at SciVerse ScienceDirect

Geotextiles and Geomembranes 34 (2012) 51e68

Contents lists available

Geotextiles and Geomembranes

journal homepage: www.elsevier .com/locate/geotexmem

Use of geosynthetics to improve seismic performance of earth structuresq

Junichi KosekiInstitute of Industrial Science, University of Tokyo, Ce201, 4-6-1 Komaba, Meguro-ku 153-8505, Japan

a r t i c l e i n f o

Article history:Received 18 September 2011Received in revised form13 February 2012Accepted 19 March 2012Available online xxx

Keywords:ReinforcementCase historyModel testRetaining wallBridge abutmentBallasted track

q This paper is the written version of the Mercer LE-mail address: koseki@iis.u-tokyo.ac.jp.

0266-1144/$ e see front matter � 2012 Elsevier Ltd.doi:10.1016/j.geotexmem.2012.03.001

a b s t r a c t

After reviewing seismic performance of earth structures based on case histories in Japan and relevantmodel test results, advantages of using geosynthetics in improving their seismic performance aredemonstrated. As one of the successful applications, geosynthetic-reinforced soil retaining walls arehighlighted, focusing on several influential factors such as facing rigidity, arrangement and properties ofreinforcements, and backfill and subsoil conditions. In addition, further applications of the reinforcementmethod using geosynthetics are introduced, which include combination with other reinforcementmethods, application to bridge abutments and piers, and application to ballasted railway tracks.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Fig. 1 shows global distribution of earthquake epicenters thattook place during a ten-year period from 1990 to 2000 withmagnitudes equal to or exceeding 4.0 and epicentral depths of50 km or less. As indicated by an arrow in the figure, Japan islocated in a very highly active zone of such seismic events.

Table 1 summarizes the peak values of horizontal groundaccelerations (PGAs) and velocities (PGVs) that were recordedduring recent major earthquakes in Japan. After the 1995Hyogoken-nanbu earthquake, the availability of strong motion datarecorded near the epicenter was improved significantly. Therefore,at some of these sites the PGA approached or exceeded 800 galsand/or the PGV exceeded 100 kines.

On the other hand, earth structures, such as embankments asschematically shown in Fig. 2a & b, have been widely employed toconstruct highways, railways, river dikes and housing lots. Inaddition, in order to reduce the area to be occupied by theconstruction of embankments and thus the volume of fill material,retaining walls (RWs) have also been frequently adopted, as sche-matically shown in Fig. 2c & d.

If we convert the horizontal seismic inertia and vertical gravityvector into a resultant pseudo-static force as schematically shown

ecture 2011.

All rights reserved.

in Fig. 3a, the direction of apparent gravity will be inclined. Then,the driving moment to trigger the sliding failure along a potentialfailure plane will be increased, as shown in Fig. 3b. Under suchcircumstances, adding reinforcements in the embankment withtheir tensile forces mobilized effectively will increase the resistingmoment, as shown in Fig. 3c.

In view of the above, by addressing the following questions inthis paper, attempts are made to share Japanese experiences on theuse of geosynthetic-reinforcement to improve seismic performanceof earth structures:

Q1: How different are the seismic performances of earthretaining structures with/without geosynthetic-reinforcement?

Q2: What are the influential factors in improving effectively theseismic performance of earth retaining structures usinggeosynthetics?

Q3: How can we extend the application of geosynthetic-reinforcement technologies to other types of earth structure?

In order to answer the above questions, the paper begins witha review of seismic performance of earth structures in Japan. Next,influential factors in improving seismic performance of retainingwalls using geosynthetics are discussed. Some of further applica-tions of geosynthetic-reinforcement are briefly reviewed as well,which are followed by conclusions.

Fig. 1. Distribution of earthquake epicenters with M >4.0 and depth <50 km recordedfrom 1990 to 2000 (modified after JMA, 2011).

a

b

c

d

Fig. 2. Schematic illustrations of embankments and retaining walls: a) embankmenton level ground, b) embankment on slope, c) RW on level ground, and d) RW on slope.

J. Koseki / Geotextiles and Geomembranes 34 (2012) 51e6852

2. Seismic performance of earth structures

In order to answer the question 1 raised in Introduction, thefollowing two sub-topics are reviewed in this chapter, whileupdating the summary made by Koseki et al. (2007):

How earth structures with/without geosynthetic-reinforcementbehaved in

- case histories, and- model tests?

2.1. Case histories in Japan

Herein, case histories of damage and reconstruction from thefollowing five recent major earthquakes in Japan are reviewed:

- 1995 Hyogoken-nanbu (Kobe) earthquake- 2004 Niigataken-chuetsu earthquake- 2007 Noto-hanto earthquake- 2007 Niigataken-chuetsu-oki earthquake- 2008 Iwate-Miyagi-nairiku earthquake- 2011 Off the Pacific Coast of Tohoku earthquake

2.1.1. 1995 Hyogoken-nanbu (Kobe) earthquakeFig. 4 shows damage to retaining walls (RWs) without reinforce-

ment that were located in severely shaken area by the January 17,1995 Hyogoken-nanbu earthquake. Conventional type RWs withouta deep foundation, such as cantilever, gravity and leaning-type ones,suffered overall tilting and/or failure of the wall body. Most of themhad to be removed and reconstructed after the earthquake.

Table 1List of recent strong motion records in Japan (modified after NIED, 2011a, b).

Earthquake Station PGA (gal) PGV(kine)

Hyogoken-Nanbu (1995) JMA Kobe (NS) 818 97JR Takatori (EW) 645 136

Tottoriken-Seibu (2000) KiK-net Hino (NS) 924 127Tokachi-Oki (2003) K-net Hiroo (EW) 970 47Niigataken-Chuetsu (2004) JMA Kawaguchi (EW) 1676 146Noto-Hanto (2007) K-net Anamizu (EW) 782 99Niigataken-Chuetsu-Oki

(2007)K-net Kashiwazaki (NS) 667 110

Iwate-Miyagi-Nairiku(2008)

Kik-net Ichinoseki-Nishi(EW)

1433 62

Off the Pacific Coastof Tohoku (2011)

K-net Tsukidate(NS)

About2700

n/a

In contrast to the above, one geosynthetic-reinforced soil (GRS)RW with full-height rigid facing, which was also located in theseverely shaken area, survived with minor residual lateral displace-ments of about 10e20 cm that are measured relative to theneighboring culvert box structure (Fig. 5). The standard proceduresfor stagedconstructionof this typeofGRSRWswitha full-height rigidfacing is illustrated in Fig. 6. It should be noted that, with this type ofGRS RWs, the connection between the facing and the reinforcementis ensured by constructing a geosynthetic-reinforced soil wall withwrapped-around wall face in stage 5 and by casting-in-place theconcrete directly on the wall face in stage 6 (Tatsuoka et al., 1996).

On the other side of the culvert structure, as shown in Fig. 7,a cantilever-type RW with bored-pile foundation suffered similaramounts of residual lateral displacement, suggesting that this walland the previous GRS RWwithout foundation exhibited almost thesame seismic resistance.

Refer toTatsuoka et al. (1995,1996,1997a,1998) and Koseki et al.(1999) for the detailed results of the damage investigation on theabove RWs and their back-analysis.

