Oligo-Miocene midcrustal subhorizontal shear zone Laurent ...web.mst.edu/~yyqkc/Ref/vietnam-ref/Jolivet_Beyssac... · Laurent Jolivet, • Olivier Beyssac, 2 Bruno Goff6, 2 Dov Avigad,

TECTONICS, VOL. 20, NO. 1, PAGES 46-57 FEBRUARY 2001

Oligo-Miocene midcrustal subhorizontal shear zone in Indochina

Laurent Jolivet, • Olivier Beyssac, 2 Bruno Goff6, 2 Dov Avigad, 3 Claude Lepvrier, • Henri Maluski, 4 and Ta Trong Thang s

Abstract. New structural observations of Oligo-Miocene deformation in north Vietnam along the Red River Shear Zone suggest that left-lateral strike-slip shear was restricted to the upper and middle crust above a horizontal shear zone. Left- lateral shear deformation is associated with low-pressure-low- temperature parageneses. High-temperature deformation is restricted to zones where the foliation has a low dip in the core of the Dai Nui Con Voi antiformal dome. These observations

complete those made earlier in the Bu Khang extensional dome farther south. Above the horizontal shear zone left-lateral

transpression was active during the first stage and changed to transtension some 33Myr ago. Purely extensional metamorphic domes (Bu Khang) or transtensional domes (Dai Nui Con Vol) were exhumed during this younger stage. Our observations plead for caution when interpreting strike-slip structures at lithospheric scale.

1. Introduction

The Cenozoic tectonic history of the Indochina Peninsula has been a keypoint in the debate about the mechanism of deformation of Asia as a consequence of the collision with India since Tapponnier et al. [1982] first proposed the extrusion model. The concept of extrusion had been first introduced by McKenzie [1972] about the westward escape of Anatolia. The first large-scale descriptions of strike-slip faults in Asia were interpreted as the result of indentation tectonics using continuum mechanics [Molnar and Tapponnier, 1975' Tapponnier and Molnar, 1976]. An analogy with some plastiscine experiments led Tapponm'er et a/. [1982] to propose that Indochina was extruded along the Red River Fault during the India-Asia collision and the opening of the South China Sea.

Behind this model is the general hypothesis that the continental lithosphere behaves as a rigid medium which strongly localizes the deformation along major strike-slip faults and that very little deformation is observed in between. This hypothesis has been challenged by several other workers who, instead, assume a viscous behavior and hence a more distributed deformation [England and McKenzie, 1982;

--• Laboratoire de Tectonique, ESA CNRS 7072, UniversItC Pierre et Marie Curie, Paris.

2 Laboratoire de Geolog•e, UMR 8538, Ecole Normale Superieure, Paris.

• Institute of Earth Sciences, The Hebrew University of Jerusaleln, Jerusalem.

4 Laboratoire de G•ochronologle Geochimie Petrologle, UMR 5567, UniversitC Montpeilier II, Montpe!!ier

s Laboratory of Geology, National University of Vietnam in Hanoi, Hanoi.

Copyright 2001 by the American Geophysical Union.

Paper number 2000TC900021. 0278-7407/01/2000TC900021 $12.00

England and Hoztseman, 1986]. The extrusion model seemed supported at the scale of the Asian continent by field evidence of large-scale strike-slip faults [Tapponmer et al., 1986' Leloztp et al., 1995] and the viscous model by the distribution of seismicity for instantaneous strain, and topography for finite strain [England and Hoztseman, 1986; England and Molnar, 1997]. Recently obtained figures give 60-80 % of crustal thickening and 20-40 % of extrusion [Mdtivier et al., 1999]. The distribution of faults can be interpreted in a way which gives a smaller importance to extrusion depending on the amount of rotation of the faults and the role played by. conjugate dextral faults [Davy and Cobbold, 1988; Dewey et al., 1989; rioliver et al., 1990: Huchon eta/., 1994; rioliver et al., 1999a]. Topography and erosion can be used as an independant data set to constrain the respective contribution of (1)crustal thickening and (2) other phenomenons such as extrusion [Mdtivier, 1996].

The Red River Shear Zone [Leloztp et al., 1995] is the major strike-slip feature of the northern part of Indochina during the Tertiary along with the Wang Chao Fault farther south [Lacassin et al., 1997]. Some authors considered that it is the only significant structure [Leloup et al., 1995]; others have described other strike-slip thults [Hztchon et al., 1994] or extensional structures [dolivet et al., 1999b] that may account for some significant deformation within Indochina. One crucial point in the extrusion model is the assumption that the strike- slip faults are deeply rooted and cut through the entire lithosphere, making them the equivalent of transform faults in the plate tectonic model. Some authors have challenged this assumption and postulated that strike-slip faults root in a midcrustal depth horizontal shear zone [Burchfi'e! et al., 1989: Wang and Bz•rchfi'el, 1997]. Other large strike-slip faults also may be shallow structures and root on subhorizontal shear zones. This has been proposed for the E1 Pilaf Fault [Passa/acqua et al., 1995] and even for the San Andreas Fault [Jones eta/., 1994].

Our study in north Vietnam indicates that the majority of the ductile structures along the Red River Shear Zone were formed at relatively low pressure-temperature conditions corresponding to the greenschist facies. These observations cast doubts on the interpretation of the Red River Shear Zone as a deeply penetrating lithospheric boundary and are better in line with the Red River Shear Zone being rooted in an horizontal shear zone at shallow crustal levels at the brittle- ductile transition.

