Strong Motion Analysis at Port Island during the Kobe Earthquake, Study notes of Engineering

Download Strong Motion Analysis at Port Island during the Kobe Earthquake and more Study notes Engineering in PDF only on Docsity!

Strong Motion at Port Island

during the Kobe Earthquake

S.P.G. Madabhushi

CUED/D-SOILS/TR285 (1995)

STRONG MOTION AT PORT ISLAND DURING THE KOBE EARTHQUAKE

S.P.Gopal Madabhushi Research Fellow (Wolfson College), Cambridge University Engineering Dept., Cambridge

Abstract

It is important to understand the propagation of the stress waves through the soil strata overlying the bed rock. Several strong motion traces were recorded during the Kobe earthquake at different observation sites monitored by CEORKA (the Committee of Earthquake Qbservation and Besearch in the &nsai Area). At the Port Island site recordings were made in the N-S, E-W and U-D directions at four different depths. The data from this site provides an excellent opportunity to investigate the modulation of stress waves as they travel from the bed rock to the soil surface. This report concerns itself with the analysis of the data recorded at this site. It was observed that there is significant attenuation of the peak ground acceleration in the upper strata of soil at this site. The N-S component and the E-W component during this earthquake are 180” out of phase as the strong motion reaches the soil surface. Lissajous figures of these two components were constructed at all the four depths. These figures suggest that the strong motion polarises in the N-W and S-E direction as the it approaches the soil surface. The arrival time of the U-D component induced by the P waves was compared with the arrival time of the horizontal components induced by the shear waves. As would be expected the P waves arrived earlier than the shear waves at the soil surface. Frequency analyses were carried out for strong motion recorded in both N-S and E-W directions. These analyses revealed that there is a strong attenuation of high frequency components while selective discrete frequencies are amplified as the stress waves propagate from the bed rock to the soil surface. One of the implications of this observation is that a single frequency earthquake actuators which can impart powerful strong motion to the centrifuge model may be very useful when modelling the dynamic soil-structure interaction problems with earthquake loading in a geotechnical centrifuge.

Data plots of CEORKA sites

KOBE-UNI V

. YAHtie Il”;~.o;,^ I.@ 8. “‘I’L -..-^ ...-.-...-..^ .-.... ...-^.^.^.^.^ .- .--..^ - 11.^ ..,I.1^ h.,., 4&.Wl-- “I: - ____:... ..._ _.. _..,. &&&&.~““‘“” Y ..-.... .. a.^ AllISNO ,,.m,~;‘i,^ I.. . “‘I’ .--......-....,_a-..^ ..-^ __-^.........^ _ _ _ _ 1.^.^ -^ - -^ .--^ . *a^ St.^ -^ Il.,,,^ ^ a “c .............. .. -*-.&““‘- S A K A I \ \ I I M/ SAKP “‘I .... ..--.....-.... ...- _ ..... .”. Giz .. TADAOKA -^ cw- “‘ n “Br

0 CII I IIAYA0. tw;;;‘;,^ 1.e I.

• 1. h “I,.* I^^ .- ..-^ .-. -_^ .-.. .-....^ - ._... .-...^ __. I- -^ .-. ...” -_-..--.- --..--.. s^ ..-..^ ISI -_.-__1^ ,I.,., .-._ “I. ..-.... --... .!.!!!““.. y.^ - a ““I^.^. _^.^ .,^ ’^ I!“‘.? ORIGIN TIME ^ 1995-01-17^1 05:46:27. \ 78 DT==O.- - - - - - - DID (~1-..-.

Fig. 1 Location of the CEORKA observation sites in the Kok City

Table 1 Recording System of CEORKA

Standard Wide Range Seiemometer Frequency Range 4Osec.^ -^ 70Hz Full Scale Low Gain 4Ocm/s Low Gain lOOcm/s (10 Volt.) High Gain l&s High Gain km/s b Act.^ 1 ,OOOcm/s2 * Recorder Channel (^3) I 6 Sampling 50,100, or^ 2OOHz Pre-Trigger 5-3Osec. Triger Type Dynamic Level, Logical Interface RS-232C(Up^ to^ 38,400bps.) Clock - Radio Wave Correctted

Wide range type systems are installed at the Kobe-Univ., the Chihaya, and the Yae sites.

Table 2 Infonnations of Sites , (^) Site Latitude fLongi:ude! Altirude( (^) Soil Conditions Kobe-L‘niv. N34.72j !E13j.2<0 j (^)! iOmj.Mesoroic grani; Kztis x34.75^ 5135X78!^ i^ ZSm(Latet P!eistocene fan deposits eA.~:~~:3s&i x34.718 $55.108 / (^) OmiTnick Holocene deposits F!.ks;sf?zi-paa 4‘3d.585 jE:3$.dTJ i (^) Om)T$ick Hoioctr,e deposits M&gswachi x34.680 jEi35.j72 (^1) im/T;?ick Hoiocene deposits YX X34.680 jE135.612I i 3miThick Holocene deposits Tovociua X34.801 fEl35.50:. i Sakai i^

3mjRiocene deposit N34.564 El35462! 2m Thin Holocene! deposits Tadaoka (^) X34.480 El35408 1 12m Thin Holocene deposits Chihaya (^) X34.439 E135. Abeno N34.636 Ei35.5 19 (Port Island) X34.670 El

28Om/ Mesozoic granit 12m Late Pleistocene deposits 4m Reclaimed land

Fig.2 Plan view of the Port Island area

IOOA

=83.Orr;

-@lOOAIr->n A c CM a 13) 1

m

p,,bf,c

I

Ug (Ma

n

Fig.3 Soil profile at the recording station in Port Island u 5sM.V I f

the acceleration is decreasing markedly in the top 16m of the soil strata. One of the reasons for this may be due to partial liquefaction resulting in a partial loss of soil stiffness. This will result in a lower transmissibility of the stress waves thus resulting in srnalIer peak accelerations near the soil surface. However, no pore pressure measurements were made at the site to confirm the partial liquefaction hypothesis. Fig.8 also shows attenuation of high frequency components in the U-D direction as the vertical accelerations travels to the soil surface. Also the vertical accelerations show a remarkable amplification as they approach the soil surface.