2.1.2. 2004 Niigataken-chuetsu earthquakeFig. 8 shows damage to highway RW and railway embankment

by the October 23, 2004 Niigataken-chuetsu earthquake (Tatsuokaet al., 2006, 2007a; Tatsuoka, 2008: Koseki et al., 2006a). Although

Embankment

mg, g=980 gal

mah

(ah=980 gal in this case)

Increase in driving moment

Inclineddirection of apparent gravity

a

b

Increase in resisting moment by tensile force mobilized in reinforcements

c

Fig. 3. Schematic illustrations on effects of horizontal inertia of embankment andtensile force mobilized in reinforcements: a) horizontal inertia of embankment, b)equivalent condition to account for horizontal inertia, and c) effects of addingreinforcements.

Fig. 4. Damage to RWs without reinforcement caused by 1995 Hyogoken-nanbuearthquake: a) and b) cantilever-type RW and c) gravity-type RW at Ishiyagawa, andd) leaning-type RW at Sumiyoshi (Tatsuoka et al., 1996).

Fig. 5. Residual displacement of GRS RW at Tanata caused by 1995 Hyogoken-nanbuearthquake (Tatsuoka et al., 1996).

J. Koseki / Geotextiles and Geomembranes 34 (2012) 51e68 53

the original structures were without reinforcement, the highwaywas reconstructed using GRS RW with segmental facing panelsmade of pre-cast concrete (Fig. 9a), while the railway on the downslope side was reconstructed using GRS RW with a full-height rigidfacing and rock bolts (Fig. 9b). Such decisions were made consid-ering ground conditions, construction time and available backfillmaterial, while adopting the same concept that the reconstructedearth structures shall exhibit improved seismic performance.Numerical verification of such improvement was made by Shinodaet al. (2009) on the railway embankment.

Figs. 10and 11 show the collapse of an embankment for newly-developed housing estates and its reconstruction using GRS RWwith segmental facing panels made of metal mesh. These panelsenable the vegetation to cover the facing (see Fig. 16a for typicaloutlook with vegetation).

2.1.3. 2007 Noto-hanto earthquakeFig. 12 shows damage to an embankment for Noto toll road by

March 25, 2007Noto-hanto earthquake. The embankment had beenconstructed by filling a valley. In this case, the fill material wasweathered tuff, which flowed down the valley for a distance

Fig. 6. Staged construction procedures for geosynthetic-reinforced soil retaining wallswith full-height rigid facing (Tatsuoka et al., 1995).

Fig. 8. Failure of highway gravity-type RW and railway embankment at Tenno causedby 2004 Niigataken-chuetsu earthquake (Koseki et al., 2006a).

Location of existing gravity-type retaining wall after E.Q.

Temporary shotcrete

Panel-type facing

Existing gravity-type retaining wall Before E.Q.

(Unit: mm)

a

J. Koseki / Geotextiles and Geomembranes 34 (2012) 51e6854

exceeding 100m.Note also that, based on the survey conducted afterthe earthquake, the ground water level was found within the fill(Japanese Geotechnical Society, 2007a).

As schematically shown in Fig. 13, the collapsed embankmentswere reconstructed using GRS RWs, while ensuring the drainageof groundand surfacewater. Thewaste soil that hadoriginally beenapart of the collapsed embankmentwas re-used after lime-treatmentfor the construction of the upper fill (Ishikawa Prefecture, 2007).

On the other hand, as typically shown in Fig. 14, the embank-ments of Anamizu road that connects to the north end of the Nototoll road could survive the earthquake with minor damage. Suchgood performance may be attributed to the use of lime-treatmentfor the fill material, while ensuring the drainage of ground waterby installing non-woven geotextile sheets.

2.1.4. 2007 Niigataken-chuetsu-oki earthquakeFig. 15 shows damage to an RW for a municipal road constructed

on a slope by the July 16, 2007 Niigataken-chuetsu-oki earthquake.In contrast, a GRS RW with segmental facing panels made of

metal mesh that was located at the foot of the slope could survivethe earthquake as shown in Fig. 16.

Fig. 7. Residual displacement of cantilever-type retaining wall with pile foundation atTanata caused by 1995 Hyogoken-nanbu earthquake (Tatsuoka et al., 1996).

2.1.5. 2008 Iwate-Miyagi-nairiku earthquakeFig. 17 shows an anchored cut-slope at a dam construction site.

The anchors with a length of 20e40 m had been installed asa countermeasure against landslide movement. By the June 14,2008 Iwate-Miyagi-nairiku earthquake, most of the tendons werebroken at their free sections, and the upper parts of the brokentendons were ejected.

On the other hand, a GRS RW with segmental facing panelsmade of metal mesh that was located on the other side of the slopesurvived the earthquake with no damage as shown in Fig. 18.

2.1.6. 2011 Off the Pacific Coast of Tohoku earthquakeFig. 19a shows collapse of an unreinforced embankment for

newly-developed housing estates, caused by the March 11, 2011 Offthe Pacific Coast of Tohoku earthquake. At an adjacent site, asshown in Fig. 19b, a GRS RW with segmental facing panels made ofmetal mesh survived the earthquake. This GRS RW had been

Geogrid

Silt stoneSand stone

1:2.0

1:0.

3

Before EQ 13

.18

m

After EQ

1:4.0

After re-construction

Collapsed gravity-type retaining wall

Rail track

Cast-in-placerigid facing

Rock bolt (L=2m)

Sand stone

b

Fig. 9. Reconstruction of failed a) highway RW (Koseki et al., 2006a) and b) railwayembankment (Morishima et al., 2005) at Tenno using GRS RWs.

Fig. 10. Failure of embankment for newly-developed housing estates at Takamachicaused by 2004 Niigataken-chuetsu earthquake.

J. Koseki / Geotextiles and Geomembranes 34 (2012) 51e68 55

constructed as rehabilitation work for past damage to the unrein-forced embankment induced by heavy rainfall.

2.2. Model tests

Herein, focusing on RWs, results from relevant model tests ontheir different seismic performances with/without geosynthetic-reinforcement are reviewed.

2.2.1. Test conditions and proceduresA series of relatively small-scale 1-g model shaking tests was

conducted on six different types of retaining walls resting on levelground as shown in Fig. 20. Thewall models were about 50 cm-highand the subsoil and backfill were modeled by very dense air-drysand layers. The reinforcement was modeled by a grid of phos-phor bronze strips (to be shown later in Fig. 29a). They were sub-jected to several sequential horizontal excitations as typicallyshown in Fig. 21 in 0.1 g increments. Refer to Koseki et al. (1998)and Watanabe et al. (2003) for the detailed test conditions.

2.2.2. Test resultsFig. 22 compares the cumulative horizontal displacements near

the top of each RW model. The horizontal axis is the seismiccoefficient defined as the peak base acceleration during eachshaking step that is normalized with the gravitational acceleration.Up to seismic coefficient of about 0.35, no significant differencecould be observed among the different wall types. However, underhigher seismic loads, the residual wall displacements accumulatedrapidly with the conventional RWs, i.e., cantilever, gravity andleaning-type ones. In contrast, the GRS RWs with a full-height rigidfacing exhibited more ductile behavior, in particular with the oneshaving extended reinforcements (types 2 & 3, Fig. 20e & f).

Fixing p

H-300*300@1.5m L=8.0m

Excavation boundary

After EQ

After temporary rehabilitation

8000

GeogridsDrainage sheet

6000

8500

32

Fig. 11. Reconstruction of failed embank

The reason for the less ductile behavior of the conventional RWscan be understood from Fig. 23. The subgrade reaction at the toe ofbase footing of the gravity-type wall increased sharply with theaccumulation of wall top displacement. It suddenly decreased,however, after showing a peak state, suggesting a local failure due toloss of bearing capacity. On the other hand, the subgrade reaction atthe heel of the base footing decreased in the beginning, followed bya slight increase with the occurrence of the local failure at the toe.

