A PROBABILISTIC ASSESSMENT OF THE SEISMIC HAZARD ANALYSIS IN IRAN

Hafez Keypour

(Thesis Supervisor: Mustafa Erdik)

ABSTRACT

This study encompasses the development of a seismic hazard map for Iran based on the historical earthquake data, geology, tectonics fault activity and a rational source regionalization. The findings are presented and compared with previous work of various researchers.

About 51OO earthquake records in the period 300 B.C. to 1991 were collected from various sources to form the basic catalogue of the earthquake. Also relevant data from catalogues of neighboring countries were correlated and combined with this catalogue. This seismologic data acquisition constitute one of the most basic and important steps of the hazard assessment.

Aspects of seismology and engineering judgments were considered for assembling seismicity data and decision upon the distribution of earthquakes in each zone. The fault locations were digitized from the relevant maps of Atomic Energy Organization of Iran [7]. Inaccuracy in the map and/or digitization does not affect the reliability of the seismic hazard analysis, since only area and strip sources are considered in the source regionalization.

For the seismic source identifications tectonic and geotectonic map, seismicity and geologic knowledge were used. Fault locations and earthquake epicenters were correlated and the regions of homogeneous seismicity were constructed. From the seismicity statistics of each zone, the annual number of earthquakes was calculated and adjusted annual rates were found. Only the attenuation relationships for bedrock are considered. For the soil sites some recommendation was presented in the last section. Ground motion attenuation relationships of Campbell [15], Joyner & Boore [24], and Intensity attenuation relationships of Erdik [19] Chandra- McWhorter-Nowroozi [18] were utilized.

The seismic hazard maps are developed using probabilistic for the seismic source identifications tectonic and geotectonic map, seismicity and geologic knowledge were used. Fault locations and earthquake epicenters were correlated and the regions of homogeneous seismicity were constructed. From the seismicity statistics of each zone, the annual number of earthquakes was calculated and adjusted annual rates were found. Only the attenuation relationships for bedrock are considered. For the soil sites some recommendation was presented in the last section. Ground motion attenuation relationships of Campbell (15], Joyner & Boore [24], and Intensity attenuation relationships of Erdik [19] Chandra­ McWhorter-Nowroozi [18] were utilized.

Procedures portraying the horizontal Peak Ground Accelerations (PGA) that corresponds to a certain annual probability of exceedance were used. The Seismic Risk III computer program adapted to PC’s was used in these analyses [11].

A grid interval of 0.125 degree latitude and 0.15 degree longitude covering the whole country is selected and the PGA at the grid points corresponding to specified probabilities of not being exceeded during specific periods were plotted. These are called Iso-acceleration maps.