2006) 207–220www.elsevier.com/locate/tecto

Tectonophysics 426 (

Borehole image analysis of the Nankai Accretionary Wedge,ODP Leg 196: Structural and stress studies

Masanori Ienaga a,1, Lisa C. McNeill b, Hitoshi Mikada c,⁎, Saneatsu Saito d,David Goldberg e, J. Casey Moore f

a Ocean Research Institute, University of Tokyo, Nakano-ku, Tokyo 104-0045, Japanb Southampton Oceanography Centre, Southampton SO14 3ZH, UK

c Department Civil and Earth Resources Eng., Kyoto University, Yoshida-hommachi, Sakyo-ku, Kyoto 606-8501, Japand IODP-MI Sapporo Office, CRIS, Hokkaido University, N21W10 Kita-ku, Sapporo 001-0021, Japan

e Borehole Research Group, Lamont-Doherty Earth Observatory, Route 9W, Palisades, NY 10964, USAf Department of Earth Sciences, University of California, Santa Cruz, CA 95064, USA

Accepted 7 February 2006Available online 30 June 2006

Abstract

Electrical images recorded with Resistivity-At-Bit (RAB) from two sites drilled during Ocean Drilling Program (ODP) Leg 196were analyzed to study the effects of subduction at the Nankai margin. For the first time in the history of scientific deep-sea drillingin ODP, in situ complete borehole images of the décollement zone were obtained. Analyses of all drilling-induced fracture dataindicated that the maximum horizontal compressive stress (SHmax) axes have an azimuth of 303°, and analyses of breakout datafrom RAB images indicated an azimuth of 310°. These azimuths approximate the convergence direction of the Philippine Sea platetowards the Eurasian plate. The frontal thrust at Site 808 was encountered at about 389 mbsf. Density, porosity, resistivity, andgamma ray data change across the frontal thrust. The décollement zone at the deformation front was identified between 937 and965 mbsf. The base of the décollement is sharply defined as the maximum extent of conductive fracturing and is marked by abruptchanges in physical properties [Mikada, H., Becker, K., Moore, J.C., Klaus, A., Austin, G.L., Bangs, N.L., Bourlange, S.,Broilliard, J., Brückmann, W., Corn, E.R., Davis, E.E., Flemings, P.B., Goldberg, D.B., Gulick, S.S., Hansen, M.B., Hayward, N.,Hills, D.J., Hunze, S., Ienaga, M., Ishiguro, H., Kinoshita, M., Macdonald, R.D., McNeill, L., Obana, S., Hong, O.S., Peacock, S.,Pettigrew, T.L., Saito, S., Sawa, T., Thaiprasert, N., Tobin, H.J., Tsurumi, H., 2002. Proc. ODP, Initial Rep., 196, College Station,TX, (Ocean Drilling Program)]. The upper boundary of the décollement is marked by several sets of conductive fractures and byhigh variability in physical properties. The décollement zone is characterized by intense brittle fracturing. These fractures areconsidered to be the consequence of cyclic stresses and high fluid pressures in this zone. We analyzed fracture dips and theirorientations at both sites and found that they are all consistent with a unique stress field model surrounding the two sites.© 2006 Elsevier B.V. All rights reserved.

Keywords: Décollement; Borehole breakout; Nankai Trough; LWD; RAB; ODP

⁎ Corresponding author. Fax: +81 75 753 4776.E-mail address: mikada@tansa.kumst.kyoto-u.ac.jp (H. Mikada).

1 Present address: SciMarkJ Inc., Kawasaki-shi, Kanagawa, 210-0012, Japan.

0040-1951/$ - see front matter © 2006 Elsevier B.V. All rights reserved.doi:10.1016/j.tecto.2006.02.018

1. Introduction

The active margin offshore Japan (Fig. 1a) hasperiodically generated large subduction earthquakes of

Fig. 1. (a) Location map shows tectonic setting of the Nankai Accretionary Prism. PSP=Philippine Sea Plate; KPR=Kyushu Palau Ridge;Fsc=Fossil Spreading Center; IBT=Izu Bonin Trench. (b) Seismic reflection profile across the Muroto transect (Moore et al., 1990). Recognizablethrust faults, Décollement zone, and top of Lower Shikoku Basin sediments are shown in broken lines. Thrusts and their conjugate faults form acomplex structure on the left side of figure. A strong seismic reflector perceived as Décollement zone (left half of figure) jumps to the top of the LowerShikoku Basin sediments where a number of normal fault type discontinuities are observed below the reflector.

208 M. Ienaga et al. / Tectonophysics 426 (2006) 207–220

the order of magnitude 8 (e.g., in 1605, 1707, 1854,1946; Ando, 1975). The Nankai Trough is one of thebest-studied convergent plate boundaries. Subductioncomplexes provide an opportunity to examine theinitiation of décollement zones, i.e., detachment planesthat separate accreted from subducted sediments andruptures during subduction earthquakes. Recent studieson subduction zones have established that fluids play amajor role in their physical and chemical evolution andseismogenic activities (Hyndman et al., 1995). Theoverall structure of the prism and the general nature ofthe deformation processes are now reasonably wellknown. However, many questions remain and thedistribution, nature, shape, and origin of the décollementzone at the base of the prism are still poorly understood.