2. Crustal Roots of the Red River Shear Zone in Indochina

The Red River Shear Zone extends from the eastern Tibetan

plateau to the Red River delta in north Vietnam (Figures. 1 and 2). It has been recognized as a letl-lateral shear zone that penetrated the ductile crust and acted at high temperature conditions [Leloup et al., 1995]. Amphibolite facies P-T conditions (- 700øC, 6 kbar) have been calculated for the peak of metamorphism along the Red River Shear Zone, and

46

JOLIVET ET AL.: MIDCRUSTAL SIIEAR ZONE IN INDOCHINA 47

Red River I I ! • 'l 110'E 1

SOUTH CHINA fault

Song Chay ........ massif .•.•':? ......... • .......... ::::"':'":'

.

Bei ,Bu Basin

Figure 1. Location map of the studied region and regional geodynamic context. Dark shaded area represent domains floored with oceanic crust.

greenschist facies conditions of- 480øC, 3 kbar were estimated for the subsequent mylonitization [Leloup and Kienast, 1993; Tran Ngoc Nam et al., 1998]. Rb/Sr and 4øAr/39Ar radiochronological data show that the Red River Shear Zone was active during the late Oligocene and early Miocene [Leloup et al., 1995; Harrison et al., 1996; Tran Ngoc Namet al., 1998; Wang et al., 1998; Mahtski et al., 2000] from 30 to 15 Ma. This timing overlaps with that of the South China Sea opening [Briais et al., 1993].

The recent discovery of an Oligo-Miocene extensional metamorphic core complex south of the Red River Fault (Bu Khang dome) (Figure 3) [dolivet et al., 1999b] questions the rigid character of the Indochina Peninsula. The 4øAr-39Ar dates on micas within the dome suggest an Oligo-Miocene age (from 33 to 21 Ma) for the extensional deformation in this area [dolivet et al., 1999b]. This recently received support from U- Pb (23-26 Ma) and Rb-Sr (10-20 Ma) ages on granitoids within the dome, suggesting that extension was contemporaneous with granitic injection from a mixed (crust and mantle) source [Nagy et al., 2000]. The Bu Khang dome is characterized by relatively high pressure parageneses indicating peak P-T conditions (above 10 kbar) [dolivet et al., 1999b]. The ubiquitous presence of kyanite in the dome makes a strong difference with the Red River-Ailao Shan shear zone where kyanite has not been found. Kyanite-bearing micaschists are described west of the Red River Ailao Shan Shear zone in

Vietnam [TranNgoc Nam, 1997] and were attributed to a Precambrian episode of metamorphism, but the Oligocene 4øAr- 3øAt ages of high-pressure micas in the Bu Khang dome lead us to question this conclusion. The possibility of an Oligocene thermal resetting of an older metamorphic stage does not seem pertinent because biotite yields older 4øAr-39Ar ages than muscovite within the extensional shear zone, although biotite has a lower closure temperature than muscovite, thus suggesting that the 4øAr-39Ar method here dates the crystallization rather than the cooling of micas. If, indeed, high- pressure metamorphism and extension in the Bu Khang dome are Tertiary, as indicated by geochronology, this part of Indochina must have been thickened considerably during the Tertiary and then thinned over low-angle normal faults that transected the crust horizontally.

In the following we first sumarize structural and petrological informations on the Bu Khang dome and then describe new observations in the Dai Nui Con Vol massif

within the Red River Shear Zone.

2.1. Bu Khang Dome

The Bu Khang dome (Figure 4) and the Phu Hoat massif form a metamorphic core complex oriented NW-SE which extends from Vietnam to Laos. Only the Vietnamese part has been studied so far [Jolivet et al., 1999b; Nagy et al., 2000].

48 .IOLIVET ET AL.' MIDCRUSTAL SHEAR ZONE IN INDOCHINA

Xue

Long Shan

Dian

Cang Shah

gneiss

granite (011gocene)

[ ] Quaternary

Phang Si Pan ranite

Po Sen

Orthogneiss

Figure 2. General geological map of the Red River Ailao Shan Shear Zone after Harrison eta/. [1992}. The Dai NuI Con Voi (DNCV) dome and Po Sen orthogneiss are located neat' the southeastern end of the Red River Shear Zone in Vietnam.

The dome shows the superposition of sedimentary and metasedimentary units (marbles and metapelites) on top of a gneissic unit which makes the core of the open antiformal structure. The uppermost unit is devoid of any significant metamorphism and is separated from the metamorphic units by an NE dipping extensional shear zone on the NW limb of the antiform. Inside the dome the peak pressure recorded in the metapelites and metabauxites increases downward. A strong retrogression toward greenschist conditions is observed within the shear zone during NE-SW stretching and top to the NE shear.

?-T conditions were obtained from metapelites and metabauxites. A first estimate was calibrated [Jolivet et al., 1999b] (Figure 5) at 600øC and 12 kbar from micaschists in the core of the dome as well as along the extensional shear zone. These calculations were made using garnet-biotite-phengite associations and were confirmed by inclusions in garnet of titanite in ilmenite and kyanite implying pressure higher than 10 kbar. These P-T conditions are also sustained by the occurrence of garnet in kyanite-corundum-zo¾site-bearing metabauxites (B. Goff• et al., manuscript in preparation, 2000) As we never found occurrences of sillimanite and/or andalusite

in any rocks in the area, we assume that these rocks remained in the stability field of kyanite during the retrograde path (Figure 5).

