Serapeum temple and the ancient annex daughter library in Alexandria, Egypt: Geotechnical–geophysical investigations and stability analysis under static and seismic conditions

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

http://www.elsevier.com/copyright

Author's personal copy

Serapeum temple and the ancient annex daughter library in Alexandria, Egypt:Geotechnical–geophysical investigations and stability analysis under static andseismic conditions

Sayed Hemeda a,⁎, Kyriazis Pitilakis b

a Department of Restoration, Faculty of Archaeology, Cairo University, Giza, Egyptb Department of Civil Engineering, Aristotle University of Thessaloniki, Greece

a b s t r a c ta r t i c l e i n f o

Article history:Received 15 July 2009Received in revised form 16 February 2010Accepted 18 February 2010Available online 4 March 2010

Keywords:Serapeum templeUnderground monumentsGeotechnical investigationMicrotremorsStatic analysisSeismic response analysis

The Serapeum Temple and the ancient annex library in Alexandria, Egypt from the Greek–Roman era,represent cultural heritage of outstanding universal values. They suffer weathering— aging as well as multiplegeotechnical and earthquake hazards. A pilot study has been carried out in order (a) to define the pathologyand the causes of deterioration and degradation (b) to study the seismic performance of this kind ofmonuments (c) to assess the global risk due to combined hazards and (d) to define the appropriate retrofittingtechniques. In the paper a general outline of the various tests, surveys and analyses is presented, highlightingthe most important issues related to the static and seismic stability of the above monuments. A particularfeature of Serapeum is the column of Diocletian, founded just above it.The paper presents the comprehensive field and laboratory surveys and tests undertaken in the site, and thenumerical analysis of the monument under static conditions and seismic conditions. The field testing programcomprises various geotechnical and geophysical field and laboratory tests aiming to define the physical,mechanical and dynamic properties of the soils and soft rock materials of the site where the monuments isfounded. Then an extensive parametric 2D and 3D numerical analyses were performed for the pillar and theSerapeum, subjected to seismic motions having different values of peak ground acceleration PGA and frequencycontent. Advanced soil and rock elasto-plastic modeling has been used throughout the different phases of thenumerical analysis. The seismic analysis combined with the static one allowed us to define the pathology of themonuments and to estimate the ultimate load that they can survive under their present conditions.

© 2010 Elsevier B.V. All rights reserved.

1. Introduction

Around the commemorative Column of Diocletian (Pompey'spillar), some monuments can be seen. On the backside, there areremains of a Serapeum or a temple of god Serapis, now heavilydamaged. It was excavated during the reigns of Ptolemy II andPtolemy III, but it was damaged due to the revolts of the Jewishpopulation in Alexandria during the reign of the Emperor Trajan (89–118 A.D). There were two galleries at the back of the temple. Bothgalleries were cut in the calcarenitic rock with high porosity (Fig. 1).

The stability conditions of the historical monuments are of crucialinterest, especially in regions like the Mediterranean Basin andparticularly Alexandria, Egypt, where the seismotectonic and weath-ering regime are active, and the geological structure is complex.Phenomena like settlement and slope movements, as well as earth-

quakes and tectonic activity contribute to the damages of thehistorical buildings. The ground water activity is also an importantfactor. Environmental factors are also important and should be takeninto account, when different protection measures are designed.

The deformation pattern of these underground monuments iscomputed using an advanced non-linear elasto-plastic model in theframework of PLAXIS. Mohr's–Coulomb and Biot model are used tomodel stress–strain conditions and consolidation. PLAXIS (PLAXISManual, 2002) is a commercial software specifically designed foradvanced geotechnical modeling (Bergado et al., 2003, Karstunenet al., 2006, Guetif et al., 2007, Berilgen, 2007, Arslan and Rpsassan-chez, 2007).

For the seismic analysis, we modeled the complex mediumassuming an equivalent plane strain conditions, applying differentproperly scaled seismic scenarios, corresponding to the seismtectonicfeatures of Alexandria (Kalamata–Greece, 1986, Erzincan–Turkey,1992, Aqaba–Egypt, 1995). Advanced soil–rock elasto-plastic model-ing has been also used. Extensive time domain parametric analysiswas performed in order to examine the response of the pillar andSerapeum. The aim is to examine the pathology of the complex and to

Engineering Geology 113 (2010) 33–43

⁎ Corresponding author.E-mail address: [email protected] (S. Hemeda).