The objective of Ocean Drilling Program (ODP) Leg196 was to clarify the nature of deformation and fluidflow in the Nankai accretionary prism. Leg 196 was the

5th deep-sea drilling investigation at the Nankai Trough,following DSDP (Deep-Sea Drilling Project) Legs 31and 87 and ODP 131 and 190. Logging-while-drilling(LWD) was carried out during Leg 196 to measure thephysical properties of the décollement zone andoverlying prism. A logging tool string was located justabove the drill bit to measure resistivity, density,porosity, and sonic velocity. In Leg 196, two siteswere drilled to obtain physical properties: one is the Site808 in the proto-thrust zone of the prism and the otherSite 1173 about 9 km seaward from the Nankai Trough.The décollement and the seaward stratigraphic equiva-lent of the décollement (referred to as “pre-décollement”here) zones were penetrated. Sites 808 and 1173 wereselected for investigation of the early stages ofdécollement evolution. The décollement zone wascharacterized seismically as a high-amplitude re-versed-polarity reflection (Fig. 1b), which might suggest

209M. Ienaga et al. / Tectonophysics 426 (2006) 207–220

a high-porosity and low-density layer in the décolle-ment. However, core samples recovered during Leg 131suggest low-porosity and high-density in the décolle-ment zone. Core recovery within the décollement zonewas 20% or less (Taira et al., 1991), and continuous andsequential data obtained by logging were required toreconstruct the structure and properties of the décolle-ment zone. This included resistivity borehole images ofthe décollement zone (Mikada et al., 2002) using LWD,which had a resolution of 10° in the horizontal aperture.

Many of the initial observations from the RAB dataset were first described by Mikada et al. (2002). Initialresults and interpretations were presented by McNeill etal. (2004) for Site 808. Here, we expand theirdiscussions and present to interpret the most significantstructural features observed in the RAB images for bothSites 808 and 1173. Two methods were used to analyzethe borehole images: characterization of fracture dip andbedding directions, and determination of in situ stressdirections from borehole breakouts (Peška and Zoback,1995). Finally, based on the results from these analyses,we introduce discussions on the role of fluid pressure inthe propagation of the décollement zone in the stressenvironment that we suggest based on results from asandbox experiment by Yamada et al. (2006). Weconclude that all observed fracture data are consistentwith a unique stress field model and that thedevelopment of the prism might have controlled thestress orientation and dip, which caused the décollementzone to advance seaward parallel to the plate motiondirection.

2. Nankai accretionary prism

The Nankai Trough marks the subducting plateboundary between the Shikoku Basin, i.e., a part of thePhilippine Sea plate, and the southwest Japan arc, i.e., apart of the Eurasian plate. At this boundary, thePhilippine Sea plate is subducting to the northwestbeneath the Eurasian plate at a rate of 0.02–0.04 m/year(Seno, 1977; Karig and Angevine, 1986), oblique to theplate margin. The convergent margin of southwesternJapan has a geological record of accretion of deep-seadeposits extending to at least the Cretaceous (Taira et al.,1988). In the area drilled in Legs 131, 190, and 196, i.e.,the Muroto Transect (Fig. 1a), the basin-to-margintransition can be divided into the undeformed ShikokuBasin and overlying trench fill, the proto-thrust zone,the imbricate thrust zone, the frontal out-of-sequencethrust zone, the large thrust slice zone, and thelandward-dipping reflector zone (Moore et al., 2001a,b). The Philippine Sea Plate entering the Nankai Trough

along the Muroto Transect is close to the axis of anextinct spreading center marked by the Kinan Sea-mounts (Okino et al., 1999). Along the Muroto transect,the prism toe has a taper angle of 4°–5° (Moore et al.,2001b). The low taper angle along this transect has beeninferred to represent either high décollement porepressures or low internal sediment strength (Screatonet al., 2002). The décollement zone is characterizedseismically as a high-amplitude and reversed-polarityreflection (Fig. 1b), which is consistent with observa-tions of high pore pressures in the sediments below thedécollement layer (Moore and Shipley, 1993).

3. Characterization of fracture and bedding

3.1. Methods

Electrical images were obtained with logging-while-drilling technology using the Resistivity-at-Bit (RAB)tool at shallow, medium, and deep depths of investiga-tion. The penetration of each depth was 0.025, 0.076,and 0.13 m, respectively. These images can be displayedusing either static or dynamic normalizations. The staticnormalization color range covers all resistivity valuesfor the logged interval, whereas the dynamic normali-zation is limited to the resistivity values over a specifieddepth interval. These two normalizations providecomplementary images of geological structures. Dy-namic normalization is used for detailed comparisons ofsedimentary structures, while static normalization ispreferred for correlating lithological or facies changes.In RAB images, bedding, fractures, and faults appear asconductive or resistive anomalies with a sinusoidalshape (Fig. 2). Fractures from RAB images can beclassified into either conductive or resistive features thatcross bedding planes. The conductive fractures wereinterpreted as fractures filled with conductive fluids.Resistive fractures are filled with non-conductive claygouge or minerals causing relative reduction in porosity.Mechanically, drilling can induce fractures by thedrilling process or by a hydraulic fracture. Consideringsmooth profiles of illite–smectite composition in theformations (Shipboard Scientific Party, 2001, 2002a,b),we think drilling induced fractures appear as conductivefractures which may reflect the present day leastprincipal stress direction (Zoback et al., 1985).

3.2. Site 1173 (reference site)

3.2.1. BeddingBedding data interpreted from the RAB images at

Holes 1173B and 1173C were compared with seismic

Fig. 2. Example of the fractures and normal fault at Site 1173 for a 10 m interval right below the stratigraphic equivalent of the décollement layer(referred as pre-décollement here). Figure shows uninterpreted image (left), interpreted RAB image (center), and fracture dips and orientations (right).