The age of the high-pressure episode is not precisely constrained. The 4øAr/39Ar dates show that the isotopic closure of the high-pressure micas occurred at 30-33 Ma and that progressive crystallization of younger micas occurred during the exhumation and the localization of shear along the major shear zone. The high-pressure episode is probably of Oligocene age, but it could also be much older.

2.2. Song Chay Dome

The Song Chay dome (Figures 6 and 7) reveals a very different tectonic timing [Leloup et al., 1999; Maluski et al, 1999, 2000]. Located to the northeast of the Red River Shear Zone, the dome is essentially made of orthogneiss overlain by a metasedimentary cover, mainly marbles. The entire column is strongly deformed and displays a flat-lying fbliation folded in a gentle antiform. The original fabric of the granite is clearly visible, especially in the core of the dome. A NE-SW or N-S trending stretching lineation is associated with top to the north sense of shear along several shear zones. Radiochronological measurements (4øAr/39Ar and fission tracks) revealed Mesozoic ages for the ductile deformation (235-240 Ma), and a final exhumation between 40 and 24 Ma. The age of the granitic protolith was determined to be- 428 Ma (U/Pb) by Le/oup et al. [1999]. No Tertiary ductile extension is thus evidenced in the area north of the Red River Shear Zone

and the outcropping HT metamorphic rocks were already cold when the Cenozoic deformation started. This makes a sharp difference with the Bu Khang dome and the Red River Shear Zone. The structural, metamorphic, and geochronological features of the Song Chay massif, which immediately flanks the Red River Shear Zone in Vietnam, suggest that the Red River Shear Zone crossed an already metamorphosed and horizontally layered continental crust shaped during the Indosinian orogeny.

2.3. Red River Shear Zone, Po Sen Gneiss

Following several other research teams [Leloup et al., 1995' TranNgocNam et al., 1998; Wang et al., 1998], we have studied several cross-sections through the gneiss of the

JOLIVET ET AL.' MIDCRUSTAL SHEAR ZONE IN INDOCIIINA 49

103 ø 00'E 23' 00'N

22" 30'N

22' 00'N

103' 30'E 104' O0'E 104' 30'E

................................ ....... ...... ............... ........... . ........ .... $;;:.:•;•3&. • • .:;;' ....... ::x .............

:•½•'•:: .:::•:;•m?•>:L ::•,"-: • ' •:.• .......

. '• :?½; :* .................

.... .

.:.

:,•:,:r: :...:.

.......

.... •:.::;..•,•:.... 8:;;:.

21' 30'N

21' 00'N

105' 00'E

. ::.:.

105' 30'E 106' 00'E .23' 00'N

22' 30'N

22' 00'N

21' 30'N

21' 00'N

20' 30'N

20 ø 00'N

19' 30'N

.• '"'"iii?•;i•::;•{ ........ •!•:,½½•!•iF*-'•"•'-•:•i•?ii::! ' . ..... ....

19 ø 00'N ...................................................................................... - .... ..:L:.. ! 103 ø 00'E 103' 30'E 104' 00'E 104' 30'E 105' 00'E 105' 30'E

'20' 30'N

20' 00'N

........

. .

,s:.:::::•'

.::• :.:

•::::....•

............. : .•

.......... .} ::. ..

..

19' OO'N 106' O0'E

Figure 3. Topography and main faults in North VietNam. The Bu Khang dome (BKD) is an Oligo-Miocene extensional metamorphic core complex. The Song Chay massif is a metamorphic core complex of Triassic age.

Red River shear zone from Bat Xat close to the Chinese border to a section through the Dai Nui Con Vol massif between Bao Yen and Bao Ha (Figures 6 and 7). In what follows we bring new data which augment but also differ from previous observations on the general characteristics of the Red River

Shear Zone. The Red River Shear Zone consists of two major parallel elongated crustal segments made of orthogneiss and paragneiss separated by the Red River Fault and some Cenozoic continental conglomerate, slightly deformed. The southwestern segment, which is the southern continuation of

50 JOLIVET ET AL.' MIDCRUSTAL SHEAR ZONE IN INDOCHINA

Que

• Radiometric age of white micas (Ms)

• Quaternary

Mesozoic sedimentary rocks

-•-• Poorly metamorphoasd pe!itas

'[•] Marbles with minor metabasites and metabauxitas (-•)

• Granite

-:::---':• Highly sheared micaschists and gneiss

""":.-•:...':• Micaschists and gneiss

• Serpentinites

Ban

Stretching Iinestion and shear sense

105 ø 105o10

i!

NE marbles and metabauxites

high temperature low temperature poorly

_ _

............... !:i,•::::•:•:i:i:i•i•!i!ii!!i!!!i!i!?:::::::?:!ii::i::i::•?:i•!i

19ø20

Figure 4. Geological map and cross section of the Bu Khang dome after dolivet et al. [1999b]. Radiometric ages of micas were obtained by 4øAr/39Ar method.

the Ailao Shan massif, is hereafter called the Po Sen massif, whereas the northeastern one is the Dai Nui Con Vol massif.

The Dai Nui Con Vol massif is bounded by the Red River Fault (Song Hong Fault) to the SW and the Song Chay Fault to the

14

12

10

_

300 I

4OO

rt alra

ilra ky qz

I

5OO

!

