Seismic refraction survey for 8 profiles was carried out at the site(Figure 2)using the 24-channel signal enhancement seismograph which was GEOMETRICS Strata View model.Each profile was 240 m long and the distance between two successive geophones was 10 m.The technique(for P-wave generation)is to shoot the profile at 10 m away from each end(normal and reverse shooting)and also from its mid-point for profiles No.1 to No.3.While from the profiles No.4 to No.8 the shooting was done behind each end with 10 m,at mid-point,at the distances of 65 m(between geophones No.6 and No.7)and 180 m(between geophones No.18 and No.19).The source of energy was the hydraulic weighting drop with a total mass weight of 100 kg.While the generation of shear waves(SH)was acquired using the horizontal geophones and a slotted striker plate by hitting one end of the plate horizontally using sledgehammer.

Figure 2.Location of the seismic profiles at Gasco site.

The seismic refraction data obtained from GASCO site was processed and analyzed using both SIP and SEISREFA programs which are complete seismic refraction processing and modeling software.The tra-vel time-distance curves and the corresponding 1-D(one-dimension)ground models were obtained(Figure 3).

Figure 3.Travel time-distance curve(a)and depth model(b)for profile No.4.

The elastic properties and the material competence coefficients for the site are deduced(Table 1 and 2)for both surface and second layers.In the tables,the ultimate bearing capacity is indicated by Q_ult.

The values for both P- and S-wave velocities are used to compute the elastic constants of the detected subsurface layers(Table 1 and 2).These constants include the density(ρ),stress ratio(S_i),Poissons ratio(σ),rigidity modulus(μ),Youngs modulus(E)and Bulks modulus(B)as follows:

(1)Stress Ratio(S_i)

The propagation of seismic waves is proportional to the differential pressure between the sedimentary overburden and the pore-filling fluids.This means that the high fluid pressure formations will have differential pressure and abnormally low seismic wave velocities.According to Abdel Rahman [1989] the stress ratio is given as:

S_i=1-2(V_S² / V_P²).(1)

The obtained values of the study area range from 0.349 to 0.397 in the surface layer(Figure 4a)and from 0.202 to 0.391 in the second layer(Figure 4b).These values of stress ratio reflect competent materials at the site.

(2)Poissons Ratio(σ)

This ratio represents the geometrical change in the shape of an elastic body.Its value is 0.5 for fluids and it approaches zero for very hard indurate rocks.Negative Poissons ratio is also recorded for very hard indurate anisotropic rocks.Poissons ratio(σ)is given in terms of P- and S-wave velocities [Abdel Rahman,1992]:

σ=[(V_P/V_S)²-2]/[(V_P/V_S)²-2].(2)

The values lie in the range from 0.258 to 0.344 in the surface layer(Figure 5a),and range from 0.168 to 0.281 in the second layer(Figure 5b),which indicate hard rocks.For the surface layer the area of profile No.3 is of the highest value.But the second layer reflects the area of profile No.1 is of the highest value.

Figure 4.Distribution of stress ratio for the surface layer(a)and second layer(b).

Figure 5.Distribution of Poissons Ratio for the surface layer(a)and second layer(b).

Table 1.Selsmic velocities and elastic properties of the surface layer at GASCO site

Table 2.Seismic velocities and elastic properties of the second layer at CASCO site

There are a number of competence scales,based on the values of P-and S-wave velocities and elastic moduli in consequence,which are used to evaluate the soundness of rocks or soil materials,including the material index(M_i),concentration index(C_i),N-value and ultimate bearing capacity(Q_ult)as follows:

(1)Material Index(M_i)

Material index(M_i)is defined as the degree of competence of the material on the basis of the elastic moduli.This index must have relation with the material composition,the degree of consolidation,fracturing and jointing,the presence or absence of fluids in pore spaces which affect the elastic moduli.The material index is given [Abdel Rahman,1989] as follows:

M_i=(1-4σ).(3)

Where σ is the Poissons ratio.The values range between-0.377 and-0.035 in the surface layer(Figure 6a),while they range from-0.124 and 0.326 in the second layer(Figure 6b),and they indicate hard rocks.For the surface layer the area of Profile No.3 is of the highest value.But for the second layer the area of Profile No.1 is of the highest value.

Figure 6.Distribution of Material Index for the surface layer(a)and second layer(b).