0013-7952/$ – see front matter © 2010 Elsevier B.V. All rights reserved.doi:10.1016/j.enggeo.2010.02.002

Contents lists available at ScienceDirect

Engineering Geology

j ourna l homepage: www.e lsev ie r.com/ locate /enggeo


determine the threshold PGA values for which the pillar and theSerapeum could survive.

2. Pathology

The Serapeum temple and ancient annex library in Alexandria areexcavated in the oolitic sandy limestone (calcareous cemented sand);it is yellowish white massive, fine to medium grained cross-beddedsandstone cemented with calcareous cement. Intersected conjugatedjoints filled with very fine friable sand saturated with water in thelower parts. This unit is underlined by loose calcareous sandstone. It isbrownishmedium to fine grained calcareous limestone over saturatedwith ground water. It overlies the El Hagif formation (Pliocene) or theolder Miocene. Surface quaternary deposits obscure actual contact(El-Fouly, 2000).

In the whole, the rocky mass within which the Serapeum templeand ancient annex library are excavated, presents alteration anddecomposition, of the visible materials above all in the arenaceouslevels and in the relation to the discontinuity stages. Among the mostimportant factors which have contributed to the present poorcondition so the rock mass and the monument is the presence ofthe water; in particular it exercise pressures in the rock joints and thefilling materials, while temporal temperature variations contribute inthe deterioration.

The presence of a fractured system of joints, with variableorientations, determines the subdivision of the rocky mass intoprismatic bodies of variable volume; in some cases the discontinuitiesare filled up with materials of poor mechanical properties, theirspatial and temporal deformation produce decompression phenom-ena in the fractured rock and reduction of the internal links of the rock

masses. Besides that erosion phenomena are added, due to the windcurrents and humidity/salts action (Fig. 1). Finally we also observedthe action of meteoric waters, which attack chemically the calcareouscomponents of the rock with a consequent decay of the mechanicalcharacteristic of the rock itself.

Additionally to the causes described above, we must considerother significant factors like the action of the atmospheric agents suchas temperature and wind, the effect produced by the roots ofvegetation forming inside the fractures and the sub verticalmorphology of the walls of the ridge.

These causes, often concomitant among themselves, provokeslowly deforming phenomena on the walls, which, by irreversibilitycharacteristics, lead to localized collapses with consequent falling ofrock blocks of various sizes.

The forms of instability that have been identified are essentially oftwo types:

a) Rock masses of various dimensions partially isolated by joints ofsub vertical discontinuity and parallel to the wall, which presentthe risk to collapse with wall base movement.

b) Partially decompressed, disjointed rocky masses divided intoblocks, presenting the risk of falling by single blocks.

The site presents intensive weathering indicated by rock surfacescaling: disintegration of construction material and intense rockmeal. Damp rock surfaces in particular for semi-sheltered parts ofthe excavation and intense honeycomb weathering, white saltefflorescence and yellowish brown iron staining can be noted atmany parts. The structural damage is represented by ceilingcracking, rock intensive erosion and surface decay. Partial collapse

Fig. 1. Serapeum temple and ancient annex library of Alexandria. Present state of the archaeological site.

34 S. Hemeda, K. Pitilakis / Engineering Geology 113 (2010) 33–43


of some parts of the ceilings and walls, rock exfoliation especiallynoted in the ceiling of the narrow entrances and tunnels, intensiveabrasion that are found at the deepest parts, and buckling of some

parts, as well as deep erosion of the low levels of these tunnels andmass wasting from its ceiling and walls of corridors. Irregular shapecracks.

Fig. 2. TGA-DTA results, thin section blue dyed X40, the blue is the porosity.

Fig. 3. Photomicrographs and (EDX) micro analysis of the tested sandy oolitic limestone samples from the Serapeum body structures.

Fig. 4. Pore size (diameter A) distribution for collected weathered rock samples.

35S. Hemeda, K. Pitilakis / Engineering Geology 113 (2010) 33–43


3. Bulk structure of rock and construction materials

Different tests have been performed to assess the durability andthe weathering effects on the bulk structure of the rock mass used asconstruction material. The following set of mineralogical analyses hasbeen performed for the soft rock and the construction materials.

– X-ray diffraction (XRD)– Thermal analyses DTA&TGA– X-ray fluorescence analysis (XRF)

– Chemical analyses

(For plaster and painting layers collected samples)

– Transmitted plane polarized light

(For the collected rock samples).

– Scanning electron microscopy (SEM), attached with EDX Micro-probe (Energy dispersive X-ray) microanalyses.

– Porous media characterization for weathered and sound rocksamples.

