Vertical seismic profiles for Bucharest

Significant damages of the built environment recorded during past seismic events have led to considering Romania’s capital city as one of the major earthquake-prone urban areas worldwide. A large number of shallow and deep boreholes and non-invasive field techniques such as seismic down-hole or MASW have been carried out in Bucharest sites only by CNRRS (now https://ccers.utcb.ro) and UTCB staff and the conclusions are presented in this paper. Shear wave velocities (V_S) have been set as the main indicator in quantifying site classification in the codes for the seismic design of structures. The comparison among site investigation results with the proposed satellite USGS V_s map for Bucharest is also presented. The end results can be considered as efficient guidelines to predict the potential effect of site conditions on similar soil types, layer sequences and properties, and might be useful for the evaluation of buildings’ safety and optimization of seismic risk management strategies.

INTRODUCTION

Destructive seismic events occurred worldwide during 20^th century: 1940 – El Centro (M_w=6.9), 1964 – Niigata (M_w=7.6), 1971 – San Fernando (M_w=6.7), 1985 – Michoacan (M_w=8.0), 1989 – Loma Prieta (M_w=6.8), 1994 – Northridge (M_w=6.7), 1999 – Kocaeli (M_w=7.6) and 21^st century: 2003 – Tokachi-Oki (M_w=8.3), 2008 – Sichuan (M_w=7.9), 2010 – Christchurch (M_w=7.1), 2010 – Chile (M_w=8.8), 2011 – Tohoku (M_w=9.0). Data demonstrated that the distribution of severe structural building damages in a specific area is often controlled by the surface geology and the effect of local soil conditions. The seismic codes, regulations and standards (Uniform Building Code, 1997; International Building Code, 2009; Building Standard Law in Japan, 2000 Romanian P100-1/2013, NEHRP 2003, ASCE 7-22, EN 1998-1: 2004) include seismic provisions regarding the consideration of site conditions. In the mentioned codes, site effects are either quantified by seismic response coefficient linked to soil category and seismicity level and/or through different spectral shapes specific for defined soil types. Generally, ground conditions refer to soil classes differentiated by qualitative criteria such as soil type and lithological profile and quantitative ones as average shear wave velocities and penetration resistance values.

Bucharest city is the most affected urban concentration by Vrancea subcrustal earthquakes, with a high density of building damages, casualties and economic loss due to its relative proximity to the seismic source and to the specificity of surface geology. Major historical seismic events generated by Vrancea source (1802: M_w=7.9; 1940: M_w=7.7 and 1977: M_w=7.4) have indicated the great influence of characteristics of soil layers on seismic motion parameters. The surface geological deposits from the Bucharest area are composed of unconsolidated alluvial layers of cohesive and cohesionless soils with significant variability in thickness and spatial distribution. The relative heterogeneity of young formations in an alluvial basin explains the peculiar site response during Vrancea strong motions.

In recent decades, due to the upgrading and extending of seismic networks, modern equipment used for data recording, storage and real-time transmission, development of specialized software for scenarios and seismic response modelling, as well as improvement of ground investigation techniques, the studies concerning local site effects assessment on Vrancea strong ground motions have substantially intensified (LUNGU et al., 2000, ALDEA et al., 2003, ARION et al., 2007, 2012, PAVEL et al., 2015).

The present paper is in line with the international practice approach by providing reliable data obtained from detailed surveys performed on different areas in Bucharest and proposing empirical correlations of specific indicators (V_S) for site characterization to be further integrated into seismic response studies.

METHODS USED FOR INVESTIGATION OF SURFACE GEOLOGY

To assess the site effects of near-surface layered structures on seismic ground response, an accurate determination of soil characteristics beneath a target site is required. Usually, site characterization in calculating seismic hazard is governed by shear wave velocities values (V_S). V_S values are also considered one of the most important input data for soil liquefaction potential or for soil-structure interaction analysis. The use of V_S has the advantage of being based on an objective measure which affects ground motion in a way that can be modelled. Conventional criteria used for earthquake engineering design purposes (BORCHERDT, 1994) are typically based on the weighted average of shear wave velocities in the upper 30 m of surface soil stratigraphy (V_S,30).

