OC EANO LOGICA ACTA , 198 1, N'
sp
Eaf----
Accretion, underplating , subduction
and tectonic evolution,
Middle America Trench,
Southern Mexico ••
results from DSDP Leg 66
Arc-trench
sy~ lt:m s
Middle America Trcnch
Accre tion
OHscrnping. I,lnderplating
Gas hydraTe
Système arc-fosse
Fosse d'Amérique Centrale
Accrétion, arrachement
Coin cemen t sous plaque
Gal hyd rat é
J . S. Walkins · . J. C. Moore ~. T. H. Shipley c. S. B. l3achman d. F . W. Beghtet c. A. BuU 1.
B. M. Oidyk '. J . K. Leggeu', N. Lundberg ~, K. J . McMillcn', N. Niitsuma l , L. E. Shcpherd J, J. F. Slephan t • H. Strado.:r',
• Exploration and Production Divis ion . Gulr Science and Tec hnology Company. P.O .
80x 2038. Pittsburgh , Pennsylvania 152 30. USA .
i> Deparlmenl of Earth Sciences. Uni versity of California, Santa Cruz, California 95060.
USA .
c Unive rs it y of CalifoTnia, San Diego, La Jalla, California 92093, USA,
d Cornell University, Ithaca , New York 14853, USA .
• Phill ips Research Center, Bartlesville , Oklahom.\ 740(}4, USA.
1 Geologyand Palaeontology Ins titute, T Übingen University, Sigwartstrasse 10.74 Tübingen ,
FRG .
1 Research Development Labortltory, Empresa Nadonlll dei Petroleo. Concon, Chi le.
h Department of Geology, Imperial College of Science and Technology, South Kensington ,
London SW72AZ, UK.
1 I ns titute of Geosciences, Shizuoka Univers ity , 8360h ya, Sh izuoka City. Japan.
1 Department of Oceanognlphy, Texas A & M Univers ity. College S tation. Texas 17843,
USA .
k Department de Géotectonique. Uni ve rsité Pierre et Marie Curie. 4. place Jussieu. 75()()5
Paris, France.
1 Geologie Bundesa nstal t. Vienna. Aus tria .
ABSTRACT
During Leg 66 eight sites were driJJed 10 form a tntnseCI across the Middle America Trenc h off
southweSlern Mexico. Cores from these sites show tha! acc retion beg:III'lpproximately 10 MY
ago and has continued to the present. Accretion began wilh offscraping folJ owed by a 2- to
4- MY episode of folding and faulting wilh uplirt rates o f 400-500 rn/MY: uplif' then s lowed to
100-200 rn/MY, and seismicall y resolved deformatio n ceased as the wedl!e <lppeared to rise
evenly. App roximalely 33 % of the sediment flux input into the subduction zone. mainly
trem; h sand and slump de posÎts, is scraped off and incorponl1ed into the toc of the lower
slope : an addilional 33 % is inilially subducted but then peeled off 10 underplale the
acc re tio nar y wedge : the remaining 33 % is s ubducted landwllrd beneath the overhangi ng Hp of
continental c rust. Although we find no direc t evidence o f Icc lonic erosion. the la rge amount of
sediment s ubducted makes tectonic e rosio n fea s ible.
Pla te reorgani zation and the onse! of s inis tml s!rike·slip raul ting roughly 23 MY ago are
thoul!h l to be respons ible for the trun caled appeam nce of geologic basement features o nshore .
Arte r strike-sl ip (llulting, the ocean transgressed the margin 22 MY ago , and the ed ge of the
conti nental crust subsided rapidly, reaching a depth al o r slighl ly I! reater than the carbonate
compensation depth 19 MY ago. Thereafter, the c rust hns risen s lowly al a rate of about
213
J, S. WATKINS el al.
90 mlMY. T he onset of accre tion 10 MY ago probabty marked the end of str ike-slip faulting
and the beginning of the oblique convergence observed today.
Oce(mol. ACIfl. 1981. Proceedings 26th International Geological
continental margins symposium. Paris. Jul y 7-17, 1980. 213-224.
RÉSUMÉ
Congre~!>.
Geology of
Accrétion , plongement de plaques, subduction et évo lution (eclOniq ue de la
tosse d ' Amérique Centra le, Mexique mér idional: rés ulta ts du Leg 66 DSD P.
Les huit puits foré s [ors du Leg 66 sont si tués sur une tnmsversa[e de la fosse d'Amérique
Centrale au large du sud-ouest du Mexique méridional. Les carottes récupé rées montrent que
["accrétion a débuté il ya 10 mA environ et a continué jusqu'à l'Actuel. L'accrétion a débuté
plU un ripage en avant. suivi pendant 2 à 4 mA d'une phase de plissement et de faillage avec
des taux de !;oulèvement de 400-500 mimA; la vitesse de soulèvement s'est ensuite ralentie
jusqu'à 100-200 mA, et la déformation responsable des séismes fi cessé lorsque le prisme s'est
aussi soulevé. 33 % environ de rapport séd imentaire dans la zone de s ubductio n, sous forme
de grès et de dépôts glissés. sont arrachés et incorporés â l'extrémité de la pente inférieure;
33 % o nt subi un début de s ubduction, mais o nt été décollés et coincés sur le prisme
d 'accrétion: les 33 % rest,lnt étant subductés sous le rebord de la croüte continentale. Bien
qu'il n'y a it pas d'évidence d'érosion tectonique, la grande quanti té de sédiment subd uc té rend
ce phénomène envisflgeable.
La réorganisation des plaques et l'insta uration (l'un système décroc hant senestre il ya 23 mA
sont probablement res ponsables de la troncature appa rente des traits géologiques d u socle
émergé. Après le décrochement. l'océa n a transgressé la marge. il ya 22 mA. c t la bordure de
la croûte continentale a subsidé rapidement. atteignant une profondeur au moins égale à la
CCD il ya 19 mA. Par la suite. la croüte s'est soulevée lentement â un taux de 90 mimA. Le
début de l'accrétion il ya 10 mA a probablement marqué la fin d u Jécrochement et le début de
la convergence oblique observable actuellement.
Ocemwl. Ac/a. 1981. Actes 26' Congrès International de Géologie, col loque Géologie des
marges continentales, Paris. 7-17 juil. 1980, 213-224.
I NT RODU CTI ON
relatcd deformation of trench !>ed iments could explain the
abrupt disappearancc of seismic reflec tors observed at the
foot of the slope in seismic trave rses across turbid ite filled
trenches ,md inner trench wlilis. Dickinson (971) suggested
that although oceanic base ment is carried down with the
descending lithosphere, lighter sediments probably arc scrapeu off against the overriding plate. These sediments
combine with ophiolitic sc raps to form mellmges. Dickinson
further observed tha t the tectonic thickness of a melange
zone depend~ on the amount of turbidite sediment fed into
Ihe subduction zone.
Von Huene (1972), with improved seismic-refJection data.
observed oceanic erust and overl ying unddormcd pelagie
sediments extending a~ much as 12 km landward of the
AleUl ian Trenc h beneath deformed slope sed iments.
Although thrust faull ~ were nOI clearly evident in the slope
seismic data, von Huene' s intcrpretation induded;l master
th rust fault deep in the se cond oceanic layer. Beck'~ (1972)
multifold seismic ~ection across the 1.. va Treneh near Bali
showed undeformed oceanic crust gently dipping beneath
the inner wall for a distance of about 40 km land ward of the
trench. Seismic fealUres in the region above oceanic crust.
though s trongl y hypcrbolated with d iffrac tio ns. suggested
land ward dipping reflect ions.
Objectives of the Leg 66 d rilling off southwestern ,\ilexico
were investigations of :
1) the accretionary mechani~m induding " test of the
imbricate th rust model of Seely e/ al. (1974) ;
2) sediment subduction and tectonic e rosion of the ovcrriding plate during the subduction process : and
3) the geologic his tor y of the margin with particular emphasis on the periods immediately before. during, and immediately after the onse! of subduction.
This parer foc uses on the processes of mas~ addition. We
u~e " accretionary prism" o r wedge 10 refer to ait material
that is tectoniC<lJly transferred from the lower to the upper
plate durins cru!> t<ll convergence. "Offscraping" describes
lIcCretioll (princi pn!ly of trem:h sed iments) at the base of the
trench stope. "Underplalins" des ignates a process of mass
addition at depth beneath the rind of off ~craped deposits. 1n
our usage. the term accretion refers to mass addition in
general. whereas ofhcraping and underplalins indicate a
particular tectonic setting for the accretionary process.