Fig. 5. Geotechnical borehole _1.

Fig. 6. Geotechnical borehole _2.



– Pore size measurement and specific surface area by nitrogen BET–TPV.

– Determination of the specific surface area (SSA).– Grain size characteristics of the weathered (salt contaminated)

rock samples.– Saturation coefficient, S.– Capillary water uptake measurements.

Based on this extended set of tests a detailed mineralogy de-scription of the rock mass is achieved together with other construc-tion materials (plaster). A detailed presentation of these tests may befound in Hemeda (2008).

The statistical analysis of the XRD and XDF results data indicatedthat the Serapeum temple and ancient daughter library are carved orexcavated in the fossiliferous sandy oolitic limestone composed ofcalcite CaCO3 (47%), quartz SiO2 (31%), halite NaCl (12%), gypsum CaSO4·2H2O (10%). The rock is yellowish and can be characterized asmedium grained with uniform relative grain size, angular to subangular grain shape with equidimensional form and rough surfacetexture. Sound pieces of rock are characterized by medium compact-ness and durability while the weathered pieces are characterized bylow compactness and durability. It must be mentioned that weather-

ing attacked strongly the rock materials, started from the surface andcontinuing inward thus loosing the mineral fabric (Fig. 2).

The thermal analyses were carried out using computerized DT.50thermal analyzer (Shimadzu Co., Kyoto, Japan). The heating rate was20 °C/min, up to 1000 C for DTA&TGA and 950 C for TMA, and undernitrogen atmosphere (30 ml/min) (Fig. 3).

The pore diameter distribution of these rock samples is, 10–20A(4%), 20–30A (9.17%), 30–50A(11.1%), 50–100A (15.847%), 100–200A(17.80%), 200–1960A (41.97 %), and nm 2.26599 E-05, and BET(m2/gr) 2.21098, TPV (ml/gr) is 0.00992, andmicro porosity (%) is 1.79327(Fig. 4). And for the adsorption/desorption isotherm diagram. Therock samples from the sound rock masses coming from the same site(SR2) have the following characteristics: pore diameter distribution10–20A (9.46%), 20–30A (12.5%), 30–50A (14.98%), 50–100A(12.72%), 100–200A (7.93%), 200–2040A (42.4%), and nm 1.5522E-05, BET(m2/gr) 1.51452, TPV (ml/gr)=0.00232, and micro porosity(%)=0.42542.

Analysis of the surface: 0.01-more than 2.000 m2/g. Adsorption/desorption isotherm, pore diameter range: 0.35–200 nm (3.5–1.000 A). Minimum pore volume: better than 2.2×10–6 ml, theinstrument (NOVA-2000 Ver 6.11) with the two analysis stations.Adsorbed nitrogen and adsorption tolerance 0.1000 mm were used.

Fig. 7. Evolution of strain with in uniaxial creep (Tests 1 and 2).

Fig. 8. Evolution of strains with time in the triaxial creep test (tests 1 and 2).



4. Geotechnical investigations

The field geotechnical investigation comprises mainly SPT testingand drilling–sampling–water table measurements. To a depth of 15 m(Figs. 5 and 6).

The laboratory tests have been carried out on intact rock in order todefine the physical and mechanical properties of the rock. The testingprogram, in about 150 core specimens (diameter of 42–44 mm2 andheight of 91 to 103 mm), comprise triaxial and creep tests. We cansummarize the results of the experimental tests as follows.

Fig. 9. Triaxial test at different confining pressures.

Fig. 10. ReMi testing results to estimated in situ Vs profile.



Fig.

11.M

axim

umgrou

nddisp

lacemen

tsin

thetunn

elof

Serape

ιum

temple,

11.2

mun

derthepilla

r.



4.1. Very slow uniaxial creep tests on rock samples

Long-term uniaxial creep tests were conducted on standardcylindrical rock samples. The sample was set between two stiff steelplates, with a steel cap set between the sample and the two plates.During each test, two high-resolution displacement sensors set in twovertical planes at a 90 angle allowed both the relative rotation of thetwo plates and their average relative displacement to be measured.

The objective of these tests is first to quantify the viscousparameters that govern the long term behavior of these undergroundstructures and, secondly, to define the necessary parameters for thenumerical modeling. Most specimens under constant axial stressshows complete creep phase: transient, steady and tertiary creepphases (Fig. 7).