(2)Concentration Index(C_i)

The concentration index(C_i)describes the degree of material concentration.The soil compaction status is considered,to a great extent,as a measure of the degree of competence for foundation and other civil engineering purposes.It depends on both elastic moduli of the soil and the pressure distribution at their

depth.The concentration index can be given in terms of velocity squared ratio [Abdel Rahman,1989] as:

C_i=(3-4 α)/(1-2 α).(4)

Where α is the velocity squared ratio(V_S²/V_P²).The obtained values in the surface layer ranges from 3.904 to 4.862(Figure 7a)and from 4.736 to

Figure 7.Distribution of Concentration Index for the surface layer(a)and second layer(b).

6.938 in the second layer(Figure 7b).This means that area of profile No.3 has the highest values for the surface layer,while the area of profile No.1 is of the highest values for the second layer.

(3)The N-value

This value is the resistance to penetration by normalized cylindrical bars under standard load,which is geophysically known as the Standard Penetration Test(SPT).N-value can be evaluated using the following formula [Imai et al.,1976; Stuempel et

Figure 8.Distribution of N-value for the surface layer.

V_S=89.9×N^0.341.(5)

Where V_S is the horizontal shear wave velocity.Material having low N-value indicates soft soil.The high values(between 133 and 551)in the surface layer(Figure 8)indicate a compacted soil at the site and hence there is no need to calculate the N-values for the second layer.

(4)Ultimate Bearing Capacity(Q_ult)

The natural and the artificial cyclic dynamic loading create additional load which is added to the building load and may cause soil liquefaction if the total loading value exceeds the ultimate bearing capacity(Q_ult)of the materialBowels J E.1984.Physical and geotechnical properties of soils.Mc Grew-Hill,London.. Q_ult can be given in terms of shear wave velocity [Abdel Rahman et al.,1992] as:

lgQ_ult=2.932(lgV_S - 1.45).(6)

The obtained values range between 3.99 and 16.534 kg/cm² in the surface layer(Figure 9a)and from 76.312 to 192.862 kg/cm²(Figure 9b)in the second layer.These values reflect that the area of profile No.3 is of the highest values for the surface and second layers.

Figure 9.Distribution of Qult for the surface layer(a)and second layer(b)

The available earthquake data around GASCO site contain two types of information:historical observations(pre-1900)of major seismic events that occurred over a period of about 4000 years and instrumental earthquakes(1900-July 1997)that collected from Maamoun et al.[1984],Ambraseys et al.[1994] and International Seismological Center(ISC).Data from Aug.1997 till Dec.2004 were gathered from the Bulletins of the Egyptian National Seismological Network(ENSN).From the distribution of the earthquake activity around the GASCO site(Figure 10)it is noticed that,GASCO site is located in the mainland of the eastern desert and surrounded by the earthquakes with magnitude ranging from less than 3 to greater than 5.

Figure 10.Earthquake activity in a 150-km circle around GASCO site.

The definition of seismic zones carried out in the present work depends on historical and instrumental earthquakes as well as the results of the previous geophysical and geological studies including the kind of seismic faulting [Papazachos et al.,1984],seismicity rate(a-value)[Papazachos,1980],b values [Hatzidimitriou et al.,1985],major trends of geological zones [Mountrakis et al.,1983],and the fault plane solutions of the major earthquakes(i.e.strike,dip and stress axes).A seismic zone is a configuration within which it is assumed that an earthquake recurrence process is considered to be spatially and temporally homogeneous.The delineation of the seismotectonic sources usually represents the major part of any seismic hazard analysis.According to this study,the seismic activities are concentrated in 4 seismic zones as follows:Central Gulf of Suez zone; Cairo-Suez District zone; Southwest Cairo zone; Beni-Suef zone.

3.2.5 Maximum Earthquake Magnitude

This step requires a determination of the maximum earthquake for each of the identified seismotectonic sources.In deterministic analysis,it is more common to define the maximum earthquake as a maximum credible earthquake [Reiter,1991].Another kind of maximum earthquake is the maximum historic earthquake with increasing 0.5 units.There are 4 seismic source zones affecting the site and each of them has its own seismic activity.

The earthquake activity in terms of focal depth and horizontal axes together with the later activity are used to determine the dimensions of the active seismic source area [Abu-Elenean,1997].The epicentral distribution of the seismic activity of the southwest Cairo seismic zone indicates that the maximum dimension of the activity is 41 km and the depth ranges from 7 to 25 km yielding a thickness of a seismogenic layer of 18 km.Thus the total seismogenic area of this zone is 738 km².According to Papazachos et al.[2004],the maximum earthquake of this zone is found to be with moment magnitude 6.31.