In case of GRS RWs, as shown in Fig. 24, the tensile forces in thereinforcements measured at three different heights increased withthe accumulation of the wall top displacement in the active direc-tion. Such a response of GRS RWs is the key feature for their goodperformance under high seismic loads. It should be noted that themobilization of tensile force was concentrated to the uppermostlong reinforcement for the type 2 GRS RW, which could effectivelyresist against the overturning of the facing. Due attention should bepaid to such concentration of tensile force in the design of GRS RW.

It should be noted that, as mentioned above, the RW modelswith/without reinforcement exhibited similar wall displacementsup to around 5 mm during the shaking steps at relatively lowexcitation levels (Fig. 22). This behavior is possibly affected by theshear deformation of subsoil layer. When such deformation occurs,as schematically illustrated in Fig. 25, the RWwould suffer residualhorizontal displacement without any slippage at the interfacebetween the base of the RWand the underlying subsoil layer. Underthe same subsoil conditions, as were the cases with the presentmodel tests, the amount of such residual displacement would notdepend largely on the difference in the RW types.

In case of GRS RWs, not only the subsoil deformation but alsothe shear deformation of reinforced backfill was observed in modeltests as typically shown in Fig. 26. In evaluating the residualdisplacements of GRS RWs, this type of shear deformation ofreinforced backfill should be considered properly. However, inmany of the current design guidelines, the reinforced backfill hasbeen modeled as a rigid body that would not undergo any sheardeformation (Koseki et al., 2006b).

3. Influential factors in improving seismic performance ofretaining walls using geosynthetics

In order to answer the question 2 raised in Introduction, thefollowing four sub-topics on the seismic performance of GRS RWsare reviewed in this chapter:

How their seismic performance is affected by

Drainage pipe

Before EQ

After temporary rehabilitation

in

Panel-typefacing

00 5575

1775 4500

5000

9000

1400

0

600

712

00=8

400

(Unit: mm)

ment using GRS RW at Takamachi.

Fig. 12. Failure of an embankment of Noto toll road (at site No. 32) caused by 2007 Noto-hanto earthquake.

J. Koseki / Geotextiles and Geomembranes 34 (2012) 51e6856

- facing rigidity,- arrangement of reinforcement,- properties of reinforcement, and- backfill and subsoil conditions?

3.1. Facing rigidity

Tatsuoka (1993) investigated and discussed in detail the effects offacing rigidityon the stabilityofGRSRWs, as summarizedbrieflybelow.

Fig. 13. a) Schematic illustrationon reconstructionof failed embankments forNoto toll road (Is

a) As the facingbecomesmore rigid, the earthpressure actingon theback face of facing increases. This large earth pressure confinesthe backfill soil immediately behind the facing, which decreasesthe deformation and increases the ultimate stability of the wall.

b) A large degree of flexibility is not necessarily a preferableproperty for completed GRS RWs, although this property isrequired to accommodate possible large deformation of thesubsoil so that a deep foundation becomes unnecessary.

hikawaPrefecture, 2007), and b) typical reconstructionworkusingGRSRW(at siteNo. 9).

Fig. 14. Minor damage to embankment for Anamizu road caused by 2007 Noto-hantoearthquake.

Fig. 16. GRS RW with at Agewa with segmental facing panels made of metal mesh.

J. Koseki / Geotextiles and Geomembranes 34 (2012) 51e68 57

The above summary a) suggests also that the geosynthetic-reinforcement technique is not necessarily a method to reduceearth pressure exerted from the backfill soil. Rather, this techniquecan take advantage of the tensile force mobilized in reinforcementsto resist against the earth pressure (and inertia force of facing incase of earthquakes).

Regarding the above summary b), one of the possible compro-mises are that walls are made as much as flexible duringconstruction, while they aremade stiff enough before they are opento service. The staged construction procedures by casting-in-placethe full-height rigid facing at the last stage as shown in Fig. 6enable us to control the flexibility in such a manner, while theyrequire in general longer construction period than GRS RWs withpre-cast segmental types of facing.

It should be noted that, in case of GRS RWswith segmental typesof facing, local instability of facing due to failure at the connectionbetween the facing and the reinforcementmay lead easily to overallfailure. Such a failure mode was observed in the 1999 Chi-chiearthquake in Taiwan as typically shown in Fig. 27a. Therefore,special attention needs to be paid to their interlocking usingmechanical connectors or shear keys, among others.

Note also that, in this particular case shown in Fig. 27, thevertical spacing of reinforcements was 80 cm, which exceeded thevalue recommended by relevant design guidelines. At the time ofthe earthquake, therefore, insufficient number of reinforcementsmay have been subjected to excessive tensile force, causing localrupture at their connection with the facing, and/or the overallfacing rigidity may have been too small to resist against theearthquake loads, causing excessive deformation of the stackedfacings and pull-out of connecting pins, as schematically illustratedin Fig. 27b.

Fig. 15. Failure of RW for municipal road at Agewa caused by 2007 Niigataken-chuetsu-oki earthquake.

On the other hand, in case of GRS RWs with full-height rigidfacing, even if local rupture or failure takes place, the tensile forcemobilized in the reinforcements would be re-distributed moreeasily, since the rigid facing is supported simultaneously by manylayers of reinforcements. This feature would enhance their redun-dancy against overall failure.

In comparing seismic performances of different facing types,due attention should be also paid to the consequences of theiroverall or local failure from a view point of repairability.

GRS RWswith full-height rigid facing that had been constructedin Japan following the procedures shown in Fig. 6 performed wellunder not only working loads but also large earthquake loads (e.g.,Fig. 5). Therefore, this particular type of GRS RWs has been adoptedfor constructing a number of important permanent earth structuresto support such as bullet train tracks and highways (Tatsuoka et al.,1997b; Tamura, 2006). As summarized in Fig. 28, their applicationin terms of total wall length exceeded 120 km as of June, 2010.

Hereafter, focusing on GRS RWs with full-height rigid facing,effects of other influential factors are discussed.

Fig. 17. Failure of anchor tendons for cut-slope at construction site of Isawa damcaused by 2008 Iwate-Miyagi-nairiku earthquake.

Fig. 18. GRS RWwith segmental facing panels made of metal mesh at construction siteof Isawa dam (by courtesy of Mr. S. Hamaya). Fig. 19. a) Collapse of unreinforced embankment caused by 2011 Off the Pacific Coast

of Tohoku earthquake, and b) undamaged GRS RW at Yamamoto town, Miyagiprefecture.

J. Koseki / Geotextiles and Geomembranes 34 (2012) 51e6858

3.2. Arrangement of reinforcement

In the model tests presented previously (Fig. 20), three types ofreinforcement arrangement were employed:

- R1 or type1 having relatively short reinforcements with anequal length,

- R2 or type 2 having partially extended reinforcements, and- R3 or type 3 having relatively long reinforcements withanother equal length.

Among the above three types, as can be seen from Fig. 22, type 3wall exhibited the smallest residual wall top displacement. Inaddition, though the total length of reinforcement of type 2 wallwas about 80% as large as that of type 3 wall, their seismicperformances in terms of the wall top displacement were similar toeach other. Such good performance of type 2 wall confirms thatpartial extension of upper reinforcement layers (if possible, pref-erably connected with another wall on the opposite side as illus-trated in Fig. 28) improves the seismic stability significantly, sincethe upper reinforcement layers can resist more effectively againstthe overturning mode of failure.