Accretion
Dickinson (1973) strengthened the argument for acc retion
with the observation that, with few exception~. widths of
arc-trench gaps are proportional to ages of the re!>pectÎve
island are-trench !>yslems. This correspondence resulted. he
The concept of accrction evolved during the la te 1960'~ and
early 1970'5. Seyfert (1%9) fir!>t pointed out that acerelion
214
RESULTS FROM DSDP LEG 66 ON THE MIDDLE AMERICA TRENCH
1
ufllm
......-'"
UtHA
---
Figur~
1
Indtx mop IJ/ DSDP L,S 66 Ir(JIIS«1 l'ho •..illg
sites. blJlhymtlry, ulld Il/CUliOIl of stinnir. dala
(dushtd Ii"'ts) ustd 10 tflimalt flux Qf Irtnc:h
IUrbidiltS alld octanic ptlagltes atld htmiptlagi·
te.r bting acatltd und/or subduc:/td. Cmrlmlr
inltr\'(11 ;$ 100 m : J()(}(J-", COlllours ruughl)'
oullint II/ rb/diu pOllds ln Ili t Irtl/c h.
"
t
:Clt~:~:--r---.
1
1
1
:
~ ,:
/
JO
40
~:
"
Subduction croslon
thoughl. from s te<ldy accretion of crustal materials by the
inner wall.
DSOP Leg 31 drilled Ihe toc of the slope landwa rd of the
Japan Trench oCt the is land of Shikoku in mid-1973. Cores
recovered s howed Iha t Pleis tocene Irench sedimen ts had
been compressed to about half their original volume and
given a d is tinct cleavage wilhin 6 km of the tre nch. Elevalion of t hese sedimenls to 300 m above the trenc h rloor
(Karig tt al. , 1975), provided furl he r evidence of accre tion.
Seely t t al. (1974) inlegrated the foregoing observations in
t heir interprelation of mullifold, common-depth-point seismie-reflection data collected off Guatemala and Oregon.
inferring thal Ihe inne r wall comprised :
1) li surficial a pron of undeformed sediments overlying:
2) il zo ne of ddormation contllining landwa rd dipping
reflee toTS (LO Rs) lIlI overlyi ng :
3) relative/y undcformed , gently landward-dipping ocellnic
crus!.
Seely et al. suggesled rOlation of LORs during accretÎon
wi th the result thllt dip increas ed with increasing elevlltion
within the seclion. They inlerpreted thi s mechanism 10
mean that older upper melange sli ees rotated upward as
younger slices were progrcssively emplaeed by unJerthrusling ( Fig. 2).
Not ail o ld trenc h systems have wide arc-trench gaps , a nd
sorne appear 10 have anomalo usly small accretionary wedges for their inferred ages. Did these systems e ve r ha ve
larger wedges ,., If so , where are t hey now ,., And if Ihey did
nOI have wedges, why no t ,., Tectonic erosion has been
s uggesled as the ans we r.
Hussong et 01. (1976) eXllmined perhaps the beSI example in
the Peru-C hile trenc h belween 8Q-12 ~S llltilude. Here, volcanic Sludies on land s how evidence o f subduc tion as old as
mid-Mesozoic. Mu hifold seis mic-reflection lmd detailed
seismic-refract ion sl udies. however, sugges t a na rrow
(l O-km ) lIccrelionary wedge compris În& low-velocity material (2.0-3.5 km/sec.) that is currently bci ng uplifted. Immcd ialely Jandward, Ihe lower s lope consis ts of a low-velocity
ap ron of undeformed sediments underlain by rocks with
velocities of 5 km/sec., velOCÎties more consislent with
t hose of continental meta-igncous rocks than wi t h deformed
trench turbidites. Hussong et (lI. lliso rou nd evidence of
normal fa ulting within the 5 kmfs ec . zone. They intcrprct
Ihese data alii indicaling nOI only tha t are sediments bcing
Sedimt nt s ubduction
The previously mentioned in veSligaiions, ahhough providing llbundan t eviJence of accretion. did not precluJe
subd uction of ft significant frac tion of the pelagiehemipelagic-Iurbidite fl ux. Von Hue ne et 01. (1980 b) nOled
that calc ulations SUgges t subduc tion of 80 % of the s edimentury in put inlO the Japan Treneh. ~I nd Ka rig and Sharrnan
(1975). in their rcvicw of t he ace retionary mec hanis m,
suggeslcd Ih:1I sorne of the sedime ntary coye r rnay reac h
cons iderable dept hs before s hearing off. Res ults of Leg 67
in (he Middle America Trench off Gualemala s trongly
suggesl sediment s ubd uc tio n (von Hue ne et a/ .. 1980 (J).
Figure 2
The imhricalt ul/der/hruSI a cc relltmar)' model of Sul)' el al. (1'.174).
Sedimtll/$ ({Ial ./yil/Il seRmtnt /0 lefl) art progressr.'e!y scrafJed uff
Ihe underlying ucea nle Cri/SI und ;ncorpurated Into l/at accreliQlI(lry
...edgt /Q tlle righl. /,Ut r/;OIr of each ne ......tdge ililhe hul/UIII of Ihe
Sltlck tlto'altS w rd fu/tlles uo'trlying ...edges ...iO, Ore resull Ilrol
major bOlllrdllries II/Id III/unoi Stra/illroplr)' (possibl)' LDRs)
/Jfl/Krusil·,tly Mupen in difJ r4/lsllltJf .
2 15
J . S. WATKINS el al.
Onshore , the Guerrero-Oaxaca massif underlies post-Paleozoic rocks (Alvarez. 1949 : Guzman, 1950: de Cserna,
1965). This massif. comprising rocks of Precambrian 10
Paleozoic age, appears 10 have undergone se veral period s of
metamorphism. A few narrow. northwesterly trending belts
contain calcareous miogeosynclinal and e ugeosynclinal
rocks of Triassic age. Jurassic seaways Iransgressed across
the massif after uplift and fraclUrin g in the Triassic .
subducted bUi also tha! the subduction process is eroding
the leading edge and unde rside of the continental cruSI and
sub<!uctÎng continental crus!. a process called subduction
e rosion by Scholl el al. (1980).
The ~texi ca n margin in the Leg 66 area resernbles the
Peru-Chile rnargin in sorne respects. L:md Sfudie s suggest
that lower Tertiary and Cretaceou;; volca ni c arcs lay much
doser to the trench than does the present volcanic axis. As
in the Peru-Chile Trench. seismic-reflec tÎon d<lta suggest an
anomalously s mall accretionary wedge considering the probable age of the subduction . Seismic e vidence of subduction
erosion. however. is less extens:ve in the Leg 66 region than
in the Per u-Chile Trendl.
Southern Mexieo experienced major orogcny in Llle e retaceous and Eocene times including emplacement of volca nic
sequençes and intrusion of batholiths. The orogeny c ulminated in folding of the massif and the overlying sedimcnta ry-volcanic sequence. Volcanie activity along the mlugin
continued through the lower Tertiary but appears to ha ve
been concentrated along the trans-Mexiean volcanic belt in
post-Oligocene time.
Offshore the upper s lope dips gently () O) for an average
distance of ro ughly ]5 km . Here , al a depth of abou t
2.000 m, the dip increases and the lower slope fall s away <II
~ relalively steep di p of slightly more than 8
Maximum
de pths of 5.0-5.3 km occ ur in Ihe tre nch a xis at a dis tance of
50-60 km from the shoreJine .
Seismic-reflection data. seismic-ref raction data. and dredge
hauls obtained during site surveyÎng for Leg 66 (S hÎpley
et ai., 1980) suggest that continental crust exte nds seaward
from the s hore for a dislance of 25-30 km (Fig. ]), or al most
Geologie his tory
In vestigations of trenehes and subduction zones generally
provide very liule information about the onse! of subduction because material involved in sedimentary and teetonic
events occurring at this time is rapidly buried to depths
inaccessible to s h<lllow coring investig<ltions, while simultaneous deformation renders s tructures seis mically indeciphemble. By the time a melange is uplifted and exposed by
erosion, multiple tectonic overprinting makes identificalion
of early history diffic uh, if nol impossible.