Using the same techniques for the central Gulf of Suez,the expected maximum earthquakes are determined,while the historical approach was used for the other seismic sources(Table 3).

Table 3.Maximum expected magnitude and hypocentral distances from each seismic source to GASCO site.

Seismic Source Maximum magnitude Hypocentral distance to the site/kmCentral Gulf of Suez Zone 5.5 139Cairo-Suez Zone 5.4 170Southwestern Cairo Zone 6.31 84Beni-Suef Zone 5.6 48

3.3 Simulation of Ground Motion at GASCO Site Using Stochastic Method 3.3.1 Input Parameters

(1)The source parameters [E(M₀,f)]

The earthquake source spectrum E(M₀,f)for the horizontal component of ground motion is given by the following:

E(M₀,f)=C(2πf)²M₀S(M₀,f).(7)

where S(M₀,f)is the displacement source spectrum and C is a constant.C=R_SVF/(4πρβ³R),with R=1 km,R_S=0.55(average shear wave radiation pattern),F=2.0(free surface effect),V=0.71(partition onto two horizontal components),ρ is the density at the source and β is the shear wave velocity at the source.

(2)The path [P(R,f),duration]

In this study,the three-segment-geometrical-spreading operator of Atkinson et al. [1995] is used.Geometrical spreading R^-1 is assumed for distance less than 70 km and R^0.0 for distances from 70 km to 130 km and R^-0.5 for greater distances.The Q model derived by Moustafa [2002] is used to represent the anelastic attenuation factor.This model is represented as:

Q=85.68f ^0.79.(8)

Although the Fourier spectrum of ground motion is not dependent on the duration,duration is a very important parameter for peak motions decrease with increasing duration.The duration is a function of the path,as well as the source:

T=T₀ + bR.(9)

Where T₀ is the source duration and bR represents the path dependent term that accounts for dispersion.Following Hanks et al. [1981],the source duration is related to the corner frequency by

T₀=f^-1_c.(10)

Empirical observations and theoretical simulations suggest that the path-dependent part of the duration can be represented by a connected series of straight-line segments with different slopes.The function of Atkinson et al. [1995] is used where the path duration is modeled as trilinear,using the transition distances 70 and 130 km for consistency with the attenuation model.The slope is 0.16 for the distance ranges between 10 and 70 km,-0.03 for the distance ranges between 70 and 130 km and 0.04 for the distance ranges from 130 to 1000 km.The slope is assumed to be zero for the distances less than 10 km.

(3)The site G(f)

The factors that control the change of the seismic ground motion amplitudes at the earths surface are impedance and diminution.Boore [2003] separated the path effect G(f)into two terms as follows:

G(f)=A(f)+ D(f).(11)

Velocity and density model is converted into site amplification A(f),using the square root of the impedance ratio between the source and the surface.For polarized shear waves,the impedance is defined as the product of the shear wave velocity,density and the cosine of the angle of incidence.The amplification is relative to the surface motion that would exist if the material were replaced with uniform material whose velocity and density equal to those at the source.

3.3.2 Results

It is noticed that,Beni-Suef seismic zone represents the nearest seismic zone to the proposed site(Table 3)and by application of stochastic simulation method,the peak ground acceleration at the site has obtained(Table 4).The simulated time history of PGA,velocity and displacement at the GASCO site for Beni-Suef seismic zones was calculated(Figure 11).

Table 4.Maximum PGA at the site for two seismic zones

Seismic zone Max.magnitude(M_W)Max.PGA/cm·s^-2Southwest Cairo 6.31 2.55Beni-Suef 5.6 9.34

Figure 11.Simulated time history of the maximum PGA,PGV and PGD at GASCO site for Beni-Suef zone

3.3.3 Response Spectra

Response spectra are defined on the basis of the response of single degree of freedom damped oscillator to the earthquake acceleration [Jennings,1983].The response spectra of an accelerogram serve the dual function characterizing the ground motion as a function of frequency and providing a tool for determining earthquake resistant design criteria.The response spectra were calculated with four selected damping values of 1%,3%,5% and 10% of the critical damping at GASCO site for Beni-Suef zone(Figure 12).

Figure 12.Response spectra for maximum PGA at GASCO site for Beni-Suef zone.