As seen from Fig. 26, the extended uppermost reinforcementlayer in type 2 wall prevented full formation of a failure plane(marked as A in Fig. 26) in the backfill aswell, which passed throughthe end of the other extended reinforcement at the middle height.

In turn, as mentioned previously on Fig. 25, the tensile forcemobilized in the reinforcements concentrated into the extendeduppermost reinforcement of type 2 wall, implying that this rein-forcement layer was the key to exhibit the above-mentioned goodperformance.

Note also that, with types 2 and 3 walls, the calculated values ofcritical seismic coefficient to induce a factor of safety equal to unityin pseudo-static limit-equilibrium stability analysis against over-turning failure were different from each other (0.55 and 0.70,respectively). Refer to Watanabe et al. (2003) for the detailedconditions of calculation. In spite of such difference, the two wallsexhibited similar seismic performances. This is possibly affected bythe shear deformation of reinforced backfill (Fig. 26), which is notconsidered in evaluating the above critical seismic coefficients.

3.3. Properties of reinforcement

As an extension of the model tests presented previously(Fig. 20), another series of 1-g model shaking tests was conductedby Nakajima et al. (2007a), where two kinds of models for thereinforcement layer as shown in Fig. 29a and b were employed. Themodel wall dimensions are the same as those of the GRS RWmodelshown in Fig. 20e.

As summarized in Table 2, their tensile stiffness per single stripor string evaluated in direct tension tests was higher with the PB(grid of phosphor bronze strips) model. On the other hand, theirtensile stiffness per unit width was higher with the PE (polyestermesh) model, since the total number of strings of this model wasmuch larger than that of strips of the PB model.

In addition, as shown also in Table 2, their ultimate pull-outresistance per unit width was higher with the PE model, duepossibly to its mesh size (¼3*3 mm) that could confine approxi-mately 10 particles of the sand material (with a mean diameter of0.23mm) that was used to prepare the backfill in themodel tests. In

Fig. 20. RW models on level ground (Watanabe et al. 2003; note that all dimensions are in cm): a) cantilever type, b) gravity type, c) leaning type, d) reinforced-soil type 1, e)reinforced-soil type 2, and f) reinforced-soil type 3.

J. Koseki / Geotextiles and Geomembranes 34 (2012) 51e68 59

contrast, the PB model had a much larger grid size (¼50*95 mm),and it exhibited larger pull-out resistance per unit width at smalllevels of pull-out displacement (up to about 1 mm, Fig. 31), duepossibly to the effects of sand particles that were pasted on thesurface of the reinforcement tomobilize the frictional resistance. Bypasting sand particles, it was expected that the pull-out propertiesof themodel reinforcementwould simulate those of geogrids underfield conditions that are interlocked with the backfill by sufficientcompaction, though no quantitative comparison has been made.

Despite the above-mentioned differences in the reinforcementproperties, the observed cumulative tilting angles and base slidingdisplacements of the GRS RW models (type 2, Fig. 20e) employingthe two types of reinforcements, respectively, were in generalsimilar to each other, as shown in Fig. 30. In these model tests, thepeak base accelerationwas 0.9 g in the first shaking step, whichwasincreased in 0.3 and 0.4 g increments in the second and thirdshaking steps, respectively. Further, the fourth shaking step wasconducted by using the same base acceleration as in the third one.

The influential factor to affect the tilting behavior of GRS RWsobserved in the present model tests would be the pull-out resis-tance mobilized at small displacement levels. As mentioned above,it was larger with the PBmodel thanwith the PE model, though theultimate resistance was vice versa (Fig. 31). The slight differencebetween the twomodels in terms of the cumulative tilting angles ascan be seen in Fig. 30a may have been affected by such pull-outproperties at small displacement levels.

1 2 3 4 5 6 7 81.0

0.8

0.6

0.4

0.2

0.0

-0.2

-0.4

-0.6

-0.8

-1.0

amax

Modified from N-S component at Kobe Marine Meteorological Observation Station during the 1995 Hyogoken-Nanbu earthquake

Bas

e ac

cele

ratio

n(G

)

Time(sec)

Fig. 21. Typical excitation time history (Watanabe et al., 2003).

On the other hand, since no rupture or pull-out failure of rein-forcements was observed, these properties would not have affectedthe tilting behavior in the present model tests.

Regarding the base sliding of GRS RWs shown in Fig. 30b, nosignificant difference was observed between the two models. Thisis possibly because the resistance against base sliding under themodel configurations employed for the present model tests is notlargely affected by the reinforcement properties but predominantlyaffected by the backfill and subsoil conditions around their inter-faces. For small displacement levels, in addition, one needs to recallthe effects of shear deformation of subsoil layer as discussedpreviously (Fig. 25), which is also independent from the rein-forcement properties.

3.4. Backfill and subsoil conditions

Not only the conditions of reinforcement arrangement andfacing rigidity as discussed above, but also the conditions of backfilland subsoil affect the seismic performance of GRS RWs.

For example, backfill soils that are not well-compacted may notmobilize sufficient pull-out resistance of reinforcements, no matterhow the reinforcements themselves are stiff and strong enough. Inaddition, poorly-compacted backfill soils may suffer excessivesettlement during their service period prior to earthquakes,resulting into local failure of reinforcements at their connectionwith the facing. Once pull-out or local failure of reinforcements

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.40

10

20

30

40

50

60

70

80

90

100

Conventional type C: cantilever G: gravity L: leaningReinforced-soil R1: type 1 R2: type 2 R3: type 3

C

L

Wal

l top

dis

plac

emen

t, d to

p(mm

)

Seismic coefficient kh

R2

R3Backfill

dtop G

R1

Fig. 22. Comparison of residual wall top displacements (Watanabe et al., 2003).

0 10 20 30 40 50 60 70 80 90 100 1100

5

10

15

20

25

30

35

1-9: Shaking step

981

234

56 7

a)

Wall top displacement, dtop (mm)

Nor

mal

stre

ss a

t bot

tom

of

bas

e fo

otin

g, σ

(kPa

)

(heel)

(toe)

Locationof loadcells

6 5 4LT7

LT5

LT6

LT4

LT7

Local failure due to loss of bearing capacity

Fig. 23. Change of subgrade reaction of gravity-type RW model (Watanabe et al.,2003).

Before shaking

After shaking

a b

Fig. 25. Schematic illustration of wall displacement induced by shear deformation ofsubsoil layer.

40.3 o

41.3 oLocation of failure planeobserved along the center

line of the sand box

Subsoil layer

20cm

80cm

45cm

Location of end of longerreinforcements when failure

planes were formed.

Backfill soil

A

B

Fig. 26. Shear deformation of reinforced backfill and formation of failure planes inunreinforced backfill of type 2 GRS RW model (modified after Watanabe et al., 2003).

J. Koseki / Geotextiles and Geomembranes 34 (2012) 51e6860

takes place, it would trigger overall instability of GRS RWs duringearthquakes as well as under working load conditions.

3.4.1. Importance of backfill soil compactionAs compared to the backfilling work without reinforcement that

is employed for conventional type RWs, the backfill soil withreinforcements can be compacted in a more effective manner, sincethe existence of reinforcement would confine the lateral spreadingof the backfill soil during compaction work. Therefore, by takingsuch an advantage, due attention shall be paid in constructing GRSRWs to compact the backfill soil sufficiently.

The above confinement of the lateral spreading of the backfillsoil would in turn mobilize initial tensile forces in the reinforce-ments. Such effective mobilization of tensile forces wouldcontribute to reduce the displacement of the facing as well.

