Estimation of ground types in different districts of Gümüşhane province based on the ambient vibrati

Abstract

Keywords: Gümüşhane; Microtremor; Predominant frequency; H/V ratio; Kg-value; Ground type

Introduction

There are various techniques that have been used in order to identify the subsoil characteristics during the strong ground motions. One of the most widely applied techniques to estimate the ground types is the microtremor horizontal to vertical (H/V) ratio. This technique was defined by Nakamura [1] in relation to borehole explorations and analysis of strong ground motion data on the different geological structures. Many researchers have used the spectral features of microtremors to estimate the predominant or natural frequency of the soil in Turkey and different sites of the world. These studies suggest a good relationship between the surface geology and the spectral ratios of microtremors [1–11]. These results show that microtremors are very suitable tools in the identification of surface geological effects on the seismic motion without detailed geology. Microtremors are always exist and known as weak ground motions based on the natural or ambient processes such as earthquakes, sea waves, wind, rivers, rain, traffic, tides, changes in atmospheric pressure, human activities, machinery, etc. Their amplitudes vary from 0.1 to 1.0 μm, periods between 0.05 and 2.0s, and these motions change the site effects [11–13]. Observations and assessments of microtremor data have some advantages over other techniques for site response characteristics due to its simplicity, inexpensive, ease of use, minimum computational time, short duration for measurements and data process [2,6]. General applications of microtremor studies can be carried out: (i) to calculate the amplification of horizontal movements in the free surface throughout earthquakes, (ii) to measure background seismic noise and to estimate the mechanical features of the earth’s subsurface, (iii) to determine the site response characteristics, (iv) to detect the subsoil types, (v) to estimate the shear wave velocity of the ground, (vi) to detect the effect of site conditions on damage distribution during the earthquakes, (vii) to estimate the predominant resonant frequency and period of the sediments, (viii) to design the seismic resistant infrastructure, (ix) to identify small-scale seismic risks and to make a convenient and reliable seismic hazard microzonation in urban areas [14–17]. The H/V spectral ratio of recorded microtremors can be estimated from ambient noise records and this technique is quite practical tool for engineers to measure the intensity of earthquake ground motion and the capacity of buildings to resist earthquakes [6]. For this purpose, we used the single station microtremor H/V spectral ratio technique in this study and aimed to investigate the seismic site response characteristics in Torul, Kürtün, Kelkit, Şiran and Köse districts of Gümüşhane province, Turkey. For this purpose, we determined the predominant frequency and H/V ratio (H/V amplitude spectrum or HVSR amplitude) of subsurface ground using measured microtremor data. Thus, a classification for ground types in different districts

of Gümüşhane was made based on the single station microtremor data analysis.

Surface Geology And Seismotectonics

OF GÜMÜŞHANE The Gümüşhane is located in the eastern part of the Pontide Orogenic belt in the northeast Turkey. The principal base rocks observed in Gümüşhane and vicinity are composed of the Palaeozoic-aged metamorphic rocks and Gümüşhane granites which rise by cutting these metamorphic rocks (One can find many details on geological ages in Taş et al., [18]). Surface geology in and around Gümüşhane is modified from the database of the General Directorate of Mineral Research and Exploration (MTA, URL-1) and given in Figure 1. Granitic formations are dominated in Gümüşhane city center and the main geological units close to the city center are formed of granite, granodiorite and quartz-diorite, Eosene-volcanic facies, undifferentiated Cretaceous, upper Cretaceous and Flysch. Torul district and surrounding area are generally covered with granite, granodiorite, quartz-diorite and Eosene-volcanic facies. The principal surface geology in Kürtün district and vicinity includes granite, granodiorite, quartz-diorite and upper Cretaceous-Volcanic facies. Geological formations of undifferentiated Eocene, Eocene flysch, mid-Eocene lutetian, andesite-spilite-porphyrite, basalt-dolerite, rhyolite-dacite and volcanic tuff-agglomerate-breccia are dominant in and around Şiran district. However, Kelkit and Köse districts are composed of the Pleistocene and Holocene-Recent structures, undifferentiated Neogene continental formations, undifferentiated Eocene, Eocene flysch, mid-Eocene lutetian floors and partially granite, granodiorite, quartz-diorite structures. The major factors that affect the current seismotectonic situation of Gümüşhane and vicinity are fracture systems and fold tectonics. Tectonic structures of Gümüşhane and vicinity were compiled from Şaroğlu et al. [19] and Bozkurt [20] and were shown in Figure 2. There are several fault segments, zones and basins located close to Köse, Kelkit, Şiran, Erzincan and Bayburt, and these tectonic structures are generally related to the NAFZ. As seen in Figure 2, these active systems can be given as Kelkit-Çoruh Fault Zone (KÇFZ), Bayburt Basin (BYB), Kelkit Basin (KLB), Kelkit Fault Segment (KLFS), Akdağ-Çayırlı Fault Zone (AÇFZ), Dağyolu Fault (DYF) and Tercan-Aşkale Fault Zone (TAFZ). KÇFZ is about 600 km long and has a left lateral strike slip fault mechanism. This zone has four segments from the southwest to the northeast: Kelkit, Çoruh, Posof and Borjomi-Kasbeg. KLFS is separated from the NAFZ with a length of about 100 km. This segment is divided into two branches around Kelkit and results in a basin [18]. TAFZ has a left lateral strike slip fault mechanism. It has a length of about 150 km and a wide of 2 to 4 km. This fault zone passes through the western part of Erzurum near the