Bucharest city is located in the central part of the Moesian Sub-plate (age: Precambrian and Paleozoic), in the Romanian Plain at the north of Danube. Over Cretaceous and Miocene deposits (having the top at about 1,000 m depth) a Pliocene shallow water deposit (~700 m thick) was settled. Later loess covered these deposits and rivers shaped the present landscape. The surface geology consists mainly of Quaternary alluvial deposits. Inferior Pleistocene deposits consist of clays, fine sands and gravels. The layers of sand and sand with gravels are also called „Fratesti” layers. Medium Pleistocene deposits consist of marls and clays with insertion of sands and clayey sands. Superior Pleistocene deposits consist of alternation of clays, sandy clays with fine and medium sands („Mostistea” sands). Holocene deposits consist of yellow clays, sandy clays, clays with gravel, boulders, gravels, and gravels with boulders. The surface geology can be divided into seven lithological formations, from surface to bottom (LITEANU, 1951): (i) backfill (thickness h up to 3 m) and (ii) sandy-clay superior deposits (loess and sand, h=3÷16 m), both formations from Holocene, and other formations from Pleistocene: (iii) „Colentina” gravel (gravel and sand, h=2÷20 m); (iv) intermediate cohesive deposits of lacustral origin (80% clay and some sand, h=0÷25 m); (v) „Mostistea” banks of sands (mainly sand, sometimes lenses of clay, h=10÷15 m); (vi) lacustral deposits (clay and sands, h=10÷60 m), and (vii) „Fratesti” gravel (gravel and sands separated by clay, h=100÷180 m).

1. Seismic down-hole method profile

The equipment for soil testing and investigation, the data acquisition and processing systems and triaxial testing equipment is located at the Seismic Risk Assessment Research Center, Technical University of Civil Engineering; formerly installed at the National Center for Seismic Risk Reduction Bucharest, Romania (NCSRR), it was donated by Japan International Cooperation Agency (JICA) through the Technical Cooperation Project on the Reduction of Seismic Risk for Buildings and Structures in Romania (2002-2009).

One of the low-strain field tests is PS Logging, a seismic down-hole technique. In the seismic down-hole method the sensor (fig. 1a) is placed at various depths in the borehole and the source of vibrations is above the sensors – usually at the surface. This technique does not require as many borings as the cross-hole method, but the waves travel through several layers from their source to the sensors. Thus, the measured travel time reflects the cumulative travel through layers with different wave velocities, and interpreting the data requires sorting out the contribution of the layers. Since S and P wave velocities are calculated from the slope of a depth/travel time curve, the velocities are not for each incremental interval but for a velocity layer that has a certain thickness (including several measuring points as average values). The P-waves are generated by hitting a wooden pile with a large wooden hammer (as shown in fig. 1b), and S-waves are generated by horizontally hitting the end of a plank with the same hammer (as shown in fig. 1c).

Figure 1: a – borehole sensor for PS logging, b – generation of P-waves, c – generation of S-waves

UTCB conducted seismic down-hole tests in more than 35 sites in Bucharest with an investigated depth of up to more than 150 m. During seismic down-hole measurements, the sensor was lowered in the borehole up to a predetermined depth investigation, being blocked on the boring wall at 1 m intervals for detecting the waves generated by the surface source. After processing field data measurements and determining V_s profiles, and after the V_s,30 values were calculated, according to the seismic codes provisions, the average shear wave velocity of the upper 30 m can be calculated with the Eq. (1). The statistical parameters of the Bucharest seismic down-hole averaged shear wave velocity are presented in Table 1.

where: d_i and V_si denote the thickness (m) and shear wave velocity of the i-th layer from the upper 30 m.

Table 1: Shear wave velocity from seismic down-hole measurements averaged on different depths (30 m, 50 m, 70 m, 100 m)

The values of shear wave velocities of each soil layer gathered from seismic down-hole measurements were grouped and statistically analyzed for estimating a potential correlation and relation with the depth of the measured shear wave velocity. Using a nonlinear regression, one can observe a strong correlation of V_S values calculated for each depth for the cohesive – Eq. (2) – and granular – Eq. (3) – soil types, reflected by coefficient correlation of about R²=0.81, respectively R²=0.86, and by standard error of the estimate: 36.22 m/s for cohesive and 32.29 m/s for granular (fig. 2). The figure reveals a relatively low increase of S-wave velocities ranging from 10% up to 15% for most of the sites. For deep depth measurements, the increase of V_S values calculated for the total investigated thickness of soil layers can reach 20-30%, so it can be mentioned that the thickness of sedimentary layers intercepted in boreholes can represent an important factor in velocity profiles, especially in case of deep alluvial deposits.

V_s = 63.943 + 80.887ln(h)           for cohesive soil layers                          (2)

V_s = 67.058 + 84.166ln(h)           for granular soil layers                           (3)

Figure 2: Bucharest – seismic down-hole measurement of shear waves velocities versus depth

2. Multi-channel analysis of surface waves (MASW) profile

An alternative technique to obtain the S-wave velocity profile at near-surface is the multi-channel analysis of surface waves (MASW) method, in which the dispersion character of Rayleigh waves is analyzed (HAYASHI et al. 2004). The method was applied to Bucharest sites (ARION et al., 2007). By using data recorded during MASW tests, the V_S profile (2D model), is obtained as in fig. 3.