A knowledge of the carly his tory would be useful in the
understandi ng of the subduction mechanism. For example.
if s ubd uc tion is a convective process, why is reJatively
young. lower density oceanic c rus t subducted in the eastern
Pacifie while much older, denser crUSI appears stable along
No rt h Atlantic ma rgins '! Know ledge of the early history
ma y shed light on (he up per slope break as weil. Karis and
Sharman (1975) observed Ihal the upper slope topographie
bre<lk usually separates a zone of uplift, including Ihe ac ti ve
pari of the: a<.;",n::t;on/iry wcd~e:. from a zone: of 5 u b~ i de:",;:c
encompassing a forearc basin. SeeJy (1979) s uggesled that
the upper s lope break represents t he topographie expression
of Ihe initial rupture in the oceanic crus t al the lime
subduction begins.
Pre-drill ing stud ies in the Leg 66 area (Shi pie y el (lI .. 1980)
suggested that the present e pisode of subduct io n and
accretion in the Leg 66 area is relalively young and thal
drilling might be able 10 reach rocks of facies and s tructures
inaccessible in other afeas. Thus a major objective of the
Les 66 drilling program was the samplins of sediments
closeSllO the continental crust boundary , sedime nts thought
to ha ve been deposited during the earlies i s tages of s ubduclion in the area.
Q
•
Figure 3
Seismic. /illl'$ Il,,o u8lr 1,1'866 sires (rmmigrated lim/' l·ecIiO/rs).
CONTINENTAL
.....
•••
...
,.
,
TRENCH -
al U TMS I li" e OM7N Ihrough si/t's 487. 486. 490. 489 (urd 493
sho"'ilrll major /eawres (a/ur Slriple)' et al. . 1980);
..
GEOLOGIe SETTI NG
The Leg 66 area of the Middle America Trench lies within a
segmcnt bounded by the T ehuantepec Rid ge 10 the souIhe<ls t and the Ri vera Fracture Zone to the northwest. Thi s
trench segment has characteri~lics OItypical of mosi other
active trench s ys tems. The trench is relati vely shallow, the
d istance from aJl trench axis to the shoreline is narro w.
continental s lrUClUres of PreC'lmbrian to Mesozoic age
appear truncated (Karig. 1974 ; Karig el (II .. 1978), and the re
is no fo rearc basin. Earthquake hypocentral dep ths do nOI
exceed 200 km and define a diffuse . shallowl y dipping
Benioff zone (Mol na r . Sykes, 1969). T he present-da y voleanic arc occurs 200-]00 km inland along an <lx is markedly
divergent from the trenc h-3xis trend . Present-dav convergence is about 7 cm/yr at the driJling locality (Mi nstcr.
Jordan. 1978).
b l UTM Sl lille fl,IX16 tlrrouglr lilt'l 488. 491. l.md 492 slro k'illll
m/lh{)r seismic Slrucrrrres (a/lU Slriplty el al.. 198{J).
216
RESUL TS FROM DSDP LEG 66 ON THE MIDDLE AMERICA TRENCH
out JO the mid.s lope break. Seismic velocities calc ula ted
from sonobuoy refractio n s tudies s how that the inferred
conti nental basement rocks gencrally have velocities of
4.0 km/sec. or greater. Local near-surfaee velocities of
3.3 km/sec. probably indieate weathering or fraeluring.
Landward dipping reflec tors (LORs) similar 10 those repor·
led by Seely et al, (1974) oceur within the upper part of the
lower slope.
Seismic velocilies in the wedge are as greal as 3.5 km/sec.
Veloeities of sed iments in t he trench axis a nd oceanic
pelagic·hemipelagic sed iments on the oçeanic c rus t seaward
of the trençh slope are 2.0 km/sec. or less . Seismicrefleç tion data near the toc of the slope and in the trench
suggest t hrus t faulting . A hemipelagiç apron, wi th typical
velocities of 1.7-1.9 km/sec.. extends from s ho re
downslope . I n par ts of the aTea, the apron extends almos! to
the base of the s lope; in other a reas. it is seis mically
resolved only on the uppe r slope and uppermost lower
s lope .
A seawa rd dipping rcflec tor. subparallel ro the seafloor and
increas ing in dep lh beneath the seafloor as the seafloor
deepens. occurs beneath much of t he lower slope . This
rdleeto r appears 10 c ross-cuI LORs. Orill ing resul ts ind icale that this reflector represents the base of a gas hydrate
zone (Moore el al.. 1979 a : Shipley el al.• 1979).
ORILLI NG RESULTS
Wc drilled eight sites along Il transe c t perpendicular to t he
Mexican margin in the vicinity of lSoN, 99"W. Figure 1
s hows locations of dr ill s ites a nd location of essential
seis mic control s uperimposed on generalized bathymetry.
Figure 4 shows generalize s tratigrap hie columns, and
Figure 5 shows a composite c ross sec tio n inferred from
seismic·refteçtion data and dr illing data.
The drill sites fall into three groups according. to drilling
obje ctives:
Trench tur bidite ponds oçcur d iseontin uously ulong the
treneh (Fig. 1 and 3). T urbidite thickness ranges fOTm
seismically unresolved « 50 m) to 625 m and averages
270 m . The pelagic-hemipelagic section immediately ubove
the oceanic crUS! and beneath the trench turbidites ranges in
thickness from seis mically unresolved to 225 m when mea·
s ured along the trenc h IIXis ; ils thiekness ave rages 11 0 m
a long a seismiç line 50 km seaward of. and parallel to. the
trench .
1) oceanic and trench reference sites dritred to sample
ocea nic pelagÏte-hemipelagite and trench tu r bid ite sections .
respectively ;
2) lower and mid·slope sites drilled to sample the acerel io·
nary prism ; and
3) upper slope s ites drilled to de termine the natu re and
extent of the continen tal crust and to document the sedi·
men tary and vertical tectonic hislory of the tegion .
-
.......
. .............
--
Figure 4
Stratigraphie co/um s sho..-iug
rn llllS 01 Leg 66 drilling (alur
Moore et al. . 19793).
•
-
0 .. .-<.. '
(.....j ---- EJ ...."
UD ....,--- _
.......
. .........,
Figure 5
Campasile structu re cross sectil)n baud
(In drilling rnullS and migra led de pth
sectians al UTMSI seismic Un es OM7N
and MX16 (Maure et al.. /979a).
~~
~~
-....... g
-~
'- .-~~'
.....
................
r
K '""",,,,,••
....,... .. . ~.,~ -- f··:::·::·~·:· -:-;: : : : : .. ~
" --,. ;'\
', ,-\',
.,.
""OCM
""'''
, . : : : : : ':":":"~~:.,'~~: : : : : : :
""'.
o
"
..,......".c'''''''
~
..-, ,
'0
"
217
,. ....
•• "
•.•••• , . , . "
.... ,.,., .. "."
"
J . s. WATKINS el si.
Siles 487 a nd 486 comprise the ocean;c-trench reference
gro up. Site 487 penetrated 11 5 m of Pleis tocene to late
Pliocene hemipelagic mud overl ying 55 m of Pliocene 10 late
Miocene brown clay before bouomi ng in b ••salt. The hemi·
pelagic mud evÎdently deri ves (rom the Mexican ma inla nd
from whence it ;5 tra nsponed d own into the trenc h a nd
thence seaward probably by density c urrent s. The transport
mechanism is s urpr isingly long ranging. Plaie reconstru ction
suggest that the o ldes! hemipelagic mud may have been
deposiled as far as 100 km (rom the treneh ax is and a t an
e levat ion 1 km above the trench fl oor.
Site 486 provided a 38-m section of predominantly fine to
very coarse sand. The unconsolidnted sand flowi ng inlo the
drÎII hole Ihreatened to seÎze the bit, thereby precluding
deeper drilling. Densily underflows p robably carricd the
sand from t he Mexican margin inl o the trench via 11
prominent submarine canyon immediately east of the transe c t. Pis ton coring d uring t he s ile survey recovered coa rse
sand from the mouth of this canyon and gravel from the a xis
in the middle slope .
slower uplif\. Site 491 appears to have been uplifted <If a ratc
of abo Ul 400 MY for ro ughly 2 MY . •Ind uplift thereafter
slo wed to abo ui 100 m/MY. Site 492 lippears 10 have been
initial y uplifted at a rate of 500 m/ MY and the reafter al a
rate of 200 rn/M Y.