210 M. Ienaga et al. / Tectonophysics 426 (2006) 207–220

reflection data (Fig. 3). Hole 1173C was drilled downto 175 mbsl to acquire sediment physical propertiesusing LWD. Data from the overlapped depth zone(124–175 mbsf) between the two holes was takenfrom Hole 1173B. Bedding dips are relatively lowthroughout the hole, with predominantly 0°–5°,although two sets of dips ranging from 5° to 35°were identified.

The very well laminated bedding of trench turbiditesis in contrast to the heterogeneous feature of bedding,i.e., varied dip angles, at the top of the Upper ShikokuBasin facies, and clearly shows a change in thesedimentation process. These bedding planes dip from0° to 20° with the majority at 0°–5° (depths 50–170 mbsf). Bedding dip direction is to the north.Bedding dips in the Upper Shikoku Basin facies andLower Shikoku Basin facies are primarily sub-horizon-tal between 170 and 725 mbsf, although bedding dipsexceeding 5° occur at 250 mbsf, 300–590 mbsf, and700–725 mbsf. Seismic depth sections suggested an

increase in bedding dips at 300–500 mbsf, therefore inagreement with RAB image interpretation. Althoughbedding dip direction could not be estimated accuratelybecause of the low dip values, the average dip directionof southwest to northwest shows that the entire sequenceis tilted westward (<5°).

3.2.2. Fractures and faultsThe number of fractures and faults are limited

throughout Holes 1173B and 1173C (Fig. 4). Site1173 is, as expected for a reference site, located ca.12 km seaward from the deformation front (Moore et al.,2001b). Intermediate fracturing was identified through-out the hole with the majority of deformation concen-trated between 300 and 450 mbsf. This zone iscorrelated with increases in bedding dip. Specificconcentrations of fractures occur at 411–417 and 633–730 mbsf. These zones also correspond to areas ofincreased deformation observed in core samples (Mooreet al., 2001a,b). Fracture dip is high (25°–80°), with the

Fig. 3. Bedding dips measured from resistivity-at-the-bit (RAB) data with depth-converted seismic reflection data, core-based lithology, and facies interpretation at Sites 808 and 1173. Holes A to G atSite 808 were drilled during Leg 131 (Taira et al., 1991), while Hole A at Site 1173 during Leg 190 (Moore et al., 2001a). It is well recognized that beddings for both sites are all close to horizontal.

211M.Ienaga

etal.

/Tectonophysics

426(2006)

207–220

Fig. 4. (a) Fracture dip and direction from resistivity-at-the-bit image interpretation at Sites 808 and 1173. (b) Stereographic projection of poles ofbedding, fractures, and faults at Sites 808 and 1173. Fractures have high dip angles around 60° but low dip angle fractures appearing below dips areobserved gradually lighter as depth at Site 808 while the orientations are always bi-modal in parallel to plate motion direction. Fracture density at Site1173 is lighter than at Site 808.

212 M. Ienaga et al. / Tectonophysics 426 (2006) 207–220

213M. Ienaga et al. / Tectonophysics 426 (2006) 207–220

majority dipping at 30°–65°. Fracture dip directionsshow orientations mainly ranging 150°–360° for the top600 m and 270°–360° for the bottom of the drilled hole,respectively.

3.3. Site 808 (deformation front)

3.3.1. BeddingBedding was discernible from RAB images at Hole

808I on Leg 196. Here, the variations in bedding dip anddirection with depth are shown in Fig. 3. Hole 808Isuccessfully penetrated both the frontal thrust zone andthe décollement zone in the toe of the accretionaryprism, allowing detailed structural analysis of thesezones and the state of deformation between them.Bedding dip angles are low throughout the hole, withmost at 0°–5°. Bedding direction appears to be fairlyrandom throughout the hole, although three discrete setsof bedding planes were identified according to theiraverage dip and orientation.

Fig. 5. Site 808 summary diagram at 930–970 mbsf showing combined resultsporosity, RAB image of the décollement, and core photos. Fractures mainly

Bedding dip above 400 mbsf (Unit A) with the verywell laminated bedding of trench turbidities ranged from0° to 25°, with the majority showing a dip of 0°–5°. Instructural Unit B at depths of 400–800 mbsf, thebedding dip ranged from 0° to 20°, with the majorityshowing a value of 0°–5°, which is almost sub-horizontal. A major change in the feature of the beddingoccurs below 800 mbsf within structural Unit C.Bedding dips are generally sub-horizontal, but increasesin the dip angle to 5°–25° were detected. The dipdirection was widely scattered in structural Unit C.

3.3.2. Natural fracturesIncreases in deformation and fracturing were

detected at Site 808 (Fig. 4) as compared with referenceSite 1173. Fracture frequency increases significantly atmajor deformation zones, such as the frontal thrust zone(∼400 mbsf), a major fracture interval at ∼560 mbsf,and the décollement zone (935–965 mbsf) (McNeill etal., 2004). A dominant fracture dip of east–northeast-to-

of Legs 131 and 196. From left to right: log resistivity, log density, logappear top and the bottom of the décollement layer.

214 M. Ienaga et al. / Tectonophysics 426 (2006) 207–220

west–southwest is present throughout the borehole,although east–west dips sometimes appear within andjust above the décollement zone. Fracture dips haverelatively high angles and range from 10° to 80°.

