Geomorphology, kinematic history, and earthquake behavior of...

Geomorphology, kinematic history, and earthquake behavior

of the active Kuwana wedge thrust anticline, central Japan

Tatsuya Ishiyama,1 Karl Mueller,2 Masami Togo,3 Atsumasa Okada,4 and Keiji Takemura5

Received 21 April 2003; revised 30 September 2004; accepted 14 October 2004; published 31 December 2004.

[1] We combine surface mapping of fault and fold scarps that deform late Quaternaryalluvial strata with interpretation of a high-resolution seismic reflection profile to developa kinematic model and determine fault slip rates for an active blind wedge thrust systemthat underlies Kuwana anticline in central Japan. Surface fold scarps on Kuwanaanticline are closely correlated with narrow fold limbs and angular hinges on the seismicprofile that suggest at least �1.3 km of fault slip completely consumed by folding in theupper 4 km of the crust. The close coincidence and kinematic link between foldedterraces and the underlying thrust geometry indicate that Kuwana anticline hasaccommodated slip at an average rate of 2.2 ± 0.5 mm/yr on a 27�, west dipping thrustfault since early-middle Pleistocene time. In contrast to classical fault bend folds the faultslip budget in the stacked wedge thrusts also indicates that (1) the fault tip propagatedupward at a low rate relative to the accrual of fault slip and (2) fault slip is partly absorbedby numerous bedding plane flexural-slip faults above the tips of wedge thrusts. An historicearthquake that occurred on the Kuwana blind thrust system possibly in A.D. 1586 isshown to have produced coseismic surface deformation above the doubly vergentwedge tip. Structural analyses of Kuwana anticline coupled with tectonic geomorphologyat 103–105 years timescales illustrate the significance of active folds as indicators of slipon underlying blind thrust faults and thus their otherwise inaccessible seismichazards. INDEX TERMS: 8005 Structural Geology: Folds and folding; 8020 Structural Geology:

Mechanics; 8107 Tectonophysics: Continental neotectonics; 9320 Information Related to Geographic Region:

Asia; 9604 Information Related to Geologic Time: Cenozoic; KEYWORDS: active fault, wedge thrust fold,

tectonic landform, coseismic folding, flexural-slip fault, central Japan

Citation: Ishiyama, T., K. Mueller, M. Togo, A. Okada, and K. Takemura (2004), Geomorphology, kinematic history, and

earthquake behavior of the active Kuwana wedge thrust anticline, central Japan, J. Geophys. Res., 109, B12408,

doi:10.1029/2003JB002547.

1. Introduction

[2] Honshu Island, Japan, is located in an intraplatetectonic environment where shortening above subductedoceanic crust is accommodated by fold-and-thrust belts thatbound, or are located within Neogene sedimentary basins[Research Group for Active Faults, 1991; Okada and Togo,2000; Ikeda et al., 2002; Nakata and Imaizumi, 2002].Following early work by Otsuka [1941] and Ikebe [1942],who related deformation of fluvial terraces to active growthof underlying folds in northeastern Japan, many other activefolds have been identified throughout the archipelago byanalysis of deformed landforms [cf. Nakamura and Ota,

1968; Nakata, 1976; Ikeda, 1983; Ikeda et al., 2002]. Otherstudies outside Japan also point to the significance of activefolds and their importance as indicators of slip on underly-ing blind thrusts [Philip and Meghraoui, 1983; Rockwell etal., 1984, 1988; Bullard and Lettis, 1993; Molnar et al.,1994; Burbank et al., 1996; Shaw and Suppe, 1996;Muellerand Suppe, 1997; Shaw and Shearer, 1999; Benedetti et al.,2000, 2003; Lave and Avouac, 2000; Champion et al.,2001; Shaw et al., 2002; Thompson et al., 2002; Dolan etal., 2003]. Tectonic landforms are thus widely recognizedfor their utility in providing finite strain markers andrecording growth of active folds. Understanding of theirstructural link with underlying thrusts and the geomorphicsignal of active folding caused by individual earthquakesare, however, still in its infancy [Kelson et al., 1996;Mueller et al., 1999; Champion et al., 2001; Hung andSuppe, 2002] and remains a focus for future work.[3] The geometry of fault-related folds has been linked to

underlying thrusts since the early work of Rich [1934].Although numerous models of fault-related folds havesubsequently been developed, end-member models arecommonly assigned to those that grow by fault bend[Suppe, 1983; Medwedeff, 1992; Medwedeff and Suppe,

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 109, B12408, doi:10.1029/2003JB002547, 2004

1Active Fault Research Center, Geological Survey of Japan, NationalInstitute of Advanced Industrial Science and Technology, Tsukuba, Japan.

2Department of Geological Sciences, University of Colorado, Boulder,Colorado, USA.

3Department of Geoscience, Hosei University, Tokyo, Japan.4Graduate School of Science, Kyoto University, Kyoto, Japan.5Institute for Geothermal Sciences, Kyoto University, Beppu, Japan.

Copyright 2004 by the American Geophysical Union.0148-0227/04/2003JB002547$09.00

B12408 1 of 20

1997] and trishear fault propagation mechanisms [Erslev,1991; Hardy and Ford, 1997; Allmendinger, 1998]. Thesemodels allow appropriate restoration of deformed stratabased on quantitative relationships between fault and foldgeometry and the amount of slip on fault planes, givenconservation of bed areas during deformation. Stratigraphicgrowth architecture also provides an independent and pow-erful means of constraining fold kinematics and their growthhistories [Suppe et al., 1992; Shaw and Suppe, 1994; Suppeet al., 1997; Allmendinger and Shaw, 2000]. Strategies forbuilding robust fault-related fold models [Suppe and Chang,1983; Woodward et al., 1989; Mount et al., 1990] are basedon well defined stratal geometry (dip and thickness) imagedin the seismic profiles, the strike and dip of strata exposed atthe ground surface and borehole data located along the crosssection. Axial surface mapping of active folds [Shaw et al.,1994; Shaw and Suppe, 1996; Novoa et al., 1998] alsoprovides a means of assessing their kinematic developmentand along-strike distribution of fault slip, based on theassumption that the underlying blind thrusts propagatelaterally as they accumulate slip [Mueller and Talling,1997].[4] We present new structural models, maps of young

deformed landforms and older strata, and a high-resolutionseismic profile acquired for this study to define the kine-matic evolution and geometry of Kuwana anticline alongthe Nobi-Ise fault zone (NIFZ), an active, intraplate, eastverging thrust belt that extends for 110 km in central Japan[Ikeda et al., 2002]. This paper is focused on the northernhalf of the NIFZ, where we constrain rates of slip onunderlying blind thrusts since early Quaternary time.Kuwana anticline has formed at the leading edge of a largersystem of active folds and extends northwest-southwest for20 km along the western margin of the Nobi and Ise plains.

The central portion of Kuwana anticline is ideal for struc-tural analysis in that its subsurface fold geometry is tightlyconstrained by a high-resolution seismic reflection profileacquired for this study [Ishiyama, 2002]. In addition,surface folding is recorded by well preserved fluvial terracesover the center of Kuwana anticline, providing an opportu-nity to document deformation mechanisms that have actedto build it since late Pleistocene time.[5] In our view, Kuwana anticline is perhaps most sig-

nificant because it appears to have accommodated shorten-ing during a historic blind thrust earthquake in A.D. 1586[Sugai et al., 1998], which is expressed as a narrow kinkband in late Holocene sediments in the Nobi plain. Thisstructure thus joins a small group of fault bend folds knownto have grown by coseismic folding during blind thrustearthquakes, suggesting an important link between other-wise hidden faults and their coseismic expression on Earth’ssurface.

2. Regional Setting of Kuwana Anticline

[6] Kuwana anticline lies at the eastern end of an array ofactive folds within the NIFZ that defines the eastern edge ofthe Kinki region of central Japan (Figure 1). The Kinkiregion is located between a Mesozoic accretionary wedge incentral and southwestern Japan and high-grade metamor-phic rocks bounded by the Median Tectonic Line (MTL) tothe south. The region is characterized by an array of northtrending thrust sheets, separated by Neogene basins [Yeatset al., 1997]. Shortening within the Kinki region appears toaccommodate right-lateral displacement on the MTL north-ward, toward the Japan Sea.[7] The northern NIFZ is defined by east vergent, thick-

skinned thrusts, based on balanced cross sections and seismic

Ise Bay

Japan Sea

Osaka Bay

Inactive MTLInactive MTLInactive MTL

Active MTLActive MTLActive MTL

Figure 2(a)Figure 2(a)

NIFZNIFZNIFZ

NIFZNIFZ

LakeLakeBiwaBiwaLakeBiwa

136˚135˚E

36˚N

35˚

34˚

137˚

50 km

KT

NG

(b)

KT

NG

OSOSOS

Active strike-slip faults

Active thrust faults,barbs on hangingwall sides

Lake

Major cities

Active faultsapproximately located

Epicenter of the 1995 Hyogo-ken Nanbu (Kobe) earthquake (MJMA7.2)

Legend

Japan Sea

EU

NA

PHS

PA

IST

L

MTL

Honshu Island

(b)Ja

pan

Tren

ch

Tokyo

Nankai Trough

130˚E 140˚35

˚45

˚N

(a)

Figure 1. (a) Regional tectonic setting of Japan. Solid lines with barbs indicate plate boundaries. Dashedlines indicate major geologic boundaries within the Japan island arc. Abbreviations are EU, Eurasian Plate;NA, North American Plate; PA, Pacific Plate; PHS, Philippine Sea Plate; ISTL, Itoigawa-ShizuokaTectonic Line; MTL, Median Tectonic Line. (b) Shaded relief map based on Geographical Survey Institute250-m digital elevation model (DEM) showing the topography and distribution of active faults within theKinki region, central Japan. Locations of active faults are from Nakata and Imaizumi [2002] and Ikeda etal. [2002]. Abbreviations are OS, Osaka; KT, Kyoto; NG, Nagoya; NIFZ, Nobi-Ise fault zone.