4.2. Triaxial creep tests on rock samples

Two rock specimens were tested under different constant axialstresses and different constant confining pressures for a period ofabout 300 h according to ASTM standard (ASTM D4406-93). Acompression machine (consolidation machine, capacity of 5000 kN)(GDS Instruments Limited, England) was used to apply constant axialload and the rock specimens were placed into a triaxial (GDS) cell toprovide constant confining pressure.

The objective of the triaxial creep tests is to determine visco-plastic parameters of the soft rock specimens under confinement. Thetime-related parameters are monitored, recorded and analyzed.

The axial strain-time curves are given in Fig. 8. From the results it isproved that there is an instantaneous transient creep state underconstant axial load and confining pressure followed by a steady creepphase. Post-test deformation increases rapidly at the beginning tothe first few hours of the test and then tends to remain constant.No specimen failed after at the end of the test duration. Crackpropagation (in the brittle field) and pore collapse (at higher stressconditions) are therefore the dominant deformation mechanisms forthe Alexandria surface rock as it has been also demonstrated.

4.3. Triaxial compressive strength

Fig. 9 presents the results of isotropic compression test performedon undistributed–intact rock samples to determine the shearresistance parameters. It is found that the sandy oolitic limestonerock samples have the following mechanical and elastic properties:c=650 kN/m2, φ=36o, E=270 to 360 MPa, ν=0.26–0.29,G=140.6 MPa, and K=272.7 Mpa.

5. Geophysical tests, ambient noise measurement

Array microtremor measurements have been carried out to definethe shear velocity profile at the archaeological site. The ReMi methodhas been used and the estimated Vs profiles are given in (Hemedaet al., 2007) (Fig. 10).

The obtained shear wave seismic velocities show a relatively highrange of shearwave velocities ranging between260 m/s and1420 m/s.However, it is clear that the ground conditionswhere the undergroundexcavations of Serapeum in Alexandria are excavated cannot beclassified as real rock at least close to the surface (Hemeda, 2008). Itsmain characteristic is the strong weathering.

6. Preliminary 2d static analysis

In the initial static analysis, the excavation is modeled assumingnon-linear soil/rock behavior and Mohr coulomb failure criterion. Thefollowing parameters are used (i) for the calcarenitic rock material:φ=36o, c=600 kN/m2, E=2.280 E+06 kN/m2, ν=0.28, andVs=805 m/s. For the granite material: φ=50o, c=900 kN/m2,

E=1.000E+07 kN/m2, ν=0.30, and Vs=1243 m/s, Fig. 11 presentsthe main results from the preliminary static analysis. The maximumground displacements in the tunnel of the Serapeum temple, 11.2 munder the pillar, is computed equal to 739.55 10−6 m, but thesidewalls of the Serapeum is under relatively high vertical compres-sion and shear stresses. The computed peak vertical effective principalcompressive stress is 2.5 103 kN/m2. While the peak shear stressequal to 974 KN/m2 (Table 1).

7. Seismic hazards in Alexandria

Alexandria is located approximately on three tectonic plateboundaries that interact with each other generating a complex systemof major and local faults close to Alexandria offshore. These faults areassociatedwith small tomoderate earthquakes. The focal mechanismsand the waveforms of the offshore events reflect the complexity ofthis tectonic zone. Because of this complexity, it was very difficult torepresent themoderate earthquake of 1998 by one source and fit bothP and SH waves.

Detailed historical earthquakes information for Alexandria isprobably missing. However the destruction of the famous Alexandrialighthouse and off-shore under-water archaeological remains inAbukir bay strongly support, either local or remote earthquakes thatdestroyed the city.

Close offshore events seem to be dominated by strong intensity inthe low frequency range (less that 4 Hz), while far field have very longduration of shaking, however their peaks are relatively weak. Thesepeaks are also of a very low frequency, which is coherent with theresponse spectra. (El-Sayed et al., 2004).

8. Seismic response analysis

In the present study, we have selected three reference earth-quakes. (i) Aqaba, Egypt,M=7.1, Ml=6.2, 1995 (ii) Erzincan, Turkey,Mw=6.9, Ms=6.8, Rrup=2 km, Re=1 km., 1992 and (iii) Kalamata,Greece, Ms=5.8 Mw=5.9, and Ml=5.5. 1986. The time histories ofthese earthquakes represent different seismotectonic settings andfrequency content; all were scaled to three peak ground accelerationvalues equal to 0.08 g, 0.16 g, and 0.24 g. The design acceleration inAlexandria according to the Egypt seismic code is 0.08 g.