In Table 4 the peak accelerations in each direction are tabulated. This table summa&es the observations made in the above paragraph on the attenuation of amplitude of the accelerations as the strong motion approaches the soil surface.

Table 4 Peak ground accelerations at the Port Island site

N-S direction @w GL-OOm 3 4 1. 2 284. GL-16m 5 6 4. 9 5 4 3. 2 GL-32m 5 4 3. 6 4 6 1. 7 GL-83m 6 7 8. 8 3 0 2. 6

E-W direction @w

U-D direction hw 5 5 5. 9

2J

on surface

1. 7'" :A-' -2. -3. 1 0 1 1 1 2 1 3 1 4 (^15 16) 1 6 1 9 20 2 1 2 7. 23 24

16m below ground level

4.

z 2.

-2. 1 0 11 1 2 1 3 1 4 1 5^ 1 6^ 1 6^ 1 9^20 2 1 22 2.3 24

32m below ground level

P o r t Islcnd - Seismograph records at different depths I

I^ TEST^ I^ Direction:^ N-S^ FIG-NO.

I MODEL^ i T I M E R E C O R D S 4

FLIGHT +

on surface

(^10 11) 1 2 IS 1 4 1 5 1 6 (^) 1 6 1 9 20 2 1 (^22) 2s

16m below ground level 5 2. (^7) c O L (^) - 2. 5 -5. I

- 7. 5 , I , 10 I I 12 1.3 14 15 1 6 1 6 1 9 20 2 1 22 23 24

32m below ground level

- 2 1 0 1 1 1 2 13 1 4 1 5 1 6 Ml' 1 6 1 9 20 2 1 22 23 24

83m below ground level

I (^) 1 0 $1 12 13 I 14 1 5 1 6 16 1 9 20 2% 22 2. 3 24

Port Island - Seismograph records at different depths I T E S T (^) D i r e c t i o n : U - D FIG.NO. MODEL^6 FLIGHT TIME^ RECORDS

4 Phase relationship between the N-S and E-W components

In this section we shall investigate the phase relationship between the N-S component and E-W component of the ground motion. In Fig.7 these two components recorded at the ground level are shown. From this figure it can be seen that the N-S component and E-W component are approximately 180” out of phase. This suggests that the peak acceleration in the North direction occurs at the same time as the peak acceleration occurs in the West direction. To study this further Lissajous figure was constructed by plotting the N-S component along the x-axis and the E-W component along the y-axis as shown in Fig.8. From this figure it can be seen that the resultant acceleration field acting on the soil body near the surface lies predominantly in the N-W and S-E direction.

In Fig.9 the N-S component and E-W component at the depth of 16m are shown. As in Fig.8 it is possible to see that the N-S component and E-W component are 180” out of phase. The Lissajous figure for this case is presented in Fig-lo. While the overall picture is presented in this figure it was observed that the two components have a phase difference of 180“ during the first half of the earthquake and the phase difference increases to 270” in the second half of the earthquake. This is reflected in the Lissajous figure as the number of points along the x-axis increases.

In Fig. 11 the N-S and E-W components recorded at a depth of 32m are shown. In this case the above components are 180” out of phase for approximately the first 2 seconds of the earthquake. The phase difference after this time is not clear. The Lissajous figure for this case is shown in Fig.12. Again the overall picture is presented in this figure. It was observed that the two components show a phase difference of 180” during the first 2 seconds or so and phase difference after this time is not clear.

Similar figures for the components at the depth of 83m are presented in Figs.13 and

14. From these figures conclusions similar to the above case may be drawn.

Comparing the Lissajous figures for all the four depths presented in Figs.8,10,12 and 14 we can observe that as the stress waves are propagating towards the soil surface the

data points plotted per complete transducer record

10 12 14 16 18 20 22

2

T ' ‘t; 20

- ! f 1’

I 10 12 14 16 18 20 22 24 i^ , (^) bl I

i. 1 /

P o r t I s l a n d - S e i s m o g r a p h n e a r g r o u n d l e v e l ii , ! FIG.NO. j

TEST / MODEL (^) i (^) TIME RECORDS 7 j! FLIGHT 1

(---- (^) :,f 1 E-W compo”e”t(m/s/s)

Duration= 10s to 24s

t

- 4 0 0

N-S component (m/s/s)

Fig.8 Lissajous plot between the N-S and E-W components

at surface level

500

r-b^30

I

E - W component (m/s/s)

Duration= 12.6s to 23s

N-S component (m/s/s)

Fig. 10 Lissajous plot between the N-S and E-W components

at 16m below ground level

u-. ii 0 f-t-. (^03)

.. z=

-. 0 (^3)..