SITE EFFECT ASSESSMENT OF SEISMOLOGICAL STATIONS OF ALBANIA THROUGH...

Romanian Reports in Physics, Vol. 63, No. 1, P. 287–301, 2011

SITE EFFECT ASSESSMENT OF SEISMOLOGICAL STATIONS OF ALBANIA THROUGH HVNR TECHNIQUE

H. RECI

Institute of Geosciences, Polytechnic University of Tirana, Str. Don Bosko, Albania, E-mail: [email protected]

Received August 30, 2009

Abstract. In this paper is presented the study of site response parameters of Albanian Seismological Stations (ASN) from ambient noise measurements and felt earthquakes. The seismological stations of Albania which use the VSAT communication technology are located generally in hard rock formations.

The aim of Methodology is to compare the response parameters of the sites (fundamental frequency, amplification) for registered seismic signals, from microtremors and the earthquakes as well.

The predominant frequency of the site of every seismological station site is determined using the methodology of Horizontal to Vertical Spectral Ratio (HVNR), of the triggered noise and several felt earthquakes in Albania territory. From the results of this study, is concluded that fundamental frequency (F0) and the amplification in this frequency (A0), are almost the same for the microtremors and seismic events. As far as for the amplification factor, we concluded that on the sites of seismological stations situated on hard rock (basement) there is not amplification. In the contrary, for the Vlora station site, the amplification factor is 5–6 at fundamental frequency 6 Hz, showing that this station is not situated on the hard rock. The microtremors are a useful technique for the site assessment effects, used in microzonation of urban areas, that complement the lack of seismic data in a region (boreholes, geology, etc.)

Key words: ambient noise; microtremors; spectral ratio; seismic microzonation; seismic

ambient noise; fundamental frequency, amplification factor.

1. INTRODUCTION

The site response of a seismic event can lead on big vibrations of the earth, which causes high damages of the structures even on the case of moderated earthquakes. During last decades, are made several theoretical and empiric studies, in order to understand the physical and the nature of this phenomenon. Particularly, is shown that the fundamental frequency usually is represented on unconsolidated lithological formations Sanchez-Sesma, 1996; Luzon et al., 2002). For this reason, the amplitude and the duration of the earth vibration on the earth surface could be much bigger than in sites situated on bedrock. Of interests, is that the impedance

H. Reci 2 288

contrast between sediment basin and the basement rock, can affect the amplitude and duration of the earth vibration during a future seismic event.

The Seismological Stations of Albania, Institute of Geosciences that use the VSAT communication technology, are installed generally above hard rock basement (Table1). A new, fully integrated digital seismographic system (Libra VSAT), using the satellite communication, is currently under implementation. The central hub data acquisition system and six VSAT remote stations are already under operation. An additional one has been just commissioned, and very soon we’ll have it installed on Berati (BERA). The digitizers used with Libra VSAT system are of TRIDENT type, 24 bits, with dynamic range of 142 dB, satellite transceiver CYGNUS. The type of sensors accepted from this digitizer, are maximally three-axial. Six remote sites are installed on SRN, PHP, KBN, BCI, PUK, and VLO and transmit continuously the data on real time. The stations operate with the following BB sensor types: CMG-40T, (SRN, BCI) and TRILLIUM 40T, (PHP, PUK, KBN, VLO, BERA). Real-time data transfer by satellite communications began at the end of July 2003 for two VSAT remote stations.

The Tirana master station (TIR) uses also Libra VSAT technology, and is operated in collaboration with MedNet project, as part of the MedNet seismic network. Continuous real time data are transmitted at INGV/Rome via satellite communication, and are then transferred to our Institute via Internet. The TIR station is equipped with a STS-2 broadband sensor.

Table 1 Seismological Stations of Albania (VSAT)

Code Name of

the station

Geographic latitude

Geographic longitude Elevation Geological

Formation Instrument Sensor

BCI B.Curri 42.3666 20.0675 500 Ultrabazic Rocks

Cygnus CMG-40T

KBN Korca 40.6236 20.7874 800 Burdigalian Marnes

Cygnus Trillium 40T

PHP Peshkopi 41.6847 20.4408 900 Limestone Cygnus Trillium 40T

SRN Sarande 39.8800 20.0005 20 Limestone Cygnus CMG-40T

TIR Tirane 41.3477 19.8650 198 Tortonian Sandstones

Cygnus STS-2

VLO Vlore 40.4686 19.4955 80 Pliocene Sandstones

Cygnus Trillium 40T

PUK Puke 42.0426 19.8926 900 Ultrabazic Rocks

Cygnus Trillium 40T

Data registered from the above mentioned seismological stations, used on the

interpretation, consists on several felt earthquakes on Albania territory. The two of them are presented in this study.