Sites 490. 489. and 493 were drilled o n the upper slope to
in ves tigate the vertical teç tonic history of the Mexican
margin und the transition from continental crUSt 10 accrelionar y prism. Sites 489 and 493 bottomed in igneous and
melamorphÎc continental c rus t after penetrating sediments
ra nging in age from Pleis tocene 10 early Miocene . Both
holes contain s ha llow-water sand overlying su bae ria l1 y
e roded basemen\. The sand a nd overlyÎng hemipelagie
sediments document a marine transgression, beginning
approximalely 22 MY ago. foll owed by rapid s ubsidence 10
depths of 3,000 m or more before the onsel of slow uplift
beginning roughly 19 MY ago.
Site 490 sediments a re anomalous wi th respect to lower a nd
middlc slope sedime nts of s ites 488 , 491 , and 492 , on one
ha nd, and with sedime nts of upper s lope sites 489 and 493.
on the other. Although drilling al site 490 penetruled deeper
than any lower o r middle slope ho les the cores s howed no
deformatÎ o n typical of lower par IS of holes 488, 491, a nd 492
(Fig. 5). At s ite 490. IU t ing a nd f racturing appear to ha ve
res ulted from faulting normal to Ihe trenc h axis. Pre-uppe r
Pliocene sediments deposited below the carbonate compensa tion dcpth al s ite 490 have been subsequenll y uplirted at a
rate of ,Ibout 90 m/ MY , w hic h is somewhal slower than
uplift of sites 488 . 491 a nd 492. We infer that site 490
sed iments were probably deposited in a relati vely stable
trans itio n zone about 10 km wide between the ed ge o f
continental basement to the north-northeas t and the a ccretiollliry wedge to the south-southwest.
T he Irenc h sed iment can be contras ted with hemipelagic
slope sediments and hemipelagic mud and rare fine sand a nd
sHI from the lower s lope bas in . No site surve y piston cores
taken from the slope recovered coarse sa nd . Trenc h sand
al so dirfers from ~he lf s,lIld in Ihat shelf sand conta ins a
s ubstantial ç arbonate compone nt (Mc Mi1Jen. Haines. 1981).
The unique c harac ter of the Ire nc h sand indiea les Ihal the
s,lnd was derived directly from sources o nshore <l nd by passed the shelf and s lope en route to depos ilion in the Irenc h.
H oles at siles 488 , 491 , Hnd 492 penetrated in the mud slope
apron and sampled the upper part of the accre tio rmry
wedge . Each s ite boUomed in s!l nd whic h Moore et al ,
(1979 b) interpreted as Irenc h sand on the b!lsis of seismic
and lithologie s imUa ri ty 10 conlemporary Irench sand . Trace
fossU a ssemblages a nd carbo nate dissol ution da ta s uggesl
that the sa nd has deposited in lower slope o r trenc h wate r
depths, thereby s trenglhening the evidenee thal the s and is
a n up lifted trenc h deposi\.
Overall, the tectollic picture is one of rapid subsidence of
the ÇQntinen ta l crus t during the period 19-22 MY ago,
fo l1owed by slow uplift thcreafter. Accretio n began a t least
10 MY ago and has continued to Ihe present (Fig. 6). T he
tecto nic pic lure between 10 lilld 19 MY is e nigmatic.
T he çoincidence of sa nd and LORs at siles 4811, 4';11, and 492
s uggests tha l t he Impedance contra si belween sa nd a nd
s urrounding s ilt and m ud is respo ns ible for the seÎsmiç
reflec tions. The reflections dip landward . but the d ips do
nO I inc rease up ~ l o pe a ~ ~ ugge ~ led by Scely el ,,1. (1 9 7~ ) .
The ages of s!l nd-bca r ing uni ts increase upslope from
Pleistocene at s ite 488, the lowermost slope site. 10 P liocene
at s ite 491. and then 10 late Miocene nt s ite 492. the
lowcrmOSI s lope sÎle. The obser ved age sequence is as
predicled by Seely el ai. (1974) and constitutes slrong
e vidence of offseraping during Pliestocene to Late Miocene
time interval. If L ORs !Ire bedding planes. as inferred in the
preced ing paragraph. then Ihe age relations represent an
appa re nt stratigraphie invers ion, probably due to unde rthrusting parallel or subpa ra llel to bedding.
DI SCUSS ION
Accretion
Uplifted Irenc h sands in the Leg 66 area su pport the
im brica le underlhrus t mechan ism of Seely el ai. (1974) a l
Jeast in the Jowe r slopc. The sllnds appea r to ha ve been
Lo we r sections of cores reço vered at these sites exhibit
su bstantial deformation ehurac terized by steep dips. local
folding and scaly clay (Fig. 4). At each site. deformation
inc rease s in intensity downhole. beginning at shal lo w depths
but in sediments of greater age at s uccessive upslope holes.
The age thres hold of significant deformalion b 0.6, J,O. and
5.8 MY (Moore ellli .. 1979 b). respectively, at s ites 488,
491. a nd 492. Defo rmat ion Îs generally restrieted to sedimentS de posited in the trench and on the lowermost slope
and thus occur~ carly in the acc rel ionary process .
Uplift history paralleh ddormational history in that maximum upl ih rates occur carly, then give way 10 a period of
MX .6
Figure 6
tille dra .. ing 01 w e of slope from UTMSI mig'lIred st'islI1 ir line
.\1XI6 sh o ...inj.t LDR s a'lIl inltrred rlmm laulu . Clwor!r :o"e
probabl)" Co nsÎ)-rs of slump dtposiu (f'lIm Shiplt)" el al. . 11J80).
2'8
RESUL TS FROM OSO? LEG 66 ON THE MIDDLE AMERICA TRENCH
upl ifted at rela ti vely rapid rates during offscraping atthe toc
of the slope. then uplifted at slower rates o nce they were
incorporated within the slo pe. The increase in age of sands
ups lope is one of the strongest argume nts for the offseraping mode!.
Wedid not penetrate thrust fault s that we could biostratigraphically document although seismic data suggest penetration of a Pleistocene thrust fault in lower slope sed imen ts a t
site 4S8. ThruSI faults are evident elsewhere in the seismic
data (Fig. 6). Failure to penetrate a demonstrable thrust
fault is no t surprisi ng in light of our discovery that LORs are
sand bodies and the fact tha t wc did not penetrate far below
the slope sediment a pron. The high sedimentation mtes in
the area require large fa ul1 offsets in order to bios tratigraphically 10 verify e.~istence of a fault.
Although our data indieate offscraping in the Leg 66 area,
sorne doubt Tcmains regarding the uni versal applicability of
the mode!. Shipley el al. (1980). in their an!llysis of
a pproximately 2000 km of mullifold seismic-reflection data
in the Leg 66 a rea. observed tha t internai refleclions are
more common in the west half of the Leg 66 area. They
further infer that refteclors occurring in discrete zo nes are
uplifted turbidites, and that their occurrence in the wes t half
of the area is assoCÎated with the presence of s ubmarine
canyons and turbidite ponds there. Shiple y et al. (in press)
examined data from the Middle Ameriça Trench off
Mex ico. Guatemala and Costa Rica. areas of intensive
investigation and eXlensive data collectio n by the Univers ity of Tuas Marine Science I ns lÎtute . Data from these
a reas support the idea tha! sandy turbidites greatly e nhance
offscra ping. Mue h. if not lall, ocean-floor pelagic and
hemipelagiç sed iments seem to be subducted .
Von Huene el al. (1980 a ) drilled into the lower slope off
Guatemala during Leg 67 and produced res ults strikingly
d ifferen! from those off southwest Mexico. I n a hole 3 km
land ward of the trench , Le.!! 67 penetrated a Cre taceous to
Pleistocene claystone and hemipelagic sequence before
bottoming in mafie and intermediate igneous rock atypical
of oceanie basalts. Seismiç data suggest several hundreds of
meters of sedimenlary rock between the bollon of the hole
and oceanic crust. Ladd et al. (1978) and Ibrah im et al.