!

!

600 700 800

NE. We use the same regional terminology as Tran Ngoc Nam et al. [1998].

Two cross sections, one between Lao Cai and Bien Lu and

one near the Bat Xat in the north, show the major characteristics of the southwestern segment. A metasedimetary cover (marbles and micaschists) rests upon a "basement" of orthogneiss, which is intruded by more or less sheared granitoids such as the Phang Si Pan granite or the Po Sen granite. The Po Sen granite is heterogeneously deformed to orthogneiss thrusted upon folded marbles and micasChists near Sa Pa.

The Po Sen orthogneiss show a NW-SE trending foliation, often steeply dipping to the east. In the west the foliation flattens near the thrust contact with the micaschists. The

steeply dipping foliation carries a horizontal stretching

Figure 5. Pressure-temperature paths of gneiss massifs along the Red River Shear Zone: Diancang Shan (DCS) from Leloup et al. [1995], Ailao Shan (ALS) from Leloup and Kienast [1993] and DNCV from D'art Ngoc Namet al [1998]. BKD was calibrated in this study. AD, andalousite; KY, kyanite; SI, sillimanite; Ms, muscovite; GT, garnet; Kfs, K feldspar; Qz, quartz; Cd, cordieritc; bi, biotite; rt, rutile; alm, almandin; Jim, ilmenite.

.IOLIVET ET AL.: MIDCRUSTAL SItEAR ZONE IN INDOCHINA 51

o CHINA

A....,,<... • z/T/t/•,.J,16ang Su Phi ' :".-'i

•... .... .: .... .. :: . ..-.' - .. ...

alert Lu ' * •:•: '•" •. • aao Yen .X.h:-.. ß ... " Phang Si •:n[I •

Bao Ha

•anh Uy•. .: :::: :-;:: * --

103ø30E 104øE 104ø30E 1050E

gneiss (Red River Shear Zone end Song Chay massif) • grenite • poorly metamo•ho• • Neogene conglomerate sedirnent•

t• iineation and shear sense 28.• 40Ar-39Ar age Fission track age

Figure 6. Geological map of the Song Chay dome and the Red River Fault zone with indications of stretching lienations, sense of shear, and isotopic and fission track ages [Jolivet et al., 1999b; Mahtski et al., 2000]. Inset' stereograms of foliation, shear planes and lineations in the Po Sen and Dai Nui Con Vol metamorphic massives.

23øN

22•40N

22ø20N

22•N

lineation. Where the foliation is less steep, the lineation strikes parallel to the main orientation of the shear zone. Clear sinistral or top to the NW kinematic indicators are observed across the whole section of the Po Sen orthogneiss.

The deformation in the Po Sen orthogneiss is heterogeneous. Strongly localized along several paralell brittle-ductile shear zones in the east, this deformation becomes more penetrative and ductile in the west where the foliation is less steep.

In the studied area the Red River Shear Zone is a

transpressional strike-slip shear zone. This is revealed at the scale of the outcrop by the association of closely packed strike- slip and reverse shear zones and regionally by the overthrusting of the Po Sen orthogneiss onto the metasediments.

The Phang Si Pan granitic massif displays a ductile deformation, quite weak and heterogeneously distributed. The cross section reveals a domal structure with several intrusive

lithologies displaying a weak foliation and an E-W trending stretching lineation. This structure is similar in appearance to

the extensional structure of the Bu Khang dome, but we cannot reach a firm conclusion with the limited observations we have.

2.4. Red River Shear Zone, Dai Nui Con Voi Massif

The Dai Nui Con Voi massif is composed of garnet- sillimanite-micaschists, gneiss and minor amphibolites, marbles, pegmatites, and migmatites. In line with previous studies, our petrological data indicate that the rocks exposed reached maximum P-T conditions of 6-7 kbar and 690øC. Our section between Bao Yen and Bao Ha reveals that rather than

monotonously steeply dipping layering, the architecture of the shear zone in this area defines an open to tight antiform. The axis of the antiformal dome and the stretching lineation are parallel to the general strike of the shear zone (N140øE). Numerous kinematic indicators show a left-lateral motion on

both limbs of the dome, but asymmetric shear fabric become less prominent at the core of the antiform where the foliation is flat lying. The presence of the antiformal structure was not taken into account in previous studies, but it is certainly important

52 JOLIVET ET AL.: MIDCRUSTAL SHEAR ZONE IN INDOCHINA .i

Bat Xat (projected)

Phang Si Pan Sa Pa Lao Cai Song Chay Riv. Xin Man Huang Su Phi

I Red River F.I Song Chay F.

W S

I Phang Si Pan Sa Pa Bat Xat (projected) Lao Cai _ _ _ / I I ...... ' =.,__=. Oai ,ui Con Voi SCF I / I J I PO Sen gneiss I=•l=•l'gneiss and micaschists \ I

granite sediments gneiss and and metasediments micaschists

Figure 7. Cross section of the Red River Shear Zone and Song Chay massif (location on Figure 6). Foliation domes of different ages crop out on either sides of the Song Chay fault. The Dai Nui Con Vol dome is Oligo- Miocene, and the Song Chay massif is Triassic. The lower part of Figtire 7 is an enlarged version of the western part of the section. It shows the geometry of foliation in the Dai Nui Con Vol dome and Posen orthogneiss.

for the understanding of the Red River Shear Zone. Shear planes measured in the field are all with strong dips and do not show the domal geometry shown by the foliation (see stereogram, Figure 6).