Figure 3: example of 2D and 1D shear wave velocity profiling for MASW method in Bucharest

For several sites where geophysical survey was conducted by UTCB through both seismic down-hole and MASW methods, a comparative analysis of V_S values corresponding to each depth interval in soil profile and average V_S,30 has been performed, as shown in fig. 4a. It can be observed that data collected from MASW application are grouped in a constant interval velocity 150-250 m/s comparing to a larger and gradually increased one obtained from the seismic down-hole technique, fig. 4b.

Figure 4: Bucharest – VS data comparison from down-hole (blue) and from MASW (orange)

Differences between V_S,30 values obtained in seismic down-hole versus the ones from MASW surveys range from 18-52%, confirming the low-quality results from MASW measurements. The use of the MASW method must be restricted to the sites where the investigated depths do not exceed 20 m or where no more than 3 soil layers have to be measured and also must be mentioned the underestimated values of the shear wave velocity.

SITE-SPECIFIC DATA FROM IN SITU INVESTIGATIONS IN BUCHAREST CITY

Considered a reference index of dynamic behaviour at small-strain levels, V_S,30 is used to classify site conditions for seismic analyses. GIS-mapping of V_S,30 values obtained from seismic down-hole measurements (Table 1) performed in the Bucharest urban area by UTCB was developed, as illustrated in fig. 5. According to EC8 and P100-1/2013, the site classification suggests that all the considered sites correspond to C-type (intermediate soil), with V_S,30 values ranging from 223 m/s up to 310 m/s.

Figure 5: Bucharest spatial distribution map of VS,30 values based on seismic down-hole measurements

Depending on the calculated V_S,30 values, two predominant subclasses are remarked. The subclass with V_S,30 values ranging from 220-260 m/s in the central part of Bucharest along Dambovita River and the southern part, while the subclass with V_S,30 values ranging from 260-300 m/s corresponds to northern, north-western and eastern parts of the city. This map may supply up-to-date knowledge for Bucharest sites applicable for engineering and other research purposes, considering that the accuracy is primarily dependent on the amount and quality of data.

V_S,30 is now one of the standard indicators for mapping seismic site conditions in most building codes of earthquake-prone countries. However, the quality and density of V_S,30 measurements vary from one region to another. Soil classification (ALLEN, T.I., and WALD, D.J., 2007) and (HEATH et al., 2020) proposed a methodology that correlates topographic slope data obtained from 30 arc-sec (SRTM30 – Shuttle Radar Topography Mission 30 arc-sec) topographic data recorded in 2000 by space shuttle Endeavor and V_S,30 values obtained from in-situ measurements from different sites. The results were extrapolated and used to create a global map for V_S,30 values, available on the USGS server. Fig. 6 presents the V_S,30 map for the Romania region created with data from the 2022 USGS website. Investigating the proposed V_S,30 map for Romania with the Bucharest measurement, we notice the differences between seismic down-hole measured values and topographical slope estimated values vary between −23% and +28% with a 14% mean value. Both methods place all Bucharest sites in ground type „C” according to Eurocode 8 ground type classification (shear wave velocity between 180 and 360 m/s), but on average the topographic slope method provides lower values, underestimating the V_S,30.

Figure 6: VS,30 map for Romania (data from USGS, 2022) and proposed VS,30 site classes 

In absence of more in situ measured V_S data, except for the Bucharest region, fig. 6 estimation may be used for ground type classification in national studies. The earthquake-induced liquefaction in Romania during the 1977 Vrancea strong earthquake (ISHIHARA and PERLEA, 1984, YOUD, 1977) occurred in Quaternary alluvial sandy deposits in the Romanian Plain. Starting from historical information on liquefaction occurrence and the existence of triggering factors related to ground conditions − high probability of liquefaction for the sites with low (< 240 m/s) values of V_S,30 identified in fig. 6, the liquefaction susceptibility is high in the case of strong earthquake on about 20% of the Romanian territory.