3.3.3. Frontal thrustThe frontal thrust zone (389–414 mbsf) represents

the most strongly deformed interval in Hole 808I RABimages and in the cores of Site 808, Leg 131 (Taira et al.,1991). The zone of the frontal thrust is characterized byclear conductive fractures (Fig. 4) with overall highresistivity fractures (McNeill et al., 2004). Fracture dipis high (30°–80°), with the majority dipping 45° to 65°.The majority of fractures show northeastern to WSWorientation and southward dipping. There is a progres-

Fig. 6. Resistivity-at-the-bit images of borehole breakouts at 1 in. (shallow)505 mbsf (McNeill et al., 2004).

sion from northward-dipping fractures in the upper partof the deformation zone to southward dipping in themiddle and lower parts of the zone (McNeill et al.,2004).

3.3.4. DécollementThe highly fragmented nature of the décollement

zone, described previously (Taira et al., 1991), made itdifficult to obtain core samples. On the other hand,detailed physical properties were acquired from LWD.Specific concentrations of highly conductive fractureswithin this interval occur at 897–965 mbsf. We definedthe décollement itself as the base of fluid layers at 937–965 mbsf (Fig. 5; McNeill et al., 2004). The base of thedécollement is sharply defined as the maximum extent

, 3 in. (medium) and 5 in. (deep) penetration from the borehole, 460–

215M. Ienaga et al. / Tectonophysics 426 (2006) 207–220

of conductive fracturing and is marked by abruptchanges in physical properties (Mikada et al., 2002).On the contrary, it is more difficult to define the top ofthe décollement. Our choice of the upper boundary ofthe décollement at 937 mbsf is marked by a set ofconductive fractures and by variability in physicalproperties (Fig. 5). Although the results obtained fromcore data at Site 808 in Leg 131 showed both upper andlower boundaries at 945–964 mbsf (Taira et al., 1991),our interpretation gave us a thickness of ca. 28 m, aresult similar to 32.6 m in the zone of the décollement atSite 1174, Leg 190 (Moore et al., 2001b).

4. In situ stress orientations and estimations

Borehole breakouts were recorded at Sites 808 usingRAB images (Fig. 6; McNeill et al., 2004). When thedifferential stress in the borehole wall exceeds thecompressive strength of the rock, failure occurs causingborehole elongation. The orientation of breakoutsconstrains the direction of the horizontal principalstress SHmin and SHmax (Zoback et al., 1985). Severalinvestigators have utilized borehole images to identifystress-induced features along the borehole and tosubsequently evaluate local stress conditions (Bell

Fig. 7. RAB images of drilling-induced tensile fractures appeared in the basaland interpreted image (right). The appearance of the tensile fracture may indicaplate motion direction.

and Gough, 1979) using the breakout data to constraindirections of stress but not their relative magnitudes.However, breakouts can yield information on stressmagnitudes, both because the presence of a breakoutindicates that the hoop stress exceeds the yield strengthof the rock and because, when the borehole is notaligned with a principal stress axis, the breakoutorientation depends on the relative magnitudes of allthree principal stresses, as well as on the orientations ofthe stress (Zoback et al., 1985).

4.1. Site 1173

Drilling induced vertical tensile fractures are identi-fied in Hole 1173B in the lowest section of the loggedinterval, i.e., in the top of the oceanic basement. Thebasement is composed of basalt between 731 and735 mbsf (Fig. 7). These drilling induced tensilefractures show echelon shapes, of which verticalfractures lie basically in NE–SWorientation. The tracesof these fractures do not cross the entire borehole but areconsistent in the direction that may indicate theexistence of horizontal differential stress in the oceanicbasement. There are discussions on the superposition oftectonic stress concentrations, pressure changes exerted

tic segment of the drilled hole. Figure shows uninterpreted image (left)te that the top of the basement is in a tensile stress field in parallel to the

Table 1Mean values and standard deviations of maximum compressedorientations measured from the RAB images at Site 808

Structuralunit

Depth(mbsf)

Fractures Boreholebreakouts+90°

Meanvalue

Standarddeviation (1σ)

Meanvalue

Standarddeviation (1σ)

A <406 320° 31° 315° 24°B 406–692 302° 43° 309° 14°C 692< 289° 37° 308° 14°All data 303° 40° 310° 17°

All orientation values indicate that the sediments at Site 808 arestrongly influenced by the direction of the plate convergence, i.e.,sediment loading from the prism.

216 M. Ienaga et al. / Tectonophysics 426 (2006) 207–220

by the drilling fluid, and thermal stresses at the boreholewall in terms of in situ stress estimations elsewhere(Moos and Zoback, 1990), but we simply use them as astress orientation indicator since there is no strength dataavailable. Fractures observed from RAB image haverelatively high dip angles and orientations rangingmainly from 270° to 360°, i.e., centered in the directionof the plate convergence. However, there are noobservable borehole breakouts in the entire sediments.

4.2. Site 808

Hole 808I data showed a large number of boreholeelongations and breakouts along the depth sectionanalyzed (Shipboard Scientific Party, 2002b; McNeillet al., 2004). The borehole breakouts are aligned in NE–SW directions similar to both conductive and resistivefracture orientations (Fig. 8), which indicates SHmax

Fig. 8. Borehole breakout azimuth from RAB image interpretation at Site 808rose diagrams (right). These breakouts are induced in the direction of SHmin.

lying normal to this direction, i.e., NW–SE direction asdemonstrated by McNeill et al. (2004). These boreholebreakouts occur at three depth intervals between 181and 389 mbsf, 407 and 692 mbsf, and 718 and 825 mbsf.