Sigma J Eng Nat Sci, Vol. 39, No. 4, pp. 374–391, December, 2021

Figure 1. Detailed near surface geology and different formations in and around Gümüşhane province (digitized from MTA, URL-1).

Figure 2. Seismic and tectonic structures in Gümüşhane and surroundings between 1970 and mid-2018. Simplified tectonics were modified from different sources such as Şaroğlu et al., [19] and Bozkurt [20].

NAFZ and includes several parallel fault segments which has a length between 2 and 20 km [20]. Gümüşhane is a very close residential area to the North Anatolian Fault Zone (NAFZ), about 80 km, and due to the proximity to the NAFZ, a strong or large earthquake on the NAFZ and surrounding area may affect this region. Especially, high-rise buildings constructed on alluvial grounds in creek beds along the Harşit stream will be affected from a possible large and/or destructive earthquakes which may occur on the NAFZ [21]. During the instrumental (past 48 years in this study) and historical periods, there are two strong events around Gümüşhane border as seen in Figure 2: January 19, 1979 (Md5.0) and August 12, 1985 (Md5.0) earthquakes (Md is duration magnitude). Also, several large and destructive earthquakes occurred in and around the NAFZ between 1970 and mid-2018 such as March 13, 1992 (Md6.5) and March 15, 1992 (Md5.3), whereas the other great events occurred in some parts of Gümüşhane near the NAFZ. Figure 2 shows the epicenter distributions of earthquakes (depth ≤ 75 km) with 1.0≤Md≤6.5 from 1970 to middle of 2018.

Sigma J Eng Nat Sci, Vol. 39, No. 4, pp. 374–391, December, 2021

Methodologies And Microtremor Data For Study Regions

The H/V Spectral Ratio Method Microtremor H/V spectral ratio technique was developed by Nakamura [1] on the different geological ground circumstances related to the borehole surveys by using the analyses of strong ground motion. Nakamura [1] supposed that vertical component of the ambient noise at the ground surface includes the properties of basement ground. Therefore, Rayleigh waves on the sedimentary rock influence the vertical component and thus, vertical component can be utilized to eliminate both of the source and Rayleigh wave effects from horizontal components [22]. In fact, the first of all, the H/V spectral ratio was provided with the strong motion measurements at different parts of Japan. Then, many researchers mentioned above showed that microtremor H/V spectral ratio of ambient noise may be utilized to describe the predominant frequency and H/V ratio of different ground structures in different regions of the world. Nakamura [1] assumes that microtremors mainly contain the shear waves, and the layer of soft soil does not magnify the vertical waves. The horizontal components of the shear waves are magnified by soft soil layer because of the multiple reflection of the waves [13]. There are four amplitude spectra described in the Fourier frequency domain. Nakamura [1] suggested that the H/V spectral ratio of the Fourier amplitude spectra of microtremors can be used to estimate the transfer function of the surface layers. Microtremor movements were described as a function of frequency: AS (ω ) =

where VS(ω) is the vertical component of the movement on the surface and VB(ω) is the vertical component of the movement on the substrate of the surface layer. Equation (2) represents the transfer function in measurement point. In this equation, HS(ω) is the horizontal component on the surface and HB(ω) is the horizontal component of the movement on the substrate of the surface layer. Source effect can be removed from observation values by dividing transfer function given in Equation (2) to source effect given in Equation (1). This rate is defined as: H S (ω ) SE (ω ) H B (ω ) H S (ω ) VB (ω ) S= = = ⋅ M (ω ) AS (ω ) VS (ω ) H B (ω ) VS (ω ) VB (ω ) H S (ω ) VB (ω ) = ⋅ =RS (ω ) × RB (ω ) VS (ω ) H B (ω )