B12408 ISHIYAMA ET AL.: ACTIVE WEDGE THRUST FOLD IN JAPAN

2 of 20

B12408

reflection profiles [Ishiyama et al., 2002a] (Figure 2). TheSuzuka and Yoro Mountains form the stripped cores of thetwo largest structures in the NIFZ and expose Triassic-Jurassic strata that are thrust eastward over a 1.5-km-thicksequence of Pliocene-Pleistocene strata deposited in the Nobibasin. The Suzuka Mountains are bounded by an emergentthrust along its eastern flank, the Fumotomura fault, while theYoro Mountains are underlain by a blind thrust that isexpressed as fold scarps along its eastern flank [Ishiyama etal., 1999; Togo, 2000; Ishiyama et al., 2002b].[8] Kuwana anticline extends from the southern termina-

tion of the central core of the Yoro Mountains (i.e., Igaianticline) to the southwest for �20 km (Figure 3). Kuwanaanticline is an east vergent, southward plunging anticlinedefined by a steeply east dipping forelimb and a moregently dipping and broader back limb. The crest of the foldvaries from a sharp anticlinal hinge at its northern end to abroad, gently dipping plunge panel at its southern end.Topography across the fold coincides closely with a decreasein structural relief from north to south, based on stratigraphicunits exposed in the core of the fold. Kuwana anticlineinterferes with two active folds (Igai and Yokkaichi anti-clines) at its northern and southern ends where they aremarked by active synclines.

3. Stratigraphic Setting of the Nobi Basin andKuwana Anticline

3.1. Pregrowth Stratigraphy

[9] We used stratigraphy characterized in four 1500-m-deep, continuously cored boreholes in the Nobi basin

[Sakamoto et al., 1986; Yoshida et al., 1991] (seelocations of boreholes in Figure 3) to define the agesof reflectors in the seismic profile [Ishiyama, 2002]. Theoldest Neogene strata in the Nobi basin include �400 mof Miocene sandstone and mudstone [Sakamoto et al.,1986; Yoshida et al., 1991] (Figure 4). Correlation ofthese strata with reflectors in the seismic profile suggeststhat these sediments underlie the Kuwana Hills, thoughthey have not been directly identified by deep drillholes.[10] Pregrowth strata of Neogene age folded by

Kuwana anticline include fluvial and lacustrine sedimentsof the late Pliocene–early Pleistocene Tokai Group[Takemura, 1985; Yoshida et al., 1991]. The Tokai Groupin the Nobi basin is �1500–2000 m thick [Committeefor Subsurface Structure of the Nobi Plain, 2001],whereas in the Kuwana Hills these strata are more than1200 m thick based on geological mapping [Yoshida,1988]. Dating of volcanic ash layers by the fission trackmethod indicates that the Tokai Group was depositedfrom �2.8 Ma to 1.0 Ma [Takemura, 1985; Yoshida etal., 1991].[11] The Tokai Group overlies Miocene strata in the Nobi

basin boreholes and is also well exposed in the core ofKuwana anticline. Seismic reflection data presented later inthis paper suggest that reflectors correlated with Miocene toPlio-Pleistocene sandstone and mudstone beneath theKuwana Hills were deposited prior to uplift of the fold.Evidence for the pregrowth nature of these strata includesthe continuity of correlative reflectors and the consistentthickness of units bounded by them across the anticline.

Figure 2. (a) Shaded relief map based on a 6-m DEM showing the locations of active thrusts in thenorthern portion of the NIFZ. Barbs indicate hanging wall sides of active thrusts. (b) Generalized crosssection across the northern NIFZ [Ishiyama et al., 2002a]. Kuwana anticline comprises the leading edgeof a larger system of active folds that deform a 1.5-km-thick sequence of Pliocene-Pleistocene stratadeposited in the Nobi basin. Abbreviations are Tt, Tokai Group; Tk, Pliocene-Pleistocene KobiwakoGroup; QTc, Chikarao Formation; QTo, Owari Group.


3 of 20

B12408

Figure 3. Geologic map of Kuwana anticline with locations of active fault and fold scarps, modifiedfrom Yoshida [1984] and Yoshida et al. [1991] (with permission from the Geological Survey of Japan).Active folds comprise structurally and topographically elevated domains, defined by uplifted lateQuaternary fluvial terraces that record structural growth since their abandonment. The location of faultand fold scarps coincides with boundaries of dip domains defined by outcrops and subsurface data.Location of seismic section [Ishiyama, 2002] with CDP (common depth point) numbers is also shown.Squares in the inset indicate locations of boreholes to depths of 1.5 km. Solid squares are those used tocorrelate strata with reflectors in the seismic section. Abbreviations for borehole locations are MT,Matsukage [Yoshida et al., 1991]; YT, Yatomi [Furusawa, 1990].


4 of 20

B12408

3.2. Growth Stratigraphy

[12] The youngest sediments in the Nobi basin include300 to 400 m of late Pleistocene–Holocene deposits thatrecord strain across Kuwana anticline (Figure 4). Growthstrata that are folded by and record growth of the anticlineare composed of coarse-grained gravel, sand and mud of theearly Quaternary Chikarao Formation [Yoshida et al., 1991]and the middle-late Quaternary Owari Group [Sakamoto etal., 1986; Yoshida et al., 1991]. Recent analysis of coresbased on lithostratigraphy, magnetostratigraphy, and teph-rostratigraphy suggests that the Owari Group has beendeposited during middle-late Quaternary and contains avolcanic ash layer dated at �0.9 Ma in its lowest part

[Sugai et al., 1999b]. A 50-m-thick sequence of Holocenestrata (Nanyo Formation) overlies the Owari Group beneaththe Nobi basin. The uppermost section of the NanyoFormation overlies a marine abrasion platform (i.e., amarine terrace) along the eastern edge of the forelimb ofKuwana anticline, and records its growth history during thelate Holocene.

3.3. Geomorphology

[13] Late Quaternary terrace deposits above strath levelsincised into transverse channels record growth of Kuwanaanticline and provide an important means of assessing itskinematic history. Mapping of these fluvial deposits wasundertaken (Figure 5) to locate active fold hinges and faultsat the surface and to define fold growth at intermediate(105 years) to short (103 years) timescales. Maps werecreated with 1:10,000 and 1:40,000 scale aerial photo-graphs taken in 1940s, prior to modern urbanization andagriculture. Modern 1:25,000 scale quadrangles with 10 mcontour intervals from the Geographical Survey Institutewere also used to create a topographic base map.[14] Mapping of fluvial terrace risers suggest that a

flight of terrace surfaces (labeled as H, M, and L) arepreserved in modern river valleys, where they are sepa-rated by an elevation difference of �20 m (Figure 5).Terraces are further separated from two or three smaller,locally spread terraces that are separated by risers �5–10 m high. Terrace deposits above strath surfaces are verythin (2–5 m) and are underlain by units of the TokaiGroup. Three groups of terraces are also differentiated bytheir lithofacies. H terraces stand at the highest elevationamong other terraces and are highly dissected by densedrainage networks. Deposits of H terraces are composedof strongly weathered gravels of chert and sandstoneclasts with thin weathering rinds that are supported bya matrix of reddish brown mud. M terraces are exten-sively widespread across the fold and are less dissectedthan H terraces. Deposits of M terraces are less weatheredand contain a matrix of coarse sand grains. In contrast,L terraces are preserved along the modern transversechannels of the Inabe and Asake River. Deposits thatcompose L terrace consist of gravels whose clasts andmatrix are unweathered.[15] Subangular gravels contained in terrace deposits

along the Inabe and Asake Rivers indicate that they weretransported shorter distances (�40 km) from the SuzukaMountains by these stream channels, rather than the Ibi andNagara Rivers where more rounded clasts have been trans-ported greater distances (�110 km) from mountains north ofthe Nobi plain.[16] The age of individual late Quaternary terraces is only

partly known. The lowermost portion of a sequence of M2terrace deposits covers the eastern edge of the east dippingforelimb and is composed of thin inner bay clays, which arecorrelated with strata of the lower Atsuta Formation (see theoutcrop location in Figure 5) [Yoshida et al., 1991]. Datingof volcanic ashes by the fission track and K-Ar methodindicates that the lower Atsuta Formation was depositedfrom �130 to 110 ka [Makinouchi et al., 2001]. The claysthin westward and onlap the forelimb �500 m from its basefarther to the east, where they lie at �29 m above sea level.The clays are overlain by 2- to 3-m-thick fluvial gravel

Figure 4. Stratigraphic column of the Nobi basin withapproximate ages of deposits based on tephrochronologyand dating by the fission track method [Yoshida et al., 1991;Sugai et al., 1999b]. Unnamed Miocene? strata arerecognized only by borehole data in the Nobi basin andare not exposed within Kuwana anticline.