We believe that with the advances in computational methods it isnow possible to predict with reasonable accuracy the seismicdemands on these geometrically complex underground monuments.

Table 1Geotechnical properties of rock and other construction materials were used in thestability static and seismic analysis of Serapeum of Alexandria.

Parameters Name Rock material Granite

Rock unit weightabove phreatic level

γunsat kN/m3 17 24.4

Rock unit weightbelow phreatic level

γsat kN/m3 18 25.9

Young's modulus Eref kN/m2 2.280E+06 1.000E+07Shear modulus Gref kN/m2 3.846E+06 1.125E+06Oedometer modulus Eoed kN/m2 2.902E+06 1.420E+06Poisson's ratio ν(nu) 0.28 0.30Cohesion cref kN/m2 600 900Friction angle φO 36 50Shear wave velocity Vs m/s 805 1243Longitudinal wavevelocity

Vp m/s 1493 3000

Uniaxial compressivestrength

UCS kN/m2 1900 6000

Bending strength Bending strengthσy kN/m2

460 700

Shear strength Τf kN/m2 464 –

Ilatancy Ψ(o) 1° 0



Fig.

12.D

eformed

meshe

s,ev

olutionof

disp

lacemen

ts,a

ccelerations

,withtimeseve

ralc

riticalp

oints.(a)Kalam

ata,

(b)Erzinc

an,a

nd(c)Aqa

baea

rthq

uake

.Inp

utmotionPG

A=

0.24

g.



Specially, computermodeling and simulations are very useful tools foridentifying regions of stress concentration where only non-invasivetechniques are allowed. Accurate quantification of stresses are alsouseful for understanding the direction of cracks propagation and forquantifying the seismic demands on whatever new materials may beintroduced in the retrofit program.

Figs. 12–15 summarize the calculated displacements for the threePGA scenarios and the three design earthquakes. In case of Aqabaearthquake, it is clear that a great part of seismic energy is dissipatedto the upper parts of the Serapeum (ground surface) even for smallvalues of PGA (Peak Ground Acceleration). Kalamata and Erzincaninput motions give much lower displacements. For the pillar abovethe Serapeum, the maximum horizontal displacement on the top ofthe pillar at the case of Aqaba earthquake scenario at PGA=0.24 g isreaching the 16 cm (in t=24 s). The maximum horizontal displace-ment at the top of the pillar was 7 cm and 6 cm for Erzincan andKalamata earthquakes respectively.

The maximum horizontal displacements at the top of theSerapeum for Aqaba earthquake at PGA=0.24 g earthquake scenariowas ux=9 cm,while the peak vertical effective principal compression

stress was 2600 kN/m2. In the case of Erzincan and Kalamataearthquakes for the 0.24 g scenario, the respective values were2550 kN/m2 and 2500 kN/m2. The maximum horizontal displacementat the top of Serapeum was 2.14 cm and 2.01 cm for Erzincan andKalamata earthquakes respectively. The maximum computed verticaldisplacement at the top of Serapeum, it was only 2.0 mm, 2.1 mm, and6.4 mm for Kalamata, Erzincan, and Aqaba earthquakes respectively.

Given the value of the static strength estimated in the laboratory(UCS or qu=2 MPa), the seismic analysis of the Serapeum complexproved that the ceiling and sidewalls, which are the most vulnerableparts of the whole complex, are rather safe for PGA values lower than0.16 g in case of the Kalamata and Erzincan earthquakes andPGA=0.14 g for the Aqaba seismic scenario.

For larger earthquakes, which are most likely to happen in theregion of Alexandria, the seismic stability of the Serapeum is notsatisfied and it is necessary to proceed to specific retrofitting works toupgrade their seismic performance. The maximum differentialhorizontal displacements of the top and the bottom of the Serapeumare of the order of 2–3 mm considering that the induced seismicground deformations are better correlated with the intensity ofdamages in underground structures, the seismic design of theSerapeum must be based on these kinematic forces.

Fig. 13. Evolution of the effective vertical compressive stresses σ/yy with time. Atseveral critical points. (a) Kalamata (b) Erzincan and (c) Aqaba earthquakes.PGA=0.24 g.

Fig. 14. Evolution of the effective vertical compressive stresses σ/yy with time at severalcritical points. (a) Kalamata (b) Erzincan and (c) Aqaba earthquakes. PGA=0.24 g.



9. Conclusions

A short overview of the static and seismic safety of the Serapeumtemple and the ancient annex library in Alexandria, Egypt ispresented.

