Macroseismic Intensity and Attenuation Laws: A Study on the Intensities of the Iranian Earthquakes of 1975-2000

Macroseismic Intensity and Attenuation laws: A Study on the Intensities of the Iranian Earthquakes of 1975-2000

M. Zaré1 and H. Memarian2

1 Assistant Professor, IIEES, Tehran, I.R. Iran, e-mail: [email protected] 2 Associate Professor, Dept. of Mining Eng., Faculty of Engineering, Tehran University, Tehran, Iran, e-mail: [email protected]

ABSTARCT The earthquake intensities in Iran are reported from the inhabitant regions. A catalog for the recent earthquakes since 1975 until 2000 is prepared. The macroseismic intensity investigations in Iran are carried out after the great and moderate earthquakes (usually after the destructive events). The intensity reports for the recent Iranian earthquakes with an acceptable level of confidence are selected to form the catalogue of the intensity values of recent earthquakes in Iran. This catalog comprises the 470 intensity values for which the information for the localities of reported or estimated intensity values and the earthquake source specifications could be found. The uncertainty of the intensity estimations are found to be +/- one degree of intensity. The attenuation of the earthquake intensities in Iran is presented as well using the developed catalogue and the coefficients are presented in this article. The attenuation rate for different regions of Iran coincides well with the attenuation of strong motion accelerations using the Iranian data coming from the same period of time (1975-2000). 1. INTRODUCTION The earthquake intensities in Iran are reported from the inhabitant regions. The sources of information on the earthquake intensities in Persia are historical books, reports and the remnants of the historical monuments and archeological locations. Ambraseys and Melville (1982) have provided a catalogue for the Iranian historical and recent earthquakes. The intensity estimations have the uncertainties as they are qualitative in its nature and meanwhile it should be investigated through the eye witnesses. Meanwhile there are some benefits to use the intensity values and

Fourth International Conference of Earthquake Engineering and Seismology 12-14 May 2003 Tehran, Islamic Republic of Iran

IranCivilCenter.com - Technical and Educational Website of Iranian Engineers

Copyright © 2003-2004 Iran Civil Center, All rights reserved worldwide. Page 1 of 8

attenuation models such as the simplicity and their availability and non-dependence to instrumental facilities. The authors have already introduced the catalogue of the Iranian earthquake intensities reported or estimated since 1975 until 2000 (Zaré and Memarian 2002). This catalogue is used in the present article in order to develop the intensity attenuation law for Iran. The macroseismic intensity investigations in Iran are carried out after the great and moderate earthquakes (usually after the destructive events), and the results are published in the form of the iso-seismal maps. In the presented study, we used the catalog of 470 values of the earthquake intensities in Iran, for which the earthquake source characteristics and the mechanisms are well known. The homogenization of the reported intensities and to estimate an intensity value to the damage reports for the moderate events with no reported intensity evaluation was performed. We have then looked for the earthquake source data for which the intensities were reported. The iso-intensity curves were plotted for the earthquake prone areas with no intensity report. In this paper, we will present first how we established the earthquake intensity catalogue based on the intensity reports and the information on source and site conditions. The attenuation laws are represented finally for different parts of the country.

2. CATALOGUE OF EARTHQUAKE INTENITIES The intensity reports for the recent Iranian earthquakes with an acceptable level of confidence are selected to form the catalogue of the intensity values of recent earthquakes in Iran (Zaré and Memarian 2002). A selected part of this catalogue is presented in Table (1) in which the intensity values greater than VII (EMS-98) are introduced for the earthquakes having a magnitude greater than M=7.0. The catalog comprises the information for the localities of reported or estimated intensity values and the earthquake source specifications. There are 470 intensity estimations (in different localities and different earthquakes) for which this data catalogue is formed. The greater events such as Tabas (16 September1978, Mw7.4) and Manjil (20 June 1990, Mw7.3) are reported by several reconnaissance group. Different iso-intensity maps are reported for such earthquakes, in which the intensity estimations contradict sometimes with each other (especially for the regions in dessert or in mountains, which are hardly accessible). These reports comprise the intensity estimations in different scales; mainly the Modified Mercalli, MSK and a 5-degrees scale developed by Ambraseys and Melville (1982) for the Iranian earthquakes. The scale chosen as the basis for the present study was the European Macroseismic Scale (EMS-98, Grunthal 1998). It is tried to invert the other scales to EMS-98: most of the MSK estimations are not essentially different than that of EMS-98; the other scales were correlated with MSK and EMS-98 using the correlation charts (Japan TC-4, 1992) and the explanatory chart of the EMS-98 (Grunthal 1998). The recent earthquakes (1975-2000) were chosen, since the details on the casualties exist in more details, and the result could be directly compared with the accelerometric studies recently performed in Iran (Zaré 1999, Zaré et al 1999). Table (1) is a selective part of the prepared catalog, which comprises the events with different magnitudes ranging form 3.0 to 7.4. To provide the database for the present study, a bibliography is performed on the reconnaissance and damage reports on the recent Iranian earthquakes (i.e. Moinfar and Naderzadeh 1990, Berberian et al 1992). The information on the geotechnical conditions in different locations of the intensity estimations is gathered and the results of the observations (whenever existed) are entered to the catalog. The surface observations are not chosen to be the basis of surface geology conditions, since the misleading are expected using such observations to classify the soil conditions (Zaré et al 1999).