5 of 20

B12408

Figure

5.

Detailedtopographic

map

ofKuwanaanticlineshowingthedistributionoffluvialterraces

andlocationsof

activefaultandfold

scarps.

Contourintervalsare50m.Structuraltrends(hingelines)ofKuwana,

Yokkaichi,andIgai

anticlines

aredepicted.Projectionlines

oflongitudinal

topographic

profilesin

Figure

6arealso

shown.Anopen

star

indicates

thelocationofboreholeofMakinouchietal.[2001].Solidstar

indicated

theoutcroplocationofthelower

Atsuta

Form

ation.The‘‘u’’and‘‘d’’denote

upthrownanddownthrownsides

ofactivefaults,respectively.


6 of 20

B12408

deposited above strath surfaces. These gravel depositscontain the On-Pm1 volcanic tephra [Awata, 1997], datedat �100 ka [Machida and Arai, 2003]. Radiocarbon anal-yses of charcoal contained in L1 terrace deposits indicatethat they were deposited at �18 ka [Kimura et al., 1984].

4. Overview of Tectonic Geomorphology ofKuwana Anticline

4.1. Tectonic Landforms

[17] Mapping of fold scarps that deform fluvial terraceswas undertaken to differentiate kinematic models and todetermine slip rates on the underlying blind thrust. Discon-tinuous preservation of terraces does not allow correlationacross the entire width of the fold. However, longitudinalprofiles of individual terraces based on 1:2500 scale topo-graphic maps with 2 m contour intervals define foldingstyles at several sites since their abandonment (Figure 6).The geometry of intermediate and higher terraces clearlymimics that of the underlying anticline, with a steeply eastdipping forelimb and a broader gently dipping back limbinclined to the west. Steeply east dipping forelimbs werealso identified as scarps on lower, younger terraces pre-served along drainage channel walls incised into the interiorof Kuwana anticline. At the middle reach of the incisedchannel in the Inabe River (5–6 km in Figure 6a and 7–9 km in Figure 6b), a gently east dipping dip panel ispreserved in higher terraces (H3 and H2) above the crest ofthe back limb.[18] While the overall character of tectonic landforms

across Kuwana anticline illuminates the styles of foldingthat act to build it, more detailed description of landformsallows the subsurface geometry of the Kuwana blind thrust

to be further constrained. Sections 4.2–4.5 therefore discusssmaller-scale structures and landforms on the forelimb, crest,and back limb of the fold.

4.2. Forelimb Geomorphology

[19] The detailed morphology of the forelimb ofKuwana anticline suggests that it is defined by a steeplyeast dipping, continuous surface near its base thatprogressively flattens upward toward its crest (see profileB-B0 on Figure 7). Photogrammetric mapping of the M2terrace folded across the forelimb of Kuwana anticline[Suzuki et al., 2002] indicates that it is offset by a seriesof 2–20 m high, north trending fault and/or fold scarps[cf. Ikeda et al., 2002] (Figure 7). Among these small-scale structures, four west facing scarps are preserved thatlie subparallel to each other and the strike of underlyingNeogene sediments, while the westernmost one dipseastward. On the basis of a fault contact exposed beneathone of the west facing scarps [Awata, 1997] we interpretthe west facing scarps as being formed by west vergentflexural-slip faults that act to accommodate bending ofthe forelimb.

4.3. Crest Geomorphology

[20] Fluvial terraces presented across the flat crest ofKuwana anticline also constrain the location of surfacefolding that is related to subsurface structure. Both H3and M2 terraces on the crest of Kuwana anticline aredeformed by small fold scarps (Figure 6a). These includean abrupt, steeply west dipping fold scarp that deforms theH3 terrace (see distance 4.5–5.0 km on Figure 6a) and abroader arch that deforms the M2 terrace farther east (seedistance 4.0 km on Figure 6a and Figure 7). A sharp, east

Figure 6. Longitudinal profiles of terrace remnants along (a) the Inabe River and (b) the Asake River.Approximate locations of projection lines used in constructing these profiles are shown in Figure 5. Solidlines indicate continuously mapped terraces, while dashed lines indicate terraces correlated based onsimilar elevations and lithofacies of terrace deposits.


7 of 20

B12408

facing scarp that separates the H3 and M2 terraces isinterpreted as an M2 terrace riser, because the scarp incisesolder deposits that do not appear to be folded or otherwiseoffset (see distance 4.2 km on Figure 6a and profile A-A0 onFigure 7).

4.4. Back Limb Geomorphology

[21] Terraces folded across the back limb also offerinsight into the kinematics of the anticline and the sliphistory on the underlying blind thrust. Longitudinal topo-graphic profiles indicate that pairs of fluvial terraces on the

Figure 7. Detailed topography of the forelimb of Kuwana anticline. Topographic map with 2-m contourinterval is redrawn from Suzuki et al. [2002]. Pairs of west verging fault and fold scarps deform theforelimb terrace (M2) that is inclined to the east. East verging fold scarp on the crest of Kuwana anticlineis also identified. Solid lines indicate continuous terrace treads. The ‘‘u’’ and ‘‘d’’ denote upthrown anddownthrown sides of active faults, respectively.


8 of 20

B12408

Figure 8


9 of 20

B12408

back limb of the anticline define broad dip domains that dipwest, or upstream. The strike of the west dipping H2 andM2’ terraces on the back limb of Kuwana anticline variesslightly, but is parallel to the sinuous leading edge of thefold in the Nobi basin. The morphology of the west dippingM2’ terrace on the back limb of Kuwana anticline is alsodefined by one, and perhaps two west facing kink bands(see Figure 6b) that deform its otherwise gently westdipping surface.[22] Local deposition of more than 10 m of fluvial sedi-

ments across the west facing fold scarp located at the baseof the gently west dipping M2’ terrace [Ota and Sangawa,1984] suggests that this fold scarp has been modified fromits original geometry (i.e., it was previously less subdued).The 10-m-thick section of terrace deposits in the synclinalaxis contrasts with a thin (1–3 m) veneer of gravels thatcommonly overlie the strath surfaces of M2’ and M1terraces in both the west and east dipping domains.

4.5. Holocene Fold Scarps

[23] Other landforms preserved at the base of the forelimbof Kuwana anticline include small fold scarps that deformlate Holocene sediments and are apparent on 1:10,000 aerialphotographs and 1:2500 scale topographic maps [Awata andYoshida, 1991; Togo, 2000; Ikeda et al., 2002] (Figures 5and 7). The map view geometry and cross-sectional mor-phology of these fold scarps indicate that they lie along theforelimb of Kuwana anticline. The scarps strike parallel tounderlying bedding in the forelimb (Figure 3), similar to anumber of active folds along the leading edge of the NIFZ,including the Yoro and Yokkaichi anticlines.[24] Borehole transects across the fold scarps on the

forelimb of Kuwana anticline indicate 10–15 m of struc-tural relief across middle-late Holocene inner bay strataburied beneath the surface of the M2 terrace [Sugai et al.,1998]. More detailed geologic sections across several scarpson the Yoro and Kuwana structure based on shallow coringsindicate that many of these fold scarps may have partlyformed during a well-recorded historical earthquake (Ten-syo earthquake in A.D. 1586: M � 7.7 [Usami, 2003])[Sugai et al., 1998, 1999a; Togo, 2000; Ishiyama et al.,2002b].

5. Subsurface Structure of Kuwana Anticline

5.1. Subsurface Structure

[25] A high-resolution seismic reflection profile wasacquired as a part of this study across Kuwana anticline toimage its subsurface geometry [Ishiyama, 2002] (Figure 8a).Reflectors imaged on the seismic profile were tied with

strata defined in boreholes in the Nobi basin [Yoshida et al.,1991] (see borehole locations in the inset of Figure 3),stratigraphic contacts on the geologic maps, and volcanictephra exposed on both the forelimb and back limb ofKuwana anticline. We first made line drawings of reflectorsbased on the original depth section, which was orientedobliquely to the mean transport direction inferred fromthe fold axis. We then subdivided the seismic section into10 segments and projected reflectors on these segmentsonto a cross section perpendicular to the mean fold axisoriented at N64�W (Figure 8b).[26] Figure 8b shows the correlation of prominent reflec-