Tran Ngoc Namet al. [1998] noticed two deformation stages along the Red River Shear Zone, one that operated in the amphibolite facies (690øC and 6-7 kbar) and a second stage that affected the rocks in the greenschist facies (480øC, below 3 kbar) associated with a vertical mylonitic shear zone. Our study reveals a similar picture, but we observe that the different deformation regimes partitioned between the limbs of the antifbrm and its core (Figure 8). In most outcrops the very clear left-lateral kinematic indicators are C' shear bands or

asymmetric pressure shadows on garnets, but our microstructural examinations showed that they are always postsillimanite. Shear bands are narrow and contemporaneous with the crystallization of biotite, muscovite, and chlorite. Sillimanite and garnets are usually broken and pulled apart along these shear bands. In quartzofeldspathic gneiss a considerable amount of cataclasis affected feldspar grains and internal deformation is ubiquitous in quartz whose grain boundaries are frequently sutured indicating inefficient recovery at low-T conditions. These observations suggest that the low-temperature left-lateral shear was predominant in the formation of the Red River Shear Zone in Vietnam.

Observations of practically intact high-temperature rocks from the flat-lying foliation zone in the center of the antiform reveals sillimanite within pressure fringes around garnets or as a foliation-forming mineral, suggesting that the flat lying foliation was formed at higher temperatures than the steeply dipping one. Figure 8 shows thin sections cut in the XZ plane

(perpendicular to the foliation and parallel to the stretching lineation). Samples taken from flat foliation outcrops in the core of the dome (Figures 8 and 9) show synsillimanite and postsillimanite shear, the former being more localized along narrow bands. Steep foliation outcrops instead show only postsillimanite shears. We thus tend to interprete the overall structure as off a rather flat high-temperature fabric dragged upward, reworked, and squeezed at low temperatures within a left-lateral shear zone. Our observations, whereby the majority of the shear structures along the Red River Shear Zone were formed in the greenschist facies, are inconsistent with the Red River Shear Zone being rooted at great depth. We suggest, instead, that shearing was localized in the upper crust and that the Red River Shear Zone was decoupled from deeper levels.

Radiochronological ages are now quite numerous in this region. The 4øAr/39Ar method on micas reveal slightly older ages in the west (between 40 Ma and 28 Ma) and younger ages in the Dai Nui Con Voi massif (between 34 and 21 Ma) [Wang et al., 1998; Mahtski eta/., 2000]. K/Ar method on hornblende separates from the Dai Nui Con Voi give ages of 26-28 Ma, and the biotite separates give ages of 24-25 Ma [Tran Ngoc Namet al., 1998]. Wang eta/. [1998] suggests that the left-lateral shearing event took place between 27 and 17 Ma. Harrison et al. [1996] showed a diachronous exhumation along the Red River Shear Zone from 24 Ma in Vietnam to 17 Ma in China from

thermochronology on feldspar. The Dai Nui Con Voi massif contains the highest temperature rocks of the left-lateral shear zone in Vietnam. These high-temperature metamorphic rocks were exhumed later than the orthogneiss farther west. Rather old ages in the Po Sen orthogneiss suggest that the transpressional deformation occurred quite early in the history

JOLIVET ET AL.: MIDCRUSTAL SHEAR ZONE IN INDOCtlINA 53

VNO 9846

post- sillimanite

•Jhear bands •

VNO 9846

postsillimanite shear band

_

VNO 9848

postsillimanite shear band

.

VNO '9848

shear bands

VNO 9851

pulled-apart sillimanite

VNO 9849

Figure 8. Drawing of thin sections of micaschists from the Dai Nui Con Vol massif. Sections were cut parallel to the stretching lineation and perpendicular to the foliation Samples extracted from low dipping foliation zones show synkinematic sillimanite fibers or prisms within pressure shadows on garnet or forming the foliation or cristallization along shear bands. Late more brittle shear bands cut through the high-temperature foliation. Samples from steep foliation zones show prekinematic sillimanite pulled apart within the foliation or the shear bands. Sample 9851 shows a mylonite with mostly a cold paragenesis with white micas, quartz, and biotite severely strained by numerous shear zones.

54 JOLIVET ET AL.: MIDCRUSTAL SHEAR ZONE IN INDOCHINA

9851 low-temperature deformation

left-lateral shear

(greenschist facies)

S/fl.

9845

s//L

high-temperature deformation (amphlbollte facies

synsllllmanite)

9846

9848

high-temperature deformation (amphibolite facies

synsilllmanite) +

low-temperature deformation left-lateral shear

(greenschist facies)

Figure 9. Position of the samples described in Figure 8 within the Dai Nui Con Voi dome. The steeper the foliation, the colder the deformation regime. Subhorizontal foliation in the core of the dome is associated with synkinematic sillimanite fibers, while steep foliation and shear bands on the limbs are postsillimanite. Vertical shear zones are associated with greenschist facies parageneses.

of left-lateral shear. Younger ages in the Dai Nui Con Voi massif between 33 Ma near the Red River Fault and 24 Ma near

the Song Chay Fault suggest a younger age for the exhumation of this massif. A normal component of displacement can be deduced along the Song Chay Fault from the juxtaposition of old (Song Chay massif) above young (Dai Nui Con Vol massif) exhumation ages. There was thus a transition in time from transpression to transtension along the Red River Shear Zone.

