inconsistencies appeared when comparing the proposed spectrum shapes with those resulting from statistical investigations of the empirical data as well as from the analysis of ESMD 2000 data  and .51 1. new data sets of strong motion recordings were used while different approaches were applied to derive the soil factor S.25 > 800m/s. the determination of appropriate soil amplification factors S implies general problems and requires further consideration.13 1. Mean values of amplification factor for different period ranges Subsoil stiff (B) Magnitude 4. Therefore. In almost the same manner as the Housner spectrum intensity  was implemented. Geological subsoil classes (A. and TD do not depend on the applied method. TC. and 3 (SC 1: Vs. The evaluated earthquake time histories showed some deficiencies.2 ≤ Ms ≤ 5. and C) consider the geological subsurface conditions.61 1. whereas class C is described by deep. as well as shallow basin structures with thicknesses of sedimentary layers between 25 and 100 m. CONCLUSIONS A selection of strong motion data regarding the recently introduced seismicity conditions was used in  to determine the shape for design spectra in the final draft of EC8 . the provisions given in ATC 35-1  and the relations of scaled and unscaled mean spectra between geological soil and rock conditions using several data sets did not induce consistent results.35 soft (C) Approach 5: Concept of geological and subsoil dependent spectra of German code The present draft of the German earthquake code DIN 4149  not only regards the influence of a globally described subsoil. but also the geological profile at a certain depth. Recommendations were given to shift the periods confining the range of constant spectral acceleration TB and TC to longer periods.01 ≤ T ≤ TB 1. In order to estimate reliable soil factors.2 4. and hence stand for the thickness of overlying sediments. It is the author`s opinion that the concept of geological and subsoil dependent spectra proposed in the German seismic code DIN 4149  is probably the best to use.7 0. and 3. In contrast.5 EC8  1.7 ≤ Ms ≤ 5.27 1. A twofold classification is carried out distinguishing between the type of soil materials of the uppermost 25 m of soil layers and their total thicknesses above geological bedrock. among which an innovative type is included using the resulting horizontal component.25 < 350m/s).65 1. The geological conditions within Germany’s earthquake regions admit six possible combinations between the different geological subsoil classes and soil condition classes.62 1.2 ≤ Ms ≤ 5. investigations on borehole profiles are strongly recommended.29 1. However.
Schwarz J. Lastannahmen. 2000. 1. 1994. Borcherdt RD. “Prediction of horizontal response spectra in Europe”. Ambraseys NN. Comparison of soil factors as provided by DIN 4149 and EC8 DIN 4149  Subsoil conditions *) A1 A2 A3 B2 B3 C3 Soil factor S 1.1. Juni 2001. Eurocode 8.und Entfernungsbereichen: Fallstudie für felsige Untergrundbedingungen”.5 resp. Draft No.V. p. Proceedings of seminar on new developments in earthquake ground motion estimation and implications for engineering design practice. (type 2) 1. Forschungsbericht im Auftrage der Deutschen Gesellschaftfür Mauerwerksbau e. 4. Bommer JJ.” Schriftenreihe der Bauhaus-Universität Weimar 2004. Habenberger J. Journal of Seismology 2002. seismic actions and rules for buildings”. DIN 4149 (2001). Zeitschrift der Bauhaus-Universität Weimar 2001. Ambraseys NN.8 resp.35 1. June 1994.: 80-91. Bommer JJ.p. Appendix B: Subsoi/ classes and spectral amplification factors. E B D C resp.. “New Developments in estimating site effects on ground motion”.8 *) change of denotation in the final version (2004) is to be expected to achieve a better correlation with EC8 REFERENCES 1. ATC 35-1. Eurocode 8. 3.0 1. “Dissemination of European Strong-Motion-Data”. 9. CD-ROM-Collection. 8. 1. Juli 1999 (1. Lang DH. Eurocode 8: Commentaries on proposed Amendments to EC 8 -Design provisions of earthquake resistance of structures.75 Subsoil conditions A B D resp. Schott C.0 1. Faccioli E. seismic actions and general requirements for structures. 15 ./2. Schwarz J. ”Design provisions for earthquake resistance of structures.5 1. wiss. Rey J. D EC8  Soil factor S. “Auswertung von Strong-Motion-Daten in den für deutsche Erdbebengebiete maßgebenden Magnituden. 2. Smit P. 371-400. Berardi R. 1996. Cotton F. “Erarbeitung von Spektren für die DIN 4149-neu unter Berücksichtigung der Besonderheiten deutscher Erdbebengebiete und der Periodenlage von Mauerwerksbauten. 25. 116. December 2003. “Auswertung von Strong-Motion-Daten in den für deutsche Erdbebengebiete maßgebenden Magnituden. Entwurf). 6: 547555 10. 6. Norm-Vorlage. 11.und Entfernungsbereichen: Fallstudien für weichen und steifen Untergrund. Rinaldis D. Bemessung und Ausführung üblicher Hochbauten”. Erdbebenzentrum am Institut für Konstruktiven Ingenieurbau der Bauhaus-Universität Weimar.”. Berge C.8 1. German version. ENV 1998-1-1 (1994). Part 1: General rules. Golbs Ch. prEN 1998-1 (2003). 5. Final Draft.25 1. “Derivation of design soil coefficients ( S) and response spectral shapes for Eurocode 8 using the European Strong-Motion Database”. 1. December 1999. Habenberger J.0 1. Earthquake Engineering and structural dynamics. “Design of structures for earthquake resistance. Schwarz J. September 1999 (Endfassung). Schott C. 7. “Bauten in deutschen Erdbebengebieten. Vol. Part 11: General rules – Seismic actions and general requirements for structures”.6 1. Simpson KA. Thesis.Table 9. Part 1-1: General rules.25 0.35 1.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.