Nakamura [1] stated that the H/V spectral ratio, RB(ω), taken in bedrock in the frequency range of interest (1–20 Hz) for engineering purposes studies, is approximately equal to 1 as given in Equation (4). In this way, the transfer function described as RS(ω) can be obtained from microtremor data measured at the surface: = RB (ω )

Two horizontal components recorded as north-south, NS (ω), and east-west, EW (ω), can be combined as a single component taking the magnitude of the vector as described in Equation (5): = H S (ω )

The ground effect is described in terms of the horizontal and vertical components at the surface. The H/V spectral ratio can be obtained by dividing horizontal component, HS(ω), to vertical component, VS(ω), as described in Equation (6): SM (ω ) =

Evaluation of the H/V ratio was achieved by using the GEOPSY software developed within the framework of the Site Effects Assessment using Ambient Excitation (SESAME) Project [23]. The first step involved in applying the H/V spectral ratio technique, recorded signals were fixed with a baseline correction and trend effects in the data were removed. Then, corrected records were filtered with a Butterworth band pass filter with a range of 0.5–20 Hz. In the third stage, several window lengths in the range of 10–40s were tested to find the suitable window size and a 20s time window length was preferred. For each window, a Cosine taper filter with 10% width was implemented to decrease border effects caused by the cutting operation. In this way, the noise at very low and high frequencies compared to the signals was removed, and only the frequency content of the microtremors remain. In the next step, the fast Fourier transform (FFT) was utilized to calculate the three component (north-south, east-west and vertical) amplitude spectrum. For each window, the signal in the time domain was converted to frequency domain. In order to smooth the computed spectrum, a smoothing type filter defined by Konno and Ohmachi [24] was used with constant number of 40. In this way, the small noises from the resulting spectrum were removed and a good approximation with clear spectrum peaks was generated by this smoothing coefficient. In the next stage, the root mean square of spectrum values for two horizontal components (north-south and east-west) was taken and the single horizontal component

was obtained. Thus, horizontal component direction was computed independently and the noise ratio on any of components was reduced. In the final step, the average of horizontal components (root mean square) was divided by the vertical component spectrum for all windows, and an average spectrum was computed with its confidence intervals to estimate the resonant frequency with respect to high amplitude peak. The H/V spectral ratio for suitable windows was obtained and then, the average of the spectra was accepted as the H/V spectral ratio for each measurement point. Representative H/V spectral ratio curves were plotted for each measurement point according to the standard criteria of SESAME [23]. As a result, the peak frequency of the H/V spectrum plot gives the dominant frequency and H/V ratio of the measured point. Figure 3 shows all these data processing steps with an example measurement point from Köse district. Study regions and measurement points for Torul, Kürtün, Kelkit, Şiran and Köse districts of Gümüşhane province were shown in Figure 4. Measurement points were placed on “Google Earth” in order to obtain the best images of the point and to see the settlements in each

Sigma J Eng Nat Sci, Vol. 39, No. 4, pp. 374–391, December, 2021

individual site. A total of 180 single station microtremor measurements were conducted to estimate the predominant frequency and H/V ratio. For this purpose, GURALP CMG-6TD three component broad band velocity seismometer was used. Measurements were recorded numerically in GCF (Guralp Compressed Format) with Scream