From the above descriptions we can draw the following conclusions: (1) the Red River Shear Zone was active at the same time as the extensional shear zone which exhumed the Bu

Khang dome. (2) The Song Chay dome was already at shallow crustal levels when the Dai Nuy Con Voi and the Po Sen orthogneiss were being deformed at depth. Only its final exhumation was achieved in the early Miocene. (3) The Mesozoic deformation in the Song Chay dome shows that a horizontal foliation was already present in the crust of the Indochina Peninsula before the inception of the Red River Shear Zone. (4) The left-lateral strike-slip deformation along steeply dipping shears seems to be everywhere at low temperature and low pressure. (5) High-temperature deformation is always associated with a subhorizontal foliation (in the Bu Khang dome and in the core of the Dai Nui Con Voi dome). (6) Domes formed before and during the left- lateral strike-slip motion. (7) The strike-slip motion was

transpressional during a first period (t?om ? 40 to 25 Ma) and transtensional afterward (fast exhumation from 24 to 17 Ma). A change in the boundary conditions occurred some 25-30 Myr ago when the system became globally extensional after a period of transpression.

3. Discussion and Conclusions: Horizontal

Shear Zone at Midcrustal Depth

The exhumation history of the Dai Nui Con Voi metamorphic rocks shows that they were first deformed at depth along an horizontal shear zone and later reworked by steeper shear zones. These observations suggest that the strike-slip shear deformation along the Red River Shear Zone did not affect the deep crust. Instead it remained localized in the upper crust and in the brittle-ductile transition above a lower crustal domain where horizontal shear zones predominate (Figures 10 and 11). Exhumation of the high-temperature rocks was the consequence of an oblique component of slip along the shear zone and/or of the extensional component of the Song Chay Fault to the east. Figure 10 shows a possible geometry at depth during the activity of the shear zone. According to this scenario the upper crust is cut by pure left-lateral faults and oblique slip extensional faults. These faults root at depth in the brittle-ductile transition zone within ductile shear zones

JOLIVET ET AL.' MIDCRUSTAL SHEAR ZONE IN INDOCItlNA 55

Dai Nui Con Voi

Red River Fault Song Chay Fault

brittle crust

brittle-ductile

transition

d•collement

ductile crust

greenschist

amphibolite facies

mylonitic zone

Figure 10. Three-dimensional diagram showing a possible geometry at depth within the brittle-ductile transition zone during the activity of the left-lateral shear zone above a midcrustal subhorizontal shear zone.

40-33 Ma

33-30 Ma

30-5 Ma

SW

•brittle crust

left-lateral t,ran spression l

I onset of left-lateral transtension

NE

Bu Khang dome Po Sen

orthogneiss RRF Dai Nuy Song Chay orthogneiss

Vol d•ne and other indosinian structures

Figure 11. Evolutive cross sections through north Vietnam from the Eocene to the Pliocene. During the transtensional deformation steep strike-slip shear zones are active in the upper and middle crust and root within a subhorizontal shear zone at midcrustal depth.

56 JOLIVET ET AL.' MIDCRUSTAL SHEAR ZONE IN INDOCHINA

active in the greenschist facies. These steep shear zones, in turn, branch into an horizontal shear zone which separates the upper and middle crust from the lower crust. High-temperature deformation may have been active along the horizontal shear zone and below. Exhumation along the eastern fault, driven by the extensional component, forces deep rocks to rise higher in the crust and to enter the brittle-ductile transition where they were reworked by greenschist shear zones, thus forming a composite elongated domal antiformal structure.

The presence of foliation domes within the strike-slip shear zone can be explained if strike-slip shear was confined to the upper part of the crust and reworked the domes which were formed below in the lower crust. The Bu Khang dome was exhumed within a purely extensional context, and the Dai N ui Con Voi in the transtensional shear zone (Figure 11).

The earlier transpressional episode in north Vietnam and the resulting crustal thickening may have been the cause for the relatively high pressures recorded in the Bu Khang dome as attested by the presence of kyanite in metapelites and garnets in metabauxites if the Oligocene age of the high-pressure event is confirmed. A simple scenario can be proposed to account for the tectonic evolution of this region. Crustal thickening occurred in the Eocene in response to the India-Asia collision and the Red River Shear Zone first formed as a transpressional strike-slip zone above the thickened lower crust. A similar geometry was proposed by Wang and Burchfiel [1997] for the Chinese portion of the Red River Shear Zone. This first episode lasted until the early Oligocene when the regime became extensional. The change from transpression to transtension can be tentatively attributed to two possible causes: either the thickened domain was extruded and reached more eastern

regions where extension had been prevailing during the whole collision, or the boundary conditions suddenly changed to extensional when the South China Sea started to rift.