(left). Distribution of borehole breakouts (center) and breakout azimuth

217M. Ienaga et al. / Tectonophysics 426 (2006) 207–220

The orientations of these borehole breakouts aresummarized in Table 1.

5. Discussions

The compressive directions (SHmax) of the observedfractures (320°) and the drilling-induced boreholebreakouts (315°) of structural Unit A differ slightlyfrom those derived in Units B and C (Table 1). However,these orientations might generally be regarded as parallelto the plate convergence within the resolution of thisstudy. Fracture orientations from RAB are consistentwith those measured in cores after rotating according topaleomagnetic declination. Stress orientations deter-mined by inversion from slickenlined fracture data(Taira et al., 1991) compare well on average with thosedetermined in this study, but also suggest that transientchanges in stress orientations occur in the process ofdécollement and thrust initiation (Lallemant et al., 1993).These fractures and borehole breakouts imaged byanalysis of RAB data have provided insight into thestructure of the logged interval. The décollement zone ischaracterized by sub-horizontal concentrations of con-ductive natural fractures at Site 808 (Maltman et al.,1993; Ujiie et al., 2003). The absence of deformationstructures below the décollement, together with isotropicmagnetic susceptibility (Moore et al., 2001a) and sonicvelocities (Yoneshima et al., 2003; Goldberg et al.,2005), suggest that the underthrust materials have under-gone virtually no strain (Maltman et al., 1993; Ujiie et al.,2003). Average resistivity trends from Leg 196 RABdata across the frontal thrust, incipient thrust, anddécollement suggest that the two former are compacted(increased resistivity), whereas the décollement is dilated(reduced resistivity) (McNeill et al., 2004).

5.1. Stress conditions

The frontal thrust at Site 808 was encountered atabout 389 mbsf. Density, porosity, resistivity, andgamma ray data all change across the frontal thrust(Shipboard Scientific Party, 2002b). We identified fourdepth zones where no drilling-induced borehole break-outs are observed, 389–407, 692–718, 825–920 mbsf,and 930 mbsf to the bottom of the logged interval at Site808, and the entire sediment interval at Site 1173. Asbriefly discussed by Hickman and Zoback (2004), theabsence of borehole breakouts may indicate zones ofstress relief either by an increase in unconfinedcompressive stress or a decrease in differential stressnormal to the hole. Since Hole 808 penetrates the frontalthrust for an interval from 389 to 414 mbsf, and

décollement zone from 940 to 960 mbsf, depth intervalsfor the 389–407 and 937–965 mbsf could be interpretedas zones of stress relief due to the development of thrust-fault and shear fracturing, respectively. With theestimated maximum compressive stress axis of 303°,all breakout data indicate that SHmax lies parallel to theplate convergence direction and that the correspondingstress is relieved at, at least, two depth zones, i.e., thefrontal thrust and décollement.

There is no observable borehole breakout at Site1173. Fractures are observed with relatively high dipangles and mainly in two orientations (Fig. 4). Thesehigh dips and twomain fracture orientations indicate thatthe principal stress directions would be close to verticalbut are slightly influenced in the direction of platemotion. These observations for Site 1173 imply thatthere is very minor horizontal differential stress. Sincethe Site 1173 is located at more than 10 km from thedeformation front, the cause of the bend of the principalstress axis from vertical may be the overload of trench fillmaterial in the seaward of the deformation front. Sincenormal displacement fractures are observed (Fig. 3), thesite appear to be located in a tensile stress field (Fig. 7)and possibly for the zone between the two Sites 808 and1173 (Bangs and Gulick, 2005; Moore et al., 2005).

Yamada et al. (2006) discussed the directions of stressfield (σ1 and σ3) at the toe of an accretionary prismfollowing his sandbox and numerical experiment. Theypredicted that the development of the accretionary prismdue to shortening and overriding of sediments above thedécollement influence maximum stress dips. Althoughtheir experiment did not consider any effects of interstitialfluid in the sediments, we may consider that horizontalshear faults could be developed along the line of principalstress directions near the décollement (Fig. 9).

Since there is no rock strength data available, it isdifficult to identify if the absence of borehole breakoutsoccurs due to strengthened rock or to stress relief.Detailed rock property analyses should be conducted forfurther quantitative investigations from the presentanalyses. The frontal thrust zone is characterized byobvious conductive fractures with overall high resistiv-ity fractures on the RAB images at Hole 808I, and couldbe a stress relieved zone filled with fluids as alreadydiscussed.

5.2. Structural development of the prism thrust faultsand décollement

The apparent heterogeneous resistivity pattern at thefrontal thrust zone and décollement zone could beexplained by the difference in interconnected fracture

Fig. 9. Model of stress field surrounding drilling sites of ODP Leg 196. Lines of principal stress (σ1) directions are from Yamada et al. (2006). Due toheavy mass loading of the prism, vertical normal trend of compaction to the sediments is distorted towards the prism. It is qualitatively indicated thatthrust type and horizontal fracturing tends to be formed above and in the décollement, respectively. A brow-up image of the décollement zone in thebottom implies that tensile fractures may be developed at the top and base of the décollement zone. An area indicated by “A” denotes a possible zoneof tensile field due to spatial changes in principal stress directions.