CONCLUSIONS

For assessing the near-surface site effects in case of strong earthquakes, it is essential to characterize the sites according to a seismic classification. In the present paper, over 30 sites located in Bucharest are characterized by using the shear wave velocity V_S,30, in order to obtain a comprehensive database to be used in site response analysis. Based on data measured by NCSRR/UTCB (geotechnical and geophysical investigations), various key parameters for dynamic behaviour analysis have been gathered. Besides soil stratigraphy, layer thickness and other important geotechnical parameters, dynamic soil parameters have been obtained by down-hole and MASW measurements. For each investigated site, V_S profiles have been determined for better structuring the local soil conditions database. Geotechnical parameters and elastic properties determined by indirect measurements through correlations from V_S reflect the large variability in thickness of stratified alluvial deposits formed by cohesive and granular soils. MASW surveys have to be used only for limited depths and only for an unsophisticated estimation of the site class.

Soil information can contribute to the development of earthquake disaster mitigation strategies and to the continuous improvement of earthquake-resistant design regulations. Data related to the seismic characterisation of ground conditions (stratigraphic profiles, densities, velocity profiles, equivalent-linear soil behaviour curves, etc.) will be integrated into a national internet-based platform SETTING (2021-2023) that will provide thematic services in the field of Earth observation, as a contribution to the European Plate Observing System EPOS. The platform will also include the directory of Romanian laboratories and institutions performing geotechnical and geophysical investigations of interest for seismology and earthquake engineering purposes. This data will accompany on the platform seismology and GPS/GNSS data. Soil information can contribute to the development of earthquake disaster mitigation strategies and to the continuous improvement of earthquake-resistant design regulations.

ACKNOWLEDGEMENTS

The authors would like to acknowledge the cooperation of Japanese specialists during JICA Project as well as the generous funding provided by the Japan International Cooperation Agency (JICA). We kindly acknowledge the support of the Building Research Institute (BRI), Tokyo Soil Research, and Oyo Corporation.The authors acknowledge the involvement of our former colleagues from NCSRR: Roxana OPREA, Aurora BUCATARU, Caterina NEGULESCU, Raluca RADOI, and Natalia POIATA. We acknowledge the cooperation of Prof. Loretta BATALI from UTCB, and of the companies: Arup, Fugro, Popp & Asociatii, Saint-Gobain. Part of the presented work received support through the SETTING Project Integrated thematic services in the field of Earth observation: a national platform for innovation, No. 108206, cofinanced from the Regional Development European Fund (FEDR) through the Operational Competitivity Programme 2014-2020.

REFERENCES

[1] AKI K. (1988), Local Site effects on Strong Ground Motion. Earthquake Engineering and Soil Dynamics II – Recent Advances in Ground Motion Evaluation, ASCE, 103-155;

[2] ANSAL A.M. (1994), Effects of Geotechnical Factors and Behavior of Soil Layers During Earthquakes. State-of-the-Art Lecture, Proceedings of the 10^th European Conference on Earthquake Engineering, (1): 467-476;

[3] ALDEA A., LUNGU D., ARION C. (2003), GIS microzonation of site effects in Bucharest based on existing seismic and geophysical evidence. 6^eme Colloque National AFPS 2003, Palaiseau, France, 8 p., CD-ROM;

[4] ALLEN T.I., and WALD D.J. (2007), Topographic slope as a proxy for global seismic site conditions (VS 30) and amplification around the globe: U.S. Geological Survey Open-File Report 2007-1357, 69 p;

[5] ARION C. (2004), Report on soil tests and investigations (laboratory and field experiments), JICA/CNRRS/BRI;

[6] ARION C., TAMURA M., CALARASU E., NEAGU C. (2007), Geotechnical in situ investigation used for seismic design of buildings. Proceedings of the 4^th ICEGE, 25-28 June, paper no. 1349, Thessaloniki;

[7] ARION C., NEAGU C., VACAREANU R., CALARASU E. (2012), In Situ Investigation for Microzonation of Bucharest Surface Geology. Proceedings of the 15^th WCEE, 24-28 September, paper no. 2034, Lisbon, Portugal. CD-ROM;

[8] ARION C., CALARASU E., NEAGU C. (2015), Evaluation of Bucharest soil liquefaction potential. Mathematical Modeling in Civil Engineering,11 (1): 5-12;

[9] BALA A., ARION C., ALDEA A. (2013), In situ borehole measurements and laboratory measurements as primary tools for the assessment of the seismic site effects. Romanian Reports in Physics, 65 (1): 285–298;

[10] BORCHERDT R.D. (1970), Effects of local geology on ground motion near San Francisco Bay. Bulletin of the Seismological Society of America, 60: 29-61;

[11] BORCHERDT R.D., GLASSMOYER G. (1994), Influences of local geology on strong and weak ground motions recorded in the San Francisco Bay region and their implications for site-specific building code provisions. U.S. Geological Survey Professional Paper 1551-A, A77–A108;