Results And Discussions

Previous studies show that two important site response characteristics (predominant frequency and amplification factor) can be estimated safely. Therefore, we used the single station microtremor data in order to identify the ground type characteristics for Torul, Kürtün, Kelkit, Şiran and Köse districts of Gümüşhane. Single station micotremor data was recorded at 12 points in Torul, 9 points in Kürtün, 48 points in Kelkit, 60 points in Şiran and 51 points in Köse. Since the site conditions were more suitable in Kelkit, Şiran and Köse, more measurements were taken in these districts. Thus, predominant frequencies and H/V ratios were obtained for each measurement points in all districts. Three components microtremor signals with resolution windows on microtremor records and the H/V spectral ratio curves with average H/V for selected points of TR7, K3, KL9, SR27

and KS38 were given as the representative examples for each region. These randomly selected examples were plotted in Figure 5. The obtained spectra include the standard deviations for all the H/V curves and the standard deviations were plotted by two dashed lines above and below of the H/V curves. All estimated values from these H/V curves as well as the other information of the points were also given in Tables 1 to 5. These selected figures show the final data processing steps (defined above) that was applied to all the microtremor measurements in the study region. These sequences of data processing implemented in the representative samples includes the interpretation of H/V curves. In these context, these primary and essential steps were involved in all the recorded microtremor data (not shown for all 180 points, only 5 measurements) in the study area. When considered the largest amplitudes of spectrum, the predominant peak in the H/V spectral rate is much more evident as a single peak with the maximum amplitude in some points (limp grounds with soft padding). However, the maximum amplitude in some points is shown in high frequencies as the multiple peaks (solid grounds). It is stated that the H/V ratios exhibit a clear peak in a soft soil structure underlying a hard rock and these peaks are constant in space and time [12]. Therefore, these peaks can be thought

Sigma J Eng Nat Sci, Vol. 39, No. 4, pp. 374–391, December, 2021

as predominant frequencies of the region. However, a second peak value can be observed in most of microtremor measurements. The peak with maximum amplitude can be false peak in the H/V curves with multiple peaks. In this situation, the frequencies estimated in the H/V graphics can be too small or too large to be expected according to the existing ground type. When examined all the values given in Tables, the maximum frequency estimated from horizontal amplitude spectrum on the H/V spectral rate curves is around 10 Hz and these results are accordance with the vertical amplitude spectrum values. For this reason, all the results were calculated by using the peak values (there are not multiple peaks in our data) in the H/V ratios. The H/V ratios were analyzed according to the SESAME [23] reliability criteria. It was seen that all results were generally found to provide reliable H/V curve criteria. However, unreliable H/V ratios were not included in the evaluations. Thus, the spectral amplitudes of Torul, Kürtün, Kelkit, Şiran and Köse were estimated on the maximum peaks which are compatible with the ground structure they were measured. As mentioned above, this study aims to contribute to the classification of the ground types from microtremor data for different districts of Gümüşhane province of Turkey. The analyses of microtremor measurements in different parts of

Figure 5. Examples of the H/V spectral ratio curves (solid black lines) and their standard deviations (dashed black lines) from different measurement points for: (a) TR7, (b) K3, (c) KL9, (d) SR27 and (e) KS38, respectively. Examples from each district were selected as randomly.

Sigma J Eng Nat Sci, Vol. 39, No. 4, pp. 374–391, December, 2021

the world show that obtained results are useful to define the subsurface soil profiles and to build the earthquake-­ resistant design. The principal subject in the geotechnical researches and in detecting of the site response properties is to define the subsoil types. There are a large number of studies such as Nogoshi and Igarashi [25] and Panah et al. [26] for the ground type classifications. Nogoshi and Igarashi (1970) suggested a general classifications of ground types for Hakodate City in Hokkaido, and Panah et al. [26] proposed a classification based on the H/V spectral ratio technique in eastern and central Iran. Predominant frequencies were used in both studies as a main factor, and these works are based on the dynamic subsurface properties such as frequency, amplification, period, alluvial thickness etc. In addition to these studies, Kanai and Tanaka [11] stated that the changes in microtremor periods are related to the subsoil type. They proposed that a relatively strong peak is recorded between the periods 0.1 and 0.6s if there is a simple stratified subsurface ground. However, if subsoil properties are complicated, more than two peaks can come in sight, which one of them is small around 0.2s and the other is large around 1.0s. A sharp peak can be seen between 0.1 and 0.2s (Z1 subsoil type) on a mountain whereas this peak can be recorded between 0.2 and 0.4s (Z2 subsoil type) for hard diluvial formation. In some records, the curves do not have regular shapes and several peaks can appear between 0.4 and 0.8s (Z3 subsoil type) on soft alluvial soils. In addition to these soil types, some peaks vary from 0.05 to 0.1s and from 1.0 to 2.0s (Z4 subsoil type) especially on soft soils. Thus, Kanai and Tanaka [11] suggested that microtremor amplitudes increase at ground surface in these periods. Since Nakamura [1] technique has been widely used for the evaluation of predominant periods of the soil, we preferred this model in order to identify the ground type classifications for the study region. These classifications by Kanai and Tanaka [11] were given in detailed in Table 6. H/V ratios give information about the stability of ground and are inversely proportional to strength of the measured area. Wave amplitude propagating in a limp ground grows in proportion to the weakness of the medium. Great H/V ratios can be obtained if the medium is weak, whereas small H/V ratios can be observed if the medium is strong [27]. However, a region with large H/V ratio cannot be said to have a limp structure and it should be supported with different techniques to confirm this result. When examining the H/V ratios given in Tables 1 to 5 and the regional variation maps plotted in Figures 6 to 10, H/V ratios are generally seen to be inversely proportional with subsoil types. In general, the largest H/V ratios were calculated for Z2 class and the smaller values were obtained for Z1 and Z3 classes. Therefore, the regions including sandy gravel, stiff sandy clay, loam or sandy alluvial deposits whose depths are 5m or greater will further enlarge the ground response compared to the other environments such as stiff rock composed of gravel, sand and other soils mainly consisting of