3 Site effect assessment of seismological stations of Albania 289

The event of 24 October 2008 with magnitude ML= 4.5 Richter felt in Tirana city. The epicenter of this earthquake was localized around 40km south-east of Tirana, at depth 4 km. Even though the epicenter of this earthquake was far from Tirana, its intensity was estimated IV according to EMS 98 scale (Fig. 1).

The earthquake of 31 January 2009, with magnitude ML=3.3 Richter and epicenter in Vlora District, which was evaluated with an intensity IV, EMS 98 on the epicenter (Fig. 2).

In this study are used microtremors registered in a time interval 3–10 min for every station.

Microzonation Studies are good tools for seismic risk assessment of urban areas. To carry out such studies, important is the estimation of site response from earth vibrations. Several empiric studies have been carried out having as purpose the seismic microzonation of urban areas. For the identification of fundamental frequency and the amplification amplitude of sediments, Kanai et al. (1954) took in consideration the Fourier Spectrum of horizontal component of microtremors. For the first time Nakamura (1989) proposed in this year, the technique of three components registration of the earth vibration in a specific site and the use of the horizontal to vertical spectral ratio technique (HVNR). The HVNR methodology is widely used to identify the fundamental frequency of the site in a specific location which in most of cases shows good relations with the geology of the region. During the last years a big number of microzonation studies have been carried out in urban and very dense populated areas using this technique.

Fig. 1 – The event of 24 October 2008 and its parameters.

H. Reci 4 290

Fig. 2 – The event of 31 January and its parameters.

2. HVNR TECHNIQUE AND ITS APPLICATION ON THE SEISMOLOGICAL STATIONS OF ALBANIA

The HVNR technique consists on the registration of three components of

earth vibration (seismic event or noise). The spectral ratio of horizontal component over the vertical one (H/V) gives the fundamental frequency and the amplification factor in this frequency. This technique was proposed earlier by Nogoshi and Igarashi (1971). Later on, the technique began to be widely used in urban areas of Japan, from (Nakamura, 1989). The initial results were experimental and very positive. In the locations where hard rock is underlain the soft sediments, the spectral H/V ratio shows a peak. These peaks are stable during the time and frequency, and can be accepted as natural (resonant) frequency of the site (Duval et al., 1994; Duval, 1996). Before the use of HVNR technique, the only way to identify the fundamental frequency of the earth was through measurement of the seismicity of a region and the comparison of it with a reference site. Through the HVNR methodology only few minutes are enough for the registration of seismic signal in a specific site, to get the resonant frequency of it. The HVNR technique holds up in the hypothesis that, the horizontal and vertical vibration in the interface between basement and the sediment are equal. Taking in consideration this


hypothesis, the site effect on the earth surface S(f), can be estimated getting the ratio of horizontal over vertical vibrations;

( )( )( )

H fS fV f

= . (1)

The Nakamura’s technique, was developed by Konno and Ohmachi (1998), extending it for a multilayered model. These authors reinforced this technique, which up to know has not a theoretical explanation. Several studies on the microzonation of urban areas show that the HVNR methodology, gives the resonant frequency on a specific site (Lermo and Chavez-Garcia, 1994; Field and Jacob, 1995; Lachet et al., 1996; Seekins et al., 1996; Coutel and Mora, 1998).

Other works on this topic have shown that this technique can be applied in the seismic wave simulation, such as sedimentary basins, (Al Yuncha and Luzon, 2000; Dravinski et al., 1996). From numeric studies of flat geological deposits, the technique of HVNR can predict very well the local fundamental frequency, if there is impedance contrast between sediments and rock basement (Luzon et al. (2001).

Many works are carried out in huge urban areas around the world, using the technique of Nakamura HVNR, such as Ibs-Von Seht and Wohlenberg (1999) in the lower Rhine (Germany), Delgado et al. (2000) in the valley of Segura river (Spain), Navarro et al. (2001) in Almeria (Spain), or Paschalis., et al. (2004) in Thessaloniki (Greece).