[12] CHAVEZ-GARCIA F.J., CUENCA J., SANCHEZ-SESMA F.J. (1996), Site Effects in Mexico City Urban Zone. A Complementary Study. Soil Dynamics and Earthquake Engineering, (15):141-146;

[13] DOBRY R., BORCHERDT R.D., CROUSE C.B., IDRISS I.M., JOYNER W.B., MARTIN G.R., POWER M.S., RINNE E.E., SEED RB. (2000), New site coefficients and site classification system used in recent building seismic code provisions. Earthquake Spectra, 16: 41–67;

[14] EPOS, the European Plate Observing System, www.epos-eu.org;

[15] EPRI (1993). Guidelines for determining design basis ground motions, Electric Power Research Institute, Palo Alto, California;

[16] http://ccers.utcb.ro/;

[17] FACCIOLI E. (1991), Seismic Amplification in the Presence of Geological and Topographic Irregularities. Proceedings of the 2^nd ICRAGEE, St. Louis, Missouri, State-of-art paper, 1779-1797;

[18] HAYASHI K., and SUZUKI H. (2004), CMP cross-correlation analysis of multi-channel surfacewave data. Exploration Geophysics, Vol 35, pp7-13;

[19] HEATH D., WALD D.J., WORDEN C.B., THOMPSON E.M., and SCMOCYK G. (2020), A Global Hybrid VS30 Map with a Topographic-Slope-Based Default and Regional Map Insets. Earthquake Spectra, vol. 36, 3: pp. 1570-1584;

[20] IDRISS I.M. (1991), Earthquake ground motions at soft soil sites. Proceedings of the 2^nd International Conference on recent advances in geotechnical earthquake engineering and soil dynamics, St. Louis, Missouri, 2265-2271;

[21] ISHIHARA K., PERLEA V. (1984), Liquefaction-associated ground damage during the Vrancea earthquake of March 4, 1977. Soils and Foundations, 24(1): 90-112;

[22] LITEANU E. (1952), Geologia zonei orasului Bucuresti. Comit. Geol., Stud. tehn. econ., E, 1, Bucuresti;

[23] LUNGU D., ALDEA A., MOLDOVEANU T., CIUGUDEAN V., STEFANICA M. (1998), Surface geology and dynamical properties of soil layers in Bucharest. In Vrancea Earthquakes. Tectonics, Hazard and Risk Mitigation, Kluwer Academic Publishers 1998, p.137-148;

[24] LUNGU D., ALDEA A., ARION C., DEMETRIU S., CORNEA T. (2000), Microzonage Sismique de la ville de Bucarest – Roumanie. Cahier Technique de l’Association Francaise du Genie Parasismique, 20:31-63;

[25] NEAGU C. (2015), Local soil condition and nonlinear soil response influence on design seismic action. Ph.D. Thesis, Technical University of Civil Engineering of Bucharest (in Romanian);

[26] PARK C.B., MILLER R.D., XIA J., IVANOV J. (2001), Characterization of geotechnical sites by Multichannel Analysis of Surface Waves (MASW) method. Proceedings of the 10^th ICSDEE, Philadelphia;

[27] PAVEL F., VACAREANU R. (2015), Investigation on site conditions for seismic stations in Romania using H/V spectral ratio. Earthquakes and Structures, 9(5), 983–997;

[28] SEED H.B., ROMO M.P., SUN J.I., JAIME A., LYSMER J. (1987), Relationships between soil conditions and earthquake ground motions in Mexico City in the event of 19.09.1985. Report No. UCB/EERC-87/15;

[29] SETTING Project (2021-2023), Integrated thematic services in the field of Earth observation: a national platform for innovation, co-financed from the Regional Development European Fund (FEDR) through the Operational Competitivity Programme 2014-202, https://setting.epos-ro.eu/;

[30] YOUD T.L. (1977), Reconnaissance report of geotechnical observations for the 4 March 1977 Romanian earthquake, extract from EERI report.

Authors:

Cristian ARION − Seismic Risk Assessment Research Center, Technical University of Civil Engineering (UTCB)

Cristian NEAGU − Dublin City Council

Elena-Andreea CALARASU − Ministry for Development, Public Works and Administration of Romania

Florin PAVEL, Alexandru ALDEA, Radu VACAREANU − Seismic Risk Assessment Research Center, UTCB

[Proceedings of the 17th Danube European Conference on Geotechnical Engineering (17DECGE), June 7-9, 2023, Bucharest, Romania – https://17decge.ro/]

…citeste articolul integral in Revista Constructiilor nr. 207 – octombrie 2023, pag. 68-70, 72-73