0 1 0 2 0 3 0 4 0 5 0 6 0 7 00

5

1 0

1 5

2 0

2 50 1 0 2 0 3 0 4 0 5 0 6 0

0

5

1 0

1 50 1 0 2 0 3 0 4 0 5 0 6 0

0

5

1 0

1 5

T y p e 3 (L = 3 5 c m )

T y p e 2 (L = 2 0 c m )

W a ll to p d is p la c e m e n t, d to p (m m )

T yp e 3 (L = 3 5 c m )

T y p e 2 (L = 4 5 c m )T y p e 1 (L = 2 0 c m )

T y p e 1 (L = 2 0 c m )

Tens

ile fo

rce

(N)

Tens

ile fo

rce

(N)

L o w e s t re in fo rc e m e n t

M id d le -h e ig h t re in fo rc e m e n t

Tens

ile fo

rce

(N) T yp e 3

(L = 3 5 c m )T y p e 2 (L = 8 0 c m )

T y p e 1 (L = 2 0 c m )

U p p e rm o s t re in fo rc e m e n t

Fig. 24. Change of reinforcement tensile force of GRS RW models (Watanabe et al.,2003).

Fig. 27. Damage to GRS RW with segmental facing caused by 1999 Chi-chi earthquake,Taiwan; a) photograph of damaged structure and b) schematic cross-section (Kosekiand Hayano, 2000).

Fig. 28. Growth statistics for GRS walls with full-height rigid facing in Japan (bycourtesy of Dr. M. Tateyama).

Fig. 29. Models of reinforcement layer (Nakajima et al., 2007a): a) phosphor bronzemodel, and b) polyester model.

a

b

Fig. 30. Comparison of residual tilting angles and base sliding displacements of wallfacing (Nakajima et al., 2007a).

1200

1600

PE σv=5kPa PB σv=5kPa

N)

(Unit: mm)

J. Koseki / Geotextiles and Geomembranes 34 (2012) 51e68 61

Note also that, by placing reinforcements at a specified verticalspacing (equal to 30 cm in case of GRS RWs with full-height rigidfacing shown in Fig. 6), one may ensure the lifting height of thebackfilling work to be equal to or smaller than this verticalspacing.

When considering the effects of shear deformation of subsoillayer as discussed previously (Fig. 25), one may understand easilythat the subsoil conditions also affect the seismic performance ofGRS RWs. The effects of subsoil conditions are further discussedbelow, referring to the review by Koseki et al. (2007) on relevantmodel test results, while focusing as well on the measures toimprove the seismic performance.

Table 2Properties of model reinforcements (modified from Nakajima et al., 2007a).

Property Secant tensile stiffness atT ¼ Tmax/2

Ultimate pull-out resistance,Tmax at sv ¼ 5 kPa

Unit per single strip(kN/ 3/strip)

per unit width(kN/ 3/m)

per unit width for buriedlength of 0.5 m (kN/m)

PB 3.5e5.7 41e66 2.96a

PE 0.31e0.36 105e121 4.48

a Strips in the air ruptured under this tensile force.

3.4.2. Model tests on RWs constructed on sloped groundAs an extension of the model tests presented previously

(Fig. 20), another series of 1-g model shaking tests was conductedby Kato et al. (2002) on RWs resting on sloped ground as typicallyshown in Fig. 32. For some models, in order to increase theirstability, large diameter nail models having a diameter of 4 cmwereadded to the backfill and subsoil and connected to the wall. At eachelevation, four nail models were employed for a total horizontaldepth of 60 cm. Refer to Nakajima et al. (2007b) for the detailedanalysis of tensile loads induced in the nail models.

Fig. 33 shows the relationship between the seismic coefficientand the cumulative horizontal displacements near the wall top. Forcomparison, the aforementioned test results on RW models of thesame type constructed on level ground (Fig. 20) are plottedtogether. The seismic stability of the respective wall model on slopewas much lower than the corresponding one on level ground.

0

400

800

Rupture of phosphor bronze

Pullo

ut re

sita

nce(

Pullout displacement(at 5cm away from wall facing) (mm)

Fig. 31. Pull-out tests on model reinforcements (Nakajima et al., 2007a).

(S): on slope(L): on level ground

Reinforced with nails(S)

Cantilever(S)

Cantilever(L)

Leaning with nails(S)

Reinforced(L)

Reinforced(S)Leaning(L)

Leaning(S)

Lat

eral

dis

plac

emen

t at w

all t

op (m

m)

Seismic coefficient kh=amax/g

Fig. 33. Comparison of residual wall top displacements (Kato et al., 2002).

(refer to Fig. 6.)Formation of shear bands 1 and 2

(S):on slope(L):on level ground

Reinforced(S)

Reinforced(L)

Ten

sile

forc

e (N

)

Lateral displacement at wall top (mm)

Formation of failure plane

Fig. 35. Comparison of tensile forces mobilized in reinforcements of GRS RW models(Kato et al., 2002).

Fig. 32. RW models with large diameter nails on sloped ground (Kato et al., 2002).

Fig. 36. Comparison of residual displacements of GRS RW models with/withoutembedded sheet pile (Nakajima et al., 2006).

J. Koseki / Geotextiles and Geomembranes 34 (2012) 51e6862

It should be noted that, as shown in Fig. 34, in case of GRS RWconstructed on slope, a full failure plane in the unreinforced backfilland the sloped subsoil was formed. After the formation of such fullfailure plane, as shown in Fig. 35, the mobilization of the tensileforces in the reinforcement was reduced, followed by rapid accu-mulation of the wall displacement (Fig. 33).

Fig. 34. Full failure plane in unreinforced backfill and sloped subsoil of GRS RW model(Kato et al., 2002).

Fig. 37. Failure of railway embankment at Tenno tunnel caused by 2004 Niigataken-chuetsu earthquake.

Fig. 38. Reconstruction of railway embankment at Tenno tunnel (Kitamoto et al., 2006).

J. Koseki / Geotextiles and Geomembranes 34 (2012) 51e68 63

On the other hand, as shown in Fig. 33, the leaning-type and GRSRWs with the large diameter nails exhibited substantially higherseismic stability than those without nails. They yielded very limitedamount of residual wall displacements even at seismic coefficientsexceeding 1.0.

4. Further applications of geosynthetic-reinforcement

In order to answer the question 3 raised in Introduction, thefollowing three sub-topics are reviewed in this chapter:

Fig. 39. a) Failure of railway embankment at Tsukanoyama caused by 2004 Niiga-taken-chuetsu earthquake and b) its reconstruction (Morishima et al. 2005).

How geosynthetics are used for

- combination with other reinforcement methods,- application to bridge abutments and piers, and- application to ballasted railway tracks?

4.1. Combination with other reinforcement methods

In the model tests shown in the previous section, GRS RWs onsloped ground exhibited substantially high seismic stability whenthey were further reinforced with large diameter nails (Fig. 32).

In addition, as shown in Fig. 36, the residual tilting angle of thefacing of GRS RW model could be effectively reduced by installinga sheet pile at the foot of the facing and connecting it to the facing(Nakajima et al., 2006).

Further, not only soil reinforcement methods but also soilimprovement methods have been combined with geosynthetic-reinforcement. For example, Izawa et al. (2009) reported

Fig. 40. Backfill settlement of bridge abutment caused by 1995 Hyogoken-nanbuearthquake (Aoki et al., 2003).