If our interpretation holds, the Red River Shear Zone would not be rooted in a strike-slip shear zone in the mantle but in an horizontal shear zone at midcrustal depth during both the transpressional and the transtensional episodes. We have no clue as to whether or not this horizontal shear zone branches

on an offset vertical strike-slip shear zone in the lower crust and mantle. In any case our observations plead for caution in interpreting crustal structures in terms of lithospheric kinematics in the continental domain. In other examples of large-scale intracontinental strike-slip faults interpreted as shallow structures, a component of oblique motion is often present, either transtensional or transpressional. Oblique transpression along the southern boundary of the Carribean plate is expressed at the surface as en echelon folds and the dextral El Pilar Fault, which accomodates only the purely strike-slip component of displacement in the upper crust. The El Pilar Fault is interpreted as a shallow structure [Passalacqua et al., 1995]. A similar interpretation has been published about the San Andreas Fault [Jones et al., 1994]. An oblique component of motion is ascertained along the Red River Shear Zone, changing with time and along strike, either transpressional or transtensional. Understanding the changing kinematics of Indochina with respect to South China requires that the oblique component is quantified.

Acknowledgments. The French members of the research team are very much indebted to their Vietnamese colleagues from the National University of Vietnam in Hanoi who provided considerable help dunng field work.

References

Bnais, A, P. Patnat, and P Tapponnler, P Updated interpretation of magnetic anomalies and seafloor spreading stages •n the South Chll•a Sea. hnpllcat•ons fol the ternroy tectonics of SE Asla,.J (ieophy,• Re.•. 98, 6299-6328, 1993

Burchfiel, B. C, Q. Deng, P Molnal, L Royden, Y Wang, P Zhang, and W Zhang, Intracrustal detachment within zones of continental

deformation, Geology. ]7, 748-752, 1989 Davy, P, and P. R. Cobbold, Indentation tectonics

•n nature and experiments Experiments scaled for gravity, Bull (;eol m.•t (Jp.•alla. 14, 129- 141, 1988

Dewey, J F., S Cande, and W C ! P•tman, Tectonic evolution of the India-Eurasia

collision zone, Eclogae GeM He/v. 82, 717- 734, 1989

England, P C., and D P McKenzie, A th•n VISCOtlS sheet model tbJ continental

deformation, (]eophy,• ./ R .4sO'on Soc, 70, 295-321 1982 ,

England, P., and P Molnar, Active defbnnat•on of Asia: Frmn kme•nancs to dynamics, Scte,ce, 278, 647-650, 1997

England, P C., and G A HouseJnan, F•n•te Stl'a•n calculations of cmmnental deformarran, lI, Apphcatlon to the In&a-Asia co!lJs•on, ./ (ieophy.¾ Re.s, 91, 3664-3676, 1986

Hm-l'•son, T. M., W Chen, P H Leloup, F J. Ryerson, and P. Tappomuel', An early Miocene transmort m deformation regime w•th•n the Red R•ver fault zone, Yunnan, and •ts s•gn•ficance for Indo~As•an tectonics, ./ (ieophvv Re,s, 97, 7159-7182, 1992

Hamson, T M, P. H Leloup, F J Ryerson, P Tapponmer, R Lacass•n, and W Chen, Diachronous •n•Uat•on of t•anstens•on along the Allao Shah - Red River shem zone, Yunnan

and Vietnam, ill 7he 7'eelonto /;volutton o/ .4sta, edited by A YJn and T M Harl'•son, pp. 208-226, Cambridge Un•v Ness, New York, 1996

Huchon, P X Le P•chon, and C Rang•n, ,

Indochina pelunsula and the co!!•s•on of Ind•a and Eurasia, Geo/ogF, 22, 27-30, 1994

Joltvet, L, P Davy, and P Cobbold, R•ght-lateral shear along the ninthwest Pacific margin and the II•dia-Euras•a colhs•on, 7betonrex; 9, 1409- 1419, 1990

Joltvet, L C Faccenna, N d'Agost•no, M Fotl•lller, and D Wo•ra!!, The k•nematlcs of marginal basins, examples flora the Tyrrhen•an, Aegean and Japan Seas, 7'ectontc.s, edited by C Mac N•ocmll and P D Ryan, pp 21-53, Geol Soc, London, 1999a

Joltvet, L, H Malusk•, O Beyssac, B GolfS, C Lepw'•el', Phan Truong The, and Nguyen Van Vuong, The Ol•go-M•ocene Bu Khang extensional gneiss dome in North V•etnam, geodynam•c •mpl•cauons, (ieo/ogy, 27, 67-70, 1999b

Jones, D L R G•aymer, Ch• Wang, T V ,

McEvilly, and A Lomax, Neogene transpressive evolution of the Cahfornla Coast Range, 7•ctonws, 13, 561-574, 1994

Lacassm, R, H Malusk•, P H Leloup, P Tapponnier, C H•nthong, K S•bhakd•, S Chuawroj, and A Charoenlavat, Temaly dlachron•c extl'us•on and detbnnat•on of western lndochlna St•ucttual and 4øAr/a9Ar ewdence fi'om NW Thailand, ./ (ieophy.s. 102, 10,013-10,037, 1997

Leloup, P H, and J R Kinhast, H•gh-temperatm'e metamorplusln •n a major strike-slip shear zone The Afiao Shm•-Red River, People's Republic of China, Earth l'/anet Sc• Left, /18, 213- 234, 1993

Leloup, P H, R Lacassln, P Tapponmer, U Scharer, Zhong Dalai, L•u Xlaohan, Zhang Llangshang, Jl Shaocheng, and Phan Trong Trmh, The A•!o Shah-Red River shear zone (Yunnan, China) Ternroy transfonn boundary of Indocluna, 7•clOnOl•tW.wc.•. 251, 3-84, 1995