(1979) s howed that seis mic refraction and reflection inte r val
velocities within the main body of the Guatemaht accre tionary wedge range from 4.1 to 6.5 km/sec, Coring in the
Leg 67 area produced relatively fresh mafic rock fragments.
T hese data s uggested 10 J. W. Ladd (pers. comm . . 1978) the
possibilil Y of ophiolite slivers within the Guatemala s lo pe.
Dickinson's (1973) o bser vation Ihat the widths of arc- tre nc h
gaps (and. by inference. the volumes of accretionary
wedges) are for the most p:tTI proportional to the ages of the
respec tive s ubduction zones suggests that (lccretion may
also proceed by a mec hanism other than offscra ping of
sandy turbidites such as obser ved at the LeS 66 site . One
other such mec hanism is the accretion of slices of oce,mic
crust, sea mounts, or o the r crustal topographic irregularitics. These Iwo model s may be end members of a sequence
with mature a ccretiona r y pris ms and mel,lOges inc luding
examples of both .
accretionary wedges with age pointed at the problem . Karis
a nd S harman (1975) indicated the possibilÎIY of consumption. Moore et al. (1979 a; b ) a nd Shipley el al. (1980 )
s uggested sediment consumption in the Leg 66 region, and
von Huene el al. (1980 al indicale Ihe necessity o f consumption in the Leg 67 area.
The Leg 66 data offers li good opportunity for calc ullltion of
the amo unt of sediment bcing consumed. The c urrent
acc retionary event began about 10 MY ago. The relative
youth of Ihis system in comparison with may other s ubduction systems minimizes errors arising from changes in
converge nce rate and direction . Data from the East Pacific
s how that the latest ridge jump and possible plate reorga nization occurred abom l0-15 MY ago, before the onset of Ihe
c urrent acc retionary event ( Handschumac her, 1976 ; Van
Andel et al. (1976). The plate system a ppears to have been
stable thereafter.
Seismic veloci ties ind icate that virtually the entire acc retionary wedge in t he Leg 66 a rea cons ists of sediments. The
velocities are s ufficiently well defined to permit good
esti mates of the porosity and hence estimates of degree of
partitioning of s ubduc ted fluid s and roc k.
Oensity of seismic da ta, both reflection (Shipley el al ..
1980 : Helsley et al., 1975; and Mooney et al .. 1975), are
suffi cienl to define mainly the geometry of the accre tio nary
wedge and of the trench turbidi te wedge and the thickness
o f the pelagic-hemipelagîc ocean crus t componen \.
Our procedure for calculating sediment consumption was as
follows :
1) Estimate thîckness of sedimentary units
a ) pelagic-hemipelagic compone nt.
b) turbidite wedge component ;
2) calculate porosity from drill core (L . Shepherd. pers.
comm.) and seismie velocities (sec Eaton, Watkins. 1970) ;
3) calculale equivalent amount of no n-porous rock :
4) assume present-day convergence ra te of 7 cmlyr (M inster, Jo rdan. 1978) applicable to last la MY ;
5) calculatc sediment input for 10 MY .
A ccretiollary wedge l'olume
1) Construct s tructure c ross section from data of S hi pie y el
al. (1 980); Moore et al. (1979 a: b). Hels lcy et al. (1975).
Mooney el al. (1975) a nd Keller et (li . (1979) (Fig. 7) ;
...
IIrl l lU I
•
i§~
"
;
Figu re 7
".
Hypollwlit'ul modet used /Q eSlimalt J'o/ume of lire IIccreliol1l1t)'
weulle. Base of crUSI nea f sire 49) IIIrd Iht Irendl is calrstrained by
seismic refracrÎ<JII data (He/sley et al., 1975 : Muane y el al., 1975):
lop af Ihe oceanic crUSI bene(l/h Iht Irel1ch and seo ward pari of Ille
...edge is CO llSlfll;ned by seismic ref/tc/iOII dllla (Shipleye t ,Il., 1980).
RemainiRY featurel' are inferred frum Cl C/Jmbin ati/Jn /Jf Ley 66
drillins dala and stismi,· re{lecf!ol1 data (Slripley el al .. 1980). 1. 11
flnd 111 represenl miRima/. probable, ///Id maximum IORd"'/lfu
fIIle", s of Ihe uccrtfionury wed!;e.
Cons umptio n
The question of consumption of sediments was rccognized
carly by von Huene (1972) who est imated li deficiency of the
vol ume of sedime nt acc reted in the Ale utian s ubduction
zone. Dir;:kinson's (1973) observation of increasi ng
arc-trenc h gaps and. by implication, of increas ing s ize of
219
J . S. WATKINS et s i.
T:a ble 1
Ca(ctIlaled /hickness
Thick ne ss
Unit
Pela,pç- hemipel agîc ............. .. ........ .. ... ............... . .. .. .. ... .. .. . .
Trench Turbidites .. .. ...... .... .......... .. .... .. .. . ..... . ... .. . ........ ..... .
Acc rct ionary Wcdgc . .... .. .. .... .. ............ .. .. .......... .. .. ....... .. ... .
Min.
Max..
Avg.
Vel o
Est.
Por. ·
Equi v.
Roç k
(m)
(m)
(m )
( kml~e c )
(% )
(m )
1.'
1.'
J.'
751
,..
30
<Il'
160
.'.'
200 -
625
.
li.
H
'
Estim:ated fro m co re recol/e red, site 486.
Enimaled from seismic velocities.
Below limit of seismic resolu tion of about 25-50 m.
From c:ure measureme nb. site 487.
..,
2) measure area of a ccrelionary wedge ;
3) ca1culate equi vale nt volume of no n-porous rock
perlkilometer of trench as in input flux. steps 2 and 3.
.,
Com/Jar~ r~Sldts
Thicknesses of pelagic-hem ipclagic and t urbidite sand units
va r y locally. Thicknesses were ave raged from dala from
se ven seismic dip lines for Irenc h turbidites and a long a
100 km seismic line 50 km seawa rd of. and parallel to , the
trench for the pclagic- hemipelagic component (see Fig. 2).
Res ults are s ummarized in Table 1.
Clearly, t rench lurbidites (antl associated downs lope trans po rted deposits) accounl for mo~ t of the sediment flu x. Il
a ppears Ihat a column equivalent to ap proximately 190 m of
non-porous rock has been fcd inlo the s ubduction zone at li
ratc of 7 cm/yr o r 70 km/MY fo r the past 10 MY . This
vol ume aggrega tes 135 km' /MY .
! •
•
"
Il is now necessa ry la eslimale the volume o f the accretio·
oa r y wedge. The principal uncertainly in the calcula lion is
the sil.e and shape of the landward parI o f the aceretionu ry
wcdge triangle. overha nging continental c rus! both obscures
reflec tÎons from depth and introduces serious errors in
refraclio n calculations. Thus, we projecl the Moho both
landward a nd seaward us ing refraclio n data from Hels ley el
al. (975) and projec t oceanic crus t la ndward us ing refte!;·
tion data from Shipley el al. (1980) ( Fig. 8). The unde lermi·
"
I:.'arl )' l'Iiocen,. "/If Y . Accre/wnur)' "'edge 1,1' bllllding :Itu"'urd,
underpfo /ing causes uplilr 01 crUSI. 'ransilion zo ne. olld i",,,rmOJI
accrt/ion ar)' "'edge :
b}
PRESENT
'0 .. y
.u
Fi~ure
.i •
• U
' iO
...
.,
8
fI)'polh elical recons/ruClion 01 acertlion·undetploling seqlltnce lor
u g 66 "tJnS«I.
'0 ••
...
..,
! •
•
,.
"
•
"
~91
ned upper l::.ndward s urface of the acc retionar y wedge we
di vide inlO three cases , 1. II. Ill. (Fig. 7). representÎng o ur
esti mate of minimum , probubJe, und muxÎmum volumes of
the accretionary wedlote . Our .. probable .. ca<;e, Il . results
"
O ..str 01 aCcrttiOIl , IQ III Y . .. Trmultion ~one .. mo y con sisl 01
stdilllen/s deposiled beluT' Ollstl olacCTe/ioll Or t arU,SI an:rtud
sedimellls. Data ITOm sitt 49(} 1(H'or Ihe prt-occrtlion m od" :
a)
"
Prtsent . Underplallng ('onlinuts fu tlt"ale margin lrom Silll
Ip 49J . Acti"e a('utlicm limiltd 10 ,icinil,. 01 Trtnch-48lJ.
c)
220
RESULTS FROM DSDP LEG 66 ON THE MIDDLE AMERICA TRENCH
fro m projection la nd ward o f the landwardm osl strong LOR ,
w hich we interpret as at o r nea r the top o f the aecre tiomlry
wedge. Case I V includes t he tra ns ition zone within the
accre tionary wedge. We conside r th is case improbable .