Table 1: A selected catalog of Iranian Earthquake Intensities (1975-2000), compiled for the estimated intensities of greater than VII and the earthquake magnitudes greater than 7.0, for the present study (Geo; site conditions, Q; Quality factor; Focal Mechanisms: ss; strike slip, Rv; reverse, Rv/SS; reveres strike slip).

Location Geo Macroseismic

epicenter

(Region)

Date ML Mb Ms Mw Focal

Depth

(km)

Epicentral

Distance

(km)

Macros

dist.

(km)

Macroseis.

Intensity

(I0)

Int.

EMS98

Focal

Mechan

ism

Q

Siah-Chashmeh Chaldoran 24/11/1976 6.8 6.1 7.3 7.0 15 32 36 VIII VII+ SS C

Kelisa Kandi Chaldoran 24/11/1976 6.8 6.1 7.3 7.0 15 37 25 VIII VII+ SS C

Zavieh-e Olia Chaldoran 24/11/1976 6.8 6.1 7.3 7.0 15 31 29 VIII VII+ SS C

Maku 1 Chaldoran 24/11/1976 6.8 6.1 7.3 7.0 15 52 31 VIII VII SS C

Arab Dizaj Chaldoran 24/11/1976 6.8 6.1 7.3 7.0 15 35 4 VIII VIII SS C

Tabas 1 Tabas 16/09/1978 7.0 6.7 7.3 7.4 11 27 0 X X Rv A

Khaneh-Rokni Tabas 16/09/1978 7.0 6.7 7.3 7.4 11 35 28 X VII Rv A

Ferdows 1 Tabas 16/09/1978 7.0 6.7 7.3 7.4 11 105 127 X VII Rv A

Esfahk 1 Tabas 16/09/1978 7.0 6.7 7.3 7.4 11 25 8 X X Rv A

Deyhuk 1 Tabas 16/09/1978 7.0 6.7 7.3 7.4 11 55 25 X VII Rv A

Boshruyeh Tabas 16/09/1978 7.0 6.7 7.3 7.4 11 76 58 X VIII Rv A

Kuli Kuli-Bonyabad 27/11/1979 6.4 6.1 7.1 7.0 25 27 0 IX IX SS B

Bonyabad Kuli-Bonyabad 27/11/1979 6.4 6.1 7.1 7.0 25 14 0 IX IX SS B

Gonabad 2 Kuli-Bonyabad 27/11/1979 6.4 6.1 7.1 7.0 25 91 53 IX VII SS B

Ghaen 1 Kuli-Bonyabad 27/11/1979 6.4 6.1 7.1 7.0 25 53 35 IX VII SS B

Ferdows 1 Kuli-Bonyabad 27/11/1979 6.4 6.1 7.1 7.0 25 125 90 IX VII SS B

Gisur Kuli-Bonyabad 27/11/1979 6.4 6.1 7.1 7.0 25 42 25 IX VIII SS B

Kerman 2 Sirch 28/07/1981 6.5 5.7 7.1 7.2 15.2 75 57 VIII VII Rv A

Anduhjerd Sirch 28/07/1981 6.5 5.7 7.1 7.2 15.2 26 21 VIII VIII Rv A

Joshan 1 Sirch 28/07/1981 6.5 5.7 7.1 7.2 15.2 23 3.5 VIII VIII Rv A

Hashtadan 1 Sirch 28/07/1981 6.5 5.7 7.1 7.2 15.2 19 0 VIII VIII Rv A

Sirch 1 Sirch 28/07/1981 6.5 5.7 7.1 7.2 15.2 31 16 VIII IX Rv A

Shahdad Sirch 28/07/1981 6.5 5.7 7.1 7.2 15.2 47 37 VIII VII+ Rv A

Manjil 1 Manjil 20/06/1990 7.3 6.8 7.7 7.3 19 9 0 X X Rv/SS A

Loshan 1 Manjil 20/06/1990 7.3 6.8 7.7 7.3 19 12 11 X VIII Rv/SS A

Rudbar 1 Manjil 20/06/1990 7.3 6.8 7.7 7.3 19 10 8 X IX Rv/SS A

Abbar 1 Manjil 20/06/1990 7.3 6.8 7.7 7.3 19 43 8 X VIII+ Rv/SS A

Ardekul Ardekul-Ghaen 10/05/1997 7.1 6.4 7.3 7.2 10 34 0 XI XI SS A

Hajiabad Ardekul-Ghaen 10/05/1997 7.1 6.4 7.3 7.2 10 22 12 XI X SS A

Khatibi Ardekul-Ghaen 10/05/1997 7.1 6.4 7.3 7.2 10 24 23 XI IX SS A

Garmab Ardekul-Ghaen 10/05/1997 7.1 6.4 7.3 7.2 10 8 38 XI VIII SS A

Baymargh Ardekul-Ghaen 10/05/1997 7.1 6.4 7.3 7.2 10 45 24 XI VIII SS A

Gezik Ardekul-Ghaen 10/05/1997 7.1 6.4 7.3 7.2 10 80 45 XI VIII+ SS A

Avaz Ardekul-Ghaen 10/05/1997 7.1 6.4 7.3 7.2 10 86 52 XI VIII+ SS A

2.1. Location of Intensity Observations The localities, where the intensities are observed or estimated, could be categorized from the great cities to the small villages. In the greater localities higher intensity variations could be observed, since the different constructions and the variations in the site conditions result in different observable intensities. In the case of the great cities, the highest intensity value has been entered to the catalog. For the small villages or farmlands (populated places that are demolished during the earthquake and some of them are then abandoned) it was sometimes difficult to find their coordinates on the recently published maps. The base map for such investigations was the