tors across the anticline, and axial surfaces and fault cutoffsthat were used to interpret the projected seismic section.Regional (i.e., original) dip is best defined in the syncline2 km from the southeast edge of the section (see right sideof Figure 8b) where strata are nearly horizontal. Theborehole used to correlate reflectors on the seismic sectionwith stratigraphic contacts of known age [e.g., Yoshida etal., 1991] is located 4 km east of the right edge of thesection. Very strong (i.e., high-amplitude) and continuousreflectors lie 1500 m below the ground surface, in thesyncline 2 km from the southeast end of the section.These prominent reflectors are interpreted as the Pliocene-Pleistocene Tokai Group, Chikarao Formation, and OwariGroup beneath the Nobi basin. The thickness of Pliocene-Pleistocene strata imaged in the seismic section correlateswell with an existing seismic profile in the southeast Nobibasin [Faculty of Science, Kyoto University et al., 1996].Subhorizontal, more discontinuous reflectors with low fre-quencies below the continuous reflectors are visible at depthof 1500–1900 m in the seismic section and correspondto Miocene strata below the Tokai Group that are alsoidentified in the nearby drill hole [Yoshida et al., 1991].[27] Between common depth point (CDP) 270 and 1050

on the projected section, high-amplitude, continuous reflec-tors are present to depths of 1200 m. These are underlain bydiscontinuous, subparallel reflectors with low frequenciesbarely visible on the profile. Geologic mapping of the backlimb between CDP 650 and 1050 [Yoshida et al., 1991]indicates that the reflectors above 1200 m depth correspondto fine-grained mudstone and sandstone of the Tokai Group.Discontinuous reflectors with low frequencies evidentbelow the strong reflectors of the Tokai Group are similarto Miocene strata imaged farther east in the Nobi basin.Although they have not been directly identified by deepdrill holes, we argue that Miocene strata may underlie theTokai Group in the core of the fold.[28] On the basis of the geometry of its forelimb and back

limb, Kuwana anticline is an asymmetric fault bend fold

Figure 8. (a) Depth-migrated seismic profile across Kuwana anticline acquired for this study [Ishiyama, 2002]. Locationis shown on Figures 3, 5, and 12. There is no vertical exaggeration. Parameters used for acquisition of seismic reflectiondata are shown in inset. (b) Structural features recognized in line drawings based on the projected depth section acrossKuwana anticline used to estimate fault trajectories of the Kuwana thrust in Figure 8c. White lines are intersections of theprojection line with the seismic line. Abbreviations are A, base of upper-middle Pleistocene Owari Group; B, base of lowerPleistocene-Pliocene Tokai Group; Kr, volocanic ash (Karegawa) identified in the borehole YT 7 km NE of the profile[Furusawa, 1990] and within the core of the fold [Yoshida et al., 1991]; FW, footwall. Horizons A and B are identified inthe borehole MK 4 km west of the profile [Yoshida et al., 1991]. (c) Interpreted, projected depth section constrained by faultbend fold theory [Suppe, 1983]. Values in parentheses are apparent limb widths measured on the section and are comparedwith estimates of theoretical fault slip in the text.


10 of 20

B12408

(Figures 8b and 8c). The back limb is significantly widerthan the forelimb (1.9 km compared to 1.1 km) and has amuch shallower dip (10� Compared to 37�). The steeper dipof the forelimb is consistent with growth of a fault bend foldabove an underlying east vergent thrust. Subhorizontalreflectors (above the thick lines in Figure 8b) in the synclineto the southeast are continuous with reflectors in thesoutheast forelimb, confirmed by the dip and depth of theKaregawa volcanic ash identified both in the forelimb[Yoshida et al., 1991] and in the Nobi basin drill hole[Furusawa, 1990] (Figures 8b and 8c). This defines adistinctive, steeply dipping synclinal axial surface thatbisects the interlimb angles, rather than a reverse fault thatoffsets strata at the base of the forelimb. Reflectors in theforelimb are parallel to each other. Although the forelimb inthe lower part of the structure is poorly imaged, continuoushigh-amplitude reflectors in the syncline, crest and backlimb suggest that the forelimb strata are contained in paralleldip panels.

[29] Two small anticlines are present along the top of theforelimb and crest of Kuwana anticline. Seismic reflectorsbetween CDP 300 and 500 indicate the presence of footwallcutoffs and west vergent secondary folds within the fore-limb. A small but distinctive anticline is also present on thecrest of the structure between CDP 550 and 640. Footwallcutoffs apparent between CDP 550 and 640 on the seismicprofile lie beneath the small anticline at 400 to 700 m depthin Tokai Group strata. Reflectors in the hanging wall of thethrust marked by the footwall cutoffs dip steeply westward.We interpret these features as defining a high-level, eastvergent blind thrust. Continuous and prominent reflectorsbelow the stratigraphic footwall cutoffs indicate that thethrust ramp flattens between horizontal reflectors at depth.[30] Moderately dipping, parallel dip panels separated by

narrow axial surfaces are indicative of simple fault bendfolds above thrust ramps [Suppe, 1983] or wedge thruststructures [Medwedeff, 1992] and imply growth by themechanism of kink band migration [Suppe et al., 1992].Wedge thrust structures are typical of frontal structures inthrust belts [cf. Morley, 1986; Couzens and Wiltschko,1996]. A key observation by which we differentiate thesetwo models is that the late Holocene fold scarp is coincidentwith the synclinal axial surface at the base of the forelimb,suggesting that it is an active axial surface. As discussed byMedwedeff [1992], active fold growth by kink band migra-tion is expected to occur at an active synclinal axial surfaceabove a wedge tip at the ground surface, whereas in a faultbend fold the synclinal axial surface at the base of theforelimb is expected to be inactive [Suppe, 1983]. Weattribute the lack of obvious growth architecture above thefold limb to a low rate of sedimentation (0.3 mm/yr) relativeto uplift, the recency of the onset of growth of this anticline,and the resolution of the seismic data. However, thecoincidence of the upward projection of the synclinal axialsurface with narrow fold scarps in Holocene depositssuggests active fold growth by kink band migration. Thelack of subsurface or surface evidence of a decollement thattransfers slip farther eastward also supports this hypothesis.Forelimbs defined by two dip domains and sharply angularinterlimb angles are also inconsistent with other mecha-nisms of forelimb growth such as trishear fault propagationfolding [Erslev, 1991; Hardy and Ford, 1997; Allmendinger,1998]. We therefore interpret Kuwana anticline as formingas an east vergent wedge thrust structure, with a higher-levelwest vergent wedge tip.

5.2. Forward Balanced Model Based on Wedge ThrustTheory

[31] Medwedeff [1992] presented methods for building abalanced forward model for wedge thrust structures, basedon fault bend fold theory [Suppe, 1983]. Figure 9 shows asequential series of balanced forward kinematic models forthe development of the wedge geometry based on quanti-tative application of fault bend fold theory. In these models,multibend fault bend folding [Medwedeff and Suppe, 1997]and an initially flat-lying decollement transfer slip upwardinto a wedge tip. The fold grows by increasing insertion ofthe wedge which lengthens the forelimb across active axialsurfaces that deform late Quaternary deposits at the groundsurface. Dip domains across the fold include a wide, steeplyeast dipping limb at deep levels and a more gently dipping

Figure 9. Sequential series of forward balanced models ofa wedge thrust fold based on fault bend fold theory. As theamount of slip increases through Figure 9a to 9c, the widthof lower dip domain lengthens, whereas the width of upperdomain does not change.


11 of 20

B12408

limb confined to a shallower level. The geometry of theforelimb of this balanced forward wedge model at the finalstage is similar to that observed in the seismic section acrossKuwana anticline (Figure 8c).

5.3. Wedge Thrust Solution for Kuwana Anticline

[32] Kuwana anticline has likely formed as a wedge thrustfold above a blind thrust with several gentle bends(Figure 8c). Back stepping of slip at 1400 m depth at the

3636˚

2727

37

˚

Figure 10


12 of 20

B12408

lowest portion of the Tokai Group then created an eastvergent wedge tip. Slip transferred eastward onto thelowermost Tokai Group may be facilitated by thinly inter-bedded tuffaceous sandstone and mudstone. The apparentforelimb dips 37�, similar to the dip of the roof thrust abovethe east vergent wedge. The model predicts that active axialsurfaces defining the dip panels are pinned to anticlinalbends of the top of the lowest thrust ramp (Figures 8c and9). Dips of strata in the hanging wall (37� on the lower dipdomain and 16� on the upper dip domain) and the axialsurfaces (72� and 82�) indicate that the lowest thrust rampdips 36� and flattens to 27� at higher levels across ananticlinal bend in the fault. Uncertainty derived from localvariation in the strike of strata in the fold seems minimal inour analysis. For example, in the extreme case where thefold axis deviates 10� from the strike of bedding, apparentfault dip on the lower ramp decreases only 0.4� from the 36�value determined by our analysis. Thus we regard uncer-tainty in ramp dips as negligible. The geometry of the backlimb of Kuwana anticline, in particular the steeply westdipping panel near its crest, requires further modification ofthe forward model shown in Figure 9. Hindward slip on theupper decollement is partially transferred upward along anupper ramp. In addition, fault bend folding above the wedgetip is partially accommodated by west vergent bedding-parallel thrusts or flexural-slip faults within the forelimb.[33] We measured the fault slip budget upward through