Table 2 s ummarizcs the case e stimates.
responsible for upl ift of the middle and upper slopc. We
belîeve underplating to be that mec han ism (Fig. 8).
If the underplati ng model is correc t. then the u pper crust of
the a ccret io nar y wedge cons is ts of trenc h turbid ites and
associated slo pe sediments. J udging from the thic kn ess o f
the zone of ac ti ve o ffscraping in the toc of the accrelio nar y
wedge (e.g .. F ig. 6), we e stimate that the th ic kness of this
crust is about 2 km a nd that its volume is abou t
60-65 km )/treneh-km. Velocities of 3 km/sec .. app ro pria te
to thi s upper zone , sugges t a n average porosity of 30 % and
an equ ivalent volume of no n-poro us rock of 45 km l/km , or
33 % o f the 1Q-MY in p ut fl ux. th is val ue. together with
previo us caJculatio ns, suggest that about one-th ird o f t he
input sed ime nt flux is subdu cted beyond the zo ne d e fined
by the acc retionar y prism in F ig ure 7. Thus, about 33 % of
the total is being ~cra ped off to fo rm a r igid .. crus t . . of the
acc ret ionary p ris m ; anothe r 33 % is initially su bducted,
then progressively "ski mmed off" and underplated on the
accretionary prism ; a nd the remaining 33 % is carried to
greater deplhs. Because there is no reason 10 d iffe re ntia le
between underplating o f the acc retionar y prism and underpla ting of the continental c rUSI. it is simpler to di vid e our
ma rgÎn into 1) an acçre ted rigid c rust ; 2) a n u nderlying zo ne
of under plating ; and 3) a lowermos t subduc ting zo ne of
pelagiles and hemipelagites (Fig . 8 cl. The bounda r y
betwee n 2) a nd 3) will progressively deepen in the se c tion as
pe lagites and hemipelagite s are s kimmed off at depth .
Table 2
ESlima/eli "o/ume
Case:
(
"
III
(V
/Jf
a~'crelio"ary
Volume oj
IIccre:tionary
wedge
(km'/km)
97
111
m
161
,,·t'lige
Equivalent
volume: of
non· porous
,~k
(km')
17
89
108
129
Amounl
subductcd
(km' )
(%)
.," "
•
43
17
20
6
Table 2 gives a ran ge o f values deri ved from varying
a ssumptio ns regardin g the s hape of the acc ret io nar y wedge.
The value we consider mos t likely , o n t he basis of our
interpretation of the available da ta, is 34 %. Percentages of
s ubduçted sediments in cases 1. Il , III , whichrange from 20
10 43 %, exceed thé pe lagic-he mipelagic input and require
subduction of a s ignificant frac l ion of trench depasits. T his
resull supports Shipley et al .'s (1980) inlerpret;-ttion th<lt
most pelagie and hemipelagic sediments entering the Middle
A merica Trench a re being s ubdu cted. o ur result is also
consistent with Leg 67 data requiring subd uc tion of a major
part o f Ihe pelagic-hemipelagic flux there (von Huene el fl l.,
Tectonic erosion
Leg 66 d ri lling provided no direct e vidence o f tec to nie
erosion. Abundant evidence o f cons umption of sed iment s.
howe ve r.suggets t ha t tec tonic erosion of the ma rgin is
mec ha nically plaus ible. Ma rgin subsidence recorded at sites
489 a nd 493 suggests thinning of the crust, and a peri od of
tectonic e rosion might ha ve preceded the o nse t o f accre tio n.
1980 a ).
Fro m a mechanieal point o f view. it is probably s igniricanl
that the input sediment fluX contains wate r volu mes roughl y
eq ua l to the vol ume of non-paro us rock. Sorne water,
espedall y in trench turbidites. is p robably squeezed o ut
durîng initial deformation , but seismic evidence of undeformed pclagic-hemipelagic layers extending 10 d cpt hs of
several kilomete rs s ugge sts tha! <1 cons iderable vol ume of
wa ter remains trapped within the sediments. The combÎned
high clay content and high water content of these sed iments
probably result in overpressurin g and a low coeffi c ient of
fric tion. These fac tors provide a good decollement s urface.
Pelagic-hemipelagic input thus a ppears to be preferentially
subducted w hile overlying sed iment may or may not be
sçraped off a nd acçreted . Sc holl and Marlow (1974) observed that pelagie sediments a re ra re wi thin melanges. and the
a bsence of t hese sediments in any o f the three holes on the
lo wer slope a nd the one on Ihe trans ition zone o f Leg 66
s upports their preferential subd uc tion .
Forear c
bas in.~
T he L eg 66 a rea is ano malous in several respec ts, including
the abse nfe of a fo rearc basin. Sorn e evidence suggests.
however, that a n incipient fo reare basin ma y be forming in
the vicinit y of site 490. Pa leobathymetric recons tru ctio n
(Fig. 9) ind icate s that the regio n of s ite s 490-492 is ris ing
faster tha n that of sites 489-490. Furthe rmore, seismic
reflec tions in the slope apTon near site 490 are til led
LA T E PLiOCEN E
...
PRE SENT
2 - 0 "'y
4 BB
Underpl a ling
4914 92
49 0
'"
,
Wc have prevÎous ly me ntioned tha t fo ldins, de watering,
fracruring, a nd th rus tin g in t he Leg 66 a rea occur early in
t he accretiona ry process. After this initial period of deforma tion the acc re tiona ry wedge rises slowly al roug hl y the
same rate as the upper slope. T he ve rtical distance between
the acçreted tur bidites a nd the downgoing ocean ic c rust
widens with time. The mechanis m o f Seely ef al. (1974)
proposed uplift b y ro tatio n of im bricate wed ges. Because
this mecha nism appears inoperative in t he Leg 66 Mea
exccpt near the foo t of the s lope, sorne other m ust be
•,
10
~m
~
VE ·4X
,
Figure 9
Paleobulhymelric rer;mUf r"ction for tire period la ie PliQct'n e II.>
presenl (0-1 MY) sh/Jwing inferreli aptifl ralU. Higher uplift r(lles il t
I/,e rtgioll of si/es relmil't ra lite rtgion af l'ius 490-489 will. if
cf/ntinutli. /ead ffl lite fOn/tOliolt of a forearc basin.
22 1
J . S. WATKINS et al.
landward similar 10 those of the Iquique Basin (Seely. 1979).
If the different rates of uplifl observed in the region of
~ites 489-490-492 persisl. a foreare basin will develop.
The rueehanis m responsible for this pattern of uplift may be
the higher raIe of underplating closer to the treneh. This
meehanism would eHuse the zone of maximum uplift to
move seaward with time. Tectonie. thermal . and isostatic
s ubs idence dominate in the landward part (Ka rig. Sharman.
1975). Seaward migration of the trench-slope break as
observed in Iquique Basin (Seely . 1979) would be an
expected consequence of underplatinll. and fo rearc basins
would widen with time. as proposed by Karig (1974).
(Seely. 1979. Fig. 8). We ha ve. however. no information
indieating close lithologie affinities of these and Leg 66
intrusive rocks. The evidence, a!tho ugh not compelling.
favors sinistral strike-slip motion.
The absence of evidence for accre tion doring the interval
from 22 to 10 MY ago may indiCllte th!lt molion was large1y
strike-slip during that lime. Handsehumaeher (1976) place~
the breakup of the EPR south of the Baja Cal ifornia a l
10 mY . Magnetic anomalies. ho wever. arc indistinct and
interpretation ambigous.
T he transition zone at s ite 490 may contain at depth
sediments shed during the 22- tp 10-MY inlerva!. A deep
penetration hole here s hould be reward ing in re vealing the
history of the Mexican margin and the recons tructio n of
Pacifie plaIe motion.