geographic atlas of Iran (topographic maps; 1:250,000 scale). The information corresponding to the localities of the observed or estimated intensities are entered in the first column of Table (1). 2.2. Site Effects The effects of the surface geology on the strong ground motions are previously studied for the stations of the accelerometric network of Iran (Zaré et al 1999). According to the mentioned study, the site conditions are classified according to their H/V ratio. This ratio is chosen based on the locations of the broadened H/V spectral ratios and the value of its amplitude. In brief, site category 1 corresponds to rock and hard alluvial sites, with Vs>800 m/s over 1st 30m depth and site amplification fundamental frequency (f0) over 15 Hz. The site category 2 relates to alluvial sites; thin soft alluviums, with 500<Vs<700 over 1st 30m depth, and 5<f0<15 Hz. The site category 3 corresponds to soft gravel and sandy sites, with 300<Vs<500 over 1st 30m depth and 2<f0<5Hz. Finally the site category 4 relates to soft soil sites; thick soft alluviums with Vs<300m/s over 1st 35m depth and f0<2Hz. This preliminary ranking is based both on studies at 50 sites where the geotechnical measurements were performed (compressional and shear wave velocity and microtremor measurements) and three component accelerograms were used to calculate the receiver function for the strong motions (Zaré et al 1999). In , we couldn’t specify the site conditions in the exact location of the intensity observation or estimation. It is however decided to include only the sites with a known H/V spectral ratio (where at least a well recorded accelerogram is obtained), and then to generalize the classification to two categories (“1” stands for hard soil conditions classes; 1 and 2, and “2” stands for soft soil conditions 3 and 4). It seems, however, that such classification is premature and it could be developed in the future researches based on the detailed site studies in the locations where the intensities are observed. The site classes (1 and 2) are given in columns 2 of Table 1. 2.3. Earthquake Source Parameters The source parameters, investigated in the present study, are the earthquake macroseismic epicenter, the data of the mainshock (UTC time), magnitude (in different scales; mb, Ms, ML and Mw, based on the available data), focal depth (using mostly the teleseismic estimations and in some cases the relocated hypocenters) and the focal mechanisms (based on the Harvard seismology Web site, 2001). The macroseismic epicenter is chosen as the location of the greatest damages (where the highest intensity values is assigned; I0). This location could be near to, or coincided with, the real epicenter (whenever the region is densely populated) or far from it (when the region is sparsely populated; in the central and eastern Iran, where the great deserts exist). In the cases that no precise data could be found from the macroseismic epicenter, a general name is applied based on the instrumental epicenter data and the felt intensities in other places. Zaré and Memarian (2002) showed the general correlation of Mw with mb and ML for the magnitudes less than 6.0, and with Ms for magnitudes greater than 6.0. The magnitudes are plotted in the upper part of figure (1) against intensities observed in different epicentral distances. Such comparison is performed between the macroseismic epicentral intensities (I0) and the observed intensities and is shown in the lwer part of figure (1).