tertiary or older layers. Considering all the study regions, the H/V ratios can be said to be consistent with the ground on which they were measured (Tables 1–5). Considering the importance in microzonatin studies, it is obvious that the calculated H/V ratios and regional variation maps for different districts of Gümüşhane can provide contribution to knowledge in establishing the reliable ground-structure interaction, to settlement planning and to determine the safe areas. As in the H/V ratios, the predominant frequency (or predominant period) values also provide information about the weakness and stability of the medium. Detailed information of total 180 single station microtremor measurements was given in Tables 1 to 5, respectively. As seen in Tables 1 to 5, predominant frequency values vary from

Conclusions

In recent years, the microtremor survey method has gained popularity in seismic microzonation studies and in the estimation of ground types. The main purpose of this study is to detect the site response characteristics for Torul, Kürtün, Kelkit, Şiran and Köse districts of Gümüşhane province, Turkey. For this purpose, we used the ratio of the horizontal to vertical components of Fourier amplitude spectra, designated as H/V spectral ratio model of Nakamura technique based on the single station microtremor data analysis. In this context, predominant frequency, H/V ratio and Kg-value were estimated, and subsurface ground types were made according to the predominant periods for a total of 180 measurement points including all districts. Predominant frequency changes between 2.86 and 9.84 Hz, H/V ratio between 1.06 and 2.03 for Torul district. In addition, predominant frequency varies from 2.59 to 9.37 Hz, H/V ratio from 1.07 to 3.89 for Kürtün district; predominant frequency from 1.27 to 9.54 Hz, H/V ratio from

Acknowledgment

This study is supported by Gümüşhane University Scientific Research Project (GUBAP, Turkey) with project no 17.F5117.02.01. I thank to the editor and ­ anonyms reviewers for their useful and constructive comments, discussions and suggestions in improving this paper.

Data Availability Statement

The authors confirm that the data that supports the findings of this study are available within the article. Raw

Sigma J Eng Nat Sci, Vol. 39, No. 4, pp. 374–391, December, 2021

Figure 9. “Google Earth” images of Şiran district for (a) Predominant frequency, (b) H/V spectral ratio and (c) Subsurface ground profiles from predominant period. The areas with Z3 ground type were mentioned in the text.

Figure 10. “Google Earth” images of Köse district for (a) Predominant frequency, (b) H/V spectral ratio and (c) Subsurface ground profiles from predominant period. The areas with Z3 ground type were mentioned in the text. data that support the finding of this study are available from the corresponding author, upon reasonable request.

Conflict Of Interest

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethics

There are no ethical issues with the publication of this manuscript.