Fig. 41. Structural failure of bridge abutment and settlement of its backfill caused by2004 Niigataken-chuetsu earthquake (JGS, 2007b).

Fig. 43. Bridge abutment model with cement-treated backfill reinforced with geogrids(Watanabe et al., 2002).

J. Koseki / Geotextiles and Geomembranes 34 (2012) 51e6864

successful development and practical application of a new methodthat combines geosynthetic-reinforcement with a fiber-mixed soil-cement wall. The benefits of a similar approach to increase thestability by a combination of geosynthetic-reinforcement andcement-treatment will be also described in the next section onbridge abutments.

The above combination of geosynthetic-reinforcement withother reinforcement or improvement methods has been adopted in

0 200 400 600 800 1000120014000

5

10

15

20

25

30

35

40

45

Conventional(Model1)

Proposed Type

Dis

plac

emen

t(m

m)

Base acceleration(gal)

Ordinary type (untreated & unreinforced backfill sand)

New type (cement-treated gravel reinforced with geogrids)

Fig. 42. Comparison of residual wall top displacements (Watanabe et al., 2002).

Japanese practice as well. Herein, some relevant case histories of itsapplications to reconstruction works of earth structures damagedby earthquakes are reviewed briefly.

4.1.1. Case histories in 2004 Niigataken-chuetsu earthquakeFigs. 37and 38 show the collapse of a railway embankment and

its reconstruction involving a combination of GRS RW and earthanchors. The benefits of a similar approach to stabilize the base ofthese structures on slope using soil nails have been confirmed inthe model tests (Fig. 31).

Fig. 39a shows the collapse of another railway embankment. Itwas reconstructedby re-using the collapsedfillmaterial (Morishimaet al., 2005). As illustrated in Fig. 39b, the fill material was improvedby adding a cement-origin stabilizer at a mixing ratio of 150 kg/m3

for the upper fill or 105 kg/m3 for the lower fill. It was furtherreinforcedwith geogrid sheets thatwere placed at a vertical spacingof 1.5mas secondary reinforcement. In order to ensure the drainage,a gravel mat was placed at the bottom of the embankment.

Fig. 44. Bridge abutment using GRS RW with cement-treated backfill gravel (Aokiet al., 2005).

Fig. 45. a) Bridge abutment model and b) comparison of residual backfill settlements(Tatsuoka et al. 2007b).

Fig. 46. Bridge pier using GRS RW with pre-loaded and pre-stressed backfill gravel(Uchimura et al., 2003).

J. Koseki / Geotextiles and Geomembranes 34 (2012) 51e68 65

4.1.2. Case histories in 2007 Noto-hanto earthquakeAs shown in Fig. 13b, the collapsed highway embankments were

reconstructed using GRS RWs, where the collapsed fill material wasre-used after lime-treatment for the construction of the upper fill(Ishikawa Prefecture, 2007).

Fig. 47. a) Original concept and b) improved version of ballasted railway tracks rein-forced with geosynthetic bags and iron bars (Kachi et al., 2010).

4.2. Application to bridge abutments and piers

In a number of past major earthquakes, bridge abutmentssuffered from several types of damage, including settlement oftheir backfill soil (Fig. 40), extensive residual displacements of thewall body, and structural failure of the wall body (Fig. 41).

In order to improve the stability of retaining walls supportingbridge girders, Watanabe et al. (2002) and Aoki et al. (2003) con-ducted a series of 1-g model shaking tests on different types ofbridge abutments. As shown in Fig. 42, the wall displacement couldbe reduced significantly by using cement-treated backfill that wasreinforced with geogrids (Fig. 43) as compared to the ordinary typewall model using unreinforced and untreated backfill sand. Thebackfill settlement could be also reduced extensively.

On the other hand, Aoki et al. (2003) revealed as well that usingcement-treated backfill without reinforcements is not enough toimprove the seismic stability of bridge abutments up to sufficientlyhigh levels. They showed that using compacted and reinforcedbackfill that is not cement-treated is not effective, either.

As reported by Aoki et al. (2005), a similar type bridge abutmentusing cement-treated backfill gravel for the GRS RW with a full-height rigid facing (Fig. 44) has been implemented in practice toconstruct an abutment for new bullet train in Kyushu Island, Japan.By employing this system, the construction cost could be saved by20% as compared to the ordinary method.

Tatsuoka et al. (2007b, 2009) and Aizawa et al. (2007) conductedanother series of 1-g model shaking tests on bridge abutmentsunder otherwise the same conditions as those on retaining wallmodels described above. The models included integral types withor without reinforcements in the backfill (Fig. 45a), where thebridge girder was firmly connected to both of the walls. They weresubjected to several sequential horizontal excitations with 20cycles of sinusoidal waves at a frequency of 5 Hz. As shown in

Fig. 48. Schematic illustrationoneffectsof soil bags (modifiedafterMatsuokaandLiu,2003).Fig. 50. Excitation time history for full-scale 1-g model shaking test (Kachi et al., 2010).

Fig. 51. Maximum horizontal displacements measured in full-scale 1-g model shakingtest (Kachi et al., 2010).

J. Koseki / Geotextiles and Geomembranes 34 (2012) 51e6866

Fig. 45b, the seismic stability could be improved significantly bycombining the integral bridge and the GRS RW systems.

With the above GRS integral bridges, in addition to the rupturestrength and pull-out resistance of reinforcements, the connectionstrength between the facing and the reinforcement was found to beone of the key issues to maintain the high seismic stability byconstraining rotation of the facing with active displacements at thefacing bottom (Hirakawa et al., 2007).

For bridge piers to support girders, pre-loaded and pre-stressedgravel backfill for GRS RW with a full-height rigid facing has beenalso implemented in practice for a railway in Kyushu Island, Japan(Fig. 46, Uchimura et al., 2003). Its high seismic stability wasconfirmed through model shaking tests (Shinoda et al., 2003), andits applicability to bridge abutments has been verified as well (Aokiet al., 2003).

4.3. Application to ballasted railway tracks

In order to reduce deformation of ballasted railway tracksduring large earthquakes, a method to reinforce their shouldersusing stacked geosynthetic bags that are filled with ballast material,as schematically shown in Fig. 47a, was developed by Kachi et al.(2010). The stacked bags were further reinforced with iron barsdriven through the bags and embedding them to the base layer.

This method can be regarded as an extension of conventionalsoil bag methods, while it employs a mesh-type bag to mobilizebetter interlocking at the interface between adjacent bags. Theopening (i.e., the aperture) of the mesh is about 25 mm, which ismuch larger than those of conventional soil bags. As schematicallyshown in Fig. 48, the tensile forces mobilized in the bag would inturn enhance the bearing capacity of the bag by adding an apparentcohesion. Refer to Matsuoka and Liu (2003), Lohani et al. (2006),Matsushima et al. (2008) and Mohri et al. (2009) for more detaileddiscussions on the advantages of soil bags.

It should be noted that, during large earthquakes, the direction ofthemajor principal stress that is mobilized in the ballasts will rotatefrom the vertical direction. Therefore, referring to the pioneeringwork by Matsushima et al. (2008), the method was improved by

Fig. 49. Full-scale 1-g model shaking test on improved version of ballasted railwaytracks reinforced with geosynthetic bags and iron bars (modified after Kachi et al.,2010).

stacking the bags in an inclined manner as shown in Fig. 47b. Inorder to resist against the overturning moment more effectively,some of the reinforcing iron bars were also inclined. In addition, inorder to increase the overall stiffness of the stacked bags, adjacentbags were connected to each other by using a U-shaped iron bar.