"brgin history
The history of the margin in the Leg 66 area is interwoven
wilh the history of t he Pacifie plate , the Cocos plate, and
perhaps the Rivera and Caribbean plates as weI!, Evidence
of the mo tions of these plats is confusing and sometimes
contradic tor y. Consequently. the history o f the Leg 66 s ite
is not clear . cspccia][y during the period from 25-10 MY.
Data from Leg 66 sites 493 and 489 suggcst the fo][owing
his tory for the continental crust part of the transec l.
SU MMARY AND CONCLUSIONS
We drilled dgh t s ite ~ a[ong a transect of the Middle
America Trenc h off southweSiern Mexico to investigale
acc retion and cons umption of sediments a nd rock during the
s ubduc tion process and to inves tigate the history of a young
subduction zone. The eight s ites included Iwo refere nce
sites on oceanic crust and in the treneh. respeclively . to
d ocument lithologies being subducted andlo r acc reted . three
sites on the lower s[ope to sample accreted sediments, and
three sites on the upper s[ope to document the his tory of the
continental crust.
22 MY
Subsidence a nd marine transgression.
19 MY
Subsidence below CC D.
< GraduaI, uniform uplift to prese nt posit ion.
Data from s ites 488. 491 and 492 on the lower slope indie!lte
the folJowing history fo r the accretionary wedge part of the
transeet.
10 MY
Earliest observed accretion.
< 10 MY progress ive aeeretion and seaward expansion of
acc retionary wedge.
Inclusion of sediment from the transition zo ne in the
ac cretionary wedge would push the onset of accretion back
tO 12-13 MY. However. the absence of LOR s in the
transition ..:one ,md the atypiciil evolution evidenced by
transition-zone sediments s uggest that this zone was neve r
part of the accrelionary wedge.
Samples collecled indicate that aceretion has been oceurrillg
in the a rea for 10 MY o r perhaps slightly longer. We
reeovered trench sa nd ranging in age from Pleistocene to
Mlocene a nd uplifted as much as 2-) km abave the presentday trench floor. and we obsefved Ihat the age of trench
sand decreases downslope. Ils the aceretionary model predicts. Landward dipping refleeto rs thou gh t in other areas to
represent Ihrust faull s were fo und here 10 result from
bedding in uplifted trenc.h turbidite sequences.
Uplifted sediments record an early period of deformalÎon
and uplift. In this period. whic h lasted 2-4 .\1Y for our
samples. rUles of uplifz were 600-800 m/MY. After this
Îniti!ll episode. most deformation ceased and uplift rates fell
to 100-200 MY.
The data s uggest that truncation and rifling , probnbly s trike
slip (Karig et al .. 1978: Malfait. Dînkleman. 1972). away of
the former cOntinenl;i1 margin (including a preexisting
accretion!lry wedge !lnd perhaps a forearc basin) occurred
during a major plate reorganization placed by Ha ndschumacher (1976) between 26 and 29 MY ago. Subsidence and
marine transgression in the Leg 66 area s uggest that this
event occurred no later than 22 MY. CeSS!ltion of su bduclion 23 MY ago. indicated by the end of volcanic ac tivity in
the Sierra Madre Occidental (Clabaugh. McDowel1. 1976).
coincides weil with the marine trans ition.
Il is Ilot clear whether the ancient s lope and s hclf slip ped
dextrally nor th wes t as part of the San Andreas developmenl
(A twater . 1970). as s uggested by s e veral other investig,ltors
(e.g. Malfai t . Dinkleman. 1972). The best evidence regarding this diehotomy may be that reconstruc tion of slippage
on the Cayman Fracture Zone of 2 to 4 cm/yr (MacDon(lld.
Holcomb. 1978) places the Guatem!lla-Honduras corner of
the Caribbean plate near the Leg 66 area 12.5-25 MY ago. or
during Ihe lime of plate reorganization s uggested by Pacific
magne tic anomalies a nd in the gap between the end of
subsidence and onset of ace retion . Also, Central Ameriean
rocks south of the Mo tagua- Polochi fault. which separates
the C!lribbean and North American pl!lles. contain granod iorite intrusive bodies of Cretaceous a nd Tertillry age
Of sediment entering the subduction zone. 33 % (mainly
treneh turbidites) is offscraped at the toc of the slope . 33 %
is initially s ubducted but then underplates the previously
ofrscraped sediments. ca using them to continue to rise with
time . and 33 % (probably induding a il the pelagitehemipelagite oceanic componen!) is subdue ted landward
be neath the continental crust. Trenc h turbidites and s lumps
accoum for 85 % of the equi 'lalent flux of non-porous rock
and the oceanic compone nt accounts fo r 15 91:. Porosites of
75 % e ns ure that large quanlities of water <Ire bcing subducted along with Ihe pelagite-hemipelagite com ponen!. The
w;Her probably causes overpressuring and pla ys a major
role in the [ubrication of the s uture. The average fate of
accretion is 8 km l/km _MY of equivalent dewatered.
non-porous rock.
Wc find no d irect evidence of tectonie erosion bUI condude
Ihat tecto nic erosion is mec ha nicall y feasible in vicw of the
large quantities of sediment being subd uc ted .
The ab~ence of !l forearc basin in the a rea of investig3tion is
probably due 10 the youth of the subduction regime. S ite 492
222
RESUL TS FROM DSDP LEG 66 ON THE MIDDLE AMERICA TRENCH
is currently rising fasler than site 49Q ; if this panern of
re lative uplifl continues a forearc basin could form.
The hîslory of Ihe reg io n documenled by sediment cores
ond inte rpreted in lîght of the his tory of the ne ighbori ng
plates is as fo llows.
23 MY Reorganizalion of the Pacifie plate ; cessalion of
volca nism in the Sierra Madre Occidental; subsequent
eastward shi!t of volcanic axis; sinistral strike-sl ip rirting
away of exisling conlinent<l.l ma rgin. Marine transgression
in Leg 66 a rea 22 MY ago followed by rapid subside nce.
19 MY Murgins sinks to , o r just below , CCD and begins
s low rise.
10 MY Baja California rift fo rms ; c hange from s trike-slip
molion 10 Oblique convergence in Leg 66 orea. Accretion
begins in Leg 66 are;1 and continues 10 presen!.
REFERENCES
AIVllru M. Jr., 1949. Unidades tectonicas de la repupliea
uni!. Bol. Suc. Geol. Mex., 14, 1-22.
M e.~i ­
!.ad" J . W., Ibrahim A. K .. 1\1 ~I\1Jl1e n K. J ., Lalhllm G. V" von
HUlne R. E .. Walklm J . S., Moore J . C., Woru l J . L. , 1978.
Tectonic o f the Middle America Trench. in : In l. S ymp. Gualtmala
Earthq"a~ and R«on struC lion Proc~u; GLlatcmala City. 1.
Atwa ter T . , 1970. Implicatiu n of plate tcetonies for the Cenowic
tec tonic evolution of we ste rn North America. Gtol. Suc. Am.
Bull., 81, m3-35J6.
Mad >O Rllld K . C" Uoloomb T . L" 19711. Inversion of maimll tÎC
anomalies .. Ad seafloor sprcading in the Cayman Trough . Earth
Plune/. Sei. L~" .• 40, 407-414.
I\1l1l1a ll B. T" Dlnkleman M . G., 1972 . CircLlm-Caribbean tec tonic
and ingeous ac tivity and the e volulio n of the Caribbe:m plate. Otol.
Soc. Am. Bull .. 83, 251-272.
Mc:.\1i1len K . J ., Haine T _ R., 1981. La te Quaterna ry sedimenl5 of
the Southern Mexico margin. in : lni/i(ll Rtports Detp StU Drillin/(
Projelr: . Vol. 66, in press.
MiOSler J . B .• Jordan T . H., 1978. Prcsent·day plate mOlions. J .
OtophY$. Res .. 83, 5331-5334.
MollUlr P" Sykt. L . R .• 1969. Tectonics of the Caribbean Ilnd
Middle Americ1l regions from focal mechanism~ and seismicity.
Oeol. Suc. Am. Bull .. 80, 1639· 16S4.
Mooney W. M . • Mt)'er K. .... UeWey C. E., l..omnlt!. C . • Lewis
B. T . R., 19" . Refl'llcted waves acrOS5 a leading edge: obse rva tions of Pacific s ho ts in soulhern Mu ico. Am. Oeuphys. Ulli()11
T,ailS .. 56, 452.