References

1. Nakamura Y. A method for dynamic characteris- for quantifying the degree of disintegration of weak tics estimation of subsurface using microtremor on rocks, engineering geology. 2016;207:115–28. [CrossRef] Sigma J Eng Nat Sci, Vol. 39, No. 4, pp. 374–391, December, 2021 391
2. Singh AP, Parmar A, Chopra S. Microtremor study Nakamura technique for the Northeast India. for evaluating the site response characteristics in Natural Hazards 2018;93:1013–27. the Surat city of western India. Natural Hazards [18] Taş N, Okumuş E, Öner Ş, Köksal C, İcat , Tanış MY 2017;89:1145–66. [CrossRef] S, ve ark. Gümüşhane il çevre durum raporu.
3. Babacan AE, Akın Ö. The investigation of soil- Gümüşhane Valiliği İl Çevre ve Orman Müdürlüğü, struc-ture resonance of historical buildings using Gümüşhane, Türkiye. 2003;179. seismic refraction and ambient vibrations HVSR [19] Şaroğlu F, Emre Ö, Kuşcu İ. Active fault map of measure-ments: a case study from Trabzon in Turkey, General Directorate of Mineral Research Turkey. Acta Geophysica 2018;66:1413–33. [CrossRef] and Exploration, Ankara, Turkey, 1992.
4. Fat-Helbary RES, El-Faragawy KO, Hamed A. Application of HVSR technique in the site effects Geodinamica Acta 2001;14:3–30. [CrossRef] estimation at the south of Marsa Alam city, Egypt. [21] Öztürk S. Regional and temporal variations of cur- Journal of African Earth Sciences 2019;154:89–100. rent earthquake activity in and around Gümüşhane: [CrossRef] A statistical assessment (inurkish T with English
5. Akkaya İ. Availability of seismic vulnerability index abstract). Gümüşhane University Journal of Science (Kg) in the assessment of building damage inVan, and Technology Institute 2017;7:25–40. [CrossRef] Eastern Turkey, Earthquake Engineering and [22] Nakamura Y. On the H/V spectrum, The 14th World Engineering Vibration. 2020;19:189–204. [CrossRef] Conference on Earthquake Engineering (WCEE),
6. Babacan AE, Ceylan S. Evaluation of soil lique- October 12–17, Beijing, China, 2018. faction potential with a holistic approach: a case [23] SESAME. Guidelines for the implementation study from Araklı (Trabzon, Turkey), Bollettino di of the H/V spectral ratio technique on ambient Geofisica Teorica ed Applicata. 2021;62:173–98. vibrations measurements, processing and inter-
7. Kanai K, Tanaka T. On microtremors. Bulletin of the pretation, SESAME European Research Project, Earthquake Research Institute 1961;39:97–114. WP12-Deliverable D23.12, Europan Commission
8. Adib A, Afzal P, Heydarzadeh K. Site effect classi- Research General Directorate. 2004; Project No. fication based on microtremor data analysis using EVG1-CT-2000-00026:62. a concentration–area fractal model. Nonlinear [24] Konno K, Ohmachi T. Ground-motion character- Processes in Geophysics 2015;22:53–63. [CrossRef] istics estimated from spectral ratio between hor-
9. Rezaei S, Choobbasti AJ. Application of the izontal and vertical components of microtremor. microtremor measurements to a site effect study. Bulletin of the Seismological Society of America Earthquake Science 2017;30:157–64. [CrossRef] 1998;88:228–41. [CrossRef]
10. Mukhopadhyay S, Bormann P. Low cost seismic microzonation using microtremor data: an example characteristics of microtremors. Journal of the from Delhi, India. Journal of Asian Earth Sciences Seismological Society of Japan 1970;23:264–80. 2004:24:271–80. [CrossRef] [CrossRef]
11. Beroya AAA, Aydin A, Tiglao R, Lasala M. Use of microtremor in liquefaction hazard mapping. MR, Motosaka M, Jafari MK, Uromieh A. Site Engineering Geology 2009;107:140–53. [CrossRef] effect classification in East-Central of Iran. Journal
12. Sanchez-Sesma FJ, Rodriguez M, Iturraran-Viveros of Seismology and Earthquake Engineering U, Luzon F, Campillo M, Margerin L, Garcia-Jerez A, 2002;4:37–46. Suarez M, Santoyo MA, Rodriguez-Castellanos A. A [27] Beker Y. Determination of soil properties by micro- theory for microtremor H/V spectral ratio: appli- tremor method in KTU campus, Master thesis, cation for a layered medium. Geophysical Journal Karadeniz Technical University, Trabzon, Turkey. International 2011;186:221–5. [CrossRef] 2011:127.
13. Pandey AK, Roy PNS, Baidya PR, Gupta AK. Estimation of current seismic hazard using web/haritalar.asp. Accessed on December 26, 2007.

Share and Cite