After confirming better performance of the improved version byhorizontal monotonic loading tests, a full-scale 1-g model shakingtest was conducted as shown in Fig. 49. The model was subjected toa severe horizontal excitation as shown in Fig. 50. The maximumresponse in terms of horizontal displacements during the excita-tion is shown in Fig. 51. The horizontal displacement at the tieposition (see Fig. 47b) was as small as 6 mm, which implies that theimproved structure can performwell even under severe earthquakeloads in preventing high-speed trains from derailment. Thus, thistechnology has been adopted for actual reinforcing works ofexisting ballasted tracks for Tokaido bullet train (or Tokaido Shin-kansen) of Central Japan Railway Company.

Kachi et al. (2010) also reported that the newly-developedmethod has good workability, as compared to an alternativemethod using a pre-cast concrete block with a mass of about200 kg, and thus requires heavy equipments. With the newmethod, on the other hand, each of the geosynthetic bags can behandled easily without using heavy equipments, since they aresimply filled with about 25 kg of ballast material and can be easilycompacted into a specified dimension of 400*400*100 mm witha plate compacter.

5. Conclusions

The contents of the present paper on the use of geosynthetic-reinforcement to improve seismic performance of earth structurescan be summarized as follows.

1) As compared to unreinforced earth structures, geosynthetic-reinforced soil retaining walls (GRS RWs) performed satisfac-torily during past large earthquakes in Japan. Their ductilebehavior under large earthquake loads was also confirmed byrelevant model tests. Thus, geosynthetic-reinforced earthstructures havebeen adopted for newconstruction of importantpermanent structures aswell as their repair and reconstruction.

2) The seismic performance of GRS RWs is affected by rigidity offacing, pull-out resistance and arrangement of reinforcements,

J. Koseki / Geotextiles and Geomembranes 34 (2012) 51e68 67

and their connection strength. Due attention shall be also paidon the effects of shear deformation of subsoil and reinforcedbackfill layers and formation of full failure plane in these layers.

3) Combination with other reinforcement or improvementmethods, such as nailing, anchoring, sheet-piling and cement-treatment, enhances further the seismic performance of GRSRWs. By employing such combination, collapsed soil materialscan be re-used effectively.

4) Due to the above advantages, the application of geosynthetic-reinforcement is extending to wider areas. For example, it hasbeen successfully applied to bridge abutments and ballastedrailway tracks in practice.

Acknowledgments

The author appreciates Tensar International Limited for spon-soring the lecture.

The author wishes to express his sincere respects to Prof. F.Tatsuoka (Tokyo University of Science) and Dr. M. Tateyama(Railway Technical Research Institute) who initiated with theauthor immediately after the 1995 Hyogoken-nanbu earthquakethe long-lasting series of researches on seismic performance of GRSRWs as well as conventional type RWs, which were referred toextensively in this paper. Prof. Tatsuoka reviewed kindly the draft ofthis paper as well.

Among many others, the author acknowledges significantcontributions of late Dr. Y. Munaf, Dr. K. Watanabe, Mr. N. Kato, Dr.S. Nakajima and Mr. K. Hong (formerly, graduate students atUniversity of Tokyo) in carrying out the researches.

The author also thanks Mr. M. Seki, Mr. T. Kachi and Mr. M.Kobayashi (Central Japan Railway Company) for extending theresearches to ballasted railway tracks.

References

Aizawa, H., Nojiri, M., Hirakawa, D., Nishikiori, H., Tatsuoka, F., Tateyama, M.,Watanabe, K., 2007. Validation of high seismic stability of a new type integralbridge consisting of geosynthetic-reinforced soil walls. In: New Horizons inEarth Reinforcement (Proc. of IS Kyushu 2007). Taylor and Francis,pp. 819e825.

Aoki, H., Watanabe, K., Tateyama, M., Yonezawa, T., 2003. Shaking table tests onearthquake resistant bridge abutment. In: Proc. of 12th Asian Regional Conf. onSoil Mechanics and Geotechnical Engineering, vol. 1, pp. 267e270.

Aoki, H., Yonezawa, T., Tateyama, M., Shinoda, M., Watanabe, K., 2005. Developmentof aseismic abutment with geogrid-reinforced cement-treated backfills. In:Proc. of 16th International Conf. on Soil Mechanics and Geotechnical Engi-neering, vol. 3, pp. 1315e1318.

Hirakawa, D., Nojiri, M., Aizawa, H., Nishikiori, H., Tatsuoka, F., Watanabe, K.,Tateyama, M., 2007. Effects of the tensile resistance of reinforcement in thebackfill on seismic stability of GRS integral bridge. In: New Horizons in EarthReinforcement (Proc. of IS Kyushu 2007). Taylor and Francis, pp. 811e817.

Ishikawa Prefecture, 2007. Damage to toll road due to Notohanto earthquake and itsrehabilitation. Official Bulletin 1, 4 (in Japanese).

Izawa, J., Ito, H., Saito, T., Ueno, M., Kuwano, J., 2009. Development of rationalseismic design method for geogrid-reinforced soil wall combined with fibre-mixed soil-cement and its applications. Geosynthetics International 16 (4),286e300.

Japanese Geotechnical Society, 2007a. Compiled Data on Damage to Highways Dueto the 2007 Noto-hanto Earthquake, 79 pp. (in Japanese).

Japanese Geotechnical Society, 2007b. Report of Damage Survey Committee on theNiigata-ken Chuetsu Earthquake, 518 pp. (in Japanese).

Japan Meteorological Agency, 2011. http://www.jma.go.jp/jma/kishou/know/whitep/2-1.html (accessed 29.08.11.) (in Japanese).

Kachi, T., Kobayashi, M., Seki, M., Koseki, J., 2010. Evaluation tests of ballasted trackreinforced with geosynthetic bags. In: Proc. of 9th International Conference onGeosynthetics, Brazil, pp. 1499e1502.

Kato, N., Huang, C.C., Tateyama, M., Watanabe, K., Koseki, J., Tatsuoka, F., 2002.Seismic stability of several types of retaining walls on sand slope. In: Proc. of7th International Conf. on Geosynthetics, Nice, vol. 1, pp. 237e240.

Kitamoto, Y., Abe, H., Shimomura, H., Morishima, H., Taniguchi, Y., 2006. Rapid andstrengthened repair construction for a seriously damaged railway embankmentduring violent earthquakes. In: Proc. 8th Int. Conf. on Geosynthetics, Yokohama,vol. 3, pp. 861e864.

Koseki, J., Munaf, Y., Tatsuoka, F., Tateyama, M., Kojima, K., Sato, T., 1998. Shakingand tilt table tests of geosynthetic-reinforced soil and conventional-typeretaining walls. Geosynthetics International 5 (1e2), 73e96.

Koseki, J., Tateyama, M., Horii, K., Munaf, Y., Kojima, K., 1999. Back analyses of casehistories and model tests on seismic stability of retaining walls. In: Proc. of 11thAsian Regional Conf. on Soil Mechanics and Geotechnical Engineering, vol. 1, pp.399e402.

Koseki, J., Hayano, K., 2000. Preliminary report on damage to retaining walls causedby the 1999 Chi-Chi earthquake. Bulletin of ERS 33, 23e34. Institute of Indus-trial Science, Univ. of Tokyo.

Koseki, J., Sasaki, T., Wada, N., Hida, J., Endo, M., Tsutsumi, Y., 2006a. Damage toearth structures for national highways by the 2004 Niigata-ken Chuetsuearthquake. Soils and Foundations 46 (6), 739e750.