Figure 1. The magnitude/intensity (above) and macroseismic-intensity/intensity (below) relationship for the

data presented in the catalogue (Zaré and Memarian 2002). As it is shown in figure (1), an earthquake magnitude M=4.0, could cause the intensities up to V. This correlation could be shown for the intensities up to VIII for M=5.0, up to IX for M=6.0, and up to XI for M=7-8. Different macroseismic intensities could cause the intensities up to that degree (i.e. for I0 =V, the intensities up to V could be observed; Figure (1)). The relationship of earthquake magnitudes and the macroseismic intensities could be shown with a linear regression which gives the best correlation coefficients. The relationship is I0= 1.45 M – 1.22 (1) The standard deviation for this relationship is 0.81 and the correlation coefficient is 0.81. Using this formula, for a magnitude 6.0, the macroseismic epicentral intensity could reach to VII+ (mean value) and VIII+ (mean plus one sigma standard deviations). 2.4. Distance The definition of distance is one of the most important factors for the attenuation formulas. The macroseismic and epicentral distances are estimated in this study. The macroseismic epicenter distance is the nearest distance between the location of the felt intensity and the macroseismic epicenter (where the maximum intensity is felt). The epicentral distance is the nearest distance between the location of the felt intensity and the instrumentally located epicenter. According to the



importance of the event and precision of the field observations, the location of the instrumentally located epicenter and the macroseismic observed intensity could be different (the greater the magnitude of the event and the more precise the observations, the locations of these two epicenters are nearer to each other). Sometimes the earthquakes occur nearby the border lines of two countries, the determination of iso-seismal curves is hard. 2.5. Intensity Values and Quality Factor The intensity values are reported either in MMI or in MSK scales. The intensity values are unified to EMS-98, which is a 12 degree European macroseismic scales developed essentially based on MSK (its old version). The bar charts are used to find the equivalent of the MMI values in the MSK and EMS-98 scales. The values in MSK could be directly taken as EMS-98 values, since there is no major difference between these two scales (at least in the scales of our observations). The Quality factors are assigned in four levels (from “A” to “D”). The quality level of “D” (the basic level of the data) was assigned, when at least a location of macroseismic intensity and an intensity value could be assigned (no more information existed). Level “C” is assigned when the quality of a magnitude and instrumental epicenter location were reported but the doubt existed on the location of the intensity observations and other source parameter. Level “B” was when all information was available from the intensity observation locations and the source parameters, but the location of the macroseismic intensity and/or the its level was uncertain. The quality level “A” introduces the best information quality. As it presented in table 1, for the magnitudes greater than 7.0, most of the data in the catalog presented in this paper have the “A” level of quality. The worse cases are the magnitudes less than 5.0, when the most uncertain information exists. 3. ATTENUATION OF INTENSITIES IN IRAN The attenuation model of the region is studied based on the developed catalogue for Iran. The data are sub-grouped to Alborz-Central Iran, Zagros and a combination of all 470 Iranian intensity data. The site conditions are considered in this study in two classes as explained previously in this article. The class one comprises the categories 1 and 2 (Zaré et al 1999) and class two corresponds to categories 3 and 4 in the same reference. The one step regression is used to develop the coefficients, using the STATISTICA software. The coefficients for the attenuation laws are estimated for the Alborz-Central Iran, Zagros and all parts of Iran. The standard deviation is assessed for each of the coefficient sets. The formula presented herein were those showed the least sigma and greatest coefficients of correlation. At first the regression is performed without the site condition coefficients. The regression is preformed for intensity (as a dependent parameter) and magnitude (Mw) and epicentral distance (to be taken as horizontal distance to the fault/source, in km) based on the following simple model. This model considers the an-elastic attenuation geometric expansion separately, as proposed by Ambraseys (1995);

I=a.M –b.X – clnX +σ.P (2) The coefficient for this attenuation formula is presented as given in table (2). Another attempt is carried out for intensity (as a dependent parameter) and macroseismic intensity (I0), and the macroseismic epicentral distance (to be taken as horizontal distance to the fault/source, in km) based on the equation (3). The coefficients for such equation are represented in table (3). I=a.I0 –b.X – clnX+ σ.P (3)



Table (2). The coefficients for the equation (2), R is the coefficient of correlation

Region a b C σ R (Correlation)

Alborz – C.Iran 1.207 0.0126 0.227 0.92 0.88

Zagros 1.148 0.0214 0.120 0.72 0.94

IRAN 1.175 0.0140 0.227 0.88 0.77

Table (3). The coefficients for the equation (3), R is the coefficient of correlation

