Vol.10, No.1 EARTHQUAKE ENGINEERING AND ENGINEERING VIBRATION March, 2011 Earthq Eng & Eng Vib (2011) 10: 143-152 DOI: 10.1007/s11803-011-0053-5 Site specic probabilistic seismic hazard analysis at Dubai Creek on the west coast of UAE Ayman A. Shama Parsons Corporation, 100 Broadway, New York, NY 10005, USA Abstract: A probabilistic seismic hazard analysis (PSHA) was conducted to establish the hazard spectra for a site located at Dubai Creek on the west coast of the United Arab Emirates (UAE). The PSHA considered all the seismogenic sources that affect the site, including plate boundaries such as the Makran subduction zone, the Zagros fold-thrust region and the transition fault system between them; and local crustal faults in UAE. PSHA indicated that local faults dominate the hazard. The peak ground acceleration (PGA) for the 475-year return period spectrum is 0.17 g and 0.33 g for the 2,475-year return period spectrum. The hazard spectra are then employed to establish rock ground motions using the spectral matching technique. Keywords: seismic hazard; PSHA; subduction zone; rock ground motion; spectral matching; logic tree; fault Correspondence to: Ayman A. Shama, Parsons Corporation, 100 Broadway, New York, NY 10005, USA Tel: +1-212-266-8585; Fax: +1-212-266-8540 E-mail: Ayman.shama@parsons.com Senior Supervisory Structural Engineer Received January 12, 2010; Accepted September 25, 2010 1 Introduction The site subject of this investigation is the proposed location for a new arch bridge at Dubai Creek on the west coast of the United Arab Emirates (UAE) at 25° 12’ 47” latitude and 55 20’ 48” longitude. The seismic design criteria for the bridge require that it satises an operational performance for the design earthquake (475-year return period), and life safety performance for the maximum considered earthquake (2,475-year return period). Ground motions for use in time history dynamic analysis of the structure were required by the design criteria for the two seismic hazards on the basis of site specic seismic hazard analysis. The main objective of this study was to develop the rock ground motions. Therefore, probabilistic seismic hazard analysis was rst conducted and counted for all the seismogenic sources that affect the site. Next, the hazard spectra obtained from the seismic hazard analysis are employed to simulate the ground motions. 2 Overview of the seismic hazard analysis 2.1 PSHA procedure Probabilistic seismic hazard analysis (PSHA) expresses the hazard in terms of the annual frequency of exceedance. Several models were developed for ground motion occurrences during the past decades. Nevertheless, the Poisson model (Cornell, 1968) is considered as a standard and was employed by the United States Geological Survey for the development of the seismic hazard maps of US. This model, which is adopted in this study, assumes that earthquakes occur independently in a xed time period wherein the probability of exceedance P(z) of a ground motion level z occurring in a design time period t at the site is related to the annual frequency of ground motion exceedance at the site γ (z) by: Pz zt () () = - - 1 e (1) The total frequency γ(z) is made up of contributions from each independent source and expressed as: () [ \ ,] , min \ z PZ z mr f m f rm rm m i i i N M R M i i i = ( ) > ( ) ( ) ∫∫ =1 dd (2) where i N = 1 = summations over all N seismic sources; m i min ( ) = the annual frequency of occurrence of earthquakes with magnitude above a minimum size m min on seismic source i; P [Z>z\m, r] is the probability that ground motion level z will be exceeded given an earthquake of magnitude m at a distance r from the site; and f m M i ( ) and f rm R M i i \ , ( ) are the probability density functions (PDF) on magnitude and distance, respectively. The rate of occurrence of earthquake M min is obtained by recurrence laws for different magnitudes. Both the truncated exponential recurrence model and the characteristic earthquake model were employed in the present study for the distribution of magnitude f m M ( ). The PDF of the truncated exponential recurrence model is expressed (McGuire and Arabasz, 1990):