Koseki, J., Bathurst, R.J., Guler, E., Kuwano, J., Maugeri, M., 2006b. Seismic stability ofreinforced soil walls. In: Proc. of 8th International Conference on Geosynthetics,Yokohama, vol. 1, pp. 51e77.

Koseki, J., Tateyama, M., Watanabe, K., Nakajima, S. 2007. Stability of earth struc-tures against high seismic loads. In: Proc. of 13th Asian Regional Conf. on SoilMechanics and Geotechnical Engineering, Kolkata, vol. 2, pp. 222e241.

Lohani, T.N., Matsushima, K., Aqil, U., Mohri, Y., Tatsuoka, F., 2006. Evaluating thestrength and deformation characteristics of a soil bag pile from full-scalelaboratory tests. Geosynthetics International 13 (6), 246e264.

Matsuoka, H., Liu, S.H., 2003. New earth reinforcement method by soilbags(“donow”). Soils and Foundations 43 (6), 173e188.

Matsushima, K., Aqil, U., Mohri, Y., Tatsuoka, F., 2008. Shear strength and defor-mation characteristics of geosynthetic soil bags stacked horizontal and inclined.Geosynthetics International 15 (2), 119e135.

Mohri, Y., Matsushima, K., Yamazaki, S., Lohani, T.N., Tatsuoka, F., Tanaka, T.,2009. New direction of earth reinforcement e disaster prevention for earthfill dam. Geosynthetics International 16 (4), 246e273 (IS Kyushu 2007Special Issue).

Morishima, H., Saruya, K., Aizawa, F., 2005. Damage to earth structures for ordinaryrailways and its rehabilitation. Foundation Engineering and Equipment (Kiso-ko) 33 (10), 78e83 (in Japanese).

Nakajima, S., Koseki, J., Watanabe, K., Tateyama, M., Kato, N., 2006. Shaking tablemodel tests on geogrid reinforced soil retaining wall with embedded sheetpile.In: Proc. of 8th International Conference on Geosynthetics, Yokohama, vol. 4,pp. 1507e1510.

Nakajima, S., Hong, K., Mulmi, S., Koseki, J., Watanabe, K., Tateyama, M., 2007a.Model tests on seismic performance of reinforced soil retaining walls by usingdifferent geo-grids. In: International Workshop on Earthquake Hazards andMitigations, Guwahati, India, pp. 319e325.

Nakajima, S., Koseki, J., Tateyama, M., Watanabe, K., 2007b. Shaking table modeltests on retaining walls reinforced with soil nailings. In: New Horizons inEarth Reinforcement (Proc. of IS Kyushu 2007). Taylor and Francis,pp. 707e712.

National Research Institute for Earth Science and Disaster Prevention, 2011a. http://www.kik.bosai.go.jp/kik/index_en.shtml (accessed 29.08.11.).

National Research Institute for Earth Science and Disaster Prevention, 2011b. http://www.k-net.bosai.go.jp (accessed 29.08.11.).

Shinoda, M., Uchimura, T., Tatsuoka, F., 2003. Improving the dynamic performanceof preloaded and prestressed mechanically reinforced backfill by using a ratchetconnection. Soils and Foundations 43 (2), 33e54.

Shinoda, M., Watanabe, K., Kojima, K., Tateyama, M., Horii, K., 2009. Seismic stabilityof a reinforced-soil structure constructed after the mid-Niigata prefectureearthquake. Geosynthetics International 16 (4), 274e285.

Tamura, Y., 2006. Lessons learnt from the construction of geosynthetic-reinforcedsoil retaining walls with full-height rigid facing for the last 10 years. In: Proc.of 8th International Conference on Geosynthetics, Yokohama, vol. 3, pp.941e944.

Tatsuoka, F., 1993. Roles of facing rigidity in soil reinforcing. In: Earth ReinforcementPractice, Balkema, vol. 2, pp. 831e870.

Tatsuoka, F., Tateyama, M., Koseki, J., Uchimura, T., 1995. Geotextile-reinforced soilretaining wall and their seismic behaviour. In: Proc. 10th Asian Regional Conf.on Soil Mechanics and Foundation Engineering, vol. 2, pp. 26e49.

Tatsuoka, F., Tateyama, M., Koseki, J., 1996. Performance of soil retaining walls forrailway embankments. Soils and Foundations, 311e324 (Special Issue of Soilsand Foundations on Geotechnical Aspects of the January 17 1995 Hyogoken-Nanbu Earthquake).

Tatsuoka, F., Koseki, J., Tateyama, M., 1997a. Performance of reinforced soil struc-tures during the 1995 Hyogo-ken Nanbu Earthquake. In: Earth Reinforcement,Balkema, vol. 2, pp. 973e1008.

Tatsuoka, F., Tateyama, M., Uchimura, T., Koseki, J., 1997b. Geosynthetic-reinforcedsoil retaining walls as important permanent structures. Geosynthetics Inter-national 4 (2), 81e136.

Tatsuoka, F., Koseki, J., Tateyama, M., Munaf, Y., Horii, N., 1998. Seismic stabilityagainst high seismic loads of geosynthetic-reinforced soil retaining structures.In: Proc. of 6th Int. Conf. on Geosynthetics, Atlanta, vol. 1, pp. 103e142.

Tatsuoka, F., Konagai, K., Kokusho, T., Koseki, J., Miyajima, M., 2006. SpecialSession on the 2004 Niigata-ken Chuetsu earthquake. In: Proc. of 16thInternational Conf. on Soil Mechanics and Geotechnical Engineering, vol. 5,pp. 3279e3287.

Tatsuoka, F., Tateyama, M., Mohri, Y., Matsushima, K., 2007a. Remedial treatment ofsoil structures using geosynthetic-reinforcing technology. Geotextiles andGeomembranes 25 (4e5), 204e220.

J. Koseki / Geotextiles and Geomembranes 34 (2012) 51e6868

Tatsuoka, F., Hirakawa, D., Nojiri, M., Aizawa, H., Tateyama, M., Watanabe, K., 2007b.A new type integral bridge comprising of geosynthetic-reinforced soil walls. In:New Horizons in Earth Reinforcement (Proc. of IS Kyushu 2007). Taylor andFrancis, pp. 803e809.

Tatsuoka, F., 2008. Recent practice and research of geosynthetic-reinforced earthstructures in Japan. Journal of GeoEngineering 3 (3), 47e67.

Tatsuoka, F., Hirakawa, D., Nojiri, M., Aizawa, H., Nishikiori, H., Soma, R.,Tateyama, M., Watanabe, K., 2009. A new type of integral bridge comprisinggeosynthetic-reinforced soil walls. Geosynthetics International 16 (4), 301e326.

Uchimura, T., Tateyama, M., Koga, T., Tatsuoka, F., 2003. Performance of a preloaded-prestressed geogrid-reinforced soil pier for a railway bridge. Soils and Foun-dations 43 (6), 33e50.

Watanabe,K., Tateyama,M., Yonezawa,T.,Aoki,H., Tatsuoka, F., Koseki, J., 2002. Shakingtable testson anew typebridge abutmentwith geogrid-reinforcedcement treatedbackfill. In: Proc. 7th International Conf. onGeosynthetics,Nice, vol.1, pp.119e122.

Watanabe, K., Munaf, Y., Koseki, J., Tateyama, M., Kojima, K., 2003. Behaviors ofseveral types of model retaining walls subjected to irregular excitation. Soilsand Foundations 43 (5), 13e27.