Kuwana anticline to test whether it grows as a wedge thruststructure, or whether other mechanisms occur that may actto control its growth (Figure 8c). We first compare the widthof the back limb to the calculated values of fault slip. On thebasis of a fault dip on the lower ramp of 36� and 750 m ofvertical structural relief on the top of the Kono Formationmeasured across the fold from geologic mapping andborehole data in the syncline, 1280 m of fault slip hasapparently been accommodated by the deeper levels of thestructure. The theoretical fault slip of 1280 m is smaller thanthe apparent width of the back limb (1850 m) between CDP750 and 1020 (Figure 8c). We interpret this difference topartly result because the ramp at depth is oblique to themean fold axis. The trend of frontal active thrusts in theNIFZ and geologic mapping (Figure 3) indicates that localvariation in the trend of the mean fold axis near theprojection line (N26�E) may deviate from the regionalE-W transport direction assumed for the fold. For this case,slip above the ramp is defined by

S ¼ U= sin q sin j� joð Þ

where S, U, q, j, and j0 are fault slip on the ramp, verticalstructural uplift, dip of the ramp, and the azimuth of thelocal mean fold axis and of the regional mean direction oftransport, respectively [Apotria et al., 1992; Lave andAvouac, 2000]. Assuming that U = 750 m, q = 36�, j =296�, and j0 = 90�, fault slip on the ramp is calculated atthis location to be �1420 m, which is closer to the apparentwidth of the back limb than our previous estimate. Wetherefore suggest that at least 1280 m of fault slip has beencompletely consumed by folding in the upper 4 km of thecrust.[34] We also compare the width of the lower forelimb to

the calculated fault slips. The upward flattening ramptrajectories produce anticlinal axial surfaces across whichfault slip is consumed, based on fault bend fold theory[Suppe, 1983]. For the 9� lower and 27� upper anticlinalfault bends, we calculate a respective loss of about 160 and280 m of fault slip. Subtraction of these amounts from theslip on the lower ramp yields a value of 840 m that shouldtheoretically be fed into the lower wedge tip as it is insertedonto the lowermost Tokai Group strata. Given the rampoblique to the mean fold axis, 930 m of slip is alternativelycalculated to be fed into the lower wedge. These theoreticalvalues of fault slip from the insertion of the wedge aresimilar to the width of the steeper, lower dip domain in theforelimb of 950 m (Figure 8c). We therefore suggest that atleast 840 m of fault slip is fed into the lower wedge tip. Thisfault slip is much larger than that of the upper wedgemeasured from the width of the steeply dipping domain atthe top of the back limb (220 m).[35] The forward balanced wedge thrust model is able to

account for the bulk of the fault slip accommodated by theshallower levels of Kuwana anticline, further suggestingthat Kuwana anticline has formed as a doubly vergentwedge thrust fold (Figure 8c). Calculation of the fault slipbudget along the inferred thrust trajectory, however, sug-gests that slip on the deeper thrust may decrease upwardinto the highest levels of the structure. In section 6.2, wediscuss how this slip may be consumed by other processesthat are not apparent in our data set.

5.4. Implications of Tectonic Geomorphology for theForward Kinematic Model

[36] Comparison of tectonic geomorphology with thewedge thrust fold model presented above provides furtherkinematic implications for the mechanisms that work tobuild Kuwana anticline (Figures 10 and 11). Axial surfacesobserved on the forelimb, crest, and back limb are inter-preted as being pinned to bends in the underlying thrust. In

Figure 10. (a) Comparison of the projected seismic profile across Kuwana anticline with a topographic profile along theforelimb terraces. Two flexural-slip faults that crosscut the forelimb and an active axial surface are recognized in the seismicsection and correspond with the three west verging fault scarps on the M2 terrace. The synclinal active axial surfaceextending from the upper wedge tip coincides with the base of the west facing fold scarp, indicating that the west vergingupper wedge produces the tectonic landforms over the crest of Kuwana anticline. Location of the east verging fold scarpthat deforms M2 and H3 terrace also corresponds with the upward projection of the upper ramp. (b) Topographic profile ofthe forelimb terrace with the interpreted active faults underneath them. The profile is the same as A-A0 in Figure 7. Note thatthe apparent structural relief on the forelimb terrace is expressed as the vertical separation on the lower dip domain of theforelimb terrace. In this profile the location of the fold scarp related to the upper ramp defines the hindward limit of thelower dip domain. (c) Method of evaluating the amount of slip on a flexural-slip fault based on the height of a surface faultscarp on a forelimb terrace.


13 of 20

B12408

particular, frontal fold scarps at the base of the forelimb arecoincident with the projection of the synclinal active axialsurface extending upward from the propagating lowerwedge tip. The upper wedge tip is also closely coincidentwith other fold scarps that deform the H3 terrace on thecrest, and a west dipping upper ramp is apparent at theground surface as the east facing scarp on the M2 terracebetween CDP 520 and 570 (Figure 10). Secondary westvergent flexural-slip faults within the forelimb terminateupward into low-amplitude west facing fault scarps on thesurface that deform the M2 terraces preserved on theforelimb (Figure 10).[37] Axial surface mapping [e.g., Shaw et al., 1994] of

Kuwana anticline based on tectonic geomorphology indi-cates that it extends for 20 km along the western edge ofthe Nobi plain. Mapping of the synclinal active axialsurface marked by frontal Holocene fold scarps suggeststhat the wedge tip is largely parallel to axial surfacesidentified in the core of the fold from deformation offluvial terraces, outcrops of bedding and the seismicprofile (Figure 12).[38] The deepest geometry of the thrust ramp may be

constrained by axial surface mapping of the back limb ofKuwana anticline (Figures 8 and 11). Assuming that narrowwest facing kink bands that deform the M2’ terrace arepinned to synclinal axial surfaces that intersect the ground at

CDP 1020 and 2 km west of the western edge of the seismicprofile, the back limb forms above two synclinal bends inthe main thrust ramp.

6. Discussion

6.1. Dip-Slip Rate of the Kuwana Fault

[39] The kinematic solution for Kuwana anticline and agecontrols from folded terrace deposits allow slip rates to bedetermined for the blind thrust beneath the fold. For a sliprate calculated solely by the limb width measured on theseismic section, the sinuosity of the seismic line introduceserrors as previously discussed. We therefore use the amountof uplift on the M2 terrace deposits across the activesynclinal axial surface at the base of the forelimb, and thedip of the shallower portion of the Kuwana thrust. Aborehole located at the mouth of the Inabe River indicatesthat inner bay clays (lower Atsuta Formation) are buried at adepth of �77–98 m [Makinouchi et al., 2001] (see loca-tions in Figures 5 and 12). Correlative clays exposed withinthe forelimb below the M2 terrace gravels stand �21–29 mabove sea level. Summing these values thus yields 98–137 m of structural relief for the �130–110 ka depositsuplifted by slip on the lower ramp. Taking a dip of 27� forthe shallower portion of the thrust, we determined a dip-sliprate of 2.2 ± 0.5 mm/yr at intermediate (105 years) time-

Figure 11. Geologic cross section of Kuwana anticline constrained by the seismic section, geologicmapping, and tectonic geomorphology of its forelimb, crest, and back limb that are linked with bends onthe underlying fault trajectory. The close links between the forelimb and back limb geomorphology andthe fault geometry support the forward wedge thrust model for Kuwana anticline. In addition, fold scarpson the back limb may define the deeper geometry of the Kuwana thrust.


14 of 20

B12408

Figure

12.

Axialsurfacemap

ofthenorthernandcentral

portionofKuwanaanticline,

based

onreinterpretationof

topographicmap

shownin

Figure

5andgeologicmappingbyYoshidaetal.[1991].Inferred

tear

faultismarked

byoffsetof

activeaxialsurfaceandsteepwestdippingbacklimbabovetheupperwedgetip.In

contrast,lateHolocenefold

scarptied

totheeastvergentwedgetipistraced

continuouslyandstrikes

parallelto

underlyingbeddingin

theforelimb.


15 of 20

B12408

scales. Error in this rate is derived both from uncertainty inages of volcanic ashes in folded strata and correlation ofclay horizons defined in boreholes.[40] Correlated clay horizons used to calculate the slip

rate are located �3 km south of the seismic line across thecentral part of Kuwana anticline. Quantitative departurefrom the predicted maximum slip rate is difficult to estimatebecause of poor constraints on the lateral variation inslip along the fold. Structural relief across the fold alsodecreases southward, as discussed in section 5.3. Thisindicates that the slip on the underlying thrust decreaseslaterally to the south. Therefore our estimate of the slip rateobtained 3 km away from the center of the fold may be aminimum value, particularly if slip rate has varied along itsestablished length at intermediate timescales.[41] If the vertical rate of uplift (1.0 ± 0.2 mm/yr) at

intermediate timescales is constant for longer (106 years)duration, we argue that structural growth of Kuwana anti-cline began at 0.6–0.9 Ma, using 750 m of verticalseparation accommodated by the deeper levels of thestructure. This age estimation is close to that of the ChikaraoFormation and the lower part of the Owari Group (Figure 4and 11), which we interpret as growth sediments in thesyncline east of Kuwana anticline.