218 M. Ienaga et al. / Tectonophysics 426 (2006) 207–220

porosity as discussed by Bourlange et al. (2003).Bourlange et al. (2003) also discussed possibleprocesses of décollement development assuming shearstress and dilation of sediment. Experiments on highlyporous, cemented clays demonstrated that cementationcould be destroyed by cyclic or repeated stresses wellbelow the yield stress (Leroueil and Vaughan, 1990).Ujiie et al. (2003) summarized that the décollement atthe Nankai Trough evolved following two deformationconditions: (1) the destruction of the porous cementedstructure probably in association with fluid pressurefluctuation and (2) clay-particle rotation and porositycollapse along sets of slip surfaces under significantshear stress. Then, they speculated that the localizationof the décollement could be caused by a combination ofincomplete mechanical destruction of porous cementedstructure and fluid pressure fluctuations. Bourlange etal. (2003) suggested that the effects of non-linear Darcyflow could augment the permeability of the décollementby a factor of 103 depending on fluid pressure. Theirspeculations would imply that the décollement wouldact as a fluid conduit more efficiently at fluid pressuresbeyond the level of the overburden vertical stress.Therefore, through a combination of these effects, itappears likely that fluid transport may occur throughconduits of shear fractures, i.e., in a horizontal directionin the décollement zone.

Sandbox experiments and subsequent numericalsimulation suggest that stress directions become strong-

ly influenced by development of the accretionary prismand that the stress directions are to be distorted landwardfrom a vertical direction (Yamada et al., 2006; Fig. 9).As shown in Fig. 9, thrust faulting and horizontalfracturing would take place in directions of shear failureexpected in the given stress orientations at Site 808 andwould release accumulated shear stress at thesestructural boundaries. In the décollement zone, shearfaulting would be developed in horizontal directions dueto the bending of the principal stress direction (Fig. 9).In the overriding sediments, shear fractures would bedeveloped in slightly upward sub-horizontal directions,while normal-fault type shear fractures could be formedas in a tensile stress field in the underthrust sediments.Considerations on dips of principal stress directions alsoindicate that the localization of décollement may takeplace in a certain stress condition shown in Fig. 9 andthat tensile fracturing may likely take place at the topand bottom of the décollement zone.

It is difficult to predict the stratigraphic layer inwhich the décollement will advance on the basis ofphysical properties, stress history, and in situ stressdistribution, but the state of stress, in particular stressdip, could be related to the initiation of the décollement.The décollement may advance due to cyclic stressesunder high fluid pressures. In such stress and overpres-sure conditions, sediments could cause brittle fracturing.Permeability measurements of clay rich sediments undercompressive stresses indicate a significant permeability

219M. Ienaga et al. / Tectonophysics 426 (2006) 207–220

reduction perpendicular to the direction of compressivestresses (Brown et al., 1994). When the pre-décollementzone approaches to the trough, the principal stressdirection would be bent landward so that horizontalshear fracturing is more likely to occur (Fig. 9). Anyraise in fluid pressure would increase opening of thesefractures. We believe that this is a preferred hypothesisof décollement development in an accretionary prismthat could be confirmed after obtaining variations inphysical properties, stress field orientations and micro-structures of the sediments as a function of distance to/from the trough.

6. Conclusions

The RAB data recorded during Leg 196 show thathigh-resolution electrical imaging is an efficient meansto record in situ tectonic features. Orientation of themain compressive stress is 303° from fractures, and310° from borehole breakouts, and all generally parallelto the direction of plate convergence (Seno, 1977;McNeill et al., 2004) within the resolution of this study.The frontal thrust at Site 808 was encountered at about389 mbsf. An incipient thrust zone at Site 808 isidentified at 680–730 mbsf. The décollement zone atSite 808 is identified at 937–965 mbsf. The top and baseof the décollement are sharply defined as the maximumextent of conductive fracturing. At Site 1173, high dipfractures whose dip orientations are mainly in twodominant orientations are observed. There is noobservable drilling induced borehole breakouts withinHole 1173B. These observations indicate that there isminor horizontal differential stress and that the bendingof the principal stress directions from the vertical axistowards horizontal and parallel to the plate motion isintensified from Site 1173 towards Site 808. Our resultsalso suggest that principal stress directions mightcontrol the development of the décollement zone.

Acknowledgements

The authors are grateful to the Ministry of Education,Science, Culture and Sports of Japan (MEXT) forfunding this research as the Grant-in-Aid for ScientificResearch for the Ocean Drilling Program (ODP), toOcean Research Institute of the University of Tokyo(ORI) and the Japan Agency for Marine-Earth Scienceand Technology (JAMSTEC) for support. The authorswould also like to thank all the crew of the JOIDESResolution for their professionalism and the members ofthe ODP Leg 196 shipboard scientific party fordiscussions. Special gratitude is given to P. Henry and

the two anonymous reviewers for discussions andsuggestions to improve the manuscript. Masanori Ienagaparticipated in this study for his undergraduate researchand greatly appreciate H. Tokuyama, M. Coffin, J. Ashi,Y. Nakamura, M. Shirai, D. Curewitz, A. Tikku, K.Suzuki, and K.T. Moe for discussions to prepare themanuscript. Hitoshi Mikada tenders a great appreciationto Yasu Yamada for discussions and for providing astress condition model of an accretionary prism.

References

Ando, M., 1975. Source mechanisms and tectonic significance ofhistorical earthquakes along the Nankai Trough, Japan. Tecto-nophys 27, 119–140.