Laboratory and fieldmeasurements proved that the strength of thecalcarenitic and sandy limestone, where the underground monumen-tal structures are excavated, is poor (RR=18b25 and RQD=15–20%). Moreover the soft rock investigated is characterized by highdeformability and low durability.

The measured shear wave seismic velocities measured with arraymicrotremor measurements show a range of shear wave velocitiesvarying between 260 m/s and 1420 m/s indicating a rather highweathering process close to the surface.

Considering all other environmental affecting factors and thespecific geometry of the site it is concluded that the low rock strengthaffects seriously the safety of the underground structure both understatic and seismic loading conditions.

The computed static surface ground displacements under the pillarabove the Serapeum, are small: maximum total vertical displacementsare between 0.7 m and 1 mm, while the peak horizontal displace-ments are of the order of 3–0.4 mm. However the sidewalls of theSerapeum are under relatively high vertical compression and shearstresses: The maximum vertical effective principal compressivestresses are 2.5 MPa higher than the UCS strength of the rock material(=2 MPa). Moreover the peak shear stress (=974 KPa), is also higherthan the measured shear strength of the rock material (=400 KPa).Also the overstress state is beyond the elastic regime. With a globalfactor of safety equal to 1.23 (b1.6) the Serapeum should not beconsidered as safe under static conditions.

The seismic analysis of these underground monumental structuresfor three seismic scenarios of different PGA values, proved that for aninputmotionwithapeakgroundacceleration (PGA)greater than0.14 g,which is rather low considering the seismic activity and the past seismichistory of the city, there are some critical supporting parts of Serapeum(i.e. the sidewalls) that are not safe. In general, the catacombs needconsiderable strengthening to survive a strong earthquake.

In conclusion the detailed analysis of the Serapeum complex inAlexandria, Egypt proved that these importantmonuments present lowsafety factors for both static and seismic loading. Consequently a wellfocused strengthening and retrofitting program is deemed necessary.

References

Arslan, H., Rpsassanchez, L., 2007. Failure analysis of the granite for a dam foundation.Environ. Geol. doi:10.1007/S00254-007-0885-5.

Bergado, D.T., Youwai, S., Teerawattanasuk, C., Visudmedanukul, P., 2003. Theinteraction mechanism and behavior of hexagonal wire mesh reinforced embank-ment with silty sand backfill on soft clay. Comput. Geotech. 30 (6), 517–534.

Berilgen, M.M., 2007. Investigation of stability of slopes under drawdown conditions.Comput. Geotech 34 (2), 81–91.

El-Fouly, A., 2000. Void investigation at Gabbari tombs, Alexandria, Egypt using groundpenetrating radar technique. ICEHM 2000. Cairo University, Egypt, pp. 84–90.September.

Guetif, Z., Bouassida, M., Debats, J.M., 2007. Improved soft clay characteristics due tostone column installation. Comput. Geotech. 34 (2), 104–111.

Karstunen, M., Wiltafsky, C., Krenn, H., Scharinger, F., Schweiger, H.F., 2006. Modeling thebehaviour of an embankment on soft clay with different constitutive models.International Journal for Numerical and Analytical Methods in Geomechanics, vol 30,Issue 10, 25 August 2006, pp 953–982.

Hemeda, S., Pitilakis, K., Bandis, S., Papayianni, I., Gamal, M., 2007. The undergroundmonuments (catacombs) in Alexandria, Egypt. Proceedings of the “4th Interna-tional Conference on Earthquake Geotechnical Engineering”, Greece 25-28 June.

Hemeda, S., 2008. An integrated approach for the pathology assessment and protectionof underground monuments in seismic regions. Application on some Greek-Romanmonuments in Alexandria, Egypt. Ph D thesis, Aristotle University of Thessaloniki,Greece.

El-Sayed, A., Korrat, I., Hussein, M., 2004. Seismicity and seismic hazard in Alexandria(Egypt) and its surrounding. Pure Appl.Geophys. 161, 1003–1019.

PLAXIS Manual, 2002. Finite element code for soil and rock analysis published anddistributed by AA Balkema Publishers, Nederland's. Comput. Geotech. 32 (5),326–339.

Fig. 15. Maximum effective vertical compressive stress (σyy) and effective shearstresses σxy on the sidewalls of Serapeum temple, for Aqaba, Erzincan, and Kalamataearthquakes, scaled to several values of PGA.


Serapeum temple and the ancient annex daughter library in Alexandria, Egypt: Geotechnical–geophysical investigations and stability analysis under static and seismic conditions

Documents