6.2. Implications for the Kinematic Evolution of aWedge Thrust Fold

[42] Fault slip decreases upward in Kuwana anticlineand cannot be completely accounted for at the highestlevels in the structure. Our analysis relies on fault bendfold theory that assumes that fault tips propagate veryrapidly in relation to the amount of slip. It is likely thatsome slip is consumed during the earliest stages of faulttip propagation if the complex fault trajectories were notachieved instantly across the two wedge tips. This infer-ence is supported by a trishear analysis [Allmendinger,1998] we undertook of the forelimb on the seismic section;the analysis provides a 35� dipping thrust solution, ap-proximately similar to that determined by fault bend foldtheory, although the angular dip panels of the forelimb,likely related to its subsequent growth as a fault bend fold,produced a poor fit based on chi-square statistics forforward models.[43] Our restoration suggests that the Kuwana thrust has

formed as a leading edge of a major thick-skinned, base-ment-involved structure, the Fumotomura thrust (Figure 2)exposed along the eastern flank of the Suzuka Mountains tothe west. A seismic profile and tectonic geomorphologyindicate that the Fumotomura thrust is an active, eastvergent, high-angle emergent thrust which has propagatedfrom the lower portion of the seismogenic crust [Ishiyama etal., 2002a]. A simple downward extrapolation of the faulttrajectories of the Kuwana thrust suggests that it soles intothe Fumotomura thrust at depths of �8–10 km [Ishiyama etal., 2002a].

6.3. Role of Flexural Slip in Fault Bend Folding AboveBlind Thrusts

[44] Flexural-slip faults play an important role inaccommodating strain above blind thrusts [Yeats, 1984;Cooke and Pollard, 1997; Roering et al., 1997; Nino etal., 1998; Bielecki and Mueller, 2002], and include

examples from Algeria [Philip and Meghraoui, 1983;King and Yielding, 1984], southern California [Rockwellet al., 1988; Klinger and Rockwell, 1989; Medwedeff,1992] and Japan [Ota and Suzuki, 1979]. Unambiguousidentification between flexural-slip fault scarps and theforward balanced solution for Kuwana anticline enablesus to estimate the amount of flexural slip accommodatedby the fold based on folding of the M2 terrace on theforelimb.[45] In the case where flexural-slip faults are responsible

for the formation of west verging fault scarps, slip on abedding plane is obtained as

S ¼ H þ S cos q tanasin q

:: : S ¼ H

sin q� cos q tana

where H, S, q, and a are the height of the fault scarp, slip onthe bedding plane, dip of the fault, and dip of the eastinclined M2 terrace respectively (Figure 10c). On the basisof the summed height of west verging scarps of 41 m(Figure 10b) and a dip of bedding of 37� and an originalgradient of M2 terrace of 3�, total west verging slipconsumed by flexural-slip faults in the forelimb since theabandonment of the M2 terrace is calculated to be �73 m,while 259 ± 43 m of slip should be fed onto the lowerdecollement based on the structural relief on the top of thelower Atsuta Formation (98–137 m) and a dip of the thrust(27�) as discussed earlier.[46] It is of interest that such large amounts of slip are

consumed by flexural-slip faults within the forelimb inthe center of Kuwana anticline. In a fault propagationfold, flexural slip commonly accommodates high shearstrains within their forelimbs [cf. Erslev and Mayborn,1997]. Modeling by Roering et al. [1997] and Nino etal. [1998] indicates that bedding-parallel slip in thesedimentary cover tends to hinder the upward propaga-tion of a low-angle fault tip because mechanical decou-pling between layers causes stress to be transferredacross wider fold limbs. Although the quantitative rela-tionship between the amount of bedding slip and the foldshape were not discussed, our field observations confirmthese mechanical interpretations of flexural-slip faultswhich may act to effectively impede upward propagationof fault tips, and may act to consume slip at successivelyhigher levels.

6.4. Coseismic Fold Scarp Associated With an HistoricEarthquake

[47] The kinematic solution for Kuwana anticline sug-gests that the late Holocene fold scarp with a height of�3 m is coincident with the upward projection of the activesynclinal axial surface formed above the east vergentwedge tip (Figures 11 and 12). In the vicinity of theseismic line, an alluvial fan deformed by the fold scarp isburied beneath the modern alluvial plain and is fed by adrainage channel that has not yet been incised below theyoungest terrace in Kuwana Hills (Figure 12). The seismicreflection data we used in our analysis does not sufficiently


16 of 20

B12408

resolve Holocene strata to compare the detailed geometryof the relatively small fold scarps with the active synclinalaxial surface. Nevertheless, axial surface mapping, ourstructural solution, and historical records, show how thefold scarp was formed by coseismic folding during ahistoric blind thrust earthquake.[48] Boreholes and excavations at the Higashi-Yuriage

site, located 2 km north of the seismic profile, revealed anearthquake occurred on the Kuwana thrust that is possiblycorrelated with large historically recorded earthquake inA.D. 1586 [Sugai et al., 1998]. Deformation from theevent is recorded by historic (�1000–700 years B.P.based on radiocarbon ages) floodplain deposits folded�3.5 m across the frontal fold scarp that coincides with

the base of the forelimb as defined on the seismic line(Figure 13).[49] Sugai et al. [1998] also argued that 12–13 m of

vertical separation across the top of a late Holocene pro-gradational delta (circa 2000–1400 years B.P. based onradiocarbon ages) composing the higher alluvial plain(Figures 12 and 13) formed during two recent historicearthquakes. We invoke a structural model that accountsfor the late Holocene uplift. The axial surface map indicatesthat the westernmost, west dipping fold scarp is coincidentwith the projection of the active axial surface extendingupward from the upper wedge tip (Figure 13). In addition,the western, east dipping fold scarp appears to be coincidentwith the upward projection of the upper ramp, which we

Figure 13. (a) Detailed topographic map of late Holocene fold scarp associated with coseismicfolding of Kuwana anticline at the Higashi-Yuriage, Kuwana City, Mie Prefecture. (b) Topographicprofile across the entire extent of the fold scarp. (c) Borehole transect across the coseismic fold scarp[Sugai et al., 1998]. The frontal coseismic fold scarp is continuous with the fold scarp across theseismic line that coincides with the upward projection of the active synclinal axis extending from thewedge tip.


17 of 20

B12408

interpret as accommodating shear above the ultimate faulttip of the Kuwana thrust. We therefore argue that folding inthis earthquake occurred in response to slip on both thelower and upper wedge tips that governed growth ofKuwana anticline (Figure 14).

7. Conclusions

[50] In conclusion, our studies show that coupled tec-tonic geomorphology with high-resolution seismic datamay be used to define the detailed kinematic evolution,geometry, and rates of slip on blind thrusts beneathQuaternary growth folds. The close coincidence betweentectonic landforms across Kuwana anticline and theunderlying thrust geometry provides confidence in assign-ing particular deformation mechanisms that act to buildfault bend folds. For Kuwana anticline, well preservedfluvial terraces folded across the forelimb and back limbsuggest that it grows above a stacked sequence of thin-skinned wedge thrusts that are consistent with the subsur-face structure of the fold imaged in a seismic section.Secondary, bedding-parallel thrusts that deform the terracesin the forelimb are interpreted to form by flexural-slipfolding that acts to accommodate strains within the fore-limb across synclinal axial surfaces and to consume faultslip. The large amount of flexural slip supports the conceptthat mechanical decoupling in the sedimentary covereffectively acts to inhibit the upward propagation of blindthrust tips. Perhaps most importantly, late Holocene foldscarps formed in the floodplain east of Kuwana anticlinecoincide with the projected surface trace of the east vergentwedge thrust tip. This suggests that coseismic fold scarpsat the ground surface may be indicators of slip onunderlying blind thrust faults, and thus their otherwiseinaccessible seismic hazards.

[51] Acknowledgments. Acquisition and processing of the seismicreflection data were a collaborative project with Hiroshi Sato at ERI,University of Tokyo. We thank Hajime Kato, Nobuhisa Matsuta, ShigeruToda, Hiroyuki Tsutsumi, Nobuhiro Ogisu, and colleagues at AichiEducational University and Kyoto University for additional assistancewith acquisition of seismic data. Daisaku Kawabata is also thanked for

providing us with a high-resolution shaded relief map, while YasuhiroSuzuki allowed us to use his detailed photogrammetric topographicmapping of Kuwana anticline. Rick Allmendinger graciously extendedthe use of his numerical Trishear code to us, for which we are grateful.Finally, our thanks go to Associate Editor Isabelle Manighetti and toJerome Lave and Robert Yeats for their constructive reviews, whichgreatly improved the manuscript. This research was funded by JSPSGrants-in-Aid for Scientific Research 12480019 to Okada, Takemura anda JSPS Fellowship to Mueller.

ReferencesAllmendinger, R. W. (1998), Inverse and forward numerical modeling oftrishear fault-propagation folds, Tectonics, 17, 640–656.

Allmendinger, R. W., and J. H. Shaw (2000), Fault nucleation and propaga-tion from fold shape: Potential implications for earthquake seismicity,Geology, 28, 1099–1102.

Apotria, T. G., W. T. Suedden, J. H. Spang, and D. V. Wiltschko (1992),Kinematic models of deformation at an oblique ramp, in Thrust Tectonics,edited by K. R. McClay, pp. 141–151, CRC Press, Boca Raton, Fla.

Awata, Y. (1997), Kuwana fault group: Geological and geomorphic survey,boring, open cutting and seismic reflection survey (in Japanese), OpenFile Rep. 300, Geol. Surv. of Jpn., Tsukuba.