Bangs, N.L.B., Gulick, S.P.S., 2005. Physical properties along thedeveloping décollement in the Nankai Trough: inferences from 3-D seismic reflection data inversion and Leg 190 and 196 drillingdata. In: Mikada, H., Moore, G.F., Taira, A., Becker, K., Moore, J.C., Klaus, A. (Eds.), Proc. ODP, Sci. Results, vol. 190/196.[Online]. Available fromWorldWideWeb: ⟨http://www-odp.tamu.edu/publications/190196SR/354/354.htm⟩. [Cited 21 May 2005].

Bell, J.S., Gough, D.I., 1979. Northeast–southwest compressive stressin Alberta: evidence from oil wells. Earth Planet. Sci. Lett. 45,475–482.

Bourlange, S., Henry, P., Moore, J.C., Mikada, H., Klaus, A., 2003.Fracture porosity in the décollement zone of Nankai accretionarywedge using Logging While Drilling resistivity data. Earth Planet.Sci. Lett. 209, 103–112.

Brown, K.M., Bekins, B., Clennell, B., Dewhurst, D., Westbrook, G.,1994. Heterogeneous hydrofracture development and accretionaryfault dynamics. Geology 22 (3), 259–262.

Goldberg, D., Cheng, A., Gulick, S.P.S., Blanch, J., Byun, J., 2005.Velocity analysis of LWD sonic data in turbidites and hemipelagicsediments offshore Japan, ODP Sites 1173 and 808. In: Mikada,H., Moore, G.F., Taira, A., Becker, K., Moore, J.C., Klaus, A.(Eds.), Proc. ODP, Sci. Results, 190/196 [Online]. Available fromWorld Wide Web: <http://www-odb.tamu.edu/publications/190196SR/352/352.htm> [Cited 25 January 2006].

Hickman, S., Zoback, M., 2004. Stress orientations and magnitudesin the SAFOD pilot hole. Geophys. Res. Lett. 31, L15S12,doi:10.1029/2004GL020043.

Hyndman, R.D., Wang, K., Yamano, M., 1995. Thermal constraints onthe seismogenic portion of the southwestern Japan subductionthrust. J. Geophys. Res. 100, 15,373–15,392.

Karig, D.E., Angevine, C.L., 1986. Geologic constraints on subduc-tion rates in the Nankai Trough. Initial Rep. DSDP, Wash. (U.S.Govt. Print. Off.) 87, 789–796.

Lallemant, S.J., Byrne, T., Maltman, A., Karig, D., Henry, P.,1993. Stress tensor at the toe of the Nankai accretionaryprism: an application of inverse methods to slickenlined faults.In: Hill, I., Taira, T., et al. (Eds.), Proc. ODP, Sci. Results,vol. 131, pp. 103–122.

Leroueil, S., Vaughan, P.R., 1990. The general and congruent effects ofstructure in natural soils and weak rocks. Geotechnique 40 (3),467–488.

Maltman, A.J., Byrne, T., Karig, D., Lallemant, S., 1993. Deforma-tion at the toe of an active accretionary prism: synopsis of resultsfrom ODP Leg 131, Nankai, SW Japan. J. Struct. Geol. 15,949–964.

220 M. Ienaga et al. / Tectonophysics 426 (2006) 207–220

McNeill, L.C., Ienaga, M., Tobin, H., Saito, S., Goldberg, D.,Moore, J.C., Mikada, H., 2004. Deformation and in situ Stress inthe Nankai Accretionary Prism from Resistivity-At-Bit Images,ODP Leg 196. Geophys. Res. Lett. 31, L02602, doi:10.1029/2003GL018799.

Mikada, H., Becker, K., Moore, J.C., Klaus, A., Austin, G.L., Bangs,N.L., Bourlange, S., Broilliard, J., Brückmann, W., Corn, E.R.,Davis, E.E., Flemings, P.B., Goldberg, D.B., Gulick, S.S., Hansen,M.B., Hayward, N., Hills, D.J., Hunze, S., Ienaga, M., Ishiguro,H., Kinoshita, M., Macdonald, R.D., McNeill, L., Obana, S.,Hong, O.S., Peacock, S., Pettigrew, T.L., Saito, S., Sawa, T.,Thaiprasert, N., Tobin, H.J., Tsurumi, H., 2002. Proc. ODP, InitialRep., vol. 196. Ocean Drilling Program, College Station, TX.

Moore, G.F., Shipley, T.H., 1993. Character of the décollement in theLeg 131 area, Nankai Trough. Proc. ODP, Sci. Results, vol. 131.Ocean Drilling Program, College Station, TX, pp. 73–78.

Moore, G.F., Shipley, T.H., Stoffa, P.L., Karig, D.E., Taira, A.,Kuramoto, S., Tokuyama, H., Suyehiro, K., 1990. Structure of theNankai Trough accretionary zone from multichannel seismicreflection data. J. Geophys. Res. 95, 8753–8765.

Moore, G.F., Taira, A., Klaus, A., Bangs, N.L., Kuramoto, S., Shipley,T.H., Alex, C.M., Gulick, S.S., Hills, D.J., Ike, T., Ito, S., Leslie, S.C., McCutcheon, A.J., Mochizuki, K., Morita, S., Nakamura, Y.,Park, J.O., Taylor, B.L., Toyama, G., Yagi, H., Zhao, Z., 2001a.Proc. ODP, Initial Rep., vol. 190. [CD-ROM]. Available from:Ocean Drilling Program, Texas A&M University, College StationTX 77845-9547, USA.