Muon J . C., Wa tkhl5 J . S. " al., 1969 a. Middle Americ1I T rench
drilled. GtOlimes. 24. 22·22 .
MOO<'e J . C. , WatklDll J . S. tl al.• 1979 b. Progressive accretio n in
the Middle America Trenc h. sou thern Mexico. Natu't. 28 1.
638-642.
S<: holl D. W" Marlow M. S. , 1974. Sedimenlary sequences in
modern Pacific trenches and the deformed c ircllm· Pacific eugeusy nclinc . in : Modern and ancient geosynclinaJ sed imentation.
edi led by R. H . Dott. R. H . Shane r. Soc. EC(J II . I'altontol. Mintro/.
Sp«. Publ .. 19. 192·211 .
S<: hoU D. W., ,'on lI~ ne R. , Vallier T. , HOIO-,U 1)., 1980. Sedimen·
lary masses and co ncepts about tec toni c processes at underthrust
ocean margins, Gtology. 8, 564-'68.
Sttly D. R., 1979. The evolLition of structlll'lli higs bordering major
(o rearc basi ns . in : OeolOMiC1l1 1100 geophysical invesligation of
continenu.i mar"ins. edited hy J. S. Watkin~ . L. Mo nt:ldert and
P. W. Dickinson. Am. A.uac. hl. Orol. Mt"' .. 29, 245·260.
SHly D. M., Vall P. K.. W.ltun G. G. , 1974. Tren ch slope mode l.
in: The ge%gy ul Conlinflrtul ma'gins, edited by C . A. BLirk and
C. L. Drake. Springer·Vcrl3g. New Yo rk, 249-260.
Sf:yflrt C. K .. 1969. Undeformed sediments in ocean;c treoc he s
wilh seafloor s preading. Nullue. 222 . 70.
Shipley T. H., Houston M. Il .. Bullln R. T . • S ..... ub t·. J ., McM iIIl n
K. J .. Ladd J . W . • Worul J . I... 1979. Seism ic e vidence for
wides pread possible gas hydrale horizons on co ntinenm l s lopes and
rises. Am. AullC. Pel. Otol. Hull .. 63 , 12. 22(14-2213.
Shlpley T . H., McMilie n K. J ., Walkins J _ S., Moore C. J ., SaOOoval·Ocooli J . If . • Worul J . L .. 1980. Continental margin and Jower
slope s tructures of the Middle America Trench neM Acapulco
(Mexico). Ma,. Oeol.. 35, 65-82.
Shlpky T . Il .. l.add J. W., BuflLlr K. T., Wa tkJns J . S .• in press .
Comparison o r struclul'll l s tyles along the Middle America TrenchMellico. GU31e maia and CoS t:. Ri cll. in : Trellr:h alld t o,tUf('
sedimtnla/ion and /tc/onles /n modem o,ld anclent subduction
to nes. edited by J . K. Leagett . Geol. Soc. London.
SNk R. H. , 1972. The oceans. the ncw fron tier in expJorllli on.
Aun. Pel. Explor. Assac" 12, 2, 5·28.
Cla baulI:h S. E., McDo IO"ell F. W., 1976. Vol canismo Creti!cÎcuCe nozoico y s u relacion con la evolucion lectonka de occ idente de
Mexico. Il,. Cong'. UR ·Am. Gtul .• Rtsumts. 35.
de Cserna Z" 1965 . Reconocimiento ,eologico en la Sierra Madre
dei Sur de Me llico ent re ChilplininiO y Acapulco. Es tado de
Querrero. Bol" /rw .. Orol.. Uni,'. Nac Autoll . Mu" 62. 76 p.
f) k klnson W. R., 1Y71. Plaie te ctonic s in geoJogie hi story. Seitn (·t.
174, I07-11J.
Dk kl lIIiOn W. R., 1Y73 . Wid ths of modern arc- trench pps propor·
tional to pas t duration of igneous IIc li vity in lIuocia ted magmatk
arcs. J . Oeophys. RtJ .• 78, 3376-))119.
!!.alon G. P .. Wl tklns J . S .. 1970. The use o f seismic refrac tion and
gravit y methods in hydrogeologi,al investigation s. in : Minins and
groundwater geophysics. edited by L. W. Morley. (Jeol. SurY. Ca,..
Econ. Orol. Rlp .. 26, 544-5611.
G u:unan E. J .• 1950. Geoloz;'1 dei noreste de Querrero. Bol. Assac.
mu. Oeol. Pel ., 2. 95-156.
Ha ndSl' humac her D. \\1., 1976. Posl· Eocene pla te tec tonics in the
caMcrn Padfic. in : The geophys ics o f the Pacifie Ocean bas in .. nd
its nmrjlin s. ed ited by G . H . Sulton. M. H . Manllhnani and R.
Mo herly. Am. GtophyJ. Un/OII Malt .• 19, ln·zoz.
Uelsley Co E. , Na tion J . 8., Me)lr R. P ., 1975. Seismic refraction
ubservations in 5-Q(Jthern Mexico. Am . Geophys. U,.ion T,ans .. .56.
452.
Il l1liSO nl! D. M., Edwards P. 8 ., Juhn~n S. It ., CampbtJl J . F. ,
SUllon G. H., 1976. C ru stal struc ture of the PerLi-C hilc Trcnch :
11-I 2"S latitude. in : The lI:euph ysics of the Pac ific Oceiln basin and
its marsin. edited by C. H. Sult on. M. H. Manjlhnani and R.
Moherly. Am. Otophys. Union ~Jo n .. t9 . 71·86.
Ib... him A. K .• l.alham G. V.. I.add J .. 1979. Seismic rdraction and
rdlec tion mea$u remcnt s in the Middle AmerÎCa Trench oHshore
Guatemala. J. OtQphys. R u" 84, 5643-,5649.
Kulg D. E ., 1'174 . 'l'ectonic eros;o" o f trenc hes. Eorth Planel. Sei.
LtII" 21. 209· 212.
Lar lll D. E., Sluirman G. t' . Ill. 1975. Sulxluction a nd IIccretion in
trenc hes. Geol. Suc. Am .. 86. 377-389.
KarlR D. E., Inll:le J . C. J r er al" 1975. Inirial Rtports of Iht Dup
Sea f)ri lling ProjtCI. Wash inllton, US Government Printing Office.
30 p .
Karlll D. E .. Ca tdwell R. K . • Moore G . F., 1\1oore 1) . G .. 1978. Late
Ceno7.o;c s LlbduClion a nd conlinent al margin Iru ncation a long the
no rthern Middle America Trenc h. Geo/. Soc. Am. Bull., 89,
265-276.
Kelllt B.. Lewis 8 . T . R. , MHder C., Hl lsley C .. Me)"er R. P .. 1979.
Ellplora lion seismol08Y stLidiu of active IInd pan ive continental
"Ul rgins. in : GeoJollica l IInd geophysicili invest igatiuns of continen·
tal margin s . edited by J . S. Walk ins. L. Mon tadert and P. W.
Dickinson. Am. Assoc. Pet. Gtol. Mtm .. 29. 443-4.5 1.
223
J . S. WATKINS el al.
\'lIn A IM:lft T. H ., Huth G. R., Moore T. C. Jr ., 197j . Cenuzoic
his lo r)' and paleo-oceanograph)' 01' the cen tral equalUrial PlIci(ic
Ocean, CtQI. Soc. Am. /lttm. , I.B , 134 .
"on Huen .. R ., 1972. Stru cture of the comin en!lIl marsin and
lectonism at the eastern ,,'eulian trcnc h. Gtal. Soc. Am. 8ull.. 83.
3613-3626.
\'on HUt' n.. R., Aubouin J . tl al.. 1980 a. Yh .. I)SDP Mid-Maerica
Trench Iranseel off G uat .. mala, Oro/. Sor. Am. 8ull.. in prus .
,-o n lIuent R. , Lll n ~lh 1'01., J"ri" asu N., Ot..adwH., 1980 b . Summar)'.
Jllpan Trench Transect. in : blitial RtpOTls of the J)eep Sen Dril/in).!
l'mitet. ' ·01. 56 a nd 57, par! 1. US GOllcmmenl t->ri nt in!; Office.
Washington D.C .
224