On the current study, several felt earthquakes have been registered and microtremors are used triggering seismic noise, in 7 stations of Albanian Seismological Network using VSAT communication technology. The registered traces of earthquakes have time duration of 3–5 minutes. Whereas, the horizontal and vertical components of microtremors, are triggered in a time interval 3–10 min for the stations BCI, PUK, SRN, PHP, TIR, VLO and KBN. Triggered noise (microtremors) is taken during midnight and midday in order to compare the results. Knowing that the data have been collected in central Station, there is not a need for the shifting of traces. Generally, the seismic signals are not affected from mechanical noises, since the stations are situated far from populated areas. The selected windows used in the interpretation with the HVNR methodology are 20–30s for both the signals from earthquakes and the microtremors as well. The Fourier transformation is carried out to the selected signal, smoothed with the Hanning window 0.7 Hz. The horizontal to vertical spectral ratio is taken using the expression;

2 2( ) ( )

( ) ,( )

NS EWH f H fS f

V f+

= (2)

H. Reci 6 292

where ( )NSH f and ( )EWH f are the North-South and East-West respectively. The used methodology consists to the application of equation 2, in successive windows along the recorded seismic traces. This procedure, gives some functions of HVNR that can be presented in the shape of H/V ratio graphics, versus frequency and time.

3. RESULTS OF HORIZONTAL TO VERTICAL SPECTRAL RATIO

Several microzonation studies aim to the estimation of the amplification and fundamental frequency for a specific site during a future expected seismic event. The Nakamura’s technique, as used in this study, shows that in general, the HVNR method leads to the same results of the fundamental frequency and the amplification factor during a seismic event and microtremors as well. A way of microzonation studies is the two-dimensional mapping (2D) of fundamental frequency and the amplification. In Fig. 3a is presented recorded seismic noise for the stations TIR and VLO, whereas in the Fig. 3b are represented the results of HVNR technique for the sites of seismological stations.

From Fig. 3b, it is clear that in the foundation of TIR Station there isn’t predominant frequency, even though a small peak of amplification up to 2 appears in the frequency 0.6 Hz. In general, for the entire frequency band, the site amplification is close to 1 (the black line). The interrupted black line shows the Standard Deviation. This result, shows that this station is situated in bedrock (Table 1), which does not give amplification. On contrary, The VLO station shows a completely different situation. There is a representation of two peak frequencies, 6Hz and 12Hz with amplification factors 4 and 5 respectively. This fact shows that the station VLO is not installed above the rock formations. The presence of two frequency peaks can be explained by the presence of two thin successive layers with soft sediments and soils. In Fig. 4 are depicted the H/V results for the stations BCI and PHP (Table 1), which are situated in hard rock formations. On the foundation of BCI station, an amplification factor is present (0.6 Hz), which could be related to another deep geological interface, which needs further discussion in future. Fig. 5, represents the H/V results for the stations PUK, SRN and KBN. All three stations are located above hard rock formations (Table 1) and as seen from Fig. 5b, there isn’t amplification on the sites of the stations. In the PUK station is presented a frequency peak, 15Hz, with amplification 3, which could be the effect of near surface soils.


a)

b)

Fig. 3 – a) Registered seismic noise; b) frequency functions of the HVNR technique (black line) for the seismological stations TIR and VLO.

H. Reci 8 294

a)

b)

Fig. 4 – a) Registered seismic noise; b) frequency functions of the HVNR technique (black line) for the seismological stations BCI and PHP.


a)

H. Reci 10 296

b)

Fig. 5 – a) Registered seismic noise; b) frequency functions of the HVNR technique (black line) for the seismological stations PUK, SRN and KBN.


a)

b)

Fig. 6 – a) Earthquake traces of 24 October 2008; b) frequency functions of HVNR Technique (black line) for the stations BCI, KBN, PUK and SRN.

H. Reci 12 298

a)

b)

Fig. 7 – a) Earthquake traces of 31 January 2009; b) frequency functions of HVNR Technique (black line) for the stations BCI, PHP, PUK, SRN and VLO.


The H/V results for the earthquake of 24 October 2008, with magnitude ML=4.4 Richter and epicenter in Elbasani district are presented in Fig. 6 whereas for the earthquake with magnitude ML=3.3 Richter with epicenter in Vlora district are presented in Fig. 7.