Awata, Y., and F. Yoshida (1991), Holocene movement of the Kuwana andYokkaichi faults, northwest of Ise Bay, central Japan (in Japanese), ActiveFault Res., 9, 61–68.

Benedetti, L., P. Tapponnier, G. C. P. King, B. Meyer, and I. Manighetti(2000), Growth folding and active thrusting in the Montello region,Veneto, northern Italy, J. Geophys. Res., 105, 739–766.

Benedetti, L. C., P. Tapponnier, Y. Gaudemer, I. Manighetti, and J. Van derWoerd (2003), Geomorphic evidence for an emergent active thrust alongthe edge of the Po Plain: The Broni-Stradella fault, J. Geophys. Res.,108(B5), 2238, doi:10.1029/2001JB001546.

Bielecki, A. E., and K. J. Mueller (2002), Origin of terraced hillslopes onactive folds in the southern San Joaquin Valley, California, Geomorphol-ogy, 42, 131–152.

Bullard, T. F., and W. R. Lettis (1993), Quaternary fold deformation asso-ciated with blind thrust faulting, Los Angeles basin, California, J. Geo-phys. Res., 98, 8349–8369.

Burbank, D., A. Meigs, and N. Brozovic (1996), Interactions of growingfolds and coeval depositional systems, Basin Res., 8, 199–223.

Champion, J. A., A. Tate, K. J. Mueller, and M. Guccione (2001), Geo-metry, numerical modeling and revised slip rate for the Reelfoot blindthrust and trishear fault-propagation fold, New Madrid seismic zone, Eng.Geol., 62, 31–49.

Committee for Subsurface Structure of the Nobi Basin (2001), Report onsurvey for subsurface structure of the Nobi Basin (in Japanese), 13 pp.,Nagoya, Japan.

Cooke, M. L., and D. D. Pollard (1997), Bedding-plane slip in initial stagesof fault-related folding, J. Struct. Geol., 19, 567–581.

Couzens, B. A., and D. V. Wiltschko (1996), The control of mechanicalstratigraphy on the formation of triangle zones, Bull. Can. Pet. Geol., 44,165–179.

Dolan, J. F., S. A. Christofferson, and J. H. Shaw (2003), Recognition ofpaleoearthquakes on the Puente Hills blind thrust fault, California,Science, 300, 115–118.

Erslev, E. A. (1991), Trishear fault-propagation folding, Geology, 19, 617–620.

Erslev, E. A., and K. R. Mayborn (1997), Multiple geometries and modes offault-propagation folding in the Canadian thrust belt, J. Struct. Geol., 19,321–335.

Faculty of Science, Kyoto University, Geo-Research Institute, HanshinConsultants Co., Ltd, JAPEX Geoscience Institute Inc., Chubu BranchOffice of Water Resources Development Public Corporation, and ChubuRegional Construction Bureau, Ministry of Construction (1996), Reflec-tion profiles of deep structure across the Kuwana fault and Ise-wan fault(in Japanese), Rep. Coord. Comm. Earthquake Predict.55, pp. 558–570,Geogr. Surv. of Jpn., Tsukuba.

Furusawa, A. (1990), Volcanic ash layers of the Tokai Group and the Amaand Yatomi Formations under the Nobi Plain, and their correlation (inJapanese with English abstract), J. Geol. Soc. Jpn., 96, 883–901.

Hardy, S., and M. Ford (1997), Numerical modeling of trishear fault-propagation folding, Tectonics, 16, 841–854.

Hung, J., and J. Suppe (2002), Subsurface geometry of the Sani-Chelungpufaults and fold scarp formation in the 1999 Chi-Chi Taiwan earthquake,Eos Trans. AGU, 83(47), Fall Meet. Suppl., Abstract T61B-1268.

Ikebe, N. (1942), Active folding with in the Echigo oil field, northeasternJapan (in Japanese), J. Jpn. Assoc. Pet. Technol., 10, 108.

Ikeda, Y. (1983), Thrust-front migration and its mechanism: Evolution ofintraplate thrust systems, Bull. Dep. Geogr. Univ. Tokyo, 15, 125–159.

Figure 14. Hypothesized schematic cross section thatexplains the fold scarp topography at this site. Coseismicslip on the doubly vergent wedge thrust produces upliftaccommodated by a pair of fold scarps, which mimics thetopography observed at the Higashi-Yuriage site.


18 of 20

B12408

Ikeda, Y., T. Imaizumi, H. Sato, M. Togo, K. Hirakawa, and T. Miyauchi(Eds.) (2002), Atlas of Quaternary Thrust Faults in Japan (in Japanese),254 pp., Univ. of Tokyo Press, Tokyo.

Ishiyama, T. (2002), Evolution of tectonic landform associated with activethrusting and active folding: Case studies on Nobi-Ise active fault zone,central Japan, thesis for Doctor of Science, 131 pp., Kyoto Univ., Kyoto,Japan, March.

Ishiyama, T., K. Takemura, and A. Okada (1999), Structural growth rate ofthe eastern margin of the Suzuka Range during Quaternary (in Japanesewith English abstract), J. Seismol. Soc. Jpn., 52, 229–240.

Ishiyama, T., K. J. Mueller, M. Togo, K. Takemura, and A. Okada (2002a),Geomorphology and kinematics of the Nobi-Ise active fault zone, centralJapan: Implications for the kinematic growth of tectonic landforms withinan active thrust belt, Eos Trans. AGU, 83(47), Fall Meet. Suppl., AbstractS11B-1140.

Ishiyama, T., M. Togo, T. Imaizumi, H. Sato, T. Nakata, T. Nohara, andT. Haraguchi (2002b), Evolution of fault-related landform associated withlate Holocene faulting on the Yoro fault, revealed by drilling survey atShizu-Shobuhara Site, Nan-no Town, Gifu Prefecture, central Japan (inJapanese with English abstract), Active Fault Res., 22, 115–126.

Kelson, K. I., G. D. Simpson, R. B. VanArsdale, C. C. Haraden, and W. R.Lettis (1996), Multiple late Holocene earthquakes along the Reelfootfault, central New Madrid seismic zone, J. Geophys. Res., 101, 6151–6170.

Kimura, I., H. Misawa, and Y. Takeuchi (1984), 14C age of a terrace depositand a humic soil layer in Tado-cho, Mie Prefecture (in Japanese withEnglish abstract), J. Earth Sci., 38, 67–69.

King, G. C. P., and G. Yielding (1984), The evolution of a thrust faultsystem: Processes of rupture initiation, propagation and termination inthe 1980 El Asnam (Algeria) earthquake, Geophys. J. R. Astron. Soc., 77,915–930.

Klinger, R. E., and T. L. Rockwell (1989), Flexural-slip folding along theeastern Elmore Ranch fault in the Superstition Hills earthquake sequenceof November 1987, Bull. Seismol. Soc. Am., 79, 297–303.

Lave, J., and J. P. Avouac (2000), Active folding of fluvial terraces acrossthe Siwaliks Hills, Himalayas of central Nepal, J. Geophys. Res., 105,5735–5770.

Machida, H., and F. Arai (2003), Atlas of Tephra in and around Japan (inJapanese), 336 pp., Univ. of Tokyo Press, Tokyo.

Makinouchi, T., S. Mori, T. Danhara, K. Takemura, and Research Commit-tee of the Ground of Nobi Plain (2001), Ages of the basal gravel ofalluvium (BG) and the lower marine clay member of late PleistoceneAtsuta Formation under the Nobi plain, central Japan (in Japanese withEnglish abstract), J. Geol. Soc. Jpn., 107, 283–295.

Medwedeff, D. A. (1992), Geometry and kinematics of an active, laterallypropagating wedge thrust, Wheeler Ridge, California, in Structural Geol-ogy of Fold and Thrust Belts, edited by S. Mitra and G. W. Fisher, pp. 3–28, Johns Hopkins Univ. Press, Baltimore, Md.

Medwedeff, D. A., and J. Suppe (1997), Multibend fault-bend folding,J. Struct. Geol., 19, 279–292.

Molnar, P., E. T. Brown, B. C. Burchfiel, Q. Deng, X. Feng, J. Li, G. M.Raisbeck, J. Shi, Z. Wu, F. Yiou, and H. You (1994), Quaternary climaticchange and the formation of river terraces across growing anticline on thenorth flank of the Tien Shan, J. Geol., 102, 583–602.

Morley, C. K. (1986), A classification of thrust fronts, AAPG Bull., 70, 12–25.

Mount, V. S., J. Suppe, and S. C. Hook (1990), A forward modelingstrategy for balancing cross sections, AAPG Bull., 74, 521–531.

Mueller, K., and J. Suppe (1997), Growth of Wheeler Ridge anticline,California: Geomorphic evidence for fault-bend folding behavior duringearthquakes, J. Struct. Geol., 19, 383–396.

Mueller, K., and P. Talling (1997), Geomorphic evidence for tear faultsaccommodating lateral propagation of an active fault-bend fold, WheelerRidge, California, J. Struct. Geol., 19, 397–411.

Mueller, K., J. Champion, M. Guccione, and K. Kelson (1999), Faultslip rates in the modern New Madrid seismic zone, Science, 286,1135–1138.