Moore, G.F., Taira, A., Klaus, A., Bangs, N.L., Kuramoto, S., Shipley,T.H., Alex, C.M., Gulick, S.S., Hills, D.J., Ike, T., Ito, S., Leslie, S.C., McCutcheon, A.J., Mochizuki, K., Morita, S., Nakamura, Y.,Park, J.O., Taylor, B.L., Toyama, G., Yagi, H., Zhao, Z., 2001b.New insights into deformation and fluid flow processes in theNankai Trough accretionary prism: results of Ocean DrillingProgram Leg 190. Geochem. Geophys. Geosyst. 2, doi:10.1029/2001GC000166.

Moore, G.F., Mikada, H., Moore, J.C., Becker, K., Taira, A., 2005.Legs 190 and 196 synthesis: deformation and fluid flow processesn the Nankai Trough accretionary prism. In: Mikada, H., Moore, G.F., Taira, A., Becker, K., Moore, J.C., Klaus, A. (Eds.), Proc. ODP,Sci. Results, vol. 190/196. [Online]. Available from World WideWeb: ⟨http://www-odp.tamu.edu/publications/190196SR/synth/synth.html⟩. [Cited 18 May 2005].

Moos, D., Zoback, M.D., 1990. Utilization of observations of wellbore failure to constrain the orientation and magnitude of crustalstresses: application to continental, Deep Sea Drilling Project, andOcean Drilling Program boreholes. J. Geophys. Res. 95,9305–9325.

Okino, K., Ohara, Y., Kasuga, S., Kato, Y., 1999. The Philippine Sea:new survey results reveal the structure and history of marginalbasins. Geophys. Res. Lett. 26, 2287–2290.

Peška, P., Zoback, M.D., 1995. Compressive and tensile failure of aninclined well bores and determination of in situ stress and rockstrength. J. Geophys. Res. 100 (B7), 12,791–12,811.

Screaton, E., Saffer, D., Henry, P., Hunze, S., Leg 190 ShipboardScientific Party, 2002. Porosity loss within the underthrustsediments of the Nankai accretionary complex; implications foroverpressures. Geology 30 (1), 19–22.

Seno, T., 1977. The instantaneous rotation vector of the Philippine SeaPlate relative to the Eurasian Plate. Tectonophysics 42, 209–224.

Shipboard Scientific Party, 2001. Site 1173. In: Moore, G.F., Taira, A.,Klaus, A., et al. (Eds.), Proc. ODP, Initial Rep., vol. 190, pp. 1–147.[CD-ROM]. Available from: Ocean Drilling Program, Texas A&MUniversity, College Station TX 77845-9547, USA.

Shipboard Scientific Party, 2002a. Site 1173. In: Mikada, H., Becker,K., Moore, J.C., Klaus, A., et al. (Eds.), Proc. ODP, Initial Rep.,vol. 196, pp. 1–97. [CD-ROM]. Available from: Ocean DrillingProgram, Texas A&M University, College Station TX 77845-9547, USA.

Shipboard Scientific Party, 2002b. Site 808. In: Mikada, H., Becker,K., Moore, J.C., Klaus, A., et al. (Eds.), Proc. ODP, Initial Rep.,vol. 196, pp. 1–68. [CD-ROM]. Available from: Ocean DrillingProgram, Texas A&M University, College Station TX 77845-9547, USA.

Taira, A., Kato, J., Tashiro, M., Okamura, M., Kodama, K., 1988. TheShimanto Belt in Shikoku, Japan: evolution of a Cretaceous toMiocene accretionary prism. Mod. Geol. 12, 5–46.

Taira, A., Hill, I., Firth, J.V., Ulrich, B., Brueckmann, W., Byrne, T.,Chabernaud, T., Fisher, A., Foucher, J.P., Gamo, T., Gieskes, J.M.,Hyndman, R.D., Karig, D.E., Kastner, M., Kato, Y., Lallemant, S.,Lu, R., Maltman, A.J., Moore, G.F., Moran, K., Olafsson, G.,Owens, W.H., Pickering, K.T., Siena, F., Taylor, E., Underwood,M.B., Wilkinson, C., Yamano, M., Zhang, J., Winkler, W.R., 1991.Proc. ODP, Initial Rep., College Station, TX (Ocean DrillingProgram), vol. 131. Texas A&M University, Ocean DrillingProgram, College Station, TX, USA.

Ujiie, K., Hisamitsu, T., Taira, A., 2003. Deformation and fluidpressure variation during initiation and evolution of the plateboundary decollement zone in the Nankai accretionary prism. J.Geophys. Res. 108 (B8), 2398, doi:10.1019/2002JB002314.

Yamada, Y., Baba, K., Matsouka, T., 2006. Analogue and numericalmodelling of accretionary prism with a decollement sediments. In:Buiter, S., Schreurs, G. (Eds.), “Numerical and AnalogueModelling of Crustal-Scale Processes”. Geological Society SpecialPublication, vol. 253, pp. 169–183.

Yoneshima, S., Endo, T., Pistre, V., Thompson, J., Campanac, P.,Mikada, H., Moore, J.C., Ienaga, M., Saito, S., the Leg 196Scientific Party, 2003. Processing leaky-compressional mode fromLWD sonic data in shallow ocean sediments: ODP sites in NankaiTrough. Proc. 6th SEGJ Int. Symp. on Imaging Technology,Society of Exploration Geophysicists of Japan, January 22–24,2003, Tokyo, Japan, pp. 45–52.

Zoback, M.D., Moos, D., Mastin, L., Anderson, R.N., 1985. Well borebreakouts and in situ stress. J. Geophys. Res. 90, 5523–5530.