From Fig. 6b, seems that the HVRN technique shows the same results with the microtremors, represented in Figs. 3, 4 and 5. The amplification factor in relation to frequency is the same with that of seismic noise and in most of stations close to 1, showing the presence of hard rock. The VLO station represents the same results with that of microtremors, the same resonant frequencies and amplification factors, with the only change that, another peak is presented in the frequency 2.2Hz, and amplification 3. In general, the site amplification in the case of earthquakes is a little bit stronger comparing to microseismic noise (Figs. 6b, 7b), but the continuity of it is almost the same for all the band of frequencies.

4. CONCLUSIONS

In this study is investigated the stability of site response on the foundations of Seismological Stations of Albania through the HVNR methodology. The microtremor and earthquake recording are used in interpretation of results. The data used are from several felt earthquakes, and random triggered noise in a time interval 3–10 min. The results of this study have shown that the amplification of the station sites in general is close to 1 that was expected, knowing that these Seismological Stations are installed in hard rock geological formations. The situation is completely different only on the VLO station where the site amplification shows two peaks, in frequencies 6Hz and 13Hz from microseismic noises, and 3 peaks with different amplification in frequencies 2.2Hz, 6Hz and 15 Hz for the earthquakes. Due to the fact that this station is not situated above hard rock, frequency peaks are present. This station has to be reinstalled at other place, above rock formation. The H/V results are almost the same for the seismic events and microseismic noise recordings. Some of the sites of the seismological stations tend to have small amplification on high frequencies, but this could be as consequence of thin sedimentary soils covering the rock formations.

As a conclusion, from results taken from this study we can say that there are practical possibilities to use the HVNR technique to the microzonation studies on urban areas in Albania.

Microtremors are a tool of microzonation studies that is widely used around the world, to predict the amplification function of a specific site during an expected earthquake on the future. The HVNR and SPAC microzonation techniques are of low economical cost. A better part of microzonation studies, has as an aim the determination of fundamental frequency and amplification factor, the realization of Vs profiles up to the depth of the basement rock. The techniques of HVNR as used

H. Reci 14 300

in this study, is a good tool for the determination of local fundamental frequency and the site amplification.

The microzonation methodologies HVNR and SPAC, in cooperation with other conventional seismic methods (reflected seismic waves, refracted seismic waves, bore holing, and other geophysical, geotechnical methods etc.), which are more costly and in some cases the implementation of them is impossible inside urban areas, are effective methodologies for the microzonation studies that can be used successfully in Albania as well.

REFERENCES

1. AL YUNCHA, Z. LUZON, F., On the Horizontal-to-vertical Spectral Ratio in Sedimentary Basins, Bull. Seismol. Soc. Am., 90, 1101–1106, (2000).

2. ALMENDROS, J., LUZON, F., POSADAS, A., Microtremors Analysis at Teide Volcano (Canary Island, Spain): Assessment of Natural Frequencies of Vibration Using Time-dependent Horizontal-tovertical Spectral Ratios, Pure Appl. Geophys., 2002.

3. COUTEL, F. MORA, P., Simulation-based Comparison of Four Site-response Estimation Techniques, Bull. Seismol. Soc. Am., 88, 30–42 (1998).

4. DELGADO, J., Lo PEZ-CASADO, C., ESTEVEZ, A. C., GINER, J., CUENCA, A., MOLINA, S., Mapping Soft Soils in the Segura River Valley (SE Spain): A Case Study of Microtremors as an Exploration Tools, J. Appl. Geophy., 45, 19–32 (2000).

5. DRAVINSKI, M., DING, G., WEN, K.-L., Analysis of Spectral Ratios for Estimating Ground Motion in Deep Basin, Bull. Seismol. Soc. Am., 86, 646–654 (1996).

6. FALLOT, P., Les Cordille´res Be´tiques, Est. Geol., 8, 83–172 (1948). 7. FIELD, E. H. JACOB, K. H., A Comparison and Test of Various Site-response Estimation

Techniques, Including Three that are not Reference-site Dependent, Bull. Seismol. Soc. Am., 85, 1127–1143 (1995).

8. IBS-VON SEHT, M., WOHLENBERG, J. (1999), Microtremor Measurements Used to Map Thickness of Soft Sediments, Bull. Seismol. Soc. Am., 89, 250–259.