Leloup, P H F Roger, M Jolt vet, R Lacass•n, T ,

Phan Trong, M Brunel, M, and D Seward, Unravelling along and cotnplex thm•nal history by •nult•system geochronology Example of the Song Chay metamorphic Dome, North Vietnam, paper presented at EUG 10, Strasbourg, France, 1999

Malusk•, H, C Lepv•e•, Tha %ong Thang, and Nguyen Duc Thang, Effect of Up-Doming Proces •n the Song Chay Mass•f V•etnam) on Ar-Ar ages, paper presented at EUG 10, Strasbourg, Flance, 1999

MalusM, H, C Lepw•er, L Jol•vet, A Carter, D •"-Røques, O Beyssac, Ta Trong Tang,

Ngfiyen Duc Thang, and D Awgad, At-At and fission track ages •n the Sollg Chay massif, Indosiman versus Cenozoic tectomcs •n

northern V•etnmn, .1 ,$'oulhea.st A.•an Earth Sc•, •n press, 2000

McKenzie, D, AcUve tectomcs •n the Med•tenanean region, (ieophy.v .] R Awron Soc. 30, 109- 185, 1972

Mfit•v•er, F, Volumes sfid•menta•res et b•!ans de masse en As•e pendant !e Cfinozmque, Ph D thes•s, Umvers•tfi Dems D•derot, Pans VII, Pm•s, 1996

Mfit•vmr, F., Y Gaudemer, P Tapponmer, and M Klein, Mass accumulation rates •n Asia during the Cenozoic, (ieophy•'. ./ Int. /37, 280-318, 1999

Molnar, P, and P Tapponn•er, Cenozoic tectomcs

JOLIVET ET AL.: MIDCRUSTAL SHEAR ZONE IN INDOCHINA .5'7

of Asia Effects of a continental collision, So'tent'e, 189, 419-426, 1975

Nagy, E A, U. Scharer, and NguyenTlung M•nh, Olg•o-Mlocene magmat•sm In central Vietnam and lmplicaUons for continental deformahon •n Indoch•na, 7'erra Nova. •n press, 2000

Passalacqua, H F Fernandez, Y Gou, and F ,

Roure, C! ustai arch•tecttn e and st• ain partitioning •n the eastern Venezuelan Ranges, •n Petroleum Ba.•tn,v o/ Soul/• America, edited by A J Tankard, R Suarez Soruco and H J Weisink, AAI'G mere. 62, 667-679, 1995

Tapponn•er, P, and P Molnai, Slipline field theoly and lmge-scale continental tectonics, Nature. 264, 319-324, 1976

Tappomuer, P, G Peitzer, A Y Le Dain, R Aimuo, and P Cobbold, Plopaganng extius•on leclolllCS ill As•a New insights fi'om s•mple experiments with plaslicine, (ieo/ogv, 10, 611- 616, 1982

Tapponmer, P, G Peltze•, and R ^rmtlo, Oil tile mechanics of the collision between India and

Asia, In ('o//t•ton 7'eclon•c•, edited by M P

Coward and A CRles, (leo/ Soc ?,'pet Pub/. 19, 115-157, 1986

Tran Ngoc Nam, The Day Nui Con Voy-Red River Shear Zone m V•etnam Deformat•onal kinematics, P-T-t patIls and tectonic •mplications, Ph D tilesis, Un•v of Tokyo, Tokyo, 1997

Tran Ngoc Nam, M Tonumi, and I' Itaya, P-T-t patIls and post-metamorphic exhumation of the Day Nui Con Vol shem zone In Vietnam, 7'eclonophyvtc.•, 290, 299-318, 1998

Wang, E, and B C Bu•chfiel, Inteq3retation of Cenozoic tectonics In the •!ght-latm al accommodation zone between the Allao Shah

Shear zone and the eastern Hnnalayan syntaxis, Int (leo/ Rev. 39, 191-219, 1997

Wang, P L C H Lo, T Y Lee, S L Chung, C Y Lan, and T Y Nguyen, Thermocluonolog•cal ewdence for the movement of the Allao Shah-Red River shear

zone: A perspective fi'om Vmtnam, (ieo/ogy, 26, 887-890, 1998

D Avigad, Institute of Eaith Sciences, The Hebrew Un•versity of Jerusalem, Jerusalem 91 904, Israel (awgad@vms ht[l• ac fi)

O Beyssac and B Goff6, Laboratoire de G•olog•e, UMR 8538, Ecole Normale Sup•r•eure, 24 rue Lhomond, 75231 Paris cedex 05, France. (beyssac@euclase ens fr, goffe@euclase ens.fi')

C Lepvner and L. Jolt vet, Laboratmre de Tectonlque, ESA 7072, Universitd P•erre et Mm-•e Curie, T26-0 El, case 129, 4 place Juss•eu, 75252 Pails cedex 05, France (claude lepvner @lgs jussleu.fr, laurent 1ol•vet@lgs lussieu fr)

H. Maluski Laboratmre de Gfochronologie G•5ocNn-ue P•Strolog•e, UMR 5567, Umvers•t•5 Montpell•er II, Place Eugene Bataillon, 34095 Montpeilier cedex, France (malusk•@dsm.umv- montp2.fi')

Ta Trong Thang Labo•atoiy of Geology, National Univeisity of Vietnam m Hahre, Hanoi, 334, nguyen Tral, Thanh Xunn Dong Da, Hanoi, Vietnam

(Received March 20, 2000, revised July 13, 2000, accepted July 31, 2000)