Nakamura, K., and Y. Ota (1968), Study of active fold in Japan: A review(in Japanese with English abstract), Quat. Res., 7, 200–211.

Nakata, T. (1976), Quaternary tectonic movements in central Tohoku dis-trict, northeast Japan, Sci. Rep. Tohoku Univ., 26, 213–239.

Nakata, T., and T. Imaizumi (Eds.) (2002), Digital active fault map of Japan(in Japanese), 60 pp., Univ. of Tokyo Press, Tokyo.

Nino, F., H. Philip, and J. Chery (1998), Role of bed-parallel slip in theformation of blind thrust faults, J. Struct. Geol., 20, 503–516.

Novoa, E., V. Mount, and J. Suppe (1998), Map-view inference of mono-clinal folds, J. Struct. Geol., 20, 339–353.

Okada, A., and M. Togo (Eds.) (2000), Active Faults in the Kinki Area,Central Japan: Sheet Maps and Inventories, 395 pp., Univ. of TokyoPress, Tokyo.

Ota, Y., and A. Sangawa (1984), Active fault in the eastern foot area of theSuzuka Range, central Japan (in Japanese with English abstract), Geogr.Rev. Jpn. A, 57(4), 237–262.

Ota, Y., and I. Suzuki (1979), Notes on active folding in the lower reachesof the Shinano River, central Japan (in Japanese with English abstract),Geogr. Rev. Jpn., 52, 592–601.

Otsuka, Y. (1941), Active rock folding in Japan, Proc. Imp. Acad. Jpn., 17,518–522.

Philip, H., and M. Meghraoui (1983), Structural analysis and interpretationof the surface deformations of the El Asnam earthquake of October 10,1980, Tectonics, 2, 17–49.

Research Group for Active Faults (Eds.) (1991), Active Faults in Japan:Sheet Maps and Inventories (in Japanese), 437 pp., Univ. of Tokyo Press,Tokyo.

Rich, J. L. (1934), Mechanics of low-angle overthrust faulting as illustratedby the Cumberland thrust block, Virginia, Kentucky, and Tennessee,AAPG Bull., 18, 1584–1596.

Rockwell, T. K., E. A. Keller, M. N. Clark, and D. L. Johnson (1984),Chronology and rates of faulting of Ventura River terraces, California,Geol. Soc. Am. Bull., 95, 1466–1474.

Rockwell, T. K., E. A. Keller, and G. R. Dembroff (1988), Quaternary rateof folding of the Ventura Avenue anticline, western Transverse Ranges,southern California, Geol. Soc. Am. Bull., 100, 850–858.

Roering, J. J., M. L. Cooke, and D. D. Pollard (1997), Why blind thrustfaults do not propagate to the Earth’s surface: Numerical modeling ofcoseismic deformation associated with thrust-related anticlines, J. Geo-phys. Res., 102, 11,901–11,912.

Sakamoto, T., Y. Takada, T. Kuwabara, and J. Itoigawa (1986), Geologyof the Nagoya-Nambu district, with geologic sheet map at 1:50,000(in Japanese with English abstract), 55 pp., Geol. Surv. of Jpn.,Tsukuba.

Shaw, J. H., and P. Shearer (1999), An elusive blind-thrust fault beneathmetropolitan Los Angeles, Science, 283, 1516–1518.

Shaw, J., and J. Suppe (1994), Active faulting and growth folding in theeastern Santa Barbara Channel, California, Geol. Soc. Am. Bull., 106,607–626.

Shaw, J. H., and J. Suppe (1996), Earthquake hazards of active blind-thrustfaults under the central Los Angeles basin, California, J. Geophys. Res.,101, 8623–8642.

Shaw, J., S. C. Hook, and J. Suppe (1994), Structural trend analysis by axialsurface mapping, AAPG Bull., 78, 700–721.

Shaw, J. H., A. Plesch, J. F. Dolan, T. L. Pratt, and P. Fiore (2002), PuenteHills blind-thrust system, Los Angeles, California, Bull. Seismol. Soc.Am., 92, 2946–2960.

Sugai, T., Y. Awata, and K. Shimokawa (1998), Paleoseismological study ofthe Kuwana fault and the Yokkaichi fault in Mie Prefecture, central Japan(in Japanese with English abstract), Interim Rep. EQ/98/1, pp. 75–90,Geol. Surv. of Jpn., Tsukuba.

Sugai, T., Y. Fusejima, Y. Awata, T. Azuma, Y. Kariya, and Y. Suzuki(1999a), Late Holocene paleoseismicity of the Yoro fault system, centralJapan (in Japanese with English abstract), Interim Rep. EQ/99/3, pp. 89–102, Geol. Surv. of Jpn., Tsukuba.

Sugai, T., Y. Sugiyama, and K. Mizuno (1999b), 900,000–year cyclostrati-graphic record from a 600 m core, GS-NB-1, the Nobi basin, centralJapan (in Japanese with English abstract), Interim Rep. EQ/99/3, pp.69–76, Geol. Surv. of Jpn., Tsukuba.

Suppe, J. (1983), Geometry and kinematics of fault-bend folding, Am.J. Sci., 283, 684–721.

Suppe, J., and Y. L. Chang (1983), Kink method applied to structuralinterpretation of seismic sections, western Taiwan, Pet. Geol. Taiwan,19, 29–49.

Suppe, J., G. T. Chou, and S. C. Hook (1992), Rates of folding and faultingdetermined from growth strata, in Thrust Tectonics, edited by K. R.McClay, pp. 105–121, CRC Press, Boca Raton, Fla.

Suppe, J., F. Sabat, J. A. Munoz, J. Poblet, E. Roca, and J. Verges (1997),Bed-by-bed fold growth by kink-band migration: Sant Llorenc deMorunys, eastern Pyrenees, J. Struct. Geol., 19, 443–461.

Suzuki, Y., S. Sano, and R. Nozawa (2002), A photogrammetrical mappingof active faults at Kuwana: Accuracy of active fault mapping by use ofaerial photographs taken in 1940’s (in Japanese with English abstract),Active Fault Research, 22, 76–82.

Takemura, K. (1985), The Plio-Pleistocene Tokai Group and the tectonicdevelopment around Ise Bay of central Japan since Pliocene, Mem. Fac.Sci. Kyoto Univ. Ser. Geol. Mineral., 51, 21–96.

Thompson, S. C., R. J. Weldon, C. M. Rubin, K. Abdrakhmatov, P. Molnar,and G. W. Berger (2002), Late Quaternary slip rates across the centralTien Shan, Kyrgyzstan, central Asia, J. Geophys. Res., 107(B9), 2203,doi:10.1029/2001JB000596.

Togo, M. (2000), Geomorphological Analysis on Surface Ruptures byReverse Faulting in Japan (in Japanese), 206 pp., Kokon, Tokyo.


19 of 20

B12408

Usami, T. (2003), Materials for Comprehensive List of Destructive Earth-quakes in Japan, 416–2001 (in Japanese), 603 pp., Univ.of Tokyo Press,Tokyo.

Woodward, N. B., S. E. Boyer, and J. Suppe (1989), Balanced Geo-logical Cross-Sections: An Essential Technique in Geological Researchand Exploration, Short Course Geol., vol. 6, 132 pp., AGU, Washing-ton, D. C.

Yeats, R. S. (1984), Active faults related to folding, in Active Tectonics,edited by R. E. Wallace, pp. 63–79, Natl. Acad. Press, Washington,D. C.

Yeats, R. S., K. Sieh, and C. R. Allen (1997), The Geology of Earthquakes,568 pp., Oxford Univ. Press, New York.

Yoshida, F. (1984), Geology of the Yokkaichi district, with geologic sheetmap at 1:50,000 (in Japanese with English abstract), 81 pp., Geol. Surv.of Jpn., Tsukuba.

Yoshida, F. (1988), Plio-Pleistocene Tokai Group between the Suzuka andYoro Mountains, central Japan (in Japanese with English abstract),J. Earth Sci., 42, 1–16.

Yoshida, F., C. Kurimoto, and M. Miyamura (1991), Geology of theKuwana district, with geologic sheet map at 1:50,000 (in Japanese withEnglish abstract), 154 pp., Geol. Surv. of Jpn., Tsukuba.

��T. Ishiyama, Active Fault Research Center, Geological Survey of Japan,

National Institute of Advanced Industrial Science and Technology, Site 7,1-1-1 Higashi, Tsukuba, Ibaraki 305-8567, Japan. ([email protected])K. Mueller, Department of Geological Sciences, University of Colorado,

Boulder, CO 80309-0399, USA. ([email protected])A. Okada, Graduate School of Science, Kyoto University, Kitashirakawa-

Oiwake-Cho, Sakyo-Ku, Kyoto 606-8502, Japan. ([email protected])K. Takemura, Institute for Geothermal Sciences, Kyoto University,

Noguchihara, Beppu, Oita 874-0903, Japan. ([email protected])M. Togo, Department of Geoscience, Hosei University, Aihara-Cho,

4342, Machida 194-0298, Japan. ([email protected])


20 of 20

B12408

Geomorphology, kinematic history, and earthquake behavior of...

Documents

Transcript of Geomorphology, kinematic history, and earthquake behavior of...