Region a b C σ R(Correlation)

Alborz-C. Iran 0.994 0.0160 0.212 0.92 0.86

Zagros 0.942 0.0152 0.188 0.88 0.77

IRAN 0.942 0.0162 0.196 0.91 0.82

To consider the site conditions the same procedure explained for equation (2) (Intensity as dependent and magnitude as independent parameter) is taken in equation (4) by adding the site coefficients (to be either 1 or 2 as explained above for hard or soft soils respectively). The coefficients for such equation are given in table (4). Another attempt similar to equation (3) is carried out for intensity (as a dependent parameter) and macroseismic intensity (I0) as independent parameter, adding the “site” parameter, in equation (5). The coefficients for such equation are represented in table (5).

I=a.M –b.X – clnX+ ci.Si +σ.P (4)

Table (4). The coefficients for the equation (4), R is the coefficient of correlation.

Region a b c d1 d2 R(Correlation)

Alborz-C. Iran 1.268 0.0142 0.256 -1.061 -1.590 0.80

Zagros 1.107 0.0188 0.227 -0.040 0.566 0.55

IRAN 1.279 0.0125 0.270 -1.274 -1.600 0.76

I=a.I0 –b.X – clnX+ ci.Si +σ.P (5)

Table (5). The coefficients for the equation (5), R is the coefficient of correlation.

Region a b c d1 d2 R(Correlation)

Alborz-C. Iran 0.719 0.0146 0.221 1.806 1.524 0.84

Zagros 0.702 0.0178 0.202 1.577 1.275 0.72

IRAN 0.746 0.0142 0.226 1.422 1.280 0.79



4. CONCLUSION The intensity of the recent earthquakes in Iran are listed in order to produce a catalog, to be used for developing the attenuation laws for the intensities in Iran. This period of time was selected since the source parameters could be found with the best qualities. The data presented in this paper indicates interesting correlation between magnitude and macroseismic epicentral intensities, as well as the relationships between the observed intensities (in different distances) and the magnitudes and/or macroseismic intensities. The attenuation model fir the earthquake intensities in Iran are presented considering the magnitude or macroseismic intensities and for the cases when we consider or not the site effect parameter. Based on the dispersion of the site effect estimations, it is recommended to use the attenuation laws which are developed in this study with no consideration of site effects. The general (form of the attenuation of intensities in Iran follows the attenuation model of acceleration for Iran (Zaré 1999). 5. REFERENCES 1. N.N. Ambraseys (1995). The Prediction of Earthquake Peak Ground Acceleration in Europe,

Earthquake Eng. Struct. Dyn. Vol.24, pp.467-490. 2. N.N. Ambraseys N.N., Melville C.P., (1982). A History of Persian Earthquakes , Cambrige

Earth Science Ser. 3. M. Berberian, M. Qorashi, J.C.P. Jakson (1992). The Rudbar-Tarom Earthquake of 20 June

1990 in NW Persia: Preliminary field and Seismological Observations, and its Tectonic Significance, Bul. of Seismological Soc. of America, Vol.82, No.4, pp.1726-1755.

4. G. Grunthal (Editor), (1998). European Macroseismic Scale 1998, Conseil de l’Europe, Cahiers du Centre Européen de Géodynamique et de Séismologie, vol. 15, 99.

5. Japan Working Group for TC4 Committee (1992). Seismic Zoning on Geotechnical Hazard, Draft, 114p.

6. Harvard University, Seismology Dept., Web page, (2001). http://www.harvard.seismology.edu/. 7. A.A. Moinfar A.A., M. Naderzadeh A. (1990). An Immediate and Preliminary Report on the

Manjil Iran Earthquake of 20 June 1990, Building and Housing Research Center, Rep. No#119, 68p.

8. M. Zaré (1999). Contribution à l’études des Mouvements forts en Iran : du Catalogue aux lois d’Atténuation, Thèse de Doctorat , Université Joseph Fourier , 237p.

9. M. Zaré, H. Memarian (2002). The Iranian Earthquake Intensity Database: 1975-2000, 12 International Association of Engineering Geology Congress, South Africa, Proceedings (published on the CD-ROM).

10. M. Zaré, P.-Y. Bard, M. Ghafory-Ashtiany (1999), Site Characterizations for the Iranian Strong Motion network, Journal of Soil Dynamics and Earthquake Engineering, Vol.18, no.2, pp.101-123.



Documents

Macroseismic Intensity and Attenuation Laws: A Study on the Intensities of the Iranian Earthquakes of 1975-2000