9. IGME (1974), Mapa Geologico de Motril, 1055 – (19–44). Esc. 1:50 000, Serv. Publi. Ministerio de Industria, p.15.

10. KANAI, K., TANAKA, T., OSADA, K. Measurement of Microtremor, I, Bull. Earthq. Res.Institute, 32, 199–209 (1954).

11. KONNO, K., OHMACHI, T., Ground-motion Characteristics Estimated from Spectral Ratio between Horizontal and Vertical Components of Microtremor, Bull. Seismol. Soc. Am., 88, 228–241 (1998).

12. LACHET, C., HAZFELD, D., BARD, P.-Y., THEODULIDIS, N., PAPAIOANNOU, C., SAVVAIDIS, A., Site Effects and Microzonation in the City of Thessaloniki (Greece) Comparison of Different Approaches, Bull. Seismol. Soc. Am., 86, 1692–1703 (1996).

13. LERMO, J. and CHA´ VEZ-GARCI´ A, F. J., Are Microtremors Useful in Site Response Evaluation?, Bull. Seismol. Soc. Am., 84, 1350–1364 (1994).

14. LUZON, F., AL YUNCHA, Z., SANCHEZ-SESMA, F. J., ORTIZ-ALEMAN, C., A Numerical Experiment on the Horizontal to Vertical Spectral Ratio in Flat Sedimentary Basins, Pure appl. Geophys., 158, 2451–2461 (2001).

15. LUZON, F., PALENCIA, V. J., MORALES, J., SANCHEZ-SESMA, F. J. Evaluation of Site Effects in Sedimentary Basins, Fısica de la Tierra, 14, 183–214 (2002).


16. NAKAMURA, Y., A Method for Dynamic Characteristics Estimation of Subsurface Using Microtremor on the Ground Surface, Report Railway Tech. Research Institute, 30, 1, 25–33 (1989).

17. NAVARRO, M., ENOMOTO, T., SANCHEZ, F. J., MATSUDA. I., IWATATE, T., POSADAS A., LUZON, F., SEO, K., Surface Soil Effects Study Using Short-period microtremor Observations in Almerıa City, Southern Spain, Pure Appl. Geophys., 158, 2481–2497 (2001).

18. *** Normativa de Construccion Sismorresistente Espanola NCSE-94., Real Decreto 2543/94, B.O.E. 33, 8 de Febrero 1995.

19. PASCHALIS, A., RAPTAKIS, D., ROUMELIOTI, Z., PTIAKIS, K. Determination of S-wave velocity structure using microtremors and SPAC method in Thessaloniki(Greece), Soil Dynamics and Earthquake Engineering, 24, 49–67 (2004).

20. SANCHEZ-SESMA, F. J., Strong ground motion and site effects. In Computer Analysis and Earthquakeresistant Structures (D. E. Beskos and S. A. Anagnostopoulos, eds.), Computational Mechanics Publications, Southampton, 1996, pp. 200–229.

21. SANZ DE GALDEANO, C., Some Geological Problems of the Betic Cordillera and Rif, Fısica de la Tierra, 4, Editorial Complutense, Madrid, 1992, pp. 11–40.

22. SANZ DE GALDEANO, C., Principal Geological Characteristics of the Betic Cordillera, Some Spanish Karstic Aquifers. Uni. de Granada, 1993, pp. 1–17.

23. SEEKINS, L. C., WENNERBERG, L., MARGHERITI, L., LIU, H. P., Site Amplification at Five Locations in San Francisco, a Comparison of S Waves, Codas, and Microtremors, Bull. Seismol. Soc. Am., 86, 627–635 (1996).

24. SOMERVILLE, P., Geotechnical Reconnaissance of the Effects of the January 17, 1995, Hyogo-ken Nanbu Earthquake, Japan, Technical Report UCB/EERC-95/01, Earthquake Engineering Research Center, University of California, Berkely, California.

25. TENG, T.-L. AKI, K., Preface to the 1994 Northridge Earthquake, Special Issue, Bull. Seismol. Soc. Am., 86, S1–S2. 1558 Z. Al Yuncha et al., Pure appl. geophys., 1996.

SITE EFFECT ASSESSMENT OF SEISMOLOGICAL STATIONS OF ALBANIA THROUGH...

Documents

Transcript of SITE EFFECT ASSESSMENT OF SEISMOLOGICAL STATIONS OF